 |
» |
|
|
|
|
|
|
|
|
|
Patch Name: PHNE_28567
Patch Description: s700_800 11.00 ONC/NFS General Release/Performance Patch
Creation Date: 03/04/29
Post Date: 03/05/14
Warning: 03/12/05 - This Critical Warning has been issued by HP.
- PHNE_27217 introduced behavior that under rare circumstances
can result in non-responsive ONC 1.2 AutoFS-managed
file systems. Any file system managed by ONC 1.2
AutoFS may experience this behavior.
- The behavior is also exhibited by superseding patches
PHNE_28102 and PHNE_28567.
- Details about this behavior are:
- The first indication of this behavior is that a command
will hang while attempting to access a non-responding
AutoFS-managed file system.
- The file system will not appear to be mounted.
- When enabled, automount logging will not show any
activity for this file system after the behavior has
occurred, regardless of how many attempts are made to
access it.
- AutoFS itself does NOT hang, so other AutoFS-managed file
systems will act properly.
- No failure messages will be displayed via dmesg(1M) or
in /var/adm/syslog/syslog.log.
- The behavior is most easily recognized by examining the
autonode an_mntflags value, using the nfsshow utility,
q4, or p4. The nfsshow utility, which is available from
HP, is the easiest tool to use. If the autonode
an_mntflags value is 0x185, then this behavior is the
cause of the non-responsive file system:
an_mntflags=MF_MNTPNT|MF_WAITING_MOUNT|MF_CHECKED|
MF_TRIGGER(0x185)
- The behavior has been observed in configurations with the
default timer setting, which is 300 seconds.
- This behavior is corrected in PHNE_28982, which is
released. To avoid this behavior, HP recommends that
PHNE_28982 be installed on systems using AutoFS.
- PHNE_28567 is included in the following Support Plus
Patch Bundle:
Sep 2003: QPK1100,B.11.00.62.4
Hardware Platforms - OS Releases:
s700: 11.00
s800: 11.00
Products: N/A
Filesets:
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP
OS-Core.CORE-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-PRG,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.KEY-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NISPLUS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-SHLIBS,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NIS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NIS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NIS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
InternetSrvcs.INETSVCS-INETD,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-64SLIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
NFS.NFS-64ALIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP
Automatic Reboot?: Yes
Status: General Superseded With Warnings
Critical:
No (superseded patches were critical)
PHNE_28136: PANIC HANG ABORT CORRUPTION MEMORY_LEAK
PHNE_28102: HANG ABORT
PHNE_27217: PANIC HANG ABORT
PHNE_26387: PANIC
PHNE_25626: HANG ABORT
PHNE_25484: ABORT
PHNE_24909: HANG
PHNE_24034: HANG ABORT CORRUPTION
PHNE_23833: PANIC HANG ABORT
PHNE_23249: HANG
PHNE_22642: ABORT PANIC
PHNE_17586: PANIC
Data Page Fault
PHNE_17247: PANIC
Data Page Fault
PHNE_16599: PANIC HANG
Assertion Failure/vn_rele Panic
System hang/semaphore deadlock
PHNE_14919: CORRUPTION
Data destroyed on Server (with warning)
PHNE_14620: PANIC
Data memory protection fault
PHNE_13208: HANG
Category Tags:
defect_repair enhancement general_release critical panic
halts_system corruption memory_leak
Path Name: /hp-ux_patches/s700_800/11.X/PHNE_28567
Symptoms:
PHNE_28567:
nfsstat
SR: 8606308270
DTS: JAGae71305
The "nfsstat -m" command does not display NFS PV3
information.
PHNE_28136:
automountd libautofs.a
SR: 8606278982
DTS: JAGae43038
An AutoFS mounted file system using the "-hosts" map
can take a long time to mount when the environment
contains dead NFS servers.
libautofs.a
SR: 8606287136
DTS: JAGae51075
The contents of a directory mounted with AutoFS
become unavailable, and the directory cannot be used,
mounted, or unmounted.
libautofs.a
SR: 8606274519
DTS: JAGae38596
After unmounting all the AutoFS mount points through
the script "nfs.client stop", the system panics if it
tries to access an unmounted directory.
libautofs.a
SR: 8606267508
DTS: JAGae31750
A process is able to write into a read-only
automounted file system.
libnfs.a
SR: 8606267227
DTS: JAGae31470
The system panics after a read/write NFS exported
file system is unmounted and then re-mounted as
read-only without re-exporting the file system.
libnfs.a
SR: 8606269865
DTS: JAGae34105
When Intergraph's DiskShare NFS server returns zero
bytes for the preferred and/or maximum transfer size
of the READ/WRITE operation, the HP-UX NFS client
panics.
libnfs.a
SR: 8606274703
DTS: JAGae38780
NFS pathconf(_PC_CHOWN_RESTRICED) and fpathconf
(_PC_CHOWN_RESTRICED) return 0 when chown is not
restricted instead of -1, the POSIX compliant result.
libnfs.a mount
SR: 8606272309
DTS: JAGae36449
Using loopback NFS mounts, the server may hang due to
a deadlock condition.
libnsl.1/a
SR: 8606288821
DTS: JAGae52752
Calls to netdir_getbyname() from a multithreaded
process may hang or terminate with SIGSEGV or SIGBUS.
libnsl.1/a
SR: 8606282498
DTS: JAGae46451
Memory/file-descriptor leak in RPC multi-threaded
server programs, if the client tries to connect using
a version number that is unsupported by the server.
The memory leak may result in the server dumping core
and terminating. This can also cause the server to
quickly run out of file descriptors.
libnsl.1/a
SR: 8606291642
DTS: JAGae55409
Memory based XDR stream implementations do not
function properly.
librpc.a
SR: 8606289500
DTS: JAGae53431
NFS/TCP client does not handle out-of-order streams
messages correctly while attempting to connect to the
server and can result in a client panic.
librpc.a
SR: 8606256016
DTS: JAGae20331
Threads accessing an NFS directory hang for a long
time with "bindresvport: loop-twice port assigned ..."
messages other than port 512 logged in syslog.log.
librpc.a
SR: 8606271824
DTS: JAGae36003
The system reports "vmunix: bindresvport: loop_twice
port assigned 512" messages.
librpc.a
SR: 8606282662
DTS: JAGae46615
The message "NFS server xxxx not responding still
trying" is seen on a client while attempting to
establish an NFS TCP connection, and the connection
is in the LISTEN state.
librpc.a
SR: 8606284964
DTS: JAGae48906
Data corruption while using XDR function
xdrmblk_getlong() with a 64-bit kernel.
nfsstat
SR: 8606275551
DTS: JAGae39627
If an NFS server listed in the /etc/mnttab file
is down or there are network problems, the
"nfsstat -m" call hangs.
rpc.lockd
SR:8606284885
DTS:JAGae48827
The number of file locks (nflocks) grows over a long
period of time, eventually causing NFS to lock up.
The only workaround is to restart NFS.
PHNE_28102:
automount
SR: 8606288394
DTS: JAGae52325
Incorrect permissions for the log file.
automount automountd
SR: 8606286563
DTS: JAGae50502
Incorrect configuration for AutoFS.
domainname domainname.1 ypinit ypserv ypxfrd
SR: 8606276334
DTS: JAGae40412
Problem with ypxfrd.
libnsl.1/a
SR:8606274555
DTS:JAGae38632
Showmount dumps the core with SIGSEGV.
libnsl.1/a
SR: 8606274502
DTS: JAGae38579
RPC does not function properly.
libnsl.1/a
SR: 8606281883
DTS: JAGae45826
The netdir_getbyname() routine using NIS can hang
under certain circumstances.
rpc.yppasswdd
SR:8606276016
DTS:JAGae40093
The rpc.yppasswdd process dies.
PHNE_27217:
automountd libautofs.a
SR: 8606264317
DTS: JAGae28647
This product update is a member of a set needed to
enable the optional HP-UX Device IDs feature. Upon
installation, the HP-UX Device IDs bundle
(DevIDs11) will install the full set of product
updates (including this one) necessary to enable
the Device IDs feature.
If the HP-UX Device IDs product (DeviceIDs11) is
not installed, this product update to HP-UX will
have no impact on your system.
automountd
SR: 8606256845
DTS: JAGae21158
The /var/adm/automount.log file is shipped incorrectly.
automountd
SR: 8606242021
DTS: JAGae09276
1. Many "dup_remove_pending: WARNING: duplicate
request cache size (210) > XID_CACHE_SIZE (200),
continuing" messages in syslog.log and
automount.log.
2. Many "Rmdir: err=247, mnt='<mount-point-dir>'"
messages in automount.log.
automountd libautofs.a
SR: 8606229377
DTS: JAGad98430
AutoFS mounted file systems never get unmounted.
automountd
SR: 8606245876
DTS: JAGae12335
System panics when the mount of an exported file
system that contains an extraneous or missing "/"
fails. The panic string is:
rm_autonode: No entry for <address>
automountd
SR: 8606234968
DTS: JAGae04152
When automountd gets a stale NFS file handle
error for a direct mount point, it removes the
corresponding AutoFS direct map entry for that
mount point from the file /etc/mnttab. Automountd
needs to be stopped and restarted to get back
the direct AutoFS mount point.
automountd libautofs.a
SR: 8606233931
DTS: JAGae03154
After the AutoFS client unmount timer expires, if the
unmount of the parent file system fails and the remount
of hierarchical mount points fails, then zero entries are
found while accessing the hierarchical mount points
under the parent file system. Once this condition occurs,
the failed hierarchical mounts will not retry mounting.
automountd
SR: 8606259713
DTS: JAGae24031
AutoFS problem.
clear_locks clear_locks.1m
SR: 8606248269
DTS: JAGae14669
If a client's /var/statmon/sm directory contents are
destroyed, future lock requests hang.
libautofs.a
SR: 8606271227
DTS: JAGae35462
The ls command will hang when users try to access
a dead server after some systems with mounted
directories are shut down.
libautofs.a
SR: 8606277106
DTS: JAGae41178
Under heavy load, file systems mounted with AutoFS
never get unmounted, even after the system load has
decreased.
libautofs.a
SR: 8606232532
DTS: JAGae01767
AutoFS mounted directories are empty.
libautofs.a
SR: 8606221909
DTS: JAGad91027
System panic occurs while AutoFS is unmounting
the idle AutoFS mount points. The panic string
is "spinlocks held with M_WAITOK".
libautofs.a
SR: 8606226778
DTS: JAGad95840
If an AutoFS mounted filesystem, using a simple
direct map, is manually unmounted, a panic may
result if that filesystem is referenced simultaneously
(e.g. by two or more ls commands).
The panic string is:
Data page fault
The panic stack trace is:
panic+0x14
report_trap_or_int_and_panic+0x80
trap+0xdb8
nokgdb+0x8
nfs3_root+0x68
hpauto_trigger_mount+0x20c
auto_access+0x30
chdirec_seg+0x5c
chdirec+0x14
chdir+0x34
syscall+0x29c
$syscallrtn+0x0
libautofs.a
SR: 8606248870
DTS: JAGae15267
The "cd" command to an AutoFS direct offset
mount directory hangs, and automountd cannot
be killed by pressing CTRL-C to stop the "cd"
command.
libnfs.a
SR: 8606147221
DTS: JAGad16564
Problem with privileged groups.
libnfs.a
SR: 8606245329
DTS: JAGae11799
A system hangs or panics when issuing a "ln"
command.
libnfs.a
SR: 8606257688
DTS: JAGae21991
Client hangs when trying to lock a remote file.
libnsl.1/a
SR: 8606264087
DTS: JAGae28412
When a 64-bit RPC application uses UNIX authentication
flavor, the RPC server does not properly decode the
client credentials.
libnsl.1/a
SR: 8606254087
DTS: JAGae18419
The rpcbind process increases in size and stops
responding to client requests. This requires
rpcbind needs to be killed and restarted.
librpc.a
SR: 8606257219
DTS: JAGae21525
Accessing a TCP AutoFS mount point causes the
system to hang.
librpc.a
SR: 8606265228
DTS: JAGae29556
Spinlock deadlock panic due to a lock being
held too long.
mountd.1m
SR: 8606249559
DTS: JAGae15949
The '-p' option for rpc.mountd described in
the man page does not function as described.
mountd.1m
SR: 8606132817
DTS: JAGad01965
rpc.mountd no longer supports the '-e' and '-n'
options but the man page for rpc.mountd still
shows both options being available.
nfs.core
SR: 8606256846
DTS: JAGae21159
The /var/spool/rexd directory is shipped incorrectly.
mount umount
SR: 8606235172
DTS: JAGae04352
The NFS mount operation returns an incorrect
error message when all ports are in use. The
error message is:
nfs mount: bindresvport: t_errno = 8
nfs mount: get_fh: Couldn't bind to a reserved port.
return = -1
nfsconfcol
SR: 8606234034
DTS: JAGae03255
The nfsconfcol script is not removing the target
data files before writing to them.
rpc.lockd rpc.statd
SR: 8606249576
DTS: JAGae15966
A client's attempt to acquire a lock may result in a
delay or a hang.
rpc.lockd
SR: 8606180770
DTS: JAGad49991
Blocked lock requests on a client system hang.
rpc.lockd
SR: 8606259831
DTS: JAGae24149
A client system hangs when trying to lock a file.
rpc.lockd rpc.statd
SR: 8606244127
DTS: JAGae10618
If rpc.lockd and rpc.statd are restarted and the
/var/statmon/sm directory is destroyed on the server,
the client's rpc.lockd takes 35 minutes to recover.
rpc.mountd
SR: 8606255006
DTS: JAGae19335
Exported directories cannot be mounted.
PHNE_27213:
{Site Specific Patch}
PHNE_26387:
libautofs.a
SR: 8606248624
DTS: JAGae15022
AutoFS causes a system panic with the panic string
"rm_autonode: No entry for <<address>>".
PHNE_25626:
automount
SR: 8606219527
DTS: JAGad88667
On a client system, the old automounter is trying to
access a non-existent entry in the NIS map and that is
generating an excessive number of packets over the
network which causes slow response time.
libautofs.a
SR: 8606221217
DTS: JAGad90351
The "ll" command hangs in the case of an automounted
file system if the server is down.
libnfs.a
SR: 8606237277
DTS: JAGae06322
NFS PV2 client does not properly show the quota exceeded
error.
libnfs.a
SR: 8606228306
DTS: JAGad97363
A file has an unusual timestamp when created with open(2)
using O_EXCL flag.
libnsl.1/a
SR: 8606204999
DTS: JAGad74175
The ttsyncd process loops, occupying much of the CPU
usage and causing the system to slow down.
libnsl.1/a
SR: 8606218377
DTS: JAGad87526
A multi-threaded server leaks file descriptors under
heavy load if clients are running on the same box.
libnss_nisplus.1
SR: 8606221102
DTS: JAGad90236
Reactivating a user account that has been disabled on a
trusted NIS+ client fails.
librpc.a
SR: 8606172908
DTS: JAGad42168
When an HANFS package is manually failed over
back-and-forth between two different cluster server nodes
within a period of time shorter than five minutes, then
on any NFS client which has the HANFS package filesystem
mounted over TCP, NFS access to that filesystem can
appear hung for a relatively long period of time -
usually more than four minutes, with no apparent response
from the server. Also the NFS client issues the error
message "NFS server XXXXXXX not responding still trying",
about every 30 seconds.
nsswitch.conf.4
SR: 8606188628
DTS: JAGad57842
The man page for nsswitch.conf does not reflect ldap as
an additional database source for nss backends.
rpc.rusersd
SR: 8606225120
DTS: JAGad94209
rpc.rusersd core dumps if more than 180 users have
logged in.
rpc.statd
SR: 8606224657
DTS: JAGad93745
File lock recovery fails when the server reboots if the
server has multiple network interfaces and the client
has mounted the file system with an interface name
different than the server's hostname.
rpcgen
SR: 8606230660
DTS: JAGad99710
A Server stub generated by rpcgen logs error messages to
/dev/console.
ypserv
SR: 8606226735
DTS: JAGad95797
The inetd is unable to resolve services.
PHNE_25484:
rpc.lockd
SR: 8606245747
DTS: JAGae12206
rpc.lockd core dumps with a SIGBUS error during heavy
NFS file locking.
PHNE_24909:
libnfs.a
SR: 8606222617
DTS: JAGad91730
If NFS loopback mounts are used, potentially all nfsd
processes may get stuck in an NFS deadlock situation.
libnfs.a
SR: 8606205013
DTS: JAGad74189
The mount option 'grpid' does not return the correct
'gid' value on NFS PV2 and PV3.
libnsl.1/a
SR: 8606205779
DTS: JAGad74954
The automountd (AUTOFS) hangs/cores while attempting
to automount exported FS and reports 100% CPU
utilization.
libnsl.1/a
SR: 8606212242
DTS: JAGad81430
When a customer specifies a username with leading
blank characters as an argument to nispasswd and
passwd(1), nispasswd modifies the username to
include those leading blanks in the NIS+ passwd
table and trusted table.
libnss_nisplus.1
SR: 8606216548
DTS: JAGad85717
A password expires unexpectedly in an NIS+
environment.
libnss_nisplus.1
SR: 8606219558
DTS: JAGad88698
During the login process, NIS+/trusted clients
force NIS+ users to modify their password when
their password is expired but fail to un-expire
the password when their password is changed.
Users will always be forced to modify their
password as soon as it has expired.
yppasswdd.1m
SR: 8606199444
DTS: JAGad68631
The yppasswd command updates fields other than the user
password field in an NIS map.
PHNE_24034:
automountd
SR: 8606152534
DTS: JAGad21864
AutoFS can mishandle unmounts when a server is down,
which may appear as bdf command anomalies.
automountd
SR: 8606196567
DTS: JAGad65770
AutoFS cannot unmount a cachefs type file system.
automountd
SR: 8606160777
DTS: JAGad30096
AutoFS can not unmount hfs or cdrom file system types.
libautofs.a
SR: 8606176017
DTS: JAGad45257
AutoFS directories are empty under heavy NFS load.
libautofs.a
SR: 8606193515
DTS: JAGad62727
Duplicate entries for direct AutoFS mounts show up in the
/etc/mnttab file.
libnfs.a
SR: 8606197003
DTS: JAGad66201
Walk through large size buffer cache causes file access
performance degradation.
libnfs.a
SR: 8606203242
DTS: JAGad72416
A process sequentially reading a large file across an NFS
mount runs extremely slow if, at the same time, there is
another process sequentially writing a large file to an
NFS mounted directory, and one or more biods are running
on the system.
libnfs.a
SR: 8606146332
DTS: JAGad15675
The I/O throughput on a locked file is low compared to
the throughput on an unlocked file.
libnsl.1/a
SR: 8606207461
DTS: JAGad76637
rpcbind crashes with signal 10, Bus error (SIGBUS).
librpc.a
SR: 8606197306
DTS: JAGad66499
An NFS client becomes hung while performing many
lock operations on files in NFS mounted directories.
It also receives "bindresvport: loop_twice port
assigned 512" messages in the message buffer.
librpc.a
SR: 8606168123
DTS: JAGad37405
An NFS/TCP client operation receives "NFS server not
responding still trying" messages while attempting to
access the server, even though the server system is up.
In this case, NFS/TCP on the server shows many TCP
connections for local address port 2049 in a CLOSE_WAIT
state when performing a "netstat -an" command.
nfs.client
SR: 8606181688
DTS: JAGad50904
On a system with active lofs mounts managed by AutoFS,
the "/sbin/shutdown" command may hang when the file
systems are unmounted.
nis.client, ypbind, ypinit, ypinit.1m, ypserv.1m
SR: 8606201620
DTS: JAGad70794
On SUN's NIS clients, it is possible to specify a list
of available NIS Servers for a given domain. HP's NIS
clients do not provide this functionality.
rpc.lockd
SR: 8606184904
DTS: JAGad54106
rpc.lockd can cause a user process to hang, if a
loopback mount is specified with an alias hostname.
rpc.lockd
SR: 8606202370
DTS: JAGad71544
NFS locks remain when killing NFS loopback mounted client
applications.
rpc.statd
SR: 8606208446
DTS: JAGad77633
NFS file lock recovery fails because rpc.statd matches
the wrong hostname.
rpcbind
SR: 8606164889 8606173448
DTS: JAGad34186 JAGad42706
Customer Symptoms:
rpcbind core dumps with SIGBUS or SIGSERV.
PHNE_23833:
automount
SR: 8606175862
DTS: JAGad45102
Mount points of automounter directories are not
visible when executing the "ls" command.
automount
SR: 8606169011
DTS: JAGad38290
The automount process hangs, making automounter
mount points inaccessable.
automountd
SR: 8606160596
DTS: JAGad29916
The /var/adm/automount.log file fills up the /var
file system.
exportfs
SR: 8606171662
DTS: JAGad40926
exportfs sometimes does not work correctly when
invalid host names are present in a host list
in the /etc/exports file.
fsirand.1m
SR: 1653311530
DTS: JAGab46804
Unable to use fsirand() on JFS file system.
libnfs.a
SR: 8606178998
DTS: JAGad48223
Unnecessary NFS read before write is causing severe
decreases in write performance.
libnfs.a
SR: 8606165859
DTS: JAGad35150
Data page fault panic while copying a file using
NFS V3 between an HP-UX client and an NT InterDrive
server where the target filesystem is full.
libnfs.a
SR: 8606178618
DTS: JAGad47845
The rename() system call always fails when trying
to rename a directory in a cachefs-mounted file
system, if the target directory name exists.
'errno' gets set to EPERM ("Not Owner").
libnfs.a
SR: 8606175781
DTS: JAGad45021
The 'du' output is eight times higher over NFS
with an HP server / Sun Client configuration when
compared to an HP server / HP Client configuration.
libnsl.1/a
SR: 8606183171
DTS: JAGad52387
UDP packets sent by remote users can cause RPC
applications to abort.
libnsl.1/a
SR: 8606186548
DTS: JAGad55752
RPC server running in TCP multithreading
mode develops a file descriptor leak when
clients are on same machine as the server.
Eventually the server fails to process
clients' requests when the process limit
is attained.
libnss_nis.1
SR: 8606179951
DTS: JAGad49173
getservbyname() using NIS returns NULL when
trying to resolve service aliases.
libnss_nisplus.1
SR: 8606172747
DTS: JAGad42007
Changing an NIS+ user's password with nispasswd() or
passwd() while the system is in trusted mode without
specifying the dot "." notation (non fully qualified
domain) results in an incorrect sub-domain being
created.
nfsstat.1m
SR: 8606173363
DTS: JAGad42623
The "man nfsstat" command does not display all of the
statistical data fields.
rpc.lockd
SR: 8606180764 8606180765
DTS: JAGad49985 JAGad49986
rpc.lockd causes user process to hang.
rpc.lockd
SR: 8606174017
DTS: JAGad43270
rpc.lockd drops the lock request from same process
for a different file.
rpc.lockd
SR: 8606185238
DTS: JAGad54440
Partial record unlocking in NFS does not work
properly.
rpc.lockd
SR: 8606179200
DTS: JAGad48424
rpc.lockd core dumps when user process is
interrupted with "control-c".
rpc.mountd
SR: 8606176384
DTS: JAGad45622
The boot process of a Linux client fails due to
rpc.mountd not allowing AUTH_NULL authentication.
ypxfr
SR: 8606164133
DTS: JAGad33433
The ypxfr command fails with the error message:
"ypxfr: bind_to_server clntudp_create error:
RPC unknown host".
PHNE_23249:
libautofs.a, libnfs.a, librpc.a
SR: 8606103740 8606159451
DTS: JAGab70789 JAGad28779
A multithreaded process can hang when performing NFS
file operations and concurrently writing core, forking,
performing debugger attachment or undergoing job
control.
PHNE_23247:
{Site Specific Patch}
PHNE_22642:
automount
SR: 8606155592
DTS: JAGad24909
Automount (automounter) coredumps when sent SIGUSR1
to force it to update the automaps.
automountd
SR: 8606160808
DTS: JAGad30127
When using AutoFS, NFS mount points do not enforce
quotas.
libautofs.a
SR: 8606135856
DTS: JAGad04988
Panic occurs when trying to umount a ufs filesystem
which is still in the process of mounting.
libautofs.a
SR: 8606159736
DTS: JAGad29062
Unexpected delay in unmounting of Autofs mounts.
libnfs.a
SR: 8606179269
DTS: JAGad48493
Poor NFS pv3 performance on 64-bit (V-class) on 11.0.
libnfs.a
SR: 8606161701
DTS: JAGad31017
When reopening an NFS file right after closing it,
file attribute related inconsistencies can occur.
libnfs.a
SR: 8606158230
DTS: JAGad27560
Unmount of an nfs file system causes a panic.
libnfs.a
SR: 8606163247
DTS: JAGad32551
Unable to remove an NFS tmp file, .nfsXXXX after
failing to create or write a file which has exceeded
the disk quota limit.
libnfs.a
SR: 8606166314
DTS: JAGad35601
Process using NFS can not be killed or stopped by
sending the process a signal (SIGKILL or SIGSTOP).
libnsl.1/a
SR: 8606140609
DTS: JAGad09968
Core dump occurs when user belongs to more than
16 groups.
libnsl.1/a
SR: 8606160400
DTS: JAGad29722
On installing PHSS_20014 and PHSS_21947, 100% CPU
is used up by 2 processes - opcmsgrd and opcdistm.
librpc.a
SR: 8606169796
DTS: JAGad39062
Data page fault panic due to bad machine name
length in NFS request.
librpc.a,nfsd
SR: 8606167053
DTS: JAGad36339
An NFS/TCP client operation receives "NFS server
not responding still trying" messages while
attempting to access the server even though the
server system is up. In this case, NFS/TCP on
the server has become inoperable due to a hang.
newkey
SR: 8606157592
DTS: JAGad26923
Newkey does not recognize the root password of the
remote machine and returns an error, "ERROR,
passwd differs from login passwd".
PHNE_22125:
libautofs.a
SR: 8606143877
DTS: JAGad13210
Using autofs, mount points become inaccessible
after a series of mount/unmount cycles.
libnfs.a
SR: 8606128739
DTS: JAGac86286
File corruption over NFS when NFS file server is
TOC'ed.
libnfs.a
SR: 8606144881
DTS: JAGad14221
HP NFS client performance is degraded as shown by
"nfsstat -c", it makes unnecessary GETATTR calls
for each read or write on files opened with
synchronous I/O flags set; and synchronous I/O
mode remains in effect for subsequent opens on an
NFS file opened once with synchronous I/O flags
set.
libnfs.a
SR: 8606147198
DTS: JAGad16541
CPU time is wasted by unnecessary calls to
compare_cred function in NFS PV3 client code.
libnfs.a
SR: 8606145970
DTS: JAGad15306
The stat() system call on NFS directories
shows a non-existent ACL.
libnfs.a
SR: 8606128493
DTS: JAGac84458
System panic when running "du -s" or
"find -name" on a 64 bit machine.
libnfs.a, librpc.a
SR: 8606160097
DTS: JAGad29423
Mount an NFS file system with non-interruptible
options remains interruptible.
libnfs.a
SR: 8606144617
DTS: JAGad13957
Using NFS, a read-only file open by one
process causes a "Permission denied"
error in an on-going write operation on
the same file by another process.
libnfs.a, librpc.a, nfsd
SR: 8606145517
DTS: JAGad14853
If NFS is stopped and restarted, the nfsd
processes do not get re-registered with rpcbind,
and rpcinfo commands fail.
libnfs.a, nfsd
SR: 8606146885
DTS: JAGad16228
nfsd unregisters a running nfsd process on
startup.
librpc.a
SR: 8606155004
DTS: JAGad24321
When a server is down, an NFS/TCP client process
e.g. a bdf command can hang.
librpc.a
SR: 8606143872
DTS: JAGad13205
NFS/TCP mounting and unmounting eventually hangs.
librpc.a
SR: 8606144478
DTS: JAGad13818
NFS/TCP connections are not torn down when the
peer sends a FIN.
nfs.core
SR: 8606159847
DTS: JAGad29173
File permission change on /etc/services file.
rpc.lockd
SR: 8606145476
DTS: JAGad14812
Lockd/statd client crash recovery does not work.
rpc.mountd
SR: 8606127276
DTS: JAGac68822
A failing mount request may return a wrong error
code.
rpc.statd
SR: 8606143409
DTS: JAGad12739
A looping problem in rpc.statd occurs between the
server and SGI clients.
ypserv
SR: 8606155843
DTS: JAGad25155
YPSERV fails to return all groups for id/group
command when concurrent access is allowed to maps.
PHNE_21376:
automountd
SR: 8606128485
DTS: JAGac84450
Automountd not triggered for mounts
anymore when autonode has NULL flag.
exportfs
SR: 8606108804
DTS: JAGab78579
When a user attempts to mount an exported
filesystem which is a symbolic link to an
actual directory or filesystem on an NFS
server, a "Permission Denied" error occurs.
This occurs after installing patch PHNE_18221.
libautofs.a
SR: 8606128485
DTS: JAGac84450
Automountd not triggered for mounts
anymore when autonode has NULL flag.
libnsl.1/a
SR: 8606134593
DTS: JAGad03728
automountd coredumps in the NIS+ environment.
libnsl.1/a
SR: 8606143750
DTS: JAGad13073
When using the TCP loopback transport a
memory leak may occur.
libnsl.1/a, rpc.nisd
SR: 8606107751
DTS: JAGab78042
When using NIS+ with TCP, as the number of open
file descriptors increases, file table overflow
errors occur in the syslog.log file and user
information is not recognized.
rpc.lockd
SR: 8606131496
DTS: JAGad00646
Failover occurs with the serviceguard NFS package because
the rpc.lockd daemon fails (terminates) during periods of
high memory usage (running out of memory).
rpc.statd
SR: 8606131366
DTS: JAGad00517
The rpc.statd daemon core dumps during high memory
utilization (when available memory runs low).
rpc.statd
SR: 8606136063
DTS: JAGad05198
rpc.statd aborted with memory leak due to unresolved
host requests.
rpcbind
SR: 8606131978
DTS: JAGad01128
Rpcbind exhibits wrong behavior on shutdown.
ypserv
SR: 8606134296
DTS: JAGad03431
ypserv becomes unavailable for the local ypbind and
for the other yp clients.
PHNE_20627:
libnfs.a
SR: 8606134214
DTS: JAGad03354
Extremely slow performance of commands or utilities
operating on large directories over NFS.
SR:8606109663
DTS:JAGab82363
Removing an NFS file causes a file named ".nfsXXXX"
(where XXXX is a pid) to be created.
SR:8606135799
DTS:JAGad04931
The i-node for the ".." directory entry of a subdirectory
is not updated on an NFS client after that subdirectory
is moved up one directory level on the NFS client,
ie - moved to "..".
SR:8606135800
DTS:JAGad04932
Output from "tail -f" on an NFS file does not reflect any
appends made to the file on the NFS server if a "umount"
is performed on the NFS file system while the "tail -f"
is being executed.
PHNE_20371:
autofs/mount
SR: none
DTS: JAGab68200
Autofs/mount can fail when the system
hostname is longer than 8 characters.
automount
SR: none
DTS: JAGab68200
Autofs/automount can fail when the system
hostname is longer than 8 characters.
automount,automountd
SR: 8606106692
DTS: JAGab76019
automount with stalemounts causing duplicate
mnttab entries & lock problems.
symptoms of automount are: stale nfs handles are
seen in mnttab after an unmount and when the
mount is repeated (in this case a cdrom...) the
mnttab contains two entries for the device. One
of which is invalid. This second (stale) handle
causes subsequent locking problems.
automountd
SR: none
DTS: JAGab68200
Autofs/automountd can fail when the system
hostname is longer than 8 characters.
automountd
SR: 8606107366
DTS: JAGab77563
Autofs exhibits wrong behavior in processing
executable maps.
automountd
SR: 8606107274
DTS: JAGab77060
automount option -proto=udp causes mounts to fail
libautofs.a
SR: 8606107122
DTS: JAGab76805
autofs - shell gets intermittent "cannot execute"
libautofs.a
SR: 8606106938
DTS: JAGab76453
automounter hangs when trying to mount cachefs
filesystem (reported on HP-UX 10.20)
libnsl.1/a
SR: none
DTS: JAGaa43721
When accessing remote files over NFS, the
NFS protocol on HP-UX only uses UDP.
libnsl.1/a
SR: none
DTS: JAGab68200
NIS and other ONC/NFS systems can fail when
libnsl does not provide support for a system
hostname longer than 8 characters.
libnsl.1/a
SR: 8606107366
DTS: JAGab77563
Autofs exhibits wrong behavior in processing
executable maps.
libstraddr.1
SR: none
DTS: JAGab68200
Libstraddr routines can fail when the system
hostname is longer than 8 characters.
make.script
SR: 8606107068
DTS: JAGab76707
ypmake and Makefile failed
on some mail.aliases file and mail.aliases map
is corrupted.
mount
SR: none
DTS: JAGaa43721
When accessing remote files over NFS, the
NFS protocol on HP-UX only uses UDP.
nfs.core
SR: none
DTS: JAGaa43721
When accessing remote files over NFS, the
NFS protocol on HP-UX only uses UDP.
nfs/mount
SR: none
DTS: JAGab68200
Nfs/mount can fail when the system
hostname is longer than 8 characters.
nfsstat
SR: none
DTS: JAGaa43721
When accessing remote files over NFS, the
NFS protocol on HP-UX only uses UDP.
nisplus/nisserver
SR: none
DTS: JAGab68200
Nisplus can fail when the system hostname
is longer than 8 characters.
rpcinfo
SR: none
DTS: JAGab68200
Rpcinfo can fail when the system hostname
is longer than 8 characters.
rpc.lockd
SR: 5003459362
DTS: JAGab12595
file locks hang when NFS server uses
fully-qualified hostname
rpc.lockd
SR: 8606113678
DTS: JAGac07192
Application hangs on client when obtaining NFS
lock on server
setoncenv
SR: none
DTS: JAGaa43721
When accessing remote files over NFS, the
NFS protocol on HP-UX only uses UDP.
ypmake
SR: 8606107068
DTS: JAGab76707
ypmake and Makefile failed
on some mail.aliases file and mail.aliases map
is corrupted.
PHNE_20344:
libnfs.a librpc.a nfs_iface.o
SR:N/A
DTS:JAGac42842
NFS version 3 only supports up to 8k read/write
transfer size.
libnfs.a
SR:8606110199
DTS:JAGab82902
Memory leak in NFS version 3 at rfs3_readdirplus().
libnfs.a
SR: 8606125471
DTS: JAGac40861
Poor NFS performance relative to SPECsfs97 benchmark.
libnfs.a
SR:8606107676 8606124555
DTS:JAGab77944 JAGac39938
NFS performance problems with NFS client
readdir cache.
libnfs.a
SR:8606126860
DTS:JAGac59701
"do_readdirplus" variable not available in the NFS
kernel.
PHNE_19759:
1.fuser -f on a file does not return pid of opener
2.compress command on a SUN client fails due to invalid
name length returned by NFS pathconf from HP Server
3.getmount_entry() returns down status for all hard
mounted NFS file systems
PHNE_19620:
automountd
DTS: JAGab72053
NFS - autofs (11.0) hangs when unmount from down server
SR: 8606103364
DTS: JAGab70009
replicated automounts fail if system parameter
maxfiles > 2048
libautofs.a
DTS: JAGab73529
autofs w/cachefs can panic in cachefs_root()
SR: 8606105077
DTS: JAGab72920
AutoFS failure with cdrom.
The failure occurs in two ways:
- Hangs if indirect mount
- Error if direct mount
libnsl.1/a
SR: 8606104454
DTS: JAGab41115
rpcinfo gets error "can't contact rpcbind: RPC:
Program not registered" from remote rpcbind on
RPCB_GETADDR/RPCB_DUMP request for list of services.
libnss_nis.1
SR: 5003465898
DTS: JAGab20796
getservbyname() is slow.
librpcsvc.a
SR: 8606103503
DTS: JAGab70329
rpc.mountd dies with SIGSEGV if rmtab is large and
showmount -a is done.
rpc.lockd
SR: 8606105833
DTS: JAGab74177
Virtual nfs servers have locking problems.
DTS: JAGab45462
rpc.lockd uses port numbers within reserved port
range, 600 to 1024.
DTS: JAGab15223
Processes waiting to lock an nfs file which has been
deleted do not get woken up, resulting in a hang.
DTS: JAGaa45544
Locks hang and rpc.lockd logs error messages about
the local and remote lock tables being inconsistent.
rpc.lockd causes lock queue corruption with loopback
mounts.
rpc.mountd
DTS: JAGab45462
rpc.mountd uses port numbers within reserved port
range, 600 to 1024.
rpc.statd
DTS: JAGab45462
rpc.statd uses port numbers within reserved port
range, 600 to 1024.
PHNE_18221:
libautofs.a
SR: 1653298828 5003458299 1653299826 5003462911
SR: 1653304972
DTS: JAGab00144 JAGab03833 JAGab11283 JAGab16597
DTS: JAGab18083
JAGab00144 - AutoFS is not triggering re-mount
JAGab03833 - AutoFS hangs even with PHNE_17200
JAGab11283 - cp(1) fails to inactive direct mount
JAGab16597 - mv command fails in AutoFS direct mnt
JAGab18083 - getcwd() fails with AutoFS filesys
libnsl.1/a
SR: 5003463992
DTS: JAGab17512
Applications fail to create a server side RPC
endpoint due to a failure to obtain a reserved port.
libnsl.1/a
SR: 5003465880
DTS: JAGab20797
A multi-threaded application which calls
getservbyname() failed.
libnss_compat.1
SR: none
DTS: JAGab03573
64-bit application using 'getgrent' fails if
nsswitch.conf uses group:compat
rpc.lockd
SR: 5003446203
DTS: JAGaa46687
rpc.lockd runs out of file descriptors
and returns the following error in
its log file:
RPC: Remote system error
rpc.mountd
SR: 5003439471 5003463331
DTS: JAGaa42795 JAGab17386
ER for share(1M) options and defect in "showmount -e"
rpc.mountd
SR: none
DTS: JAGaa44896 JAGaa24142
rpc.mountd - server allows mounts not on access list
rpc.pcnfsd
SR: 5003460030
DTS: JAGab13451
rpc.pcnfsd does not accept '~' for
file names.
rpc.statd
SR: 5003446195
DTS: JAGaa46688
rpc.statd fails on fopen when more than
260 entries are in /var/statmon/sm.
PHNE_17586:
1. Huge directory reads take too long and client system
performance degrades as NFS IO activity increases.
2. System panics with data page fault when 0 len UDP
pkt received by NFS server system.
3. Root users are denied access to valid root exported
file systems on 11.0 server.
4. Root users on NFS PV3 client who do not have file
permissions for read/write are allowed to open file
successfully but denied access when reading or
writing the file.This behaviour is inconsistent
with non-root users who are denied access at file
open time.
5. NFS requests from anonymous NT clients are rejected.
6. Large file reads on a 32 bit NFS client does not work,
file copy completes without errors but no data is copied
to the target file.
7. When copying large file from local disk to NFS
file system,more bytes than necessary are copied to
the file and the resulting file size is bigger.
8. Buffering of data on NFS file is permanently
turned off by the client system even after releasing
all file locks and closing all file descriptors.
9. Exportfs prints debug messages in syslog.log.
10. When a file is opened with O_CREAT|O_TRUNC|O_EXCL
flags set and the file does not already exist,open is
expected to succeed but for NFS V3 files,open seems to
fail with EACCES.
11. When file has Mandatory lock mode set,HP NFS V3
allows file read/write access when no locks exist on
the file,but NFS V2 denies file read/write access.
Also,NFS V3 client hangs application when locks exist
on the file.This behaviour is inconsistent with SUN's,
where NFS V3 and V2 access is denied consistently at
file open time when file has mandatory lock mode set.
12. Data page fault panic in mvfs due to NFS server
initialization corrupting global memory at boot up.
PHNE_17247:
1. Data page fault panic due to NFS server
initialization corrupting global memory at boot up.
2. Large file reads on a 32 bit NFS client does not work,
file copy completes without errors but no data is copied
to the target file.
3. When copying large file from local disk to NFS
file system,more bytes than necessary are copied to
the file and the resulting file size is bigger.
PHNE_17200:
Makefile
SR: none
DTS: JAGaa73221
When rpc.mountd tries to match a hostname
in netgroup, the given hostname has to be
the exact same as it is stored in the
netgroup.
automount
SR: 5003450916
DTS: JAGaa93190
automount loops using replicated server
automount
SR: 5003362707
DTS: JAGaa25720
automount fails to mount all exported
file systems.
automountd
SR: 1653267377
DTS: JAGaa24193
autofs holding lock on /etc/.mnttab.lock.
automountd
SR: none
DTS: JAGaa95626
Logging information is hard to
read.
automountd
SR: none
DTS: JAGaa40567
executable map processing may grant
caller additional privileges.
libautofs.a
SR: 5003445601
DTS: JAGaa46003
NFS - autofs hangs when manual unmounts used
nfs.core
SR: none
DTS: JAGaa43075
rexd does not start since it cannot find
/var/spool/rexd on system.
nis.server
SR: none
DTS: JAGaa94166
nis.server script won't run correctly unless the
new namesvrs is installed at the same time.
rpc.mountd
SR: 5003454371
DTS: JAGaa94274
rpc.mountd deadlock with automount
rpc.mountd
SR: 5003447003
DTS: JAGaa45940
NFS - rpc.mountd MOUNT performance is very slow
rpc.nispasswdd
SR: 1653286955
DTS: JAGaa44956
when rpc.nispasswdd is running with "-g"
option, it corrupts the credential of a
user who has a credential and changes
password.
rpc.pcnfsd
SR: 5003405118 5003336198
DTS: JAGaa26176 JAGaa25473
1) PC users fail to authenticate when trusted
system is enabled.
2) pcnfsd stops responding to all requests when
a request is made to printer that is down.
rpc.pcnfsd
SR: 1653293043
DTS: JAGaa93189
User authentication fails for users who
do not use a shell that ends with "sh".
rpc.rstatd
SR: 5003442574
DTS: none
The amount of memory associated with the
rpc.rstatd process continues to grow
without ever being returned to the system.
rpc.statd
SR: 5003445858
DTS: JAGaa45938
rpc.statd core dumps with SIGSEGV.
rpcbind
SR: none
DTS: JAGaa92935
rpcb_unset() does not remove all registration
entries from rpcbind.
rpcgen
SR: none
DTS: JAGaa57297
rpcgen generated makefiles fail to build targets
rpcgen
SR: 1653291484 1653291500
DTS: JAGaa57208 JAGaa57209
rpcgen generated file fails to compile.
stdhosts
SR: none
DTS: JAGaa94167
stdhosts command fails on lines longer
than 256 char.
ypinit
SR: none
DTS: JAGaa41361
ypinit has warning message.
ypmake
SR: none
DTS: JAGaa73221
When rpc.mountd tries to match a hostname
in netgroup, the given hostname has to be
the exact same as it is stored in the
netgroup.
ypserv
SR: 5003448159
DTS: JAGaa46689
ypserv is running out of file descriptors
due to not releasing them.
ypserv
SR: none
DTS: JAGaa41159
ypserv core in svc_freeargs ( a rpc routine)
after nslookup for a period of time.
ypxfr_1perday
SR: none
DTS: JAGaa43641
vhe_list should not be in the script.
PHNE_17101:
autofs
SR: none
DTS: none
Autofs fails to start on a 64-bit OS
unless PHNE_15638 has already been
installed.
nfsstat
SR: 5003444745
DTS: JAGaa44646
nfsstat -m exited with an error that it
could not open the mount table
PHNE_16599:
1. NFS server system panics with assertion failure
in vn_rele sometimes during NFS file locking/unlocking.
2. Too much rnode memory allocated resulting in gradual
shortage of free memory resource on the system.
3. NFS Version 3 mounted directories may appear to be
empty on the NFS client even though the directories exist
on the server file system.
4. A touch command to change the modification time stamp
of NFS file may cause system hang.
5. Some NFS clients may experience poor NFS Read
performance with HP servers when using NFS Version 3 mounts.
6. Invalid file system id returned to Statfs of
NFS Version 2 mounted file set.
PHNE_16470:
Makefile
SR: none
DTS: INDaa30982
vhe_list is no longer supported by NIS.
It needs to be removed from ypinit,
ypmake and /var/yp/Makefile.
automount
SR: none
DTS: JAGaa26320, JAGaa26354
Made debug logging change and checked
in the change for setmntent for read locks.
automount
SR: none
DTS: JAGaa40906
automount tries to remount already mounted directory
automount
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
automountd
SR: 5003431353, 5003434654
DTS: none
1. duplicate entries in /etc/mnttab error
message in syslog
2. AutoFS will not unmount LOFS mounts
exportfs
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
libautofs.a
SR: none
DTS: JAGaa24180
1. HOME env set to autofs directory on 64-bit 9808
system hangs gpm help.
2. Autofs hangs when running Netscape.
3. Autofs hangs when remounting hierarchical mounts.
libnsl.1/a
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
libnsl.1/a
SR: 5003429845
DTS: INDaa31432
Compile of application using archived libraries
files with duplicate symbol __nss2herrno
libnsl.1/a
SR: 5003429456
DTS: INDaa31422
rpc.mountd stops running
libnsl.1/a
SR: none
DTS: JAGaa24798
CDE logins fail when user belongs to more then
16 groups.
libnsl.1/a
SR: none
DTS: INDaa30681
The performance of a 64-bit application
slows down when it uses the 32-bit cache
information in the ypbind file.
libnss_nisplus.1
SR: none
DTS: JAGaa40026
getxxxent leaks memory when using NIS+
libnss_nisplus.1
SR: 5003418210
DTS: JAGaa24152
Module ngetprpwent is missing
from libnss_nisplus.
librpcd_util.a
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
librpcsoc.1
SR: none
DTS: INDaa30814
on command fails with the messages
on: failed to reach <host> at address <host address>
mknetid
SR: none
DTS: INDaa30115
mknetid cores when a user is in more that 20 groups
namesvrs
SR: none
DTS: INDaa30072
nis.server script fails if IBM is NIS master to
HP slave.
nfs.client
SR: none
DTS: JAGaa27109
Non-busy NFS mounts are frequently not
unmounted during nfs.client stop.
nfs.server
SR: none
DTS: JAGaa24199
local file systems are mounted when NFS client restarts
nfs_umount
SR: none
DTS: JAGaa27067
/sbin/fs/nfs/umount -aF nfs does not unmount
all NFS file systems.
nfsstat
SR: none
DTS: INDaa31388
Executable hangs.
nis.server
SR: none
DTS: INDaa30072
nis.server script fails if IBM is NIS master to
HP slave.
on
SR: 5003420992
DTS: INDaa30864
on fails when home directory is mounted via
autofs.
rpc.lockd
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.lockd
SR: none
DTS: JAGaa40591
Lock failure caused data corruption
rpc.lockd
SR: none
DTS: JAGaa26538
logged messages causes file system to fill
rpc.lockd
SR: 1653254474, 5003417998
DTS: JAGaa26354, JAGaa26221, JAGaa26257
The rpc.lockd tracing was improper format.
Loss of UDP NFS file locking packets resulted
in a growing message queue. rpc.lockd processes
control C (NLM_CANCEL) incorrectly when the server is
a SUN system.
rpc.mountd
SR: none
DTS: JAGaa24186
Believes there is a problem
in the way the daemon checks to see if it is up.
rpc.mountd
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.mountd
SR: 1653266148
DTS: INDaa31024
valid clients in the NFS server's export list
can not mount the NFS server.
rpc.pcnfsd
SR: 1653272484
DTS: JAGaa26373
Validity checks not performed on printer and
system name.
rpc.rexd
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.rquotad
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.rquotad
SR: none
DTS: INDaa31358
Executable hangs.
rpc.rstatd
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.rusersd
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.sprayd
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.statd
SR: none
DTS: JAGaa26538
logged messages causes file system to fill
rpc.statd
SR: 4701392027
DTS: JAGaa26354, JAGaa26276
The debug logging needs change.
rpc.walld
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
rpc.yppasswdd
SR: none
DTS: INDaa31524
If a password contains a ":", it cannot be changed
via NIS.
rpc.yppasswdd
SR: none
DTS: INDaa31380
rpc.yppasswdd does not update NIS maps when ptmp
and maps on diff file system.
rpcbind
SR: 1653270413
DTS: INDaa31358
logging data is difficult to read.
ypbind
SR: none
DTS: INDaa31412
ypbind cache file sometimes is not created
due to the value in an uninitialized time
stamp. The performance of NIS degrades
when cache file is absent.
ypbind
SR: none
DTS: INDaa31381
ypbind cache file cannot be created
due to a missing compilation flag.
The performance of NIS degrades
when cache file is absent.
ypinit
SR: none
DTS: INDaa30982
vhe_list is no longer supported by NIS.
It needs to be removed from ypinit,
ypmake and /var/yp/Makefile.
ypmake
SR: none
DTS: INDaa30982
vhe_list is no longer supported by NIS.
It needs to be removed from ypinit,
ypmake and /var/yp/Makefile.
ypmake
SR: none
DTS: INDaa31440
When a new map is made on the Master with
NOPUSH=1 option, on the Slave, transferring
the new map fails sometimes.
ypserv
SR: none
DTS: INDaa29252
ypserv stops answering some queries for
entries outside domain on a slow system.
ypxfr
SR: 5003418160
DTS: JAGaa24144
ypmake on the YP master reports a failure
when contacting a slave server
ypxfr
SR: none
DTS: none
Incorrect parameter declaration for the
function ypxfrd_getdbm.
PHNE_16017:
1. The setting of NFS file/directory modification and access
time stamps is inconsistent.
2. An HP NFS server does not permit NFS file/directory time
stamps to be set from a non-HP NFS client.
3. Copying a file from a 64-bit client causes the file
timestamp to be incorrectly set to Dec 31, 1969.
4. Poor NFS performance over 100BT.
PHNE_15638:
automountd:
1. No autofs support on HP-UX.
mount:
1. No autofs support on HP-UX.
umount:
1. No autofs support on HP-UX.
rup:
error exit code incorrect
libnsl:
rpcbind has memory leak
bindreserveport fails
rpc.statd:
gethostbyname called incorrectly.
rpc.lockd:
No support for shares and non-monitored locks in NLM
version 4
automount:
Messages written to syslog file without verbose option
nfs.client:
rpc.lockd does not reclaim locks after a server reboots.
PHNE_14919:
1. NFS write performance may be slow, and may be
particularly poor when large files are being closed or
synchronized.
2. NFS Server files may be accessed and possibly destroyed
when their permissions should not allow access.
3. NFS Remounts may fail with a "device busy" error.
4. Locking large files (> 2 GBytes) may encounter errors.
PHNE_14800:
NFS Performance as measured by the SPECsfs benchmark
gives an inadequate number of IOPS on the initial 11.0
system release.
PHNE_14772:
N/A
PHNE_14746:
rpcgen:
1. rpcgen displays an incorrect usage string
2. C++ applications fail to compile due to non-existent
header file. <sysent.h>
nfs.core:
1. rpcbind cannot receive messages on ticotsord device
rpc.nisd:
1. NISPLUS server can not resolve hostname using DNS for
NIS client.
(SR/DTS: 5003-395731/INDaa28803)
2. NIS+ clients sometimes get "NIS+ unreachable" message
when in fact NIS+ server(rpc.nisd) is up and running.
(SR/DTS: 5003-395558/INDaa29132)
nisaddent:
1. nisaddent adds empty passwd fields on a trusted system.
(DTS: INDaa28982)
nisserver:
1. nisserver creates unnecessary environment variables -
NIS_MASTER and NIS_SLAVE, and adds them into
/etc/rc.config.d/namesvrs file.
(SR/DTS: 5003-395582/INDaa29138)
2. nisserver script does not set up NIS+ server correctly
if
- NIS compat mode is not set, and if,
- nisserver was previously executed with NIS
compat mode set.
(SR/DTS: 5003-395582/INDaa29138)
nisclient:
1. nisclient has illegal grep syntax for hp
(DTS: INDaa28804)
2. nisclient -u script reports success even when
chkey fails
(SR/DTS: 5003-397422/INDaa29242)
nispopulate:
1. nispopulate has illegal grep syntax for hp
(DTS: INDaa28804)
libnss_nisplus.1:
1. 64-bit gethostbyname() returns 0.0.0.0 for the IP
address of a valid host stored in the NISPLUS table
hosts.org_dir.
(DTS: INDaa29686)
nisdefaults:
1. nisdefaults does not indicate if a principal is not
authenticated.
(SR/DTS: 5003-396655/INDaa29212)
nisstat:
1. nisstat -H hostname does not work.
(SR/DTS: 5003-395541/INDaa29137)
automount:
1. Old automount daemon conflicts with New automount
command
2. Automount daemon does not use NIS master map
rpc:
1. no entry for rpc.nisd_resolv
2. no entry for automountd
nfsconf:
1. No support for AutoFS
nfs.client:
1. No support for AutoFS
libstraddr.1:
1. 64-bit application can not send request on loopback
transport.
2. 64-bit library does not exist
libnsl.1:
1. Can not generate new thread after the 65th has been
created.
2. Can not create RPC MT Servers.
3. netdir_getbyname() returns different error message
for non-existent hosts.
4. 64-bit Cannot use loopback transports over RPC.
5. Can not create 64-bit RPC MT Servers.
6. 64-bit netdir_getbyname() returns different
error message for non-existent hosts.
libnsl.a:
1. Can not generate new thread after 65th has been
created
2. Can not create RPC MT Servers
3. netdir_getbyname() returns different error message
for non-existent hosts.
4. 64-bit application cannot use loopback transports
over RPC.
5. Can not create 64-bit RPC MT Servers.
6. 64-bit netdir_getbyname() returns different
error message for non-existent hosts.
vhe_altlog:
1. Obsoleted product vhe is still available.
vhe_getlist:
1. obsoleted product vhe is still available.
vhe_list:
1. obsoleted product vhe is still available.
vhe_mounter:
1. obsoleted product vhe is still available.
vhe_script:
1. obsoleted product vhe is still available.
config:
1. Loopback transport configuration files not moved
to correct location
unconfig:
1. Loopback transport files left after deinstall.
netconfig:
1. Customer configured information lost in /etc/netconfig
hosts:
1. Customer configured information lost in
/etc/net/[ticlts|ticots|ticotsord]/hosts
services:
1. Customer configured information lost in file
/etc/net/[ticlts|ticots|ticotsord]/services
rpc.rstatd:
1. When run from the command line, rpc.rstatd dies after 10
minutes.
keyserv:
1. keyserv registers a UDP port with rpcbind
nis_cachemgr:
1. nis_cachemgr registers a UDP port with rpcbind
(DTS: INDaa29627)
rpcbind:
1. Can not send request over loopback transport devices
to rpcbind
rpc.mountd:
rpc.mountd denies access to valid netgroup.
PHNE_14620:
1. System panics with a data memory protection fault
2. NFS unmounts may take several minutes with large
memory systems
PHNE_13208:
1.Processes hung when doing NFS file IO
2.NFS sub-system failed to respond under 2 way NFS IO stress
3.Sometimes returned invalid error during NFS Read/Write
Defect Description:
PHNE_28567:
nfsstat
JAGae71305
If device ID's are enabled, and the file systems
are configured such that the index for NFS PV3
file systems is a value other than 3, the
information for NFS PV3 file systems will not be
displayed when the "nfsstat -m" command is
executed.
Resolution:
Added a function call to get the file system
index instead of setting it to a constant value.
PHNE_28136:
automountd libautofs.a
JAGae43038
The AutoFS daemon automountd serializes mount and
unmount requests. When a mount request for a dead
NFS server is received it takes 75 seconds for
automountd to fail the mount request. During this
time no other mount or unmount requests can be
processed, even if the requests are for a live NFS
server.
Resolution:
The number of mount requests sent from the kernel
to automountd has been reduced which will decrease
the chance of automountd getting a request for a
dead NFS server. Also, no mount requests will be
sent to automountd without first checking to see
if the NFS server is up.
libautofs.a
JAGae51075
Threads in the AutoFS code are not synchronized
properly while one thread is trying to remove a
directory and another thread is sending a mount
request for the same directory. This causes AutoFS
to create a fresh directory with null flags.
Therefore AutoFS is unable send a mount request
for the newly created directory.
Resolution:
Measures have been taken to provide a proper
synchronization between the threads in AutoFS that
are accessing the same directory.
libautofs.a
JAGae38596
When the script "nfs.client stop" is run, it unmounts
all the AutoFS mount points. While unmounting the
AutoFS mount points, AutoFS does not verify whether or
not a directory entry under such an AutoFS mount point
is busy and therefore it frees the internal data
structures of the corresponding AutoFS mount point.
The system panics when some process tries to access a
directory entry under such an AutoFS mount point
because AutoFS' internal data structures no longer
exist.
Resolution:
AutoFS verifies whether or not the directory entries
under an AutoFS mount point are busy. If busy, the
mount point is not unmounted.
libautofs.a
JAGae31750
AutoFS is allowing a process to write into a
read-only file system. This is because AutoFS
is not verifying whether the underlying file
system is read-only or not.
Resolution:
AutoFS is now verifying whether the underlying
file system is read-only or not. If the underlying
file system is read-only then AutoFS returns
an error.
libnfs.a
JAGae31470
The NFS server does not see any changes made to an
exported file system. Because of this, making an
exported read-write mount point read-only through
remounting will result in a system crash in the NFS
server stack.
Resolution:
Modified the NFS server to check whether an exported
file system permits alterations before each attempt
to modify it, and aborts all modification requests on
a read-only file system.
libnfs.a
JAGae34105
Intergraph's DiskShare NFS server returns zero bytes
for the preferred and/or maximum transfer size of
READ/WRITE, which makes the HP-UX NFS client panic.
The correct behavior should be to 1) return I/O error
message when the preferred and/or maximum size has
a value of zero, and 2) allow normal operation when
the maximum transfer size has a non-zero value.
Resolution:
When an NFS server returns zero or MAX_LARGE_FILES
for a READ/WRITE operation's transfer size, the
HP-UX client now responds correctly instead of
panicking.
libnfs.a
JAGae38780
The functions pathconf(_PC_CHOWN_RESTRICTED) and
fpathconf (_PC_CHOWN_RESTRICTED) of NFS are
returning the wrong value if chown is not
restricted.
Resolution:
Created a kernel variable that allows toggling
between the existing behavior and the POSIX
conformance. This fix is disabled by default. To
enable the fix, set the NFS kernel global variable
pathconf_chown_restricted_posix to 1 by patching
the vmunix file and rebooting, or by patching the
kernel image in memory. To patch the vmunix file
(survives a reboot), use this command:
echo 'pathconf_chown_restricted_posix?W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
To patch the in-memory kernel (does not survive a
reboot), use this command:
echo 'pathconf_chown_restricted_posix/W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
A kernel regeneration will require these steps to
be done again.
libnfs.a mount
JAGae36449
This deadlock lock happens only in loopback mounts.
This happens while adding an entry into directory
name lookup cache (DNLC) for a file. The NFS server
attempts to hold a vnode lock that is already held.
This leads to a deadlock situation.
Resolution:
Avoid DNLC caching of removed NFS files in NFS
loopback mounts.
libnsl.1/a
JAGae52752
Mutex variables are not initialized which causes
the calling process to hang or terminate.
Resolution:
Functions for initializing the mutex variables
were added.
libnsl.1/a
JAGae46451
File descriptors are not closed when the server
detects an RPC_PROGVERSMISMATCH error.
Resolution:
Modified libnsl to close open file descriptors and
free transport handles when an RPC_PROGVERSMISMATCH
error is returned to the client.
libnsl.1/a
JAGae55409
XDR stream manipulation is not handled correctly
in some cases.
Resolution:
Modified the XDR code to correct the problem.
librpc.a
JAGae53431
An NFS/TCP operation on an NFS/TCP client can
cause the NFS client to panic. Problem occurs when
an NFS server requests an orderly release of the
connection while an NFS client is setting the
TCP_NODELAY option. The panic occurs when the
orderly release request from the NFS server arrives
after the NFS client has sent the TCP_NODELAY
request to the transport but before the transport
has replied.
Resolution:
The NFS/TCP client code was modified to ignore
orderly release messages when waiting for a reply
to a TCP_NODELAY request.
librpc.a
JAGae20331
An error condition returned while binding to a
reserved port is not handled in the code.
Resolution:
The error condition returned while binding to a
reserved port is now handled properly.
librpc.a
JAGae36003
The system reports "vmunix: bindresvport:
loop_twice port assigned 512" many times in
syslog.log, even though there are reserved ports
available.
Resolution:
Provide a kernel variable that by default disables
these messages. To re-enable these messages, the
kernel parameter 'loop_twice_msg' must be set to
one, by patching the vmunix file and rebooting, or
by patching the kernel image in memory. To patch
the vmunix file (survives a reboot), use this
command:
echo 'loop_twice_msg?W 0d1' |
adb -w /stand/vmunix /dev/kmem
To patch the in-memory kernel (does not survive a
reboot), use this command:
echo 'loop_twice_msg/W 0d1' |
adb -w /stand/vmunix /dev/kmem
A kernel regeneration will require these steps to
be done again.
librpc.a
JAGae46615
While binding a TCP connection, an uninitialized
variable causes the connection to incorrectly be
in the LISTEN state.
Resolution:
Initialized the variable to fix the state of the
connection.
librpc.a
JAGae48906
When copying 32-bit XDR long variables to 64-bit
kernel variables, the upper bits are left the way
they were before the copy.
Resolution:
Added code to make sure the upper 32 bits of 64-bit
variable data is now properly set.
nfsstat
JAGae39627
When "nfsstat -m" is executed from the client, a
statfs call is made across the wire to each server
listed in the /etc/mnttab file. The statfs call waits
for a response from the server, so if no response is
received, the client hangs.
Resolution:
With the HP-UX Device IDs product (DeviceIDs11i)
installed, the device id information is available in
the /etc/mnttab file on the client and the statfs
call is not made across the wire.
rpc.lockd
JAGae48827
A certain sequence of shared (read) locking and
partial unlocking on overlapping regions by multiple
processes over NFS causes the server's lockd to skip
unlocking a portion of the range in the kernel,
thereby losing track of these locks. These locks
accumulate over a period of time and consequently
exhaust all the available lock structures in the
kernel, at which time the kernel can no longer give
out any locks.
Resolution:
The code in lockd dealing with unlocking regions in
the kernel for overlapping shared (read) locks has
been rewritten. The server lockd now correctly
identifies non-overlapping regions and issues
unlock requests whenever necessary.
PHNE_28102:
automount
JAGae52325
Incorrect permissions for the log file.
Resolution:
Automount has been modified to set the proper
permissions for the default log file.
automount automountd
JAGae50502
Incorrect configuration of AutoFS.
Resolution:
The AutoFS code has been modified to address the
problem.
domainname domainname.1 ypinit ypserv ypxfrd
JAGae40412
ypxfrd picks up files with the wrong pathname.
Resolution:
Code changes were made to check for the wrong pathname.
Some restrictions regarding invalid domain and map
names are now enforced. The domain and map names of "."
and ".." and those containing "/" are now rejected.
libnsl.1/a
JAGae38632
Showmount dumps the core with SIGSEGV.
Resolution:
Code has been added to address the problem.
libnsl.1/a
JAGae38579
RPC does not function properly.
Resolution:
RPC has been modified to correct the problem.
libnsl.1/a
JAGae45826
When netdir_getbyname() calls ypmatch a hang can result
under certain circumstances.
Resolution:
The logic in ypmatch has been modified.
rpc.yppasswdd
JAGae40093
The rpc.yppasswdd process dumps the core.
Resolution:
rpc.yppasswdd has been modified to correct the problem.
PHNE_27217:
automountd libautofs.a
JAGae28647
This product update contains minor enhancements
required to enable the HP-UX Device IDs feature.
When AutoFS unmounts a filesystem, it obtains
the device id from the filesystem server. If
that server is slow, or not responding, there
may be a delay of 15 seconds or more. On systems
with many AutoFS managed filesystems and many
slow or non-responding servers, the cumulative
delay may be quite significant.
Resolution:
HP-UX is implementing Device IDs in the
/etc/mnttab file. In concert with changes in
libc and File Systems, AutoFS will now make use
of the Device IDs and not query each server when
performing unmounts.
automountd
JAGae21158
The /var/adm/automount.log file is shipped incorrectly.
Resolution:
The /var/adm/automount.log file is now shipped properly.
automountd
JAGae09276
1. If there are multiple mount requests for a
non-existent server, then the automount daemon
tries to move these pending requests to cache.
When this operation causes the cache size to
exceed a certain threshold value, the automount
daemon prints "dup_remove_pending: WARNING:
duplicate request cache size (210) >
XID_CACHE_SIZE (200), continuing" to the
syslog.log file.
2. When a mount request fails, the automount
daemon first removes the child node, and then
tries to remove the parent node. If the parent
node is mounted or if there are other files in
parent directory, the automount daemon logs
"Rmdir: err=247, mnt='<mount point-directory>'"
to the /var/adm/automount.log file.
Resolution:
1. The trace level must now be set to 10 (instead
of greater than 0) to log the message
"dup_remove_pending: WARNING: duplicate request
cache size (210) > XID_CACHE_SIZE (200),
continuing" to /var/adm/automount.log. It will be
logged to /var/adm/syslog/syslog.log only if the
verbose option is enabled.
2. The trace level must now be set to 10 (instead
of greater than 0) to log the message
"Rmdir: err=247, mnt='<mount point-directory>'"
to /var/adm/automount.log.
automountd libautofs.a
JAGad98430
The AutoFS code would incorrectly set the
MF_MOUNTED flag for an AutoFS autonode. This
could happen during an AutoFS force remount or
an AutoFS unmount operation.
Resolution:
Change AutoFS code to correctly set the MF_MOUNTED
flag for AutoFS force remount and AutoFS unmount
operations.
automountd
JAGae12335
When an automount daemon's mount request fails,
it counts the number of slashes in the exported
file system (e.g. /dir/) to remove the directories
that are created at the time of mounting. In the
/etc/exports a missing leading "/" entry
(e.g. export/dir/subdir) or an entry that contains
an extra "/" (e.g. /export/dir/) can result in
automountd counting too many "/"s. Since the
automount daemon counts the slashes from the
exported file system (/etc/xtab), the automount
daemon counts one extra slash. This causes the
automount daemon to unlink an extra autonode, which
is supposed to be unlinked by AutoFS. AutoFS tries
to unlink the autonode that has already been
unlinked by the automount daemon, which causes the
panic.
Resolution:
The automount daemon now ignores the trailing
slashes, and thus does not try to unlink the
autonode that AutoFS unlinks.
automountd
JAGae04152
The unmount module of automountd is unable to
find the difference between the mount points with
ESTALE errors and the mount points that are to be
unmounted. As a result, apart from unmounting the
mount points that are to be unmounted, mount
points with ESTALE errors are also unmounted.
Because of this, corresponding direct AutoFS
entries for those mount points are removed from
the file /etc/mnttab.
Resolution:
In automountd, the unmount module, which finds
the mount points that are to be unmounted, is
modified so that it is able to find the difference
between the mount points with ESTALE errors and the
mount points that are to be unmounted.
automountd libautofs.a
JAGae03154
The parent file system has inconsistent hierarchical
mount points when the automount daemon cannot remount
them at the time the unmount of the parent file system
fails. The unmount of the parent file system can fail
if a process is accessing that file system at the same
time the automount daemon is trying to unmount it. The
automount daemon could not remount the hierarchical
mount points due to unavailability of the remote mount
daemon (i.e. the remote mount daemon is not running).
Resolution:
The automount daemon has been modified so that it
informs the AutoFS client whether or not it could
remount the hierarchical mount points when the unmount
of the parent file system fails. AutoFS has been
modified to act based on the return value from the
automount daemon. AutoFS sends a remount request for
the autonode if the automount daemon could not remount
the hierarchical mount points under the parent file
system.
automountd
JAGae24031
AutoFS problem.
Resolution:
Fixed the AutoFS problem.
clear_locks clear_locks.1m
JAGae14669
If the /var/statmon/sm contents are destroyed on the
client, the client's rpc.statd will not know which
servers to notify that the client has crashed and
recovered. Since the server never receives
notification of the client system reboot, it will
continue holding onto locks for this client that are
now "stale". The server will also send any replies to
new lock requests to the old lockd port number on the
client. In most cases, the new instance of lockd on
the client will not be listening on the same port as
the previous instance so it will not receive these
lock replies. The result is that future lock requests
from this client will hang, causing the applications
issuing these lock requests to hang.
Resolution:
The clear_locks(1M) command is provided to allow a
root user to forcibly notify the NFS server's
rpc.lockd to discard all locks and cached port
information for a specified client. This tool should
only be used in cases where a client system experienced
a state change (i.e. re-installed or had rpc.lockd or
rpc.statd killed and re-started) and was not able to
successfully notify the server system using the normal
lockd/statd recovery mechanism.
libautofs.a
JAGae35462
A hang occurs when an ls attempts to access directories
from a dead server. AutoFS will send mount requests for
all the autonodes under the same parent directory
(e.g.: /net) when any one of the subdirectory tries
(e.g.: /net/serverx) has failed during a mount request.
If there are a lot of entries (100+) under the same
parent directory, then it will take a long time to
respond to the ls command.
Resolution:
AutoFS does not send out a mount request if the
file system was already mounted.
libautofs.a
JAGae41178
There is a race condition in the AutoFS kernel code
where two threads executing the same code can cause
one of the threads to wait forever. The thread that
waits forever causes the AutoFS mounted file system
to be busy, preventing it from ever being unmounted.
Resolution:
The AutoFS kernel code was modified to prevent this
race condition. The race condition is avoided by
changing the conditions for which a thread will
block when executing this particular AutoFS kernel
code.
libautofs.a
JAGae01767
The AutoFS code could send a mount request for an
AutoFS mount point that already had a mount in
progress. This led to the creation of corrupt
AutoFS autonodes which showed up as empty
directories.
Resolution:
Modified AutoFS to check for mounts in progress
before sending a mount request.
libautofs.a
JAGad91027
The system is running short of memory while AutoFS
is trying to get the memory to form the unmount
list. AutoFS is holding the spinlock and M_WAITOK
is being passed as an argument to the memory
allocation function. When the system is running
short of memory, the memory allocation function
does not allow a spinlock to be held with M_WAITOK,
and the system panics.
Resolution:
While AutoFS is trying to get the memory to form
the unmount list, M_NOWAIT is passed instead of
M_WAITOK as an argument to the memory allocation
function.
libautofs.a
JAGad95840
After a manual unmount, two accesses take place
simultaneously. One access is forced to initiate
a mount, and before that mount completes, the other
access dereferences a pointer that hasn't been set
yet. This causes a data page fault which results in
the panic.
Resolution:
Add a condition to check for a mount-in-progress
which makes the second access follow an existing
code path to wait for the first mount to complete.
libautofs.a
JAGae15267
For direct offset mounts, the "cd" command calls
auto_lookup in the AutoFS kernel for the offset
directory. At this moment, the parent directory
is at the mounting status and waiting for this
lookup to be completed. But the auto_lookup routine
sleeps when the parent directory is mounting, so the
"cd" command hangs until the user presses CTRL-C.
The CTRL-C kills the kernel thread and leaves
automountd in an unknown status, so users cannot
kill automountd.
Resolution:
The auto_lookup routine was modified to check if
the calling process is automountd and the autonode
has child nodes linked to it (which means it is a
direct offset mount). If it is direct offset mount,
then auto_lookup does not sleep and thus the lookup
completes.
libnfs.a
JAGad16564
Problem with privileged groups.
Resolution:
Fixed problem with privileged groups.
libnfs.a
JAGae11799
When a server does not support a PV3 link
system call from a client, the server returns
an error message which hangs or panics the
client system.
Resolution:
When the server returns the error message for
the PV3 link system call, the client now
releases a semaphore to avoid the deadlock/hang.
libnfs.a
JAGae21991
If a kill signal is delivered to a thread sending
an unlock request on a single CPU machine, the
thread enters a tight loop in order to clean up
the locking information on the server. This causes
the thread to hold the CPU and the machine hangs.
Resolution:
When the interrupt error (EINTR) returns to the
kernel lock manager due to signals, the kernel
lock manager makes the next RPC call
uninterruptible, which causes threads to sleep
during the RPC calls and releases the CPU.
libnsl.1/a
JAGae28412
When a 64-bit RPC application uses UNIX authentication,
the client credentials are decoded incorrectly because
the XDR stream pointer is not positioned properly in
the XDR stream. Thus 64-bit RPC applications fail.
Resolution:
Modified libnsl to decode the UNIX credentials properly
for 64-bit applications that use UNIX authentication.
libnsl.1/a
JAGae18419
On the server, a transport handle is created for
each client portmap request. If the request is
invalid, rpcbind ignores the request and does not
destroy the transport handle. Thus, the size of
rpcbind increases. Eventually rpcbind reaches a
limit and stops responding.
Resolution:
Rpcbind was modified to destroy the transport
handle when it finds that the connection oriented
rpc request is invalid.
librpc.a
JAGae21525
Access to a TCP AutoFS mounted directory hangs
because the unmount from the previous access
failed. This was due to the connect address
structure not being fully initialized during
binding to the port when the connection was
established.
Resolution:
Modified the code to completely initialize the
connect address structure.
librpc.a
JAGae29556
System panics if all the entries in duplicate
cache are marked DUP_INPROGRESS, when a new
duplicate cache request arrives after reaching
cache maximum limit.
Resolution:
Duplicate cache requests will return an error
if cache reaches maximum limit and all entries
are marked DUP_INPROGRESS.
mountd.1m
JAGae15949
The '-p' option for rpc.mountd described in
the man page is obsolete.
Resolution:
Modify the man page to indicate that the '-p'
option for rpc.mountd is obsolete.
mountd.1m
JAGad01965
The '-e' and '-n' options for rpc.mountd described
in the man page are no longer supported.
Resolution:
Change the man page to remove references to options
'-e' and '-n'.
nfs.core
JAGae21159
The /var/spool/rexd directory is shipped incorrectly.
Resolution:
The /var/spool/rexd directory is now shipped properly.
mount umount
JAGae04352
During an NFS mount operation, the client performs
a bind operation. When all ports are in use, the
error message returned is "system error" (t_errno = 8)
instead of "address busy" (t_errno = 23).
Resolution:
Return the error message "address busy" instead of
"system error" when all ports are busy.
nfsconfcol
JAGae03255
The nfsconfcol script is not removing the target
and temporary data files before writing data to
them. If the script aborts, the files may be wrong.
Resolution:
Remove the target and temporary data files before
writing to them.
rpc.lockd rpc.statd
JAGae15966
To obtain the port information of client's rpc.lockd,
a server needs to communicate with the client's
portmapper. A file descriptor is allocated on the
server and it polls to check whether replies arrive
from the client's portmapper. For this polling process,
the file descriptor should be registered with the
polling list. However, file descriptors with values
that are multiples of 32 do not get registered, and
thus the server does not get the port information.
Resolution:
Fixed the routine that registers the file descriptors.
rpc.lockd
JAGad49991
The client currently does not retransmit the blocked
lock requests. As a result, if the NLM_GRANTED_MSG sent
by the server is lost and it doesn't retry the request,
the blocked lock requests on the client will wait
forever.
Resolution:
The client lockd is modified to retransmit the blocked
lock requests. If the NLM_GRANTED_MSG is lost and the
server does not retry then the retransmission of the
blocked lock requests will prevent the hangs on the
client.
rpc.lockd
JAGae24149
When rpc.lockd starts after a crash, it sends a
crash notification to rpc.statd. If rpc.statd is
not yet registered with rpcbind, rpc.lockd does
not re-try the crash notification, and crash
recovery does not take place. This causes apparent
hangs to clients trying to lock files.
Resolution:
Modified the rpc.lockd code to retry the crash
notification to rpc.statd 20 times, due to the
delay in rpc.statd registration.
rpc.lockd rpc.statd
JAGae10618
If /var/statmon/sm directory is destroyed on server
side, rpc.lockd does not notify clients that the
server has crashed and recovered. If a client tries
to contact the server, it cannot get the new port
number until 35 minutes have elapsed since the last
contact with the server's rpc.lockd. This is because
when the time value is checked to verify if the port
number is valid, the time value is converted to
microseconds, and this results in an overflow
condition that causes the delay.
Resolution:
When verifying if the port number is valid, only
compare the time value in seconds, and thus the
overflow and the delay do not occur.
rpc.mountd
JAGae19335
While performing multiple concurrent exportfs
operations, the /etc/xtab file's modification
time remains the same. This leads to incorrect
mount and showmount output, and at least one
of the exported directories cannot all be mounted.
Resolution:
hanged the logic so that rpc.mountd considers
both the modification time and the file size of
the /etc/xtab file to update the output of the
mount and showmount commands.
PHNE_27213:
{Site Specific Patch}
PHNE_26387:
libautofs.a
JAGae15022
There was a race condition in the AutoFS code which
could result in an autonode hierarchy being removed
twice. When this happens the system panics.
Resolution:
The code which could cause the hierarchy to be
removed a second time has been removed.
PHNE_25626:
automount
JAGad88667
The performance problem is caused by the automounter
generating a tremendous number of NIS packets over the
network. This is caused by a series of infinite
recursive calls to access a non-existent entry in the
NIS map.
Resolution:
Infinite recursive calls were stopped by implementing
a function that checks whether the file causing the
problem was already opened.
libautofs.a
JAGad90351
If "control-c" is pressed within 75 seconds of issuing
an "ll" command across an automounted file system when
the server is down, the command is interrupted properly.
However, if another "ll" command is performed at least
75 seconds after the first "ll" command, the command
hangs forever. In both cases, a "server not found"
message should be returned and terminate the "ll"
command if the command is not manually terminated with
"control-c" in 75 seconds.
Resolution:
Remove the stale autonode from the autonode list in
autofs after a forced remount.
libnfs.a
JAGae06322
An NFS client does not return an error message when a
user's write(cp) operation exceeds its assigned quota
limit on a server's PV2 mounted file system.
Resolution:
If an NFS client exceeds the quota limit assigned on a
server's PV2 mounted file system during a write
operation, it will get a quota exceeded error message.
libnfs.a
JAGad97363
An unusual timestamp is set for files created with
exclusive mode flag O_EXCL on pv3 mounted file systems
for Linux or Network Appliance servers. These servers do
not support exclusive mode for creating files and the
client does not correct the timestamp.
Resolution:
The client resets the unusual timestamp returned by the
server to the correct 'server time'.
libnsl.1/a
JAGad74175
When the NIS+ server becomes busy and an NIS+ client
contacts the server, the client should create another
handle to contact the server since the server is invalid.
When this repeats more than a few iterations, the client
uses the same handle to contact the server again and
again, even though the server is busy. This is because
the checking for a valid server is done only if the
number of failures is less than the hard-coded limit.
Resolution:
Check whether or not the NIS+ server handle is valid
before checking for the number of failed iterations.
libnsl.1/a
JAGad87526
When an RPC server receives the last packet from the
client, it should flag the connection as dead and close
the file descriptor. The XPRT flag was set on the parent,
but checking is performed on the child's XPRT flag. The
SVC_FAILED flag was copied to the parent's XPRT flag and
not to the child's XPRT flag. The server develops a file
descriptor leak and eventually fails to process the
client requests when the process limit is reached.
Resolution:
Change to copy the parent XPRT flag to the child XPRT
flag once the connection is dead.
libnss_nisplus.1
JAGad90236
The user account being reactivated remains locked on
trusted NIS+ clients.
Resolution:
NIS+ locks are now detected in the PAM module instead
of the NIS+ backend library.
librpc.a
JAGad42168
If an HANFS package has been failed back to its original
server in a period of less than five minutes after it
was failed over to a new node, and a given client has
been trying to access the package over TCP the entire
time, then the TCP connection the client will try to
communicate over will still exist on the original server
host, but its state will be stale relative to the
client's TCP connection endpoint. This results in
multiple retransmissions over TCP, timeouts in
client-side RPC, and RPC call retries by NFS.
Resolution:
Have the client's RPC monitor its TCP connections more
intelligently, and initiate brand new TCP connections
over a different port when it receives indication that
transport has failed due to a peer reset by the server.
nsswitch.conf.4
JAGad57842
Additional source support for nss backends from the ldap
directory server has been released since March 2000. The
man page for nsswitch.conf needs to have ldap added as an
additional source for nss backends.
Resolution:
The ldap information is added into the nsswitch.conf man
page.
rpc.rusersd
JAGad94209
rpc.rusersd supports a maximum of 180 users. rpc.rusersd
code did not check for this limit and thus caused the
program to abort when memory outside of the structures
that kept track of the user information was accessed.
Resolution:
In rpc.rusersd code, a check is made on the number of
users whose information is read and stored. If that
number exceeds 180, further reading of user information
is stopped.
rpc.statd
JAGad93745
The server's statd process sends a notify message with
its hostname only. If the client mounted the file
system using an interface name different than the
server hostname, then it will reject the notify message
since its monitor entry does not match the server
hostname. As a result the client file lock recovery
will not take place.
Resolution:
The server's statd process was modified to build a list
of all local interface names during startup and send
notify message to all the names in the list when a
reboot occurs. The client accepts the notify message
matching with its monitor entry and notifies the lockd.
The lockd recovers all the locks on the server.
rpcgen
JAGad99710
Excessive error messages are logged to the console
device.
Resolution:
Open the null special file instead of the console and
redirect the stderr and stdout to the null special file.
ypserv
JAGad95797
An HP-UX server does not allow a blank entry in the map,
whereas a Solaris server does. When the /etc/services
map contains blank lines, a Solaris server creates blank
entries in the map. When the master server is Solaris
and the slave server is HP-UX, then the slave server
clients are unable to resolve the services after the
blank entry in the map.
Resolution:
The check for a blank entry is removed on the server
side.
PHNE_25484:
rpc.lockd
JAGae12206
rpc.lockd can core dump, with PHNE_24034 or
PHNE_24909 installed, if a local variable has not
been initialized.
Resolution:
Changed routine to initialize the local variable.
PHNE_24909:
libnfs.a
JAGad91730
Under certain conditions (memory pressure, file
system full), when a system is configured as an NFS
client and server using NFS-loopback mounts, a
deadlock condition can occur. NFS buffers remaining
in buffer cache that have active vnodes are re-used.
Resolution:
Dependency on PHKL_25525.
When writing dirty buffers, added "B_ASYNC" to ensure
buffer is eventually invalidated.
libnfs.a
JAGad74189
The mount option 'grpid' does not return the correct
'gid' value for the 'touch' command on HFS files sytems,
or for the 'mkdir' command on HFS and VxFS file systems.
Resolution:
Modified NFS client code to set the correct 'gid' value
when the file system is mounted with the 'grpid' mount
option.
libnsl.1/a
JAGad74954
The problem occurs when /etc/exports contains either
a large number of -access options for an exported
file system or a large number of exported file systems.
The recursive nature of XDR calls causes the process
to be unable to extend thread stack size, which leads
to the hanging/coring of automountd.
Resolution:
Increase the default RPC thread stack size at time
of thread creation.
libnsl.1/a
JAGad81430
The function called by passwd and nispasswd to
search for an entry in any NIS+ table skips any
leading blank characters in the search value.
Then, passwd/nispasswd obtains the new passwd from
the user and updates the NIS+ tables with the
username containing leading blanks.
Resolution:
Modified the function to include the leading blanks
in the search value.
libnss_nisplus.1
JAGad85717
The shadow password field is modified when a
user changes the password in an NIS+ trusted
environment. The new value causes getpwent
to assume that password aging is enabled.
This also occurs when a new NIS+ user is
added with SAM. This is because certain
values in the shadow password field are
interpreted incorrectly.
Resolution:
If an invalid character is encountered in the
shadow password field, set the field value so
that password aging is disabled.
libnss_nisplus.1
JAGad88698
Default system attributes are not assigned to
NIS+ users that log on to trusted NIS+ clients
for the first time.
Resolution:
The NIS+ name service switch library,
libnss_nisplus.1, was modified to set system
default attributes for NIS+ users that have
never logged on to the trusted NIS+ client.
yppasswdd.1m
JAGad68631
There are some command-line options for yppasswdd:
"-nogecos -noshell -nopw -nohome". If these options are
used, then those fields may not be changed remotely
using passwd(1). The -nohome option was not documented
in the yppasswdd man page.
Resolution:
Added the -nohome option to the yppasswdd man page.
PHNE_24034:
automountd
JAGad21864
During unmount operations, automountd incorrectly returns
a success error code to the kernel, which causes the
kernel to remove autonode data structures. Subsequent
events then cause autonodes to be created with partial
data, which results in the bdf command anomaly.
Resolution:
Fix automountd so it returns the correct error code to
the kernel during unmount operations.
automountd
JAGad65770
The unmount command failed for a mounted cachefs type
file system when trying to unmount with AutoFS
automountd. The cachefs unmount command was missing a
parameter that unmount included.
Resolution:
Added the new parameter for the cachefs unmount command.
automountd
JAGad30096
The AutoFS automountd caused the unmount command to fail
for mounted hfs and cdrom type file systems.
Resolution:
Added a new parameter for the hfs and cdrom unmount
commands.
libautofs.a
JAGad45257
Due to a race condition in AutoFS, autonodes are getting
created with NULL mount flags. The affected AutoFS mount
points show empty directories. A system reboot is
required to clear the condition.
Resolution:
Modified the code to eliminate the race condition.
libautofs.a
JAGad62727
For direct mounts, AutoFS did not wait for mounts to
complete when the direct mount point was accessed. This
triggered a second mount, creating a duplicate entry in
the /etc/mnttab file.
Resolution:
Added code in AutoFS to wait for mounts in progress for
direct mount points.
libnfs.a
JAGad66201
File access performance is slow on systems with large
memory and buffer cache.
Resolution:
Restore the performance version of the NFS code to walk
through buffer cache. The problem described in
JAGad32551 is also fixed.
libnfs.a
JAGad72416
If there are biods running in the system, asynchronous
reads are done by the kernel on behalf of the reader
process, and asynchronous writes are done on behalf of
the writer process. The problem with this is that
asynchronous write requests will be enqueued faster than
asynchronous read requests, thus monopolizing the biods,
and hence causing starvation for the asynchronous read
requests.
Resolution:
Modified the NFS read logic to avoid doing read-ahead
(asynchronous reads) if the biods are "too busy".
However, this fix is disabled by default. In order to
enable it, the kernel parameter
'async_read_avoidance_enabled' must be set to one, by
patching the vmunix file and rebooting, or by patching
the kernel image in memory. To patch the vmunix file
(survives a reboot), use this command:
echo 'async_read_avoidance_enabled?W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
To patch the in-memory kernel (does not survive a
reboot), use this command:
echo 'async_read_avoidance_enabled/W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
A kernel regeneration will require these steps to be
done again.
libnfs.a
JAGad15675
The code does synchronous I/O when the file is locked.
This means that the biods are bypassed and the I/O is
performed by the calling thread.
Resolution:
Do not set the RNOCACHE flag for a whole-file lock.
Also, flush the buffer cache for each lock/unlock
request. Thus, fresh data is read at lock time and all
data is written before releasing the lock. Note that
for partial locks, the behavior remains unchanged. This
fix is disabled by default. In order to enable it, the
kernel flag 'nfs_new_lock_code' needs to be turned on,
by either patching the vmunix file and rebooting, or by
patching the kernel image in memory. To patch the
vmunix file (survives a reboot), use the following
commands:
echo 'nfs_new_lock_code?W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
To patch the in-memory kernel (does not survive a
reboot), use the following commands:
echo 'nfs_new_lock_code/W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
A kernel regeneration will require these steps be done
again.
libnsl.1/a
JAGad76637
rpcbind and other RPC/XDR applications can be adversely
affected with SIGBUS.
Resolution:
Added code to address the problem.
librpc.a
JAGad66499
The system is running out of reserved ports.
Hundreds of processes are trying to gain access to
such ports (ie, perform a lock through lock manager).
Since the lock invocation (via a lockf or fcntl
system call) tries to use reserved ports 512 to 1023
to communicate with the local lock manager, these
ports are taken by the first lockers. If these are
all busy then subsequent lock invocations just loop
trying to get the next available reserved port. The
loop to get the next available reserved port is a
tight hard loop with a one-second delay between
activity to check whether any of the 512 reserved
ports are available. These lock invocations are the
ones actually starving the processes that first
obtained the reserved ports and they are sending
the messages to the message buffer. Thus, a virtual
deadlock occurs.
Resolution:
A backoff for the delay between successive attempts
at obtaining a reserved port is now provided. The
delay is progressive in that it starts at 1 second
and is doubled until it is 32 seconds. This allows
for those processes with a reserved port to run and
complete, and therefore give up their reserved port
to the processes waiting for one.
librpc.a
JAGad37405
NFS/TCP incorrectly handles the case where incoming
messages are being processed for a stream, as that
stream is being closed by the nfsd associated with TCP.
In this case, replies to messages being processed do not
get sent down the stream due to the close. However,
the stream cannot complete the close until all the
messages have had replies sent. Thus, the nfsd
associated with TCP hangs in its close.
Resolution:
NFS/TCP changed to not wait for all messages to have
replies sent before closing a stream. No data is lost
since messages are retransmitted if the client didn't
receive a reply.
nfs.client
JAGad50904
There is a window of time between the "Stop NFS client
subsystem" and "Unmount file systems" steps during
system shutdown. If there are active lofs mounts managed
by AutoFS, and an autofs unmount request is sent during
this window, shutdown will hang at the "Unmount file
systems" step. This is because automountd is killed at
the "Stop NFS client subsystem" step and thus cannot
process mount and unmount requests. The unmount request
pends forever.
Resolution:
In the nfs.client script, prior to killing automountd,
increase the AutoFS unmount timeout to a very large
value so that no detrimental unmount requests are
initiated before shutdown concludes.
nis.client, ypbind, ypinit, ypinit.1m, ypserv.1m
JAGad70794
NIS clients do not have a mechanism to bind to a
particular NIS server.
Resolution:
Modified the ypinit script to include a new option
"-c" for construction of an NIS servers list. The NIS
servers list is constructed by invoking "ypinit -c"
and is stored in /var/yp/binding/<domain_name>/ypservers.
The list of NIS servers should be in the order of
preference that ypbind can try to bind.
The fix for JAGad70794 is not enabled unless the ypinit
script is copied from /usr/newconfig/usr/sbin to
/usr/sbin/. To backup the current ypinit script and
enable the new ypinit script, do the following steps:
mv /usr/sbin/ypinit /usr/sbin/ypinit.bak
cp -p /usr/newconfig/usr/sbin/ypinit /usr/sbin/ypinit
When disabling the fix for JAGad70794 and migrating
back to the previous ypinit script, the file
/var/yp/binding/<domain_name>/ypservers will need
to be deleted:
mv /usr/sbin/ypinit.bak /usr/sbin/ypinit
rm /var/yp/binding/<domain_name>/ypservers
rpc.lockd
JAGad54106
rpc.lockd doesn't use NLM messages in the case of a
loopback mount with a hostname and directly modifies
msg_q entries. In the case of a loopback mount with
an alias hostname, the lockd mistakenly treats it as
a loopback mount with a hostname and modifies the
msg_q entries which causes the process to hang, as
lockd doesn't find the entry when NLM messages are
received.
Resolution:
Added checking for whether the loopback mount is with
a hostname or alias hostname. Do not modify msg_q
entries in the case of a loopback mount with an alias
hostname.
rpc.lockd
JAGad71544
When one process is holding a lock on a file, another
process is waiting for the lock on the same file, and the
two processes are killed at the same time, the lock
remains held by one of the processes.
Resolution:
In the CANCEL request handling routine local_cancel(),
code was added to delete the lock upon a CANCEL request
if the lock is found in the granted queue. This prevents
the lock from remaining after the process is killed.
rpc.statd
JAGad77633
The server's rpc.statd process sends the notify message
to the client's rpc.statd process with the server's
hostname. When the client's rpc.statd tries to match
the hostname with its monitor entry it checks for four
possible names used to mount the server: short name,
long name, alias name, and IP address. The current
algorithm to check the short name and long name in one
step matches the wrong monitor entry, thus causing
problems in file lock recovery.
Resolution:
Separated the short name and long name checking into
two steps, checking only for short name and then
checking for long name.
rpcbind
JAGad34186 JAGad42706
Customer Symptoms:
rpcbind core dumps with SIGBUS or SIGSERV.
rpcbind core dumps with SIGBUG or SIGSEGV due to memory
corruption in the internal linked list of registered
programs. This list is abnormally terminated by memory
corruption and when the invalid node is accessed,
rpcbind can core. This has been noticed on several
occassions including NIS+ startup and random buffer
overflow under heavy load.
Resolution:
The core of memory corruption is in netbufdup().
Changed the malloc to valloc() to increase the required
buffer allocation.
PHNE_23833:
automount
JAGad45102
When automounter triggers new mounts a symbolic
link is added to the automounter vnode. After the
link was added the modification time of the vnode
was not being updated. Since the modification
time of the vnode never changed the rddir cache
was never invalidated.
Resolution:
Add code in the automounter to update the
modification time of the automounter vnodes
whenever new symbolic links are added.
automount
JAGad38290
If an automounter map contains a mount point
which is a symbolic link that points to a
non-existant directory, automounter will hang.
Resolution:
During automounter initialization, add a check
which will ignore these invalid map entries and
log an error message to syslog.
automountd
JAGad29916
Automountd will log "Rmdir" error messages to
/var/adm/automount.log even if automountd was
started without the "-T" option.
Resolution:
Add code to check for tracing before logging the
error message.
exportfs
JAGad40926
Invalid host names in a root option's host list are
not handled correctly. They cause invalid entries
to be created in the root access host list for the
corresponding filesystem. The same type of defect
occurs for the 'rw' option.
Resolution:
Changed the logic that processes the 'root' and
'rw' options, in order to completely ignore invalid
host names.
fsirand.1m
JAGab46804
The manpage does not mention that fsirand is not
supported on a JFS file system.
Resolution:
Added a line to the "fsirand" man page stating
that the fsirand command is only supported on HFS
file systems.
libnfs.a
JAGad48223
Server reads are occurring unnecessarily when a
client is only doing writes.
Resolution:
Removed the read operation that occurred before
the write operation.
libnfs.a
JAGad35150
NFS incorrectly handles a return from a write
operation having a 0 status (NFS3_OK) and a
count of bytes written that is not reflective
of the number of bytes actually written (that is,
a large negative number). By not rejecting such
values, it attempts to use the count of bytes
written to offset into memory causing the panic.
A negative count of bytes written is allowed
because it is cast as signed.
Resolution:
NFS interprets the count of bytes written as
unsigned, thus eliminating any negative values.
libnfs.a
JAGad47845
Function nfs3rename() is implementing the
"unlink-open-file" semantics for the target
name (when it exists), without checking that it
should only do so if the target is not a directory.
Resolution:
Modify function nfs3rename() to only implement the
"unlink-open-file" semantics if the target name
is not a directory.
libnfs.a
JAGad45021
The calculation for determining the size of a file
was using an incorrect constant value.
Resolution:
NFS PV3 calculates the size occupied by a file.
This calculation was based on MAXBSIZE, which is
8192 instead of DEV_BSIZE which is 1024. So it
was eight times too high. Changed this to base
the calculation on DEV_BSIZE.
libnsl.1/a
JAGad52387
Remote users can cause denial of RPC services.
Resolution:
In __svcauth_sys(), change the local variables
strlen, gid_len and i from type integer to
unsigned integer.
libnsl.1/a
JAGad55752
File descriptors used by the TCP connections
are released too soon so that they confuse the
polling thread to create more file descriptors.
The wrongly created file descriptors are never
closed.
Resolution:
In svc_vc_getargs(), comment out the call to
svc_args_done(). This will keep the file
descriptors until the RPC service routine is
totally complete.
libnss_nis.1
JAGad49173
getservbyname() first searches the NIS map
servi.bynp, which can only resolve official
service names. getservbyname() never gets to
call YP_ALL as a last resort to parse through
the other NIS map services.byname for the
service alias.
Resolution:
If getservbyname() cannot resolve the service
name passed to it, call YP_ALL and parse through
the NIS map services.byname.
libnss_nisplus.1
JAGad42007
When in trusted mode, the passwd change should
always create a fully qualified domain name even
if one was not passed into the command.
Resolution:
New code was added to libnss_nisplus.1 to fully
qualify domain names if they are passed non-fully
qualified.
nfsstat.1m
JAGad42623
The nfsstat man page does not have a description
for the badverfs, cantconn, nomem, interrupts, and
cantsend fields.
Resolution:
Added these fields to the nfsstat man page.
rpc.lockd
JAGad49985 JAGad49986
rpc.lockd rejects NLM_GRANTED_MSG when it is
received out of order with NLM_LOCK_RES
(blocked). rpc.lockd ignores NLM_GRANTED_MSG
when it is received out of order with
NLM_UNLOCK_RES.
Resolution:
Modified the processing algorithm for NLM_
GRANTED_MSG. When NLM_GRANTED_MSG is
received ahead of NLM_LOCK_RES(blocked),
it is treated as if NLM_LOCK_RES(granted)
was received. When NLM_GRANTED_MSG is
received ahead of NLM_UNLOCK_RES it is
treated as if NLM_UNLOCK_RES was received.
rpc.lockd
JAGad43270
rpc.lockd puts the lock requests in msg_q while
waiting for a response from the server. If a
process lock request for file 'f1' is
waiting in msg_q, subsequent lock requests from
same process are ignored even if it is for
different file. This can happen with
multi-threaded processes where two threads can
send lock requests for two different files.
Resolution:
Added checking for the file handle in determining
the uniqueness of lock. This enables lockd to
distinguish between requests from same process
for different files.
rpc.lockd
JAGad54440
When a client tries to unlock the first part of
a locked area in an NFS file, the whole lock is
removed in the server.
Resolution:
Fixed the delete_reclock() function to use the
proper argument when calling the
delete_kernel_lock() function.
rpc.lockd
JAGad48424
rpc.lockd logs messages to a log file by
calling the LOG_MSG function. The LOG_MSG
function is being called with incorrect
parameters in lockd when it is logging messages
related to "control-c" interrupt processing.
Resolution:
Corrected the parameters being passed to the
LOG_MSG function.
rpc.mountd
JAGad45622
When booting a diskless Linux client, the
process fails when mounting the root file system.
This happens because the mount request uses the
AUTH_NULL authentication type and rpc.mountd does
not allow this. The mount request fails with a
"permission denied" error.
Resolution:
Allowed rpc.mountd to accept mount requests with
AUTH_NULL authentication type.
ypxfr
JAGad33433
After a map transfer, ypxfr sends a YPPROC_CLEAR
to the local ypserv if ypxfr is not invoked with
a "-c" option. If the local host name is greater
than 8 characters then the call to bind_to_server()
fails with the error "ypxfr: bind_to_server
clntudp_create error: RPC unknown host".
Resolution:
Increased the size of the array local_host_name
from 8 to MAXHOSTNAMELEN characters.
PHNE_23249:
libautofs.a, libnfs.a, librpc.a
JAGab70789 JAGad28779
Threads in a multithreaded process sleeping
interruptibly while holding a non-interruptible
resource are susceptible to deadlock.
Resolution:
NFS threads are placed in a "nonstop" mode in
sections of code where non-interruptible resources
are being held during interruptible sleeps. There
are four other patches which must be installed to
enable the "nonstop" mode: PHKL_23406, PHKL_23407,
PHKL_23408, and PHKL_23409. These patches will be
released in the Spring of 2001.
PHNE_23247:
{Site Specific Patch}
PHNE_22642:
automount
JAGad24909
The function (which loads a direct map file)
involved in the call sequence for handling the
SIGUSR1 signal, was being called with incorrect
arguments which would result in a "segmentation
fault" and coredump when executed.
Resolution:
The function call which was causing the coredumps
was modified to reference the proper argument set.
automountd
JAGad30127
The automountd code simply was not parsing the
quota option out of the options string.
Resolution:
Added a hasmntopt call to check for the quota
option and set the quota flag as needed.
libautofs.a
JAGad04988
Race condition between mount and umount with ufs
filesystems.
Resolution:
Check VFS_MOUNTING flag in make_unmount_list and
get_hierarchical_mounts.
libautofs.a
JAGad29062
When an unmount request comes in to the automountd,
it stats each file system to find the entry with
the matching file system id. The stat command updates
the reference times of the autonodes, preventing them
from unmounting when they should.
Resolution:
AutoFS now uses an existing macro to make sure that
the calling process is not automountd before updating
the reference times of autonodes.
libnfs.a
JAGad48493
Extremely poor performance on a 64-bit (V-class) 11.0
from NFS pv3. All processes accessing NFS were in
_psema_miss except for one that was running in
cpu. This is the same problem this customer had
as described in JAGad16541 (8606147198) NFS
performance problem due to bad compare_cred()
logic in nfs3_vnops.c.
Resolution:
The current calculation sizeof(*cr1) -
sizeof(cr1->cr_ref) - sizeof(cr1->cr_pag) yields
100 bytes. However, cr1->cr_pag is an 8-byte
structure in a 64 bit OS. Since it is 8-byte
alligned, there are 4 bytes between cr1->cr_ref
and cr1->cr_pag that were unaccounted for.
The fix is to #define a size of 96 and use it
in the comparison.
libnfs.a
JAGad31017
The NFS client does not get new attributes during
close-to-open when a file is not cached.
Resolution:
The NFS client now gets new attributes from the
server at file open time if close-to-open
consistency (i.e. MI_NOCTO is overridden at
mount time) is desired.
libnfs.a
JAGad27560
Panic occurs when vfs lock is improperly unlocked
and an attempt to unlock it occurs.
Resolution:
Code modified to check the return value of
vfs_lock and if it is busy, sleep for some time
and try to acquire the lock again.
libnfs.a
JAGad32551
v_count was not decremented properly in the
NFS_BLKINVALFREE macro. In the case of removing
the file, instead of sending the remove rfscall
to the server, the client renames the file to
another name and the file remains.
Resolution:
The fix is to call the routine binvalfree() instead
of the NFS_BLKINVALFREE macro in order to decrement
v_count properly.
libnfs.a
JAGad35601
By masking SIGKILL and SIGSTOP it is not possible to
kill an NFS client process if the server goes down.
Also, SIGKILL and SIGSTOP are non-maskable signals in
HP-UX, hence they should not be masked.
Resolution:
Unmask SIGKILL and SIGSTOP in masked_sigs in
the routine, nfs_subr.c.
libnsl.1/a
JAGad09968
The function authunix_create() calls abort()
if more than 16 groups are passed in args (gids).
Resolution:
Replaced abort() code with return(NULL). This
prevents code from dumping.
libnsl.1/a
JAGad29722
A thread walks off its stack due to a very large
stack variable (struct pollfd svc_pollset) in
svc_run.c, thereby resulting in the corruption
of unrelated memory.
Resolution:
1) Removed the explicit initialization of fd_set
svc_fdset in rpc_comdata.c.
2) Changed the array sizes of svc_pending_fds and
svc_pollset to kernel tunable parameter
maxfiles_lim, and dynamically allocated both
arrays.
3) Reduced FD_SETSIZE in svc_vc.c's
__svc_nisplus_purge_since() to 1024.
librpc.a
JAGad39062
NFS incorrectly handles a machine name length
that is negative by not rejecting such a value
and then attempting to use it as an offset into
an array. A negative machine name length is
allowed because it is declared as signed.
Resolution:
NFS interprets the machine name length as unsigned,
thus eliminating any negative values.
librpc.a,nfsd
JAGad36339
NFS/TCP incorrectly handles the case where a new
thread cannot be created by the NFS TCP kernel
daemon (nfsktcpd). In such cases, it allows data
to come in to be processed though there is no
thread to do such. On the close of the data
stream, the nfsd associated with TCP hangs
waiting for the incoming data to be processed.
Resolution:
NFS/TCP immediately closes the data stream when
its associated thread cannot be created.
newkey
JAGad26923
In changing a public/secret key pair for the
privileged user of a given hostname on the master
server with "newkey", this command fails with
error message, "ERROR, passwd differs from
login passwd". "Newkey" prompts for the root
password of the given hostname, but compares
the given password with the root password of
the master server.
Resolution:
The prompt to ask the root password of the given
hostname is inserted if the password supplied is
different from that of the master server.
PHNE_22125:
libautofs.a
JAGad13210
The root cause involves an atypical use of a for
loop in the force_remount() function of the module
auto_subr.c. Using the for loop to traverse a
linked list, the increment or looping expression
assigned to a pointer variable was a variable that
had previously been assigned to the next item in
the list. The problem here is that the value of
the next pointer may be altered by a different
thread during the execution of the for loop's body.
When this happened, the next loop iteration was
then processing an item that had been released
from use and no longer part of the list. This
situation manifests itself in the above symptom.
Resolution:
Change the code so the for loop's increment
expression does the incrementing.
libnfs.a
JAGac86286
While testing a large (25MB or more) file
transfer from an NFS client to an NFS server,
the NFS server is TOC'ed during the copy.
The client hard mounts the NFS server, so
the copy command will hang waiting for the
NFS file server to come back up. When the
NFS server comes back up, the copy
continues and eventually completes. When
you do a cksum on the local file and the
remote file, the values are different.
Data is missing from the remote file.
Resolution:
Record the write verifier before
collecting the buffers instead of
sending time. Every time when the buffer
is not used, just clear all of its links.
libnfs.a
JAGad14221
HP NFS makes 2 GETATTR RPC calls for each
synchronous read and 1 GETATTR RPC call for
each synchronous write on a file opened with the
O_RSYNC|O_SYNC flags set, while SUN NFS makes 1
GETATTR RPC call per synchronous read and 0
GETATTR RPC calls per synchronous write. An rnode
for an NFS file with the RNOCACHE flag set gets
put on the rnode free list after the file is
closed, and a subsequent open of the same NFS
file will pick up the same rnode from the free
list with RNOCACHE flag set, causing I/O mode to
be synchronous for this subsequent open even
though the O_RSYNC|O_SYNC flags were not set.
Resolution:
Extraneous GETATTR calls in the NFS synchronous
read/write path have been removed. The synchronous
flag RNOCACHE for an rnode is now turned off
during NFS file close to eliminate persistent
synchronous I/O mode across system calls.
libnfs.a
JAGad16541
The compare_cred function was unnecessarily
checking the cr_ref and cr_pag of the cred
structure. It was also redundantly checking
the groups' array.
Resolution:
Removed checking of cr_ref, cr_pag and redundant
check of groups' array.
libnfs.a
JAGad15306
The tar command complains about "Optional
ACL entries". An ll command shows some "+"
signs in the mode bits. A second ll command
shows no more "+" signs. This is caused by
uninitialized bit field va_aclv.
Resolution:
Since NFS does not support ACL's, initialize
va_aclv to 0.
libnfs.a
JAGac84458
A 64-bit machine may panic when executing
a command over nfs hits the path that
calls makenfsnode or makenfs3node. This
happens only on a 64-bit machine and is
caused by an assignment in these routines
that causes non 8-byte alignment.
Resolution:
Remove this assignment that causes non 8-byte
alignment in makenfs3node and makenfsnode.
libnfs.a, librpc.a
JAGad29423
NFS is not working properly for a mount option
specified as interruptible or non-interruptible.
An example, non-interruptible is interruptible
in NFS/UDP.
Resolution:
Fixed the mount intr/nointr option so that it
behaves as expected.
libnfs.a
JAGad13957
NFS incorrectly checks the filemode
flags when opening a file for read-only
access. Because of this, it mistakenly
views the access as writable and
modifies the write credentials. This
causes the "Permission denied" error
in the write operation.
Resolution:
NFS correctly checks the filemode flags
when opening a file. In particular, it
checks the filemode against FWRITE
instead of O_WRONLY and O_RDWR.
libnfs.a, librpc.a, nfsd
JAGad14853
When NFS is stopped and re-started, the nfsd
transport table can reach a state where the table
is full and the new nfsd daemons start faster
than the old nfsd daemons stop. The only way to
get the table to a correct state is to reboot the
system.
Resolution:
Fix the code that controls the nfsd transport
table.
libnfs.a, nfsd
JAGad16228
When NFS is stopped and restarted, and then
additional nfsd processes are started, the
nfsd processes which were started first are
no longer registered with rpcbind. This
causes the rpcinfo command to fail.
Resolution:
Modify the return errno from EINVAL to EIDRM
which will be used in nfsd.
librpc.a
JAGad24321
If a client mounts an NFS file system with an
interruptible mount over TCP when the server is
down, the client is expected to try for a period
of time and quit with an RPC time out error
message. This does not happen in RPC/TCP path,
and the client is hung.
Resolution:
The RPC/TCP code was fixed so it does not sleep
on cv_wait() for a soft interruptible mount.
librpc.a
JAGad13205
This was traced to a 64-bit only issue. On a
64-bit system, a particular arithmetic expression
always resulted in a non-negative result. The
subsequent test for a negative result always
failed therefore sending actual negative results
down the wrong code path. Repeated occurrences
eventually caused the system to hang.
Resolution:
A variable in the arithmetic expression was
changed to type int so the arithmetic could
correctly result in a negative number.
librpc.a
JAGad13818
If for some reason the peer wants to tear the
TCP connection, it sends us a FIN, but then
nothing happens. The TCP connection remains in
the CLOSE_WAIT state and we keep sending NFS
requests on this connection. Eventually the
local inactivity timeout will pop which will
remove this connection. This problem happens on
64-bit machines only.
Resolution:
Changed data types from unsigned long to
unsigned int so that messages such as FIN are
not dropped.
nfs.core
JAGad29173
After installing PHNE_20371 or PHNE_21376 on a
system with a default umask which does not allow
universal read permission, the /etc/services file
permissions are set per the default umask and
non-root processes cannot read the file.
Resolution:
Modify the "/sbin/init.d/nfs.core" script to
preserve original permissions and ownership of
"/etc/services" in the case where it operates on
"/etc/services" to add a service entry for
NFS/TCP, by using "cp -p" on "/etc/services" to
create a working copy for editing, and then
using "mv" to overwrite the original
"/etc/services" with the changed version.
rpc.lockd
JAGad14812
When a client crashes and does recovery after
locking a remote file, the server does not grant
the lock held by the crashed client to other
clients waiting for the lock.
Resolution:
In recovery, server releases the lock information
held by the crashed after waking up a client
waiting for the lock.
rpc.mountd
JAGac68822
The mount function returns the ENOENT error, without
checking the existence of the file in the mount table,
when the realpath() function does not find the path
entry for the file.
Resolution:
The function call "mount_enoent_error" is added to
check the existence of the file or the existence
of a the file name entry in the mount table and
return the error appropriately, when the realpath()
in the mount function fails with the ENOENT error.
rpc.statd
JAGad12739
When the server receives SM_NOTIFY protocol,
the send_notice() in the rpc.statd is looking
for a client from the monitor list.
Then the looping problem between clients and
server occurs which is caused by the excessive
amount of time spent in gethostbyname()
in send_notice()
Resolution:
The lookup procedure in send_notice() is changed.
Instead of calling gethostbyname(),
IP addresses are compared.
ypserv
JAGad25155
When more then two processes execute the
"id/group" command simultaneously, the ypserv
fails to return group information to all
processes.
Resolution:
A local data structure is defined where a
buffer is allocated to keep the nextkey or
firstkey.
PHNE_21376:
automountd
JAGac84450
Some mount points become inactive and cannot be
re-mounted (they are seen as empty directories).
For example, if /sys is an indirect map, we may
end up with ls /sys which shows a directory
"srv_spool", but ls /sys/srv_spool returns an
empty directory (it should trigger an NFS mount).
Resolution:
Two problems were resolved - one in automountd
and one in libautofs.a (see below). The
automountd process was modified to detect
duplicate requests in a more reliable fashion.
exportfs
JAGab78579
During the mount process rpc.mountd on the NFS
server resolves the requested mount path to its
actual path, then it verifies export permission
for this path against the list of export entries.
Previously rpc.mountd would resolve each path
entry in the exports list before verification
comparison, now it no longer does; and exportfs
exports directory paths exactly as specified to
it without resolving linked paths.
Resolution:
exportfs resolves symbolic links to real
directory paths and exports the real directory
paths.
libautofs.a
JAGac84450
Some mount points become inactive and cannot be
re-mounted (they are seen as empty directories).
For example, if /sys is an indirect map, we may
end up with ls /sys which shows a directory
"srv_spool", but ls /sys/srv_spool returns an
empty directory (it should trigger an NFS mount).
Resolution:
Two problems were resolved - one in automountd
(see above) and one in libautofs.a.
The libautofs.a routines were modified to
wakeup all threads sleeping on the mount
regardless of whether the first requestor has
been interrupted, and to preserve partially
mounted autonodes when an interrupt occurs to
allow the mount to complete.
libnsl.1/a
JAGad03728
Since the default stack size for pthread is 64 K bytes,
the large stack usage in functions like nis_cast()
ultimately causes the pthread stack overflow.
Resolution:
Moved the automatic variables which take up a large
stack space to the heap area by malloc'ing the variables.
libnsl.1/a
JAGad13073
The library functions svc_vc_create() and
svc_vc_xprtcopy() incorrectly made duplicate calls
to t_alloc which causes the memory leak.
Resolution:
Removed the second t_alloc() call from both functions.
libnsl.1/a, rpc.nisd
JAGab78042
rpc.nisd forks a child to handle existing connections
when it does checkpointing (nisping -Ca). The child
may close some of these connections. But the descriptors
still stay open in the parent process.
Resolution:
Added a mechanism to clean up the idle file descriptors
periodically in the svc_run() forever loop. The clean up
functions are defined in the libnsl/rpc.
rpc.lockd
JAGad00646
There are routines that call "xmalloc()" but do not
check the value returned. A "NULL" pointer is
returned from "xmalloc()" when no memory is available.
A "NULL" pointer value will cause a "core dump" if used
to access data.
Resolution:
Check the return value of xmalloc() in prot_nlm_share(),
prot_nlm4_share(), new_sxx() and new_sxx4().
Log the reason of failure in these routines.
rpc.statd
JAGad00517
There are routines that call "xmalloc()" but do not
check the value returned. A "NULL" pointer is
returned from "xmalloc()" when no memory is available.
A "NULL" pointer value will cause a "core dump" if used
to access data.
Resolution:
Check the return value of xmalloc() in insert_name() and
log the reason of failure in this function.
rpc.statd
JAGad05198
When an unknown host requests a file lock, rpc.statd
in the server site can not add this client to the
monitor queue as the hostname was not resolved.
So the lock request would be failed. But the unknown
host keeps sending lock requests, which causes the
memory region of rpc.statd to be used up and finally
abort.
Resolution:
The insert_mon() in rpc.statd is called to add a client
to the monitor queue. In case that this routine fails,
free the memory area allocated before this routine is
called.
rpcbind
JAGad01128
Rpcbind writes warm-start RPC service registration
information on shutdown to files in an unsafe way.
Resolution:
Rpcbind modified to handle shutdown file writes properly.
ypserv
JAGad03431
ypserv closes its own UDP socket when it receives the
YPPROC_ALL request.
Resolution:
Added the check in the ypall function to make sure that
the only TCP server handle gets destroyed and not the
UDP server handle.
PHNE_20627:
libnfs.a
JAGad03354
The NFS server "readdirplus" directory read routine was
packing data into its RPC directory information buffer
beyond its 8Kb size, resulting in corrupt directory data
being returned to the RPC layer causing delays in NFS
server response to NFS client readdirplus calls.
Resolution:
The NFS server calculation of directory information size
during packing for response to "readdirplus" calls was
corrected.
JAGab82363
On an NFS client an extra buffer and buffer cache entry
was allocated by read/write path routines when reading
NFS files with size a multiple of the NFS read window;
when such files were removed on the NFS client the extra
buffer was not removed and continued to hold a reference
count on the file vnode causing the vnode subsystem to
think the file was kept open by NFS and the kernel to
create an arbitrary file entry, ".nfsXXXX" for the
vnode.
Resolution:
The NFS client read/write path routines were modified
so that if it appears that an NFS file is going to be
read past EOF, then the cached attributes for the file
are validated and a re-check for read past EOF is done;
which prevents allocation of an extra buffer cache entry
for files with size a multiple of the NFS read window.
JAGad04931
On an NFS client when a directory move was done, the dnlc
entry for the ".." directory entry became invalid but was
not being removed.
Resolution:
The NFS client rename routines were modified to remove
the invalid dnlc entry for ".." for a directory when a
directory move was done.
JAGad04932
On an NFS client the read/write path routines would
return immediately without validating cached file
attributes if a read past EOF was attempted; so, any
unmount attempted flushed the buffer cache entries causing
the NFS client to not know that file size had changed and
a "tail -f" command on the NFS client showed no change in
the file even though it was being appended to on the NFS
server.
Resolution:
The NFS client read/write path routines were modified
so that if it appears that an NFS file is going to be
read past EOF, then the cached attributes for the file
are validated followed by a re-check for read past EOF;
so the NFS client will always see an updated size for
a given file and a "tail -f" command on an NFS file
would follow any appends to the file done on the NFS
server.
PHNE_20371:
autofs/mount
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
automount
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
automount,automountd
automount with stalemounts causing duplicate
mnttab entries & lock problems.
Resolution:
Automount and automountd code was revised to
handle this situation. we now check the mnttab
for 'stale' entries and then remove them.
automountd
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
automountd
Automountd processes executable maps incorrectly.
Resolution:
Automountd modified to handle executable maps
correctly.
automountd
When an autofs map entry specifies the
mount option "proto=udp", an attempt to
mount that entry fails to mount and prints
a message "udp protocol not allowed".
Resolution:
Allow the specification "proto=udp".
User modules: autod_main.c, autod_nfs.c
libautofs.a
When an autofs link is part of a shell pipe,
it can intermittently fail to generate a mount
request. It happens primarily on MP boxes.
Resolution:
Add SIGINTR/SIGUNINTR() calls to mask lower-priority
signals (e.g., SIGCLD) to prevent a race to interrupt
an daemon request while in progress. High priority
signals such as SIGINT and SIGHUP are unaffected.
Modules affected:
auto_vnops.c
auto_subr.c
libautofs.a
autofs controlling cachefs indirect map
may hang on mount or unmount.
Resolution:
Improve detection logic to allow automountd
detection logic to detect mount calling mount
or umount calling umount.
Change applied to AUTOFS_PROCESS_IS_AUTOMOUNTD
macro in kernel.
libnsl.1/a
NFS protocol on HP-UX was originally designed
only to use UDP.
Resolution:
Provide support for the NFS protocol over
TCP.
libnsl.1/a
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
libnsl.1/a
Automountd processes executable maps incorrectly.
Resolution:
Libnsl modified to provide RPC functionality to
support the correct handling of executable maps
by autofs/automountd.
libstraddr.1
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
make.script
When mail.aliases contains no space around ":",
makemap, which is called by ypmake fails.
Resolution:
changed sed in Makefile to substitute a space
in place of a colon.
mount
NFS protocol on HP-UX was originally designed
only to use UDP.
Resolution:
Provide support for the NFS protocol over
TCP.
nfs.core
NFS protocol on HP-UX was originally designed
only to use UDP.
Resolution:
Provide support for the NFS protocol over
TCP.
nfs/mount
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
nfsstat
NFS protocol on HP-UX was originally designed
only to use UDP.
Resolution:
Provide support for the NFS protocol over
TCP.
nisplus/nisserver
When the system hostname is longer than 8
characters "uname -a" reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace "uname -a" command for system hostname with
"hostname".
rpcinfo
When the system hostname is longer than 8
characters uname() reports the nodename as
"unknown", and ONC/NFS subsystems using this
call fail because the local hostname cannot
be determined.
Resolution:
Replace uname() calls for system hostname with
gethostname().
rpc.lockd
rpc.lockd is not able to correctly identify
the case where a lock is local or remote and
take the appropriate action.
Resolution:
Made the changes to remote_data(), the
function that is used to determine if
data being locked/unlocked is at remote side
or not. Put in additional checking when
strcmp() against two host names not the same,
if short name of two hosts are the same,
comparing IP addresses to determine the
result. Also fixed not able to perform wakeup
after doing the local unlock (fix is in
overlap()).
rpc.lockd
The root cause to this problem was that
rpc.lockd retrieves/caches the port number of
remote end after dealing with the LOCK, UNLOCK,
TEST and CANCEL requests, however it only
adds the monitor entry for LOCK requests.
It should monitor that remote end when caches
info for the requests because it would use the
cached info later without knowing that remote
end state changes and its cached port number
is not up-to-date any more.
Resolution:
Made the change to call add_mon() that
regardles of the KLM request we add the
monitor entry.
setoncenv
NFS protocol on HP-UX was originally designed
only to use UDP.
Resolution:
Provide support for the NFS protocol over
TCP.
ypmake
When mail.aliases contains no space around ":",
makemap, which is called by ypmake fails.
Resolution:
changed sed in Makefile to substitute a space
in place of a colon.
PHNE_20344:
libnfs.a librpc.a nfs_iface.o
Reads and writes for NFS version 3 are currently
limited to 8kB.
Resolution:
Enhanced NFS version 3 to support the full 32k
read/write block size.
libnfs.a
rfs3_readdirplus() calls kmem_alloc() even when
the allocation request size is zero, which causes
a minimum of 32 bytes of memory to be allocated
each time.
Resolution:
rfs3_readdirplus() modified to check for and avoid
calling kmem_alloc() when the allocation request
size is zero.
libnfs.a
NFS server code path flow through readdirplus() is
a SPECsfs97 performance bottleneck.
Resolution:
The readdirplus() code path was optimized to get
a much better IOPs number for SPECsfs97.
libnfs.a
On the NFS client an incorrect algorithm for gauging
"memory pressure" and an "nfs_do_purge_rddir_cache"
with a default setting of 1, cause the readdir cache
to be purged too often, reducing performance.
Resolution:
Replaced the MEMORY_PRESSURE macro with a call to
thrashing(). We are interested in purging the
client cache and reusing client cache entries if
and only if we are under high system load. Memory
pressure was not an accurate indicator but the
thrashing call is exactly the correct indicator.
libnfs.a
The HP-specific kernel variable "do_readdirplus" was
removed from NFS server kernel code in a previous
NFS kernel patch.
Resolution:
"do_readdirplus" and support code added back into NFS
readdirplus() code.
PHNE_19759:
1.fuser -f on a NFS file on the client does not show
pids of openers of this file due to missing update
to vnode fid on the client (JAGab68768,JAGab54207) .
Resolution:
The file attribute cache is updated with fid
received from server but is not copied to
the vnode field.This leads to fuser file id
comparision failure.This has been fixed.
2.Invalid name length is returned by NFS pathconf
from HP Server on regular files because HP's local file
system does not support Pathconf on regular files
and when SUN's compress command generates Pathconf
on regular files, server replies with name length
of zero ( JAGab20794) .
Resolution:
On the server,Pathconf() oprn to local FS now
specifies the file directory,rather than the
file itself.This works.
3.getmount_entry() returns down status for all hard
mounted NFS file systems (JAGab69071) ,
Resolution:
An invalid logical bitwise operation of the
mount flags resulted in an invalid mount
status return from a kernel function.
This has been fixed.
PHNE_19620:
automountd
JAGab72053
When an existing autofs mount of a remote NFS
server times out, and the server is unavailable,
the unmount attempt from autofs results in an
indefinite suspension of autofs service.
Resolution:
Put code into autod_mount.c to "ping"
the host before attempting a stat() call.
If the "ping" times out, abandon this
unmount request.
JAGab70009
On a system which has a system setting
of maxfiles > 2048, if replicated automounts
are used, the automountd function nfs_cast()
will return EBADF and fail to mount.
Resolution:
Limit the size of the select() block
to FD_SETSIZE entries or less.
Routine modified: nfs_cast.c::nfs_cast()
-----
libautofs.a
JAGab73529
autofs with cachefs under stress testing
can panic in cachefs_root(). Reported
originally on 10.20.
Resolution:
Additional autonode state testing
in autofs::hpauto_trigger_mount()
prior to calling VFS_ROOT().
Module affected is auto_vnops.c
JAGab72920
When autofs is used with a non-NFS mount
point (such as cdrom), it call fail to
mount when it uses the "mount" command.
Resolution:
Fix detection logic for automountd versus
normal client lookup. Detects if either
process is "automountd" or it is a "mount"
command spawned by "automountd".
Kern modules: hpautofs.h, hpautofs.c
User modules: autod_mount.c
-----
libnsl.1/a
JAGab41115
The address "merging" on a host performed by
rpcbind when it receives and executes its
RPCB_GETADDR procedure can fail if the host
has multiple network addresses.
Resolution:
Modify the function "inet_netdir_mergeaddr"
in libnsl/nss/netdir_inet_sundry.c which
performs address merging to use a better
algorithm by calling a newly added function
get_best_match(), which uses the matching
algorithm from ONC+ v2.2 to perform address
matching.
-----
libnss_nis.1
JAGab20796
getservbyname() looks up a HP non-supported
map "services.byservicename and calls
yp_all when the lookup fails.
yp_all would get the whole map over the
network which is time consuming.
Resolution:
Add a lookup to servi.bynp to avoid the yp_all call.
-----
librpcsvc.a
JAGab70329
If the number of active remote mounts becomes
sufficiently large (approx 23000+), rpc.mountd
will terminate with SIGSEGV when "showmount -a"
is requested of the server.
Resolution:
Modify routine librpcsvc.a::xdr_mountlist()
to traverse a linked list without recursion.
-----
rpc.lockd
JAGab74177
An 11.0 client fails to lock files on a
virtual NFS server if the answer to the
getport call from NLM comes from a different
IP address than the virtual IP address of
the server (i.e., from the address of the
current physical server).
Resolution:
Fix rewriting of address in routine
recv_from_pmap() to:
- Preserve the cached address (if any)
- Use the client handle address rather
than the address of the client reply
when generating a new cached address.
Module affected is pmap.c.
JAGab45462
Using the reserved port number is a common
symptom to all of our RPC services because
we check all possible port numbers within
reserved port range, 600 to 1024.
Resolution:
Use XTI-based API, svc_tli_create() which
triggers kernel to provide the random port
number automatically.
JAGab15223
The lockf() system call, when issued
on a remote file, can permanently hang
if the remote file gets removed.
Resolution:
Within wakeup(), the function to be called
when a delete is successful or when an
exclusive lock is downgraded to a shared lock,
before wakeup taking place make sure that
nfs_fcntl() succeeds or the errno set by
nfs_fcntl() is equal to ESTALE, i.e.
Stale NFS file handle.
JAGaa45544
The root of the problem is when the NFS server
is mounting one of it's local filesystems to
itself using a server hostname other than the
local hostname. This causes both the client-side
and server-side portions of rpc.lockd to
add/delete/enqueue/dequeue entries from the
single set of queues. The end result is that the
client-side rpc.lockd code ends up dequeuing
entries that the server enqueued and vice-versa.
This is what leads to the "lock table
inconsistent" messages.
Resolution:
Within add_wait(), before declaring a reclock
is a match to an existing blocking lock, make
sure that the server is the same as well.
Within wakeup(), removing remote_clnt() to
send a callback to a blocked loopback NFS lock.
-----
rpc.mountd
JAGab45462
Using the reserved port number is a common
symptom to all of our RPC services because
we check all possible port numbers within
reserved port range, 600 to 1024.
Resolution:
Use XTI-based API, svc_tli_create() which
triggers kernel to provide the random port
number automatically.
-----
rpc.statd
JAGab45462
Using the reserved port number is a common
symptom to all of our RPC services because
we check all possible port numbers within
reserved port range, 600 to 1024.
Resolution:
Use XTI-based API, svc_tli_create() which
triggers kernel to provide the random port
number automatically.
PHNE_18221:
libautofs.a
=== JAGab00144 - AutoFS is not triggering re-mount ===
Running AutoFS with very short node timeout (e.g.
"automount -t0") can cause failures to access files
which are present.
=== JAGab03833 - AutoFS hangs even with PHNE_17200 ===
AutoFS with a short node timeout with a script
using manual umounts in a tight loop can hang autofs
after 10-15 minutes.
=== JAGab11283 - cp(1) fails to inactive direct mount ===
When a command is given to copy a file to the
top of an unmounted direct mount point, 'cp' fails
with "Operation not supported".
== JAGab16597 - mv command fails in AutoFS direct mnt ==
When the 'mv' command is used to rename a file
from the current directory and it is a direct mount
point, it will fail if the direct mount is not already
active (mounted), with error ENOSYS.
=== JAGab18083 - getcwd() fails with AutoFS filesys ===
A 64-bit mode compiled program which calls getcwd()
will fail when it is run after connecting to an
AutoFS indirect mountpoint.
Resolution:
[JAGab00144, JAGab03833]
Adjust the logic used to detect special requests
from the "automountd" daemon. Include tests to
avoid race conditions between autofs_proc and file
lookups. Set a minimum hold time for autonodes so
that they are not recycled immediately after being
created.
[JAGab11283]
Add a pathconf() handler for direct mounts.
[JAGab16597]
Add direct mount handler to auto_access() and
auto_rename(). With changes, 'cd' triggers mount.
[JAGab18083]
In auto_readdir2(), if 64-bit capable, force 8-byte
alignment of autofs directory entries.
libnsl.1/a
netdir_options(), fails to find the next available
reserved port due to a logic failure that causes the
search to terminate before all reserved ports have
been checked.
Resolution:
Modify the logic so that all ports will be tested
before netdir_options() fails because there are no
reserved ports available.
libnsl.1/a
getservbyname() calls _yp_new_unbind() which
closes socket after calling clnt_destroy()
which has already closed the socket.
Resolution:
Remove the close socket call after
clnt_destroy().
libnss_compat.1
The read_line function in libnss_compat tries to return
-1 when fgets() in readline() fails. Howerver, 64-bit
read_line() returns a size_t value which is unsigned
that confuses the code that handles the returned value.
Resolution:
change read_line to return int instead of size_t.
rpc.lockd
The problem is that the number of files
that rpc.lockd can open is not set to
the maximum for a process. Related to
that problem is the fact that the error
returned when no more files can be
opened by rpc.lockd is too inspecific.
Resolution:
rpc.lockd now allows for the maximum
number of open files for processes on
a system. Also, rpc.lockd returns
a more specific error when no more
files can be opened.
rpc.mountd
== JAGaa42795 - ER - share(1M) functionality ==
Customer has requested additional exportfs(1M)
mount functionality available for entries in "access="
list. Options include "-" (negate), "@netmask" and
".domain". These options are similar to those offered
by the Solaris share(1M) command.
== JAGab17386 - rpc.mountd allows systems to mount ==
== which are not in access list ==
The original defect of allowing hosts to mount
which were not in the access list had been fixed
in JAGaa44896. Another defect was present in the
"showmount -e" functionality, which would cause
it to incorrectly report what was allowed.
Resolution:
[JAGaa42795]
Added support for options "-", ".domain" and
"@netmask" in 'access=' lists.
[JAGab17386]
Fixed "showmount -e" functionality to report
consistently with post-JAGaa44896 behavior.
rpc.mountd
rpc.mountd uses access lists setup by exportfs
to control access. In the case of
/dir root=hosta:hostb,access=hosta
it is not intended that "hostb" automatically
gain mount access just because it is in the
"root=" list. This is reserved for hosts
in the "access=" list, if one is specified.
In addition, an empty access list:
/dir access=
should open up access to everyone, per Sun
ONC reference behavior.
Resolution:
Fix access list code to closely follow behavior
of Sun ONC+ reference implementation.
rpc.pcnfsd
The problem is that '~' was removed from
the acceptable input characters list
for file names in rpc.pcnfsd.
Resolution:
rpc.pcnfsd now accepts '~' within
file names.
rpc.statd
The problem is that the number of files
that rpc.statd can open is not set to
the maximum for a process. Related to
that problem is the fact that the error
returned when no more files can be
opened by rpc.statd is too inspecific.
Resolution:
rpc.statd now allows for the maximum
number of open files for processes on
a system. Also, rpc.statd returns
a more specific error when no more
files can be opened.
PHNE_17586:
1. NFS client system slows down considerably when reading
huge directories and as the number of active buffers
in buffer cache increase.
Resolution:
Optimized NFS buffer caching,file sync operations
and semaphore handling etc for performance.
2. When a remote application sends a UDP pkt with zero
length to the NFS port of a 11.0 server,the system panics
with data page fault in the NFS/RPC code.
Resolution:
Missing check for invalid pointer value has been added.
3. The server denies access to valid root users on the
client due to failure in address comparision resulting
from uninitialized address fields and address masks in
the exports root list of the server.The failure appears
to be at random depending on the unintialized value
stored in the address fields.
Resolution:
The address field and address mask of the exports lists
is now initialized appropriately at the server.
4. At file open time,NFS PV3 client does not
check file access permissions for root user
instead allows open to succeed,but when reading/writing
the file,server denies access for read/write.When a non-root
user opens the file,access permissions are checked.
Resolution:
NFS PV3 client now examines the file access bits
returned by the server for both root and non-root users.
5. NT client sends the NFS request to HP server
with uid set to -2 and gid value set to an invalid
value of -1,NFS server does not map the uid/gid
values correctly.
Even though anon mapping for a exported file system
is set to -1,anonymous clients are granted access.
Resolution:
NFS server now maps gid value of -1 appropriately when
uid value is set to -2.Also,NFS server correctly handles
anon mapping of -1 by denying anonymous client access.
6. When copying a large file (>2GB) from NFS file
system to local disk on a 32 bit client system,
cp command completes without errors but no data
is copied.
Resolution:
A local variable used in computation of file size/
read offset was maintained 32 bit instead of 64 bit
causing the file transfer size computation to fail.
This has been fixed.
7. When copying a large file from local disk to NFS
file system,more bytes than the size of the source
file are written.The resulting file size is the
original file size rounded off to align on the
NFS file system blocksize boundary[default is 8K].
Resolution:
The problem is a missing type cast in an expression
that computes the number of remaining bytes to be
copied to the last file block.The type cast has been
fixed.
8. NFS client system turns off data caching on the
file when it is first locked.Later when no locks exist
on the file and no active opens exist on the file
,caching remains turned off,impacting read/write
performance.
Resolution:
After file locks have been released and no active
opens exist on the file,NFS read/write data caching
is now turned on.
9. Exportfs prints debug messages in syslog.log.
Resolution:
Turned off logging of exportfs debug messages.
10. Specifying O_TRUNC flag with O_EXCL|O_CREAT flags
during file open/create causes HP NFS V3 client to try and
set the file size to zero via a Setattr operation ,
even before file permissions have been set.This causes
Create operation to fail.
Resolution:
A work around to the problem is to not specify O_TRUNC
with O_CREAT|O_EXCL during file open ,because it is
really redundant.The solution is that NFS V3 client
should not try and set the file size to zero for
exclusive open/create requests.
11. NFS V3 server allows file access when Mandatory lock
mode 2664 is set on a file,whereas NFS V2 server
denies read/write access. This behaviour is different
from SUN's where file access is denied at file open
time.
Resolution:
NFS Server and client now deny access to file when
Mandatory lock mode is set 2664,for both NFS V3
and V2 clients at file open time.
12. During system boot up,system panics with data page
fault in mvfs.The NFS server initialization function
caused memory corruption of mvfs global data during
initialization of the Write verifier.
Resolution:
Write verifier is initialized by copying the
system time. This copy resulted in memory
corruption due to writing beyond valid
range. This has been fixed.
PHNE_17247:
1. During system boot up,system panics with data page
fault in mvfs.The NFS server initialization function
caused memory corruption of mvfs global data during
initialization of the Write verifier.
Resolution:
Write verifier is initialized by copying the
system time. This copy resulted in memory
corruption due to writing beyond valid
range. This has been fixed.
2. When copying a large file (>2GB) from NFS file
system to local disk on a 32 bit client system,
cp command completes without errors but no data
is copied.
Resolution:
A local variable used in computation of file size/
read offset was maintained 32 bit instead of 64 bit
causing the file transfer size computation to fail.
This has been fixed.
3. When copying a large file from local disk to NFS
file system,more bytes than the size of the source
file are written.The resulting file size is the
original file size rounded off to align on the
NFS file system blocksize boundary[default is 8K].
Resolution:
The problem is a missing type cast in an expression
that computes the number of remaining bytes to be
copied to the last file block.The type cast has been
fixed.
PHNE_17200:
Makefile
innetgr expects input hostname either
the exact match in the netgroup map or
it switches it into lower case and tries
a further match.
Resolution:
build netgroup.byhost with lower case
key so that innetgr can evaluate host
name case-insensitively since it tries
the match with lower-case input host
name eventually.
automount
automount will get stuck in a loop if a mount
point on a replicated server does not exist
and the automount maps refer to the mount point
via an indirect map.
Resolution:
Change code to skip entries marked "ignore".
Change summary:
/ux/core/onc/src/auto/automount/auto_mount.c
Modify function find_server() to skip
entries which have already been marked
with flag "mfs_ignore".
automount
The automount process removes mount points
from its internal tables when it can not
remount a unmounted mount point.
Resolution:
Allow additional remount attempts before
eventually giving up and removing the
mount point.
automountd
automountd can wait indefinitely while holding a lock
on "/etc/mnttab". This can affect other programs
such as bdf.
Resolution:
stat() of a NFS mount point can be forced to wait
indefinitely, so do not hold a file lock on "/etc/mnttab"
during this operation. Also conform to a more recent
interface to "mntent" (setmntent() and endmntent()).
Change summary:
~onc/src/autofs/auto_mnttab.c
- make get_devid() externally callable
- mkmntlist(): remove get_devid() call
- getmntlist(): remove fs[un]lock_mnttab() calls
- getmntlist(): use setmntent() interface
~onc/src/autofs/autod_mount.c
- add prototype for get_devid()
- do_unmount1(): add get_devid() call
~onc/src/autofs/fslib.c
- Remove fslock_mnttab(), fsunlock_mnttab()
~onc/src/autofs/fslib.h
- Remove prototypes for fs[un]lock_mnttab()
automountd
Logging information does not have
any timestamp data.
Resolution:
Include timestamp information on internal
logging.
automountd
executable maps are incorrectly processed.
Resolution:
provide internal mechanism for running
executable maps.
libautofs.a
Module "kern/autofs/auto_subr.c", function
"force_remount()" is used to remount autonodes
which have been manually unmounted. This routine
flags a "busy" condition (MF_INPROG) on nodes
before working on them. It could set the condition
on more nodes than it will later clear, causing
other other access to wait forever for the "busy"
node to be remounted. A second problem is:
although MF_TRIGGER can be set on multiple
autonodes, force_remount() does not check beyond
the first one found, resulting in remounts
sometimes not happening.
Resolution:
Change logic in "force_remount()" to set MF_INPROG
only on nodes which it is checking (and will later
clear). Also, to prevent multiple nodes with
MF_TRIGGER set from confusing force_remount(),
loop through all instances of nodes with MF_TRIGGER
set and attempt to remount all unmounted ones.
nfs.core
/var/spool/rexd is no longer created.
On cold installs the directory /var/spool/rexd
is not in existance. Nothing checks to see if
it does not. If not, then rexd fails to start.
Resolution:
Nfs.core script is modified to check and create
/var/spool/rexd if not there on nfs startup.
nis.server
Previous fix required that nis.server and namesvrs
be installed as a pair.
Resolution:
Code changed to remove this dependancy.
rpc.mountd
rpc.mountd in an HA configuration can deadlock
with automount while attempting to resolve a
mount request containing the mountpoint for an
automount map entry on the local host. Using a
secondary host address, when it is not detected
as a loopback, if the function realpath() is
called with the requested path traversing an
automount map point, the mountd process lookup()
activates automounter which must call mountd
to complete the map -- deadlock.
Resolution:
Add code specific to mountd to refuse to mount
any path which passes through an automount or
autofs controlled directory.
Change summary:
/ux/core/onc/src/nfs/mountd/rpc.mountd.c
- mount(), umount(), checkexports() changed
from calling realpath() to realpath_local()
- adapted [from libc] function realpath_local()
- adapted [from libc] function pathcanon_local()
- add function generate_exclude_list()
- add function is_excluded_path()
- add function path_has_prefix()
rpc.mountd
Routine to check access lists, in_access_list(), can
call compare_ip_addr() routine many times, resulting
in many calls to gethostbyname().
Resolution:
Call new routine compare_hostnames_hp(), which does a
string compare to determine host match if it can.
Otherwise it calls compare_ip_addr() as always.
rpc.nispasswdd
rpc.nispasswdd was loading the value from
an uninitialized variable to the public key
field when changing password for a user who
has a credential.
Resolution:
change the public key only when a pair of
new key is generated and leaving the
public key untouched when reencrypted the
private key.
rpc.pcnfsd
1) The wrong user verification mechanism was
being used.
2) pcnfsd is blocked obtaining information from
the down printer.
Resolution:
1) For trusted systems use the correct authentication
routines.
2) Allow printer requests to timeout instead of waiting
forever for a reply. The default value is 10 seconds
but can be changed by specifing a value for lpstat_timer
in the /etc/pcnfsd.conf file.
rpc.pcnfsd
Additional shells are configured in the
/etc/shells file. rpc.pcnfsd is not
using the contents of this file to verify
if the users shell is allowed.
Resolution:
Code to read the /etc/shells file was
conditionally compiled off. Adding the
appropriate define will activate this code
allowing rpc.pcnfsd to compare the users
shell with the information in /etc/shells
rpc.rstatd
For each request, rpc.rstatd would always obtain
memory to store transport specific information.
When the request completed, this memory was not
freed.
Resolution:
If memory was obtained, release it before completing
the request.
rpc.statd
The problem is that the function used to
compare internet addresses in rpc.statd
does not work at all and causes any number
of problems including this core dump.
Resolution:
The function to compare internet addresses
in rpc.statd was rewritten to properly
compare an incoming internet address with
the list of addresses of remote systems.
rpcbind
rpcbind failed to remove an entry when the
netid field in the unset request was NULL
Resolution:
If the netid field is NULL, don't use
that field has part of the match criteria.
rpcgen
rpcgen, generated a makefile using syntax from
another vendors make implementation
Resolution:
rpcgen generates hp-ux make syntax.
rpcgen
pcgen incorrectly generated a struct
pthread_mutexattr.
Resolution:
generate the correct posix pthread_mutexattr
structure.
stdhosts
the read buffer in stdhosts has length
for 256 char.
Resolution:
increase the lenght of the buffer to
1024 char.
ypinit
ypinit has warning message due to
the change of sendmail.
Resolution:
change umask to 022.
ypmake
innetgr expects input hostname either
the exact match in the netgroup map or
it switches it into lower case and tries
a further match.
Resolution:
build netgroup.byhost with lower case
key so that innetgr can evaluate host
name case-insensitively since it tries
the match with lower-case input host
name eventually.
ypserv
When ypserv receives YPPROC_ALL request, it
forks a child process to handle the request.
The tcp channel between the client and
the ypserv parent process is not closed
until the client finishes and call
clnt_destroy. If client did not get the
chance to exit normally, the tcp connection
does not get cleaned up.
Resolution:
ypserv closes the tcp channel to the client
after it forks a child process.
ypserv
ypserv writes into some memory that it
did not allocate which causes segmentation
violation in svc_freeargs.
Resolution:
allocate memory for size of hostname and a
null terminator and copy over the hostname
from keydat structure instead of adding null
terminator into the hostname of keydat structure
which did not have any extra space.
ypxfr_1perday
vhe_list is not supported. It should
not be in the map list.
Resolution:
remove vhe_list from ypxfr_1perday.
PHNE_17101:
autofs
PHNE_16470 did not deliver the kernel
configuration file that allows the 64-bit
AutoFS library to be incorporated into the
64-bit kernel. If PHNE_15638 had been
previously installed, no problem will be
seen.
nfsstat
nfsstat requires but does not have root
ownership.
PHNE_16599:
1. The system panics in vn_rele due to the NFS kernel file
locking routines on the server attempting to deallocate a
vnode that has already been deallocated , during NFS file
locking, unlocking.The panic is also likely when rpc.lockd
is killed and when local locks and NFS blocked locks
co-exist.
2. An NFS client may allocate too much memory resource
for the rnodes due to allocating ever increasing number
of rnodes on the client system that may never be freed.
3. Sometimes NFS Version 3 mounted directories appear to be
empty on the client even though these directories exist
on the server file system.
4. When a non-root user who is not the file owner,
issues a touch command to change the modification time
stamp of an NFS file,the NFS client system may hang due
to not releasing the file system semaphore.
5. Some NFS clients may fail to read from buffer cache,
instead read over the wire causing Read performance
degradation with HP servers,apparently due to
a redundant update to the file modification time
stamp in a NFS V3 commit operation.
6. A invalid file system id is being returned by
the NFS client interface to Statfs of a NFS Version 2
mounted file set.
PHNE_16470:
Makefile
Remove support for vhe_list in ypinit,
ypmake and /var/yp/Makefile
automount
The date in the automounter log file was not
proper.
automount
automount ignores all NFS mounts that don't export
the filesystem with a beginning "/" character.
automount
logging data is formatted poorly.
automountd
1. syslog was getting filled with duplicate
messages that were not meaningful for direct mounts.
2. The check for non-zero rdevid field was not valid
since it was sometimes 0 and that was valid in the case
of loopback mounts.
exportfs
logging data is formatted poorly.
libautofs.a
1. readdir call was not working on 64 bit autofs.
2. Autofs was not handling interrupts
correctly when the clnt_call was made.
3. When unmounts were done for hierarchical maps, autofs
was trying to unmount autonodes which caused the hang to
happen.
libnsl.1/a
logging data is formatted poorly.
libnsl.1/a
__nss2herrno is also defined in libc.2/a.
The definition in libnsl.1/a should not
exist.
libnsl.1/a
A memory leak in netdir_getbyname() eventually
will cause rpc.mountd to dump core with a
SIGBUS
libnsl.1/a
RPC only supports 16 groups and aborts if the
user belongs to more.
libnsl.1/a
The ypbind daemon writes down a 32-bit data
structure in /var/yp/binding as a cache
file. The 64-bit application reads the file
into a 64-bit data structure.
libnss_nisplus.1
The nis_result structure was not released
when there is no entry found by
nis_first_entry or nis_next_entry.
libnss_nisplus.1
Module ngetprpwent is missing
from libnss_nisplus.
librpcd_util.a
logging data is formatted poorly.
librpcsoc.1
Socket base RPC code was incorrectly determining
the number of open files causing select() to
fail
mknetid
There is an incorrect boundary checking when
mknetid fills out the group list for a user.
namesvrs
The HP box has several maps which are
non-existent on the IBM machines. This causes
the nis.server script to fail as the maps are
hard coded into the script.
nfs.client
The proper command for unmounting all file systems
based on file system type is umountall.
nfs.server
nfs.server script issues a mount -a instead of a
mount -aF nfs, causing local file systems to mount.
nfs_umount
Incorrectly parsing the /etc/mnttab table
causes some NFS file systems not to be
unmounted.
nfsstat
Incorrect locking scheme on /etc/mnttab
nis.server
The HP box has several maps which are
non-existent on the IBM machines. This causes
the nis.server script to fail as the maps are
hard coded into the script.
on
on incorrectly parses the /etc/mnttab table
looking to see if the current directory is
NFS mounted
rpc.lockd
logging data is formatted poorly.
rpc.lockd
IB3, J2240 nfs stress test lock failed
cause data corruption
rpc.lockd
rpc.lockd logs messages even though the -l
option was not specified.
rpc.lockd
1. The tracing produced by rpc.lockd was changed.
2. Messages were not getting deleted from the
queue when NFS packets were lost.
3. rpc.lockd behaved inconsistently in the NLM_CANCEL
case when the client or server was a SUN machine
instead of HP. The protocol was violated in the
rpc.lockd code.
rpc.mountd
The code used to stat the tmp
file for determining the daemon up could possibly be
exploited.
rpc.mountd
logging data is formatted poorly.
rpc.mountd
domainname wildcard symbol was being incorrectly
interprated by the netgroup look up routine.
rpc.pcnfsd
Validity checks not performed on printer and
system name.
rpc.rexd
logging data is formatted poorly.
rpc.rquotad
logging data is formatted poorly.
rpc.rquotad
Incorrect locking scheme on /etc/mnttab
rpc.rstatd
logging data is formatted poorly.
rpc.rusersd
logging data is formatted poorly.
rpc.sprayd
logging data is formatted poorly.
rpc.statd
rpc.lockd logs messages even though the -l
option was not specified.
rpc.statd
Changed the debug logging for rpc.statd.
Fixed the gethostbyname problem.
rpc.walld
logging data is formatted poorly.
rpc.yppasswdd
rpc.yppasswdd has an unnecessary checking for ":"
in the old password entered by the user.
rpc.yppasswdd
"mv" system call should be used when "rename"
fails on moving files between different file
systems.
rpcbind
logging data is formatted poorly.
ypbind
The uninitialized timestamp in the domain
structure sometime is bigger than the current
time which makes ypbind daemon thinks the
current bind is not a new bind. It skips
creating the bind file in that situation.
ypbind
The compilation flag for cache is
missing in the Makefile.
ypinit
Remove support for vhe_list in ypinit,
ypmake and /var/yp/Makefile
ypmake
Remove support for vhe_list in ypinit,
ypmake and /var/yp/Makefile
ypmake
customer wants an option to clear map
in the memory when ypmake NOPUSH=1
ypserv
The amount of timeout given to the async
service could be negative on a slow system
which caused select() to function unexpectedly.
ypxfr
ypxfr on the slave process is dumping core
when trying to use a sock_addr over a XTI
connection.
ypxfr
Function parameter declarations don't match
the actual parameter list.
PHNE_16017:
1. When a utime() operation with a NULL timestamp is
performed on an NFS file/directory, the NFS client uses
the system time of the client to set file/directory
modification and access times with SETATTR operations.
However, WRITE operations uses the system time of the
server to set file/directory modification and access
times. This causes inconsistencies when the client and
server systems are in different time zones or when the
clocks on the client and server systems are not
synchronized.
2. A SUN NFS client attempting to perform a utime()
operation with a NULL timestamp on an HP NFS server
file/directory is rejected due to a permissions error
even though the process has write permissions but is
not the file/directory owner.
3. A comparison was being performed between an unsigned
64-bit long constant and an unsigned 32-bit integer
variable which caused the comparison to fail in a certain
case that should have succeeded. This in turn caused an
invalid client access time to be sent to the server as
the access time for the file being copied.
4. NFS performance over 100BT is poor.
PHNE_15638:
automountd:
1. Enhancement for autofs.
mount:
1. Enhancement for autofs.
umount:
1. Enhancement for autofs.
rup:
exit code returns wrong value
libnsl:
rpcbind hangs on udp tranfers when mbuf overflows.
Incorrect error logic
rpc.statd:
The return value from gethostbyname was being checked
incorrectly.
rpc.lockd:
For PC clients talking to NLM version 4, the support
for shares, non-monitored locks and freeall was not
included.
automount:
Check for the verbose flag was missing when writing to
syslog file in some cases.
nfs.client:
rpc.lockd and rpc.statd needed the filesystems to be
exported so that the remote clients could recover the
locks after the server reboots.
PHNE_14919:
1. HP-UX executes a read before writing to the EOF of a
NFS mounted file, when the transfer size is less that 8K.
Additionally, the HP-UX buffer cache interaction with
general NFS PV3 IO was poor when synchronizing the file.
2. NFS Client code allowed file metadata to be modified
when Server's permissions should prevent access, possibly
causing data destruction.
3. The attempt to remount a mount point is incorrectly
assuming that file system data should be retrieved from
the Server, rather than from the current mount structures.
4. The file lock protocol implementation would sometimes
not correctly handle large file locking.
PHNE_14800:
Several NFS/RPC features found in the 11.0 system
release need to be tuned in order to yield adequate
SPECsfs benchmark numbers.
PHNE_14772:
N/A
PHNE_14746:
rpcgen:
1. The usage string contains a [Y not {Y
2. Code generated by rpcgen fails to compile when using
C++, because of a reference to an invalid include file,
/usr/include/sysent.h
nfs.core
1. libstraddr split across two lines.
2. Check for loopback transport here instead
of configure script
rpc.nisd:
1. Using -Y -B option of rpc.nisd fail to start daemon,
rpc.nisd_resolv.
2. rpc.nisd logs messages into syslog.log.
These messages are not written into the file descriptor
that belongs to syslog, instead syslogd writes
them into an fd that belongs to rpc.nisd thus blocking
the server from receiving the requests from the
clients.
Eventually, the clients give up and issue "NIS+
unreachable" error message.
nisaddent:
1. nisaddent command puts empty passwd for password
entries that do not have corresponding entries in the
tcb files on a trusted system.
nisserver:
1. Syntax error causes unnecessary environment
variables being added to /etc/rc.config.d/namesvrs
file.
2. nisserver script does not set correct value for
$EMULYP environment variable. This cause the
NIS+ server not being set up correctly if
- NIS compat mode is not set, and if,
- nisserver was previously executed with NIS
compat mode set
nisclient:
1. nisclient has illegal grep syntax for hp
2. nisclient -u script reports success even when
chkey fails
nispopulate:
1. nispopulate has illegal grep syntax for hp
libnss_nisplus.1:
1. 64-bit gethostbyname() returns 0.0.0.0 for the IP
address of a valid host stored in the NISPLUS table
hosts.org_dir.
nisdefaults:
1. nisdefaults does not indicate if a principal is not
authenticated.
nisstat:
1. nisstat requires fully qualified NIS+ hostnames.
automount:
1. autofs and automount command have name conflict
2. automount does not try to read the auto.master map
if auto_master map is not present on the system.
rpc:
1. no entry for rpc.nisd_resolv
2. no entry for automountd
nfsconf:
1. Can only start automount
nfs.client:
1. Can only start automount
libstraddr.1:
1. Failure to load 64-bit library correctly.
1. No 64-bit support for loopback transports.
libnsl.1:
1. Default threads are not detached.
2. No support for MT RPC dameon.
3. netdir_getbyname() returns ND_OK instead
of ND_NOHOST.
4. No 64-bit support for loopback transports
5. No 64-bit support for RPC MT Servers
6. netdir_getbyname() returns ND_OK instead
of ND_NOHOST.
libnsl.a:
1. Default threads are not detached.
2. No support for MT RPC daemon.
3. netdir_getbyname() returns ND_OK instead
of ND_NOHOST.
4. No 64-bit support for loopback transports
5. No 64-bit support for RPC MT Servers
6. netdir_getbyname() returns ND_OK instead
of ND_NOHOST.
vhe_altlog:
1. vhe_altlog was not removed from system
vhe_getlist:
1. vhe_getlist was not removed from system
vhe_list:
1. vhe_list was not removed from system
vhe_mounter:
1. vhe_mounter was not removed from system
vhe_script:
1. vhe_script was not removed from system
config:
1. Files are delivered to /usr/newconfig/... but are
not delivered to the correct directory.
unconfig:
1. Files are not removed when patch is swremoved.
netconfig:
1. /etc/netconfig should be delivered into
/usr/newconfig/etc/netconfig
hosts:
1. /etc/net/[ticlts|ticots|ticotsord]/hosts should be
delivered to
/usr/newconfig/etc/net/[ticlts|ticots|ticotsord]/hosts
services:
1. /etc/net/[ticlts|ticots|ticotsord]/services should be
delivered to
/usr/newconfig/etc/net/[ticlts|ticots|ticotsord]/services
rpc.rstatd:
1. When started from the command line, rpc.rstatd keeps
dying after 10 minutes and needs to be executed from
the command line again.
keyserv:
1. Insecure communication between client/server over UDP.
nis_cachemgr:
1. Should not be using UDP for client/server
communication.
rpcbind:
1. rpcbind does not support loopback transport provider
devices
rpc.mountd:
1. getdomainname() was being called incorrectly.
PHNE_14620:
1. A misaligned pointer used in RPC causes a fault
2. NFS unmount schemes are very inefficient
PHNE_13208:
1. Some processes doing NFS Writes would not die on a
800/MP 2 way system ,when 50 or more processes
concurrently access same file over NFS.
2. Sometimes ,NFS sub-system hung when NFS was stressed
with IO's between 2 800/4 way MP systems .
3. A different error value than expected sometimes
returned to application ( rare occurance).
Enhancement:
No (superseded patches contained enhancements)
PHNE_27217:
This product update contains minor enhancements
required to enable the HP-UX Device IDs feature.
JAGae14669 allows a root user to forcibly notify
the NFS server's rpc.lockd to discard all locks
and cached port information for a specified client.
Additional enhancements were delivered in a patch
this one has superseded. Please review the Defect
Description text for more information.
PHNE_24034:
JAGad15675 improves the performance of I/O
throughput on a locked file.
JAGad70794 allows HP's NIS clients the option of
specifying a list of available NIS servers for a
given domain.
SR:
1653045047 1653236562 1653249250 1653249425 1653254235
1653254474 1653256941 1653266148 1653266189 1653266289
1653267377 1653270413 1653270488 1653272484 1653272781
1653275149 1653286955 1653291484 1653291500 1653293043
1653298828 1653299826 1653301143 1653304972 1653311530
4701374694 4701379610 4701383729 4701392027 4701401935
4701415216 4701420604 4701425918 5003336198 5003362707
5003379289 5003395541 5003395558 5003395574 5003395582
5003395731 5003396655 5003397422 5003401406 5003405118
5003405548 5003407155 5003407643 5003409086 5003409862
5003411082 5003412072 5003417998 5003418160 5003418210
5003420992 5003423590 5003429456 5003429845 5003431353
5003434654 5003435339 5003435537 5003437285 5003437459
5003438572 5003439471 5003439646 5003442574 5003444745
5003445601 5003445858 5003446005 5003446195 5003446203
5003446211 5003447003 5003447946 5003448159 5003450338
5003450916 5003451724 5003454371 5003458224 5003458299
5003459362 5003460030 5003461210 5003462911 5003463331
5003463992 5003465807 5003465880 5003465898 8606102788
8606103364 8606103503 8606103740 8606104454 8606104479
8606105077 8606105833 8606106692 8606106938 8606107068
8606107122 8606107274 8606107366 8606107676 8606107751
8606108110 8606108804 8606109663 8606110199 8606111538
8606113678 8606124555 8606125471 8606125899 8606126860
8606127276 8606128485 8606128493 8606128739 8606131366
8606131496 8606131978 8606132817 8606134214 8606134296
8606134593 8606135799 8606135800 8606135856 8606136063
8606140609 8606143409 8606143750 8606143872 8606143877
8606144478 8606144617 8606144881 8606145476 8606145517
8606145970 8606146332 8606146885 8606147198 8606147221
8606152534 8606155004 8606155592 8606155843 8606157592
8606158230 8606159451 8606159736 8606159847 8606160097
8606160400 8606160596 8606160777 8606160808 8606161701
8606163247 8606164133 8606164889 8606165859 8606166314
8606167053 8606168123 8606169011 8606169796 8606171662
8606172747 8606172908 8606173363 8606173448 8606174017
8606175781 8606175862 8606176017 8606176384 8606178618
8606178998 8606179200 8606179269 8606179951 8606180764
8606180765 8606180770 8606181688 8606183171 8606184904
8606185238 8606186347 8606186548 8606188628 8606193515
8606196567 8606197003 8606197306 8606199444 8606201620
8606202370 8606203242 8606204999 8606205013 8606205779
8606207461 8606208446 8606212242 8606216548 8606218377
8606219527 8606219558 8606221102 8606221217 8606221909
8606222617 8606224657 8606225120 8606226735 8606226778
8606228306 8606229377 8606230660 8606232532 8606233931
8606234034 8606234968 8606235172 8606237277 8606242021
8606244127 8606245329 8606245747 8606245876 8606248269
8606248624 8606248870 8606249559 8606249576 8606254087
8606255006 8606256016 8606256845 8606256846 8606257219
8606257688 8606259713 8606259831 8606264087 8606264317
8606265228 8606267227 8606267508 8606269865 8606271227
8606271824 8606272309 8606274502 8606274519 8606274555
8606274703 8606275551 8606276016 8606276334 8606277106
8606278982 8606281883 8606282498 8606282662 8606284885
8606284964 8606286563 8606287136 8606288394 8606288821
8606289500 8606291642 8606308270
Patch Files:
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP:
/usr/conf/lib/libhp-ux.a(nfs_iface.o)
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP:
/usr/conf/lib/libnfs.a
/usr/conf/master.d/nfs
/usr/conf/lib/librpc.a
/usr/conf/lib/libautofs.a
/usr/conf/master.d/autofs
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP:
/usr/conf/lib/libhp-ux.a(nfs_iface.o)
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP:
/usr/conf/lib/libnfs.a
/usr/conf/master.d/nfs
/usr/conf/lib/librpc.a
/usr/conf/lib/libautofs.a
/usr/conf/master.d/autofs
OS-Core.CORE-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,
v=HP:
/usr/share/man/man1m.Z/mount_nfs.1m
NFS.NFS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/newconfig/etc/rpc
/usr/newconfig/etc/vhe_list
/usr/newconfig/usr/sbin/vhe/vhe_getlist
/usr/newconfig/usr/sbin/vhe/vhe_altlog
/usr/newconfig/usr/sbin/vhe/vhe_mounter
/usr/newconfig/usr/sbin/vhe/vhe_script
/usr/sbin/automount
/usr/lib/netsvc/fs/automount/automount
/usr/lib/netsvc/fs/autofs/automount
/usr/lib/netsvc/fs/autofs/automountd
/usr/bin/rup
/usr/bin/on
/sbin/init.d/nfs.client
/sbin/fs/autofs/mount
/sbin/fs/autofs/umount
NFS.NFS-PRG,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/bin/rpcgen
/usr/lib/libnsl.a
/usr/lib/librpcsvc.a
NFS.KEY-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/keyserv
/usr/sbin/newkey
NFS.NFS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/rpc.statd
/usr/sbin/rpcbind
/usr/sbin/rpc.lockd
/usr/sbin/setoncenv
/usr/sbin/clear_locks
/usr/sbin/nfsconfcol
/usr/newconfig/etc/net/ticlts/hosts
/usr/newconfig/etc/net/ticlts/services
/usr/newconfig/etc/net/ticots/hosts
/usr/newconfig/etc/net/ticots/services
/usr/newconfig/etc/net/ticotsord/hosts
/usr/newconfig/etc/net/ticotsord/services
/usr/newconfig/etc/rc.config.d/nfsconf
/usr/newconfig/etc/netconfig
/usr/bin/nfsstat
/usr/bin/rpcinfo
/sbin/init.d/nfs.core
/sbin/fs/nfs/umount
/sbin/fs/nfs/mount
NFS.NISPLUS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/nis_cachemgr
/usr/sbin/rpc.nisd
/usr/sbin/rpc.nispasswdd
/usr/lib/nis/nisserver
/usr/lib/nis/nispopulate
/usr/lib/nis/nisclient
/usr/lib/nis/nisstat
/usr/lib/nis/nisaddent
/usr/lib/libnss_nisplus.1
/usr/bin/nisdefaults
NFS.NFS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/rpc.mountd
/usr/sbin/exportfs
/usr/sbin/rpc.rexd
/usr/sbin/rpc.rquotad
/usr/sbin/rpc.pcnfsd
/usr/sbin/nfsd
/usr/lib/netsvc/rstat/rpc.rstatd
/usr/lib/netsvc/rusers/rpc.rusersd
/usr/lib/netsvc/rwall/rpc.rwalld
/usr/lib/netsvc/spray/rpc.sprayd
/sbin/init.d/nfs.server
NFS.NFS-SHLIBS,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/libnsl.1
/usr/lib/libstraddr.1
/usr/lib/librpcsoc.1
/usr/lib/libnss_nis.1
/usr/lib/librpcsvc.1
NFS.NIS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/ypxfr
/usr/lib/libnss_compat.1
/usr/bin/domainname
NFS.NIS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/netsvc/yp/ypbind
/sbin/init.d/nis.client
NFS.NIS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/sbin/init.d/nis.server
/usr/lib/netsvc/yp/rpc.yppasswdd
/usr/lib/netsvc/yp/ypserv
/usr/newconfig/usr/sbin/ypinit
/usr/newconfig/var/yp/ypmake
/usr/newconfig/var/yp/Makefile
/usr/newconfig/var/yp/ypxfr_1perday
/usr/sbin/mknetid
/usr/sbin/stdhosts
/usr/sbin/ypxfrd
NFS.NFS-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/share/man/man1m.Z/automount.1m
/usr/share/man/man1m.Z/automountd.1m
/usr/share/man/man1m.Z/exportfs.1m
/usr/share/man/man1m.Z/setoncenv.1m
/usr/share/man/man1m.Z/nfsd.1m
/usr/share/man/man1m.Z/fsirand.1m
/usr/share/man/man1m.Z/nfsstat.1m
/usr/share/man/man1m.Z/ypinit.1m
/usr/share/man/man1m.Z/ypserv.1m
/usr/share/man/man1m.Z/yppasswdd.1m
/usr/share/man/man1m.Z/biod.1m
/usr/share/man/man1m.Z/ypbind.1m
/usr/share/man/man1m.Z/ypxfrd.1m
/usr/share/man/man1m.Z/clear_locks.1m
/usr/share/man/man1m.Z/mountd.1m
/usr/share/man/man1m.Z/rpc.mountd.1m
/usr/share/man/man4.Z/exports.4
/usr/share/man/man4.Z/xtab.4
/usr/share/man/man4.Z/nsswitch.conf.4
/usr/share/man/man4.Z/switch.4
/usr/share/man/man1.Z/domainname.1
/usr/share/doc/RelNotesNFS_TCP.txt
InternetSrvcs.INETSVCS-INETD,fr=B.11.00,
fa=HP-UX_B.11.00_32/64,v=HP:
/usr/newconfig/etc/rc.config.d/namesvrs
NFS.NFS-64SLIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/pa20_64/libnsl.1
/usr/lib/pa20_64/libstraddr.1
/usr/lib/pa20_64/libnss_nisplus.1
/usr/lib/pa20_64/libnss_compat.1
/usr/lib/pa20_64/libnss_nis.1
/usr/lib/pa20_64/librpcsvc.1
NFS.NFS-64ALIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/pa20_64/libnsl.a
/usr/lib/pa20_64/librpcsvc.a
what(1) Output:
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP:
/usr/conf/lib/libhp-ux.a(nfs_iface.o):
nfs_iface.c 11R2 + PHNE_20344
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP:
/usr/conf/lib/libnfs.a:
nfs_fcntl.c 11R2 + PHNE_16599
klm_lckmgr.c PHNE_27217
klm_kprot.c $Revision: 1.1.105.3 $
nfsm.c 11R2 - ONC+ Performance Patch
nfs_xdr.c 11R2 + PHNE_16017
nfs_vnops.c PHNE_28136
nfs_vfsops.c 11R2 + PHNE_28136
nfs_subr.c PHNE_23249
nfs_strerror.c $Revision: 1.1.105.2 $
nfs_server.c 11R2 + PHNE_28136
nfs3_server.c 11R2 + PHNE_28136
nfs_dispatch.c 11R2 + PHNE_22125
nfs_kstat.c PHNE_19759
nfs_export.c PHNE_17586
nfs_common.c PHNE_23833
nfs_client.c PHNE_24034
nfs3_xdr.c $Revision: 1.1.106.3 $
nfs3_vnops.c 11R2 + PHNE_28136
nfs3_vfsops.c 11R2 + PHNE_28136
hpnfs_vnops.c 11R2 + PHNE_25626
rpc_misc.c PHNE_20344
hpnfs_srvrinit.c 11R2 - PHNE_19759
hpnfs_kmem.c 11R2 + PHNE_20344
hpnfs_cv_stub.c $Revision: 1.2.105.5 $ X.10SAC.25
hpnfs_coreinit.c 11R2 - ONC+ Performance Patch
hpnfs_clntinit.c 11R2 + PHNE_16599
/usr/conf/master.d/nfs:
$Revision: 1.1.118.3 $
/usr/conf/lib/librpc.a:
xdr_mblk.c PHNE_28136
tcp_svc.c PHNE_22642
svc_gen.c 11 R2 + PHNE_22125
svc_cots.c PHNE_27217
svc_clts.c PHNE_27217
svc_authu.c PHNE_22642
svc.c 11R2 - ONC+ Performance Patch
rpcmod.c PHNE_24034
rpc_tcp.c PHNE_22642
rpc_str.c PHNE_19759
rpc_misc.c PHNE_20344
onc_mutex.c PHNE_19759
mt_rpcinit.c PHNE_19759
hprpc.c PHNE_20344
clnt_gen.c PHNE_28136
clnt_cots.c PHNE_28136
clnt_clts.c PHNE_23249
/usr/conf/lib/libautofs.a:
hpautofs.c PHNE_27217
auto_vnops.c PHNE_28136
auto_vfsops.c PHNE_28136
auto_subr.c PHNE_28136
/usr/conf/master.d/autofs:
PHNE_17101
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP:
/usr/conf/lib/libhp-ux.a(nfs_iface.o):
nfs_iface.c 11R2 + PHNE_20344
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP:
/usr/conf/lib/libnfs.a:
nfs_fcntl.c 11R2 + PHNE_16599
klm_lckmgr.c PHNE_27217
klm_kprot.c $Revision: 1.1.105.3 $
nfsm.c 11R2 - ONC+ Performance Patch
nfs_xdr.c 11R2 + PHNE_16017
nfs_vnops.c PHNE_28136
nfs_vfsops.c 11R2 + PHNE_28136
nfs_subr.c PHNE_23249
nfs_strerror.c $Revision: 1.1.105.2 $
nfs_server.c 11R2 + PHNE_28136
nfs3_server.c 11R2 + PHNE_28136
nfs_dispatch.c 11R2 + PHNE_22125
nfs_kstat.c PHNE_19759
nfs_export.c PHNE_17586
nfs_common.c PHNE_23833
nfs_client.c PHNE_24034
nfs3_xdr.c $Revision: 1.1.106.3 $
nfs3_vnops.c 11R2 + PHNE_28136
nfs3_vfsops.c 11R2 + PHNE_28136
hpnfs_vnops.c 11R2 + PHNE_25626
rpc_misc.c PHNE_20344
hpnfs_srvrinit.c 11R2 - PHNE_19759
hpnfs_kmem.c 11R2 + PHNE_20344
hpnfs_cv_stub.c $Revision: 1.2.105.5 $ X.10SAC.25
hpnfs_coreinit.c 11R2 - ONC+ Performance Patch
hpnfs_clntinit.c 11R2 + PHNE_16599
/usr/conf/master.d/nfs:
$Revision: 1.1.118.2 $
/usr/conf/lib/librpc.a:
xdr_mblk.c PHNE_28136
tcp_svc.c PHNE_22642
svc_gen.c 11 R2 + PHNE_22125
svc_cots.c PHNE_27217
svc_clts.c PHNE_27217
svc_authu.c PHNE_22642
svc.c 11R2 - ONC+ Performance Patch
rpcmod.c PHNE_24034
rpc_tcp.c PHNE_22642
rpc_str.c PHNE_19759
rpc_misc.c PHNE_20344
onc_mutex.c PHNE_19759
mt_rpcinit.c PHNE_19759
hprpc.c PHNE_20344
clnt_gen.c PHNE_28136
clnt_cots.c PHNE_28136
clnt_clts.c PHNE_23249
/usr/conf/lib/libautofs.a:
hpautofs.c PHNE_27217
auto_vnops.c PHNE_28136
auto_vfsops.c PHNE_28136
auto_subr.c PHNE_28136
/usr/conf/master.d/autofs:
$Revision: 1.1.106.1 $
OS-Core.CORE-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,
v=HP:
/usr/share/man/man1m.Z/mount_nfs.1m:
None
NFS.NFS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/newconfig/etc/rpc:
rpc: 11R2-4
/usr/newconfig/etc/vhe_list:
vhe_list: 11R2-3
/usr/newconfig/usr/sbin/vhe/vhe_getlist:
vhe_getlist: 11R2-2
/usr/newconfig/usr/sbin/vhe/vhe_altlog:
vhe_altlog: 11R2-2
/usr/newconfig/usr/sbin/vhe/vhe_mounter:
vhe_mounter: 11R2-2
/usr/newconfig/usr/sbin/vhe/vhe_script:
vhe_script: 11R2-2
/usr/sbin/automount:
automount: 11R2-2
/usr/lib/netsvc/fs/automount/automount:
automount.h: PHNE_25626
auto_main.c: PHNE_28102
automount.h: PHNE_25626
auto_look.c: PHNE_25626
automount.h: PHNE_25626
automount.h: PHNE_25626
auto_node.c: PHNE_25626
automount.h: PHNE_25626
auto_mount.c: PHNE_28102
automount.h: PHNE_25626
auto_all.c: PHNE_17200
/usr/lib/netsvc/fs/autofs/automount:
automount: PHNE_20371
ns_files.c: PHNE_17200
auto_mnttab.c: PHNE_27217
auto_subr.c: PHNE_20371
fslib.c: PHNE_17200
/usr/lib/netsvc/fs/autofs/automountd:
automountd: PHNE_27217
key_generic.c: PHNE_20371
autod_mount.c: PHNE_28136
autod_nfs.c: PHNE_28102
nfs_cast.c: PHNE_19620
ns_files.c: PHNE_17200
auto_mnttab.c: PHNE_27217
auto_subr.c: PHNE_20371
fslib.c: PHNE_17200
/usr/bin/rup:
rup.c 11R3-1 1.3 90/07/25 4.1NFSSRC Copyr 1990 Sun
Micro
rup: $Revision: 1.2.211.1 $ $Date: 96/10/09 11:21:52
$
/usr/bin/on:
on.c: PHNE_16470
/sbin/init.d/nfs.client:
nfs.client: PHNE_24034
/sbin/fs/autofs/mount:
mount: PHNE_20371
/sbin/fs/autofs/umount:
umount: 11R3-3
NFS.NFS-PRG,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/bin/rpcgen:
rpc_main.c: PHNE_17200
rpc_clntout.c $Revision: 1.2.211.1 $
rpc_cout.c $Revision: 1.2.211.1 $
rpc_hout.c $Revision: 1.2.211.1 $
rpc_parse.c $Revision: 1.2.211.1 $
rpc_scan.c $Revision: 1.2.211.1 $
rpc_svcout.c PHNE_25626
rpc_tblout.c $Revision: 1.2.211.1 $
/usr/lib/libnsl.a:
libnsl: PHNE_20371
common/stubs.c: PHNE_24909
common/read_config.c: PHNE_20371
nis_cast.c: PHNE_21376
nis_rpc.c: PHNE_25626
nis_subr.c: PHNE_24909
netdir_inet.c: PHNE_28136
netdir_inet_sundry.c: PHNE_20371
svc_vc.c: PHNE_22642
clnt_perror.c: PHNE_28102
key_call.c: PHNE_20371
rpc_comdata.c: PHNE_22642
rpcb_clnt.c: PHNE_20371
svc.c: PHNE_28136
svc_auth_sys.c: PHNE_27217
svc_run.c: PHNE_28136
svc_vc.c: PHNE_27217
ti_opts.c: PHNE_20371
xdr_array.c: PHNE_28102
xdr_mem.c: PHNE_28136
xdr_rec.c: PHNE_24034
yp_bind.c: PHNE_28102
yp_match.c: PHNE_28102
/usr/lib/librpcsvc.a:
librpcsvc.a: PHNE_19620 $Revision: 1.1.211.2 $ $Dat
e: 1999/10/12 11:35:14 $
NFS.KEY-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/keyserv:
key_generic.c: 11R2-2
keyserv.c: $Revision: 1.2.211.1 $ $Date: 96/10/09 11
:05:04 $
setkey.c: $Revision: 1.2.211.1 $ $Date: 96/10/09 11:
05:06 $
detach.c: $Revision: 1.2.211.1 $ $Date: 96/10/09 11:
04:55 $
mout.c 5.3 (Berkeley) 3/2/87
pow.c 5.2 (Berkeley) 3/2/87
util.c 1.2 88/05/16 4.0NFSSRC SMI; from 5.1 (Berkele
y) 4/30/85
madd.c 5.2 (Berkeley) 3/2/87
mdiv.c 5.2 (Berkeley) 3/2/87
mult.c 5.2 (Berkeley) 3/2/87
/usr/sbin/newkey:
newkey.c PHNE_22642
newkey $Revision: 1.2.211.2 $
chkey_common.c: $Revision: 1.2.211.1 $ $Date: 96/10/
09 11:04:41 $
update.c: $Revision: 1.2.211.2 $ $Date: 97/06/04 15:
28:50 $
generic.c: $Revision: 1.2.211.1 $ $Date: 96/10/09 11
:04:43 $
madd.c 5.2 (Berkeley) 3/2/87
mdiv.c 5.2 (Berkeley) 3/2/87
mult.c 5.2 (Berkeley) 3/2/87
pow.c 5.2 (Berkeley) 3/2/87
util.c 1.2 88/05/16 4.0NFSSRC SMI; from 5.1 (Berkele
y) 4/30/85
mout.c 5.3 (Berkeley) 3/2/87
NFS.NFS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/rpc.statd:
sm_svc.c PHNE_25626
sm_proc.c PHNE_25626
sm_statd.c PHNE_25626
cache.h: PHNE_27217
udp.c: PHNE_27217
cache.h: PHNE_27217
pmap.c: PHNE_27217
/usr/sbin/rpcbind:
rpcbind.c: 11r2 + PHNE_21376
warmstart.c: 11r2 + PHNE_21376
check_bound.c: $Revision: 1.2.211.11 $ $Date: 97/01/
07 21:56:25 $
pmap_svc.c: $Revision: 1.2.211.15 $ $Date: 97/03/20
13:35:56 $
rpcb_stat.c: $Revision: 1.2.211.5 $ $Date: 96/11/06
15:56:13 $
rpcb_svc.c: $Revision: 1.2.211.7 $ $Date: 96/12/17 1
0:43:25 $
rpcb_svc_4.c: $Revision: 1.2.211.11 $ $Date: 96/12/1
7 10:43:46 $
rpcb_svc_com.c: PHNE_24034
/usr/sbin/rpc.lockd:
prot_main.c: PHNE_20371
prot_lock.c: PHNE_28136
prot_pklm.c: PHNE_19620
prot_pnlm.c: PHNE_27217
prot_msg.c: PHNE_27217
prot_proc.c: PHNE_25484
prot_priv.c: PHNE_22125
prot_share.c PHNE_21376
cache.h: PHNE_27217
pmap.c: PHNE_27217
cache.h: PHNE_27217
udp.c: PHNE_27217
sm_monitor.c PHNE_27217
prot_libr.c: PHNE_20371
svc_dg1.c PHNE_21376
prot_freeall.c 1.1 88/08/05 NFSSRC4.0 1.3 88/02/07 C
opyr 1984 Sun Micro
prot_share.c 1.2.211.2 NFSSRC4.0 1.4 97/06/17 HP-UX
/usr/sbin/setoncenv:
setoncenv: PHNE_20371
/usr/sbin/clear_locks:
clear_locks.c PHNE_27217
/usr/sbin/nfsconfcol:
nfsconfcol: PHNE_27217
/usr/newconfig/etc/net/ticlts/hosts:
hosts: 11R2-2
/usr/newconfig/etc/net/ticlts/services:
services: 11R2-3
/usr/newconfig/etc/net/ticots/hosts:
hosts: 11R2-2
/usr/newconfig/etc/net/ticots/services:
services: 11R2-3
/usr/newconfig/etc/net/ticotsord/hosts:
hosts: 11R2-2
/usr/newconfig/etc/net/ticotsord/services:
services: 11R2-3
/usr/newconfig/etc/rc.config.d/nfsconf:
nfsconf: 11R2-2
/usr/newconfig/etc/netconfig:
netconfig: 11R2-2
/usr/bin/nfsstat:
nfsstat.c: PHNE_28567
/usr/bin/rpcinfo:
rpcinfo: PHNE_20371
rpcinfo.c $Revision: 1.2.211.4 $ $Date: 96/10/25 14:
55:43 $
/sbin/init.d/nfs.core:
nfs.core: PHNE_27217
/sbin/fs/nfs/umount:
umount: PHNE_16470
bindresvport.c: PHNE_27217
mount_xdr.c: $Revision: 1.2.211.4 $ $Date: 96/10/25
14:53:56 $
/sbin/fs/nfs/mount:
mount.c: PHNE_28136
mount_xdr.c: $Revision: 1.2.211.4 $ $Date: 96/10/25
14:53:56 $
xdr_nlm.c: $Revision: 1.2.211.4 $ $Date: 96/10/25 14
:53:59 $
bindresvport.c: PHNE_27217
NFS.NISPLUS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/nis_cachemgr:
nis_cachemgr: 11R2-2
/usr/sbin/rpc.nisd:
nis_main.c PHNE_21376
/usr/sbin/rpc.nispasswdd:
npd_svcsubr.c: PHNE_17200
/usr/lib/nis/nisserver:
nisserver.sh PHNE_20371
/usr/lib/nis/nispopulate:
nispopulate.sh 11R2-02/27/98
/usr/lib/nis/nisclient:
nisclient.sh 11R2-02/26/98
/usr/lib/nis/nisstat:
nisstat.c 11R2-2/26/98
/usr/lib/nis/nisaddent:
nisaddent.c 11R2-02/26/98
/usr/lib/libnss_nisplus.1:
gethostent.c 11R2-02/27/98
getnetent.c.c 11R2-02/27/98
getpwnam.c PHNE_24909
nisplus_common.c PHNE_16470
ngetprpwent.c PHNE_25626
/usr/bin/nisdefaults:
nisdefaults.c 11R2-2/26/98
NFS.NFS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/rpc.mountd:
rpc.mountd: PHNE_27217
/usr/sbin/exportfs:
exportfs: PHNE_23833
/usr/sbin/rpc.rexd:
rpc.rexd: PHNE_16470
/usr/sbin/rpc.rquotad:
rpc.rquotad: PHNE_16470
/usr/sbin/rpc.pcnfsd:
pcnfsd_misc.c: PHNE_18221
pcnfsd_print.c: PHNE_18221
rpc.pcnfsd: PHNE_18221
pcnfsd_v1.c: PHNE_17200
pcnfsd_v2.c: PHNE_17200
rpc.pcnfsd - April 1, 1993
/usr/sbin/nfsd:
nfsd daemon PHNE_22642
/usr/lib/netsvc/rstat/rpc.rstatd:
rpc.rstatd: PHNE_17200
/usr/lib/netsvc/rusers/rpc.rusersd:
rpc.rusersd: PHNE_25626
/usr/lib/netsvc/rwall/rpc.rwalld:
rpc.rwalld: PHNE_16470
/usr/lib/netsvc/spray/rpc.sprayd:
rpc.sprayd: PHNE_16470
/sbin/init.d/nfs.server:
nfs.server: PHNE_16470
NFS.NFS-SHLIBS,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/libnsl.1:
libnsl: PHNE_20371
common/stubs.c: PHNE_24909
common/read_config.c: PHNE_20371
nis_cast.c: PHNE_21376
nis_rpc.c: PHNE_25626
nis_subr.c: PHNE_24909
netdir_inet.c: PHNE_28136
netdir_inet_sundry.c: PHNE_20371
svc_vc.c: PHNE_22642
clnt_perror.c: PHNE_28102
key_call.c: PHNE_20371
rpc_comdata.c: PHNE_22642
rpcb_clnt.c: PHNE_20371
svc.c: PHNE_28136
svc_auth_sys.c: PHNE_27217
svc_run.c: PHNE_28136
svc_vc.c: PHNE_27217
ti_opts.c: PHNE_20371
xdr_array.c: PHNE_28102
xdr_mem.c: PHNE_28136
xdr_rec.c: PHNE_24034
yp_bind.c: PHNE_28102
yp_match.c: PHNE_28102
/usr/lib/libstraddr.1:
libstraddr: PHNE_20371
/usr/lib/librpcsoc.1:
librpcsoc: PHNE_16470
/usr/lib/libnss_nis.1:
getservent.c PHNE_23833
/usr/lib/librpcsvc.1:
librpcsvc.a: PHNE_19620 $Revision: 1.1.211.2 $ $Dat
e: 1999/10/12 11:35:14 $
NFS.NIS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/sbin/ypxfr:
ypxfr: PHNE_23833
ypserv_net_secure.c 1.2 92/06/05 4.1NFSSRC Copyr 199
0 Sun Micro
/usr/lib/libnss_compat.1:
compat_common.c PHNE_18221
/usr/bin/domainname:
domainname.c: PHNE_28102
domainname: $Revision: 1.2.211.2 $ $Date: 96/10/10 1
3:32:28 $
NFS.NIS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/netsvc/yp/ypbind:
ypbind.c PHNE_24034
/sbin/init.d/nis.client:
nis.client: PHNE_24034
NFS.NIS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/sbin/init.d/nis.server:
nis.server: PHNE_17200
/usr/lib/netsvc/yp/rpc.yppasswdd:
rpc.yppasswdd.c: PHNE_28102
/usr/lib/netsvc/yp/ypserv:
ypserv_map.c: PHNE_28102
ypsrv_proc.c PHNE_21376
ypserv_dbmtrak.c: PHNE_25626
rpc_as.c PHNE_16470
ypserv.c 1.4 92/06/05 4.1NFSSRC Copyr 1990 Sun Micro
ypserv_ancil.c 1.4 90/07/23 4.1NFSSRC Copyr 1990 Sun
Micro
ypserv_map.c 1.3 92/06/05 4.1NFSSRC Copyr 1990 Sun M
icro
yp_cache.c: $Revision: 1.1.211.1 $ $Date: 96/10/09 1
1:26:17 $
ypserv_dbmtrak.c 1.1 90/07/24 4.1NFSSRC Copyr 1990 S
MI
ypserv_net_secure.c 1.2 92/06/05 4.1NFSSRC Copyr 199
0 Sun Micro
ngethostbyname.c 1.2 92/04/08 4.1NFSSRC
/usr/newconfig/usr/sbin/ypinit:
ypinit: PHNE_28102
/usr/newconfig/var/yp/ypmake:
ypmake: PHNE_20371
/usr/newconfig/var/yp/Makefile:
make.script PHNE_20371
/usr/newconfig/var/yp/ypxfr_1perday:
ypxfr_1perday: PHNE_17200
/usr/sbin/mknetid:
mknetid.c PHNE_16470
getname.c 1.3 90/07/24 4.1NFSSRC Copyr 1990 Sun Micr
o
/usr/sbin/stdhosts:
stdhosts.c PHNE_17200
/usr/sbin/ypxfrd:
ypxfrd_subr.c: PHNE_28102
ypserv_net_secure.c 1.2 92/06/05 4.1NFSSRC Copyr 199
0 Sun Micro
NFS.NFS-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/share/man/man1m.Z/automount.1m:
None
/usr/share/man/man1m.Z/automountd.1m:
None
/usr/share/man/man1m.Z/exportfs.1m:
None
/usr/share/man/man1m.Z/setoncenv.1m:
None
/usr/share/man/man1m.Z/nfsd.1m:
None
/usr/share/man/man1m.Z/fsirand.1m:
None
/usr/share/man/man1m.Z/nfsstat.1m:
None
/usr/share/man/man1m.Z/ypinit.1m:
None
/usr/share/man/man1m.Z/ypserv.1m:
None
/usr/share/man/man1m.Z/yppasswdd.1m:
None
/usr/share/man/man1m.Z/biod.1m:
None
/usr/share/man/man1m.Z/ypbind.1m:
None
/usr/share/man/man1m.Z/ypxfrd.1m:
None
/usr/share/man/man1m.Z/clear_locks.1m:
None
/usr/share/man/man1m.Z/mountd.1m:
None
/usr/share/man/man1m.Z/rpc.mountd.1m:
None
/usr/share/man/man4.Z/exports.4:
None
/usr/share/man/man4.Z/xtab.4:
None
/usr/share/man/man4.Z/nsswitch.conf.4:
None
/usr/share/man/man4.Z/switch.4:
None
/usr/share/man/man1.Z/domainname.1:
None
/usr/share/doc/RelNotesNFS_TCP.txt:
None
InternetSrvcs.INETSVCS-INETD,fr=B.11.00,
fa=HP-UX_B.11.00_32/64,v=HP:
/usr/newconfig/etc/rc.config.d/namesvrs:
namesvrs: PHNE_16470
NFS.NFS-64SLIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/pa20_64/libnsl.1:
libnsl: PHNE_20371
common/stubs.c: PHNE_24909
common/read_config.c: PHNE_20371
nis_cast.c: PHNE_21376
nis_rpc.c: PHNE_25626
nis_subr.c: PHNE_24909
netdir_inet.c: PHNE_28136
netdir_inet_sundry.c: PHNE_20371
svc_vc.c: PHNE_22642
clnt_perror.c: PHNE_28102
key_call.c: PHNE_20371
rpc_comdata.c: PHNE_22642
rpcb_clnt.c: PHNE_20371
svc.c: PHNE_28136
svc_auth_sys.c: PHNE_27217
svc_run.c: PHNE_28136
svc_vc.c: PHNE_27217
ti_opts.c: PHNE_20371
xdr_array.c: PHNE_28102
xdr_mem.c: PHNE_28136
xdr_rec.c: PHNE_24034
yp_bind.c: PHNE_28102
yp_match.c: PHNE_28102
/usr/lib/pa20_64/libstraddr.1:
libstraddr: PHNE_20371
/usr/lib/pa20_64/libnss_nisplus.1:
gethostent.c 11R2-02/27/98
getnetent.c.c 11R2-02/27/98
getpwnam.c PHNE_24909
nisplus_common.c PHNE_16470
ngetprpwent.c PHNE_25626
/usr/lib/pa20_64/libnss_compat.1:
compat_common.c PHNE_18221
/usr/lib/pa20_64/libnss_nis.1:
getservent.c PHNE_23833
/usr/lib/pa20_64/librpcsvc.1:
librpcsvc.a: PHNE_19620 $Revision: 1.1.211.2 $ $Dat
e: 1999/10/12 11:35:14 $
NFS.NFS-64ALIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
/usr/lib/pa20_64/libnsl.a:
libnsl: PHNE_20371
common/stubs.c: PHNE_24909
common/read_config.c: PHNE_20371
nis_cast.c: PHNE_21376
nis_rpc.c: PHNE_25626
nis_subr.c: PHNE_24909
netdir_inet.c: PHNE_28136
netdir_inet_sundry.c: PHNE_20371
svc_vc.c: PHNE_22642
clnt_perror.c: PHNE_28102
key_call.c: PHNE_20371
rpc_comdata.c: PHNE_22642
rpcb_clnt.c: PHNE_20371
svc.c: PHNE_28136
svc_auth_sys.c: PHNE_27217
svc_run.c: PHNE_28136
svc_vc.c: PHNE_27217
ti_opts.c: PHNE_20371
xdr_array.c: PHNE_28102
xdr_mem.c: PHNE_28136
xdr_rec.c: PHNE_24034
yp_bind.c: PHNE_28102
yp_match.c: PHNE_28102
/usr/lib/pa20_64/librpcsvc.a:
librpcsvc.a: PHNE_19620 $Revision: 1.1.211.2 $ $Dat
e: 1999/10/12 11:35:14 $
cksum(1) Output:
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP:
2760757419 1960 /usr/conf/lib/libhp-ux.a(nfs_iface.o)
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_32,v=HP:
332390567 367368 /usr/conf/lib/libnfs.a
2516941901 4605 /usr/conf/master.d/nfs
1653623388 159604 /usr/conf/lib/librpc.a
1805634050 50300 /usr/conf/lib/libautofs.a
2861396597 4175 /usr/conf/master.d/autofs
OS-Core.CORE2-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP:
1291178035 5176 /usr/conf/lib/libhp-ux.a(nfs_iface.o)
NFS.NFS-KRN,fr=B.11.00,fa=HP-UX_B.11.00_64,v=HP:
3673442387 794968 /usr/conf/lib/libnfs.a
4102962794 4596 /usr/conf/master.d/nfs
1858045543 320962 /usr/conf/lib/librpc.a
4170834966 109012 /usr/conf/lib/libautofs.a
3540743214 4187 /usr/conf/master.d/autofs
OS-Core.CORE-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,
v=HP:
43366572 5284 /usr/share/man/man1m.Z/mount_nfs.1m
NFS.NFS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
178743537 1005 /usr/newconfig/etc/rpc
2514079349 246 /usr/newconfig/etc/vhe_list
3724157041 1153 /usr/newconfig/usr/sbin/vhe/vhe_getlist
1382004015 287 /usr/newconfig/usr/sbin/vhe/vhe_altlog
1771038307 219 /usr/newconfig/usr/sbin/vhe/vhe_mounter
3795005447 855 /usr/newconfig/usr/sbin/vhe/vhe_script
2563193047 475 /usr/sbin/automount
406111600 98304 /usr/lib/netsvc/fs/automount/automount
2843290802 45056 /usr/lib/netsvc/fs/autofs/automount
63136450 86016 /usr/lib/netsvc/fs/autofs/automountd
4085928117 20480 /usr/bin/rup
1904872655 32768 /usr/bin/on
2973211391 13020 /sbin/init.d/nfs.client
1592713053 20480 /sbin/fs/autofs/mount
3672883531 16384 /sbin/fs/autofs/umount
NFS.NFS-PRG,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
777890110 86016 /usr/bin/rpcgen
3442766314 819244 /usr/lib/libnsl.a
1614146856 27632 /usr/lib/librpcsvc.a
NFS.KEY-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
1910340694 36864 /usr/sbin/keyserv
1710430142 40960 /usr/sbin/newkey
NFS.NFS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
990409921 53248 /usr/sbin/rpc.statd
1770503456 77824 /usr/sbin/rpcbind
2911504197 122880 /usr/sbin/rpc.lockd
3408786521 16384 /usr/sbin/setoncenv
2959663890 20480 /usr/sbin/clear_locks
2829063665 15730 /usr/sbin/nfsconfcol
1360146993 226 /usr/newconfig/etc/net/ticlts/hosts
94885058 310 /usr/newconfig/etc/net/ticlts/services
3109028859 226 /usr/newconfig/etc/net/ticots/hosts
2727090760 310 /usr/newconfig/etc/net/ticots/services
2684034731 229 /usr/newconfig/etc/net/ticotsord/hosts
3786729237 313 /usr/newconfig/etc/net/ticotsord/services
475334471 2978 /usr/newconfig/etc/rc.config.d/nfsconf
3462203210 982 /usr/newconfig/etc/netconfig
2198200846 28672 /usr/bin/nfsstat
4252304490 36864 /usr/bin/rpcinfo
4200239617 6738 /sbin/init.d/nfs.core
2456726965 24576 /sbin/fs/nfs/umount
2325941212 45056 /sbin/fs/nfs/mount
NFS.NISPLUS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
2680669693 36864 /usr/sbin/nis_cachemgr
2616298117 151552 /usr/sbin/rpc.nisd
2271579757 40960 /usr/sbin/rpc.nispasswdd
3256901992 35995 /usr/lib/nis/nisserver
4034347018 31250 /usr/lib/nis/nispopulate
3454596372 32562 /usr/lib/nis/nisclient
4139046671 20480 /usr/lib/nis/nisstat
1365299230 65536 /usr/lib/nis/nisaddent
2657676018 57344 /usr/lib/libnss_nisplus.1
4068397568 24576 /usr/bin/nisdefaults
NFS.NFS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
4006507184 49152 /usr/sbin/rpc.mountd
3125821939 32768 /usr/sbin/exportfs
3800425561 40960 /usr/sbin/rpc.rexd
3935809057 20480 /usr/sbin/rpc.rquotad
2994811083 49152 /usr/sbin/rpc.pcnfsd
3320616446 36864 /usr/sbin/nfsd
1953546584 24576 /usr/lib/netsvc/rstat/rpc.rstatd
1206267629 24576 /usr/lib/netsvc/rusers/rpc.rusersd
685839461 20480 /usr/lib/netsvc/rwall/rpc.rwalld
8728557 20480 /usr/lib/netsvc/spray/rpc.sprayd
2829252411 8097 /sbin/init.d/nfs.server
NFS.NFS-SHLIBS,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
1661167438 679936 /usr/lib/libnsl.1
4114447388 16384 /usr/lib/libstraddr.1
662587888 36864 /usr/lib/librpcsoc.1
1364318230 40960 /usr/lib/libnss_nis.1
224321950 36864 /usr/lib/librpcsvc.1
NFS.NIS-CORE,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
455483673 45056 /usr/sbin/ypxfr
771024879 28672 /usr/lib/libnss_compat.1
296822547 16384 /usr/bin/domainname
NFS.NIS-CLIENT,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
647083143 36864 /usr/lib/netsvc/yp/ypbind
1523077957 8172 /sbin/init.d/nis.client
NFS.NIS-SERVER,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
4228164065 12573 /sbin/init.d/nis.server
2267274460 28672 /usr/lib/netsvc/yp/rpc.yppasswdd
3900887027 65536 /usr/lib/netsvc/yp/ypserv
772735613 15377 /usr/newconfig/usr/sbin/ypinit
3443849567 17560 /usr/newconfig/var/yp/ypmake
4097736587 13517 /usr/newconfig/var/yp/Makefile
3461546495 568 /usr/newconfig/var/yp/ypxfr_1perday
3869673866 20480 /usr/sbin/mknetid
4293094210 16384 /usr/sbin/stdhosts
3948655941 24576 /usr/sbin/ypxfrd
NFS.NFS-ENG-A-MAN,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
644419266 13338 /usr/share/man/man1m.Z/automount.1m
1265682756 1564 /usr/share/man/man1m.Z/automountd.1m
2793478962 5845 /usr/share/man/man1m.Z/exportfs.1m
3680814790 1103 /usr/share/man/man1m.Z/setoncenv.1m
3921735893 1703 /usr/share/man/man1m.Z/nfsd.1m
1962081458 1033 /usr/share/man/man1m.Z/fsirand.1m
3710451762 2546 /usr/share/man/man1m.Z/nfsstat.1m
1096190839 2378 /usr/share/man/man1m.Z/ypinit.1m
574622833 5472 /usr/share/man/man1m.Z/ypserv.1m
2132424219 2686 /usr/share/man/man1m.Z/yppasswdd.1m
3921735893 1703 /usr/share/man/man1m.Z/biod.1m
574622833 5472 /usr/share/man/man1m.Z/ypbind.1m
574622833 5472 /usr/share/man/man1m.Z/ypxfrd.1m
318597495 1332 /usr/share/man/man1m.Z/clear_locks.1m
3772558319 1730 /usr/share/man/man1m.Z/mountd.1m
3772558319 1730 /usr/share/man/man1m.Z/rpc.mountd.1m
313707377 5017 /usr/share/man/man4.Z/exports.4
313707377 5017 /usr/share/man/man4.Z/xtab.4
2726941245 7165 /usr/share/man/man4.Z/nsswitch.conf.4
2726941245 7165 /usr/share/man/man4.Z/switch.4
3057682343 1180 /usr/share/man/man1.Z/domainname.1
148930347 7088 /usr/share/doc/RelNotesNFS_TCP.txt
InternetSrvcs.INETSVCS-INETD,fr=B.11.00,
fa=HP-UX_B.11.00_32/64,v=HP:
166483852 3774 /usr/newconfig/etc/rc.config.d/namesvrs
NFS.NFS-64SLIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
60846646 653168 /usr/lib/pa20_64/libnsl.1
816237266 13392 /usr/lib/pa20_64/libstraddr.1
1293063650 60008 /usr/lib/pa20_64/libnss_nisplus.1
1162139978 24336 /usr/lib/pa20_64/libnss_compat.1
283808479 39656 /usr/lib/pa20_64/libnss_nis.1
1269408194 29904 /usr/lib/pa20_64/librpcsvc.1
NFS.NFS-64ALIB,fr=B.11.00,fa=HP-UX_B.11.00_32/64,v=HP:
724029847 1458560 /usr/lib/pa20_64/libnsl.a
1188584970 45886 /usr/lib/pa20_64/librpcsvc.a
Patch Conflicts: None
Patch Dependencies:
s700: 11.00: PHCO_22314 PHKL_22589 PHCO_23117
PHNE_22566 PHKL_20202
s800: 11.00: PHCO_22314 PHKL_22589 PHCO_23117
PHNE_22566 PHKL_20202
Hardware Dependencies: None
Other Dependencies:
Fix for JAGad91730 depends on Filesystems patch
PHKL_25525.
On systems with PCI interfaces PHKL_20170
should also be installed.
Fix for JAGab70789/JAGad28779 depends on
PM patches PHKL_23406, PHKL_23407,
PHKL_23408, and PHKL_23409.
Supersedes:
PHNE_14746 PHNE_13208 PHNE_14620 PHNE_14772 PHNE_14800 PHNE_14919
PHNE_15638 PHNE_16017 PHNE_16470 PHNE_16599 PHNE_17101 PHNE_17200
PHNE_17247 PHNE_17586 PHNE_18221 PHNE_19620 PHNE_19759 PHNE_20371
PHNE_20344 PHNE_20627 PHNE_21376 PHNE_22125 PHNE_22642 PHNE_23247
PHNE_23249 PHNE_23833 PHNE_24034 PHNE_24909 PHNE_25484 PHNE_25626
PHNE_26387 PHNE_27213 PHNE_27217 PHNE_28102 PHNE_28136
Equivalent Patches: None
Patch Package Size: 3160 KBytes
Installation Instructions:
Please review all instructions and the Hewlett-Packard
SupportLine User Guide or your Hewlett-Packard support terms
and conditions for precautions, scope of license,
restrictions, and, limitation of liability and warranties,
before installing this patch.
------------------------------------------------------------
1. Back up your system before installing a patch.
2. Login as root.
3. Copy the patch to the /tmp directory.
4. Move to the /tmp directory and unshar the patch:
cd /tmp
sh PHNE_28567
5. Run swinstall to install the patch:
swinstall -x autoreboot=true -x patch_match_target=true \
-s /tmp/PHNE_28567.depot
By default swinstall will archive the original software in
/var/adm/sw/save/PHNE_28567. If you do not wish to retain a
copy of the original software, include the patch_save_files
option in the swinstall command above:
-x patch_save_files=false
WARNING: If patch_save_files is false when a patch is installed,
the patch cannot be deinstalled. Please be careful
when using this feature.
For future reference, the contents of the PHNE_28567.text file is
available in the product readme:
swlist -l product -a readme -d @ /tmp/PHNE_28567.depot
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHNE_28567.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions:
The fix for JAGae40412 requires the ypinit script
to be copied from /usr/newconfig/usr/sbin to /usr/sbin/.
To backup the current ypinit script and enable the new
ypinit script, do the following steps:
mv /usr/sbin/ypinit /usr/sbin/ypinit.bak
cp -p /usr/newconfig/usr/sbin/ypinit /usr/sbin/ypinit
To resolve the deadlock due to filesystem locks being
held (JAGad91730), PHKL_25525 must also be installed
along with this patch. If PHKL_25525 is not installed,
the fix will not be enabled.
The fix for JAGad15675 is disabled by default. In order to
enable the fix, the kernel parameter 'nfs_new_lock_code'
must be changed from the default value of 0 to 1. This can
be done by either patching the vmunix file and rebooting,
or patching the kernel image in memory. To patch the vmunix
file (survives a reboot), use the following command:
echo 'nfs_new_lock_code?W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
To patch the in-memory kernel (does not survive a reboot),
use the following command:
echo 'nfs_new_lock_code/W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
A kernel regeneration will require these steps be done
again to enable the fix.
The fix for JAGad72416 is disabled by default. In order to
enable the fix, the kernel parameter
'async_read_avoidance_enabled' must be changed from the
default value of 0 to 1. This can be done by either
patching the vmunix file and rebooting, or patching the
kernel image in memory. To patch the vmunix file (survives
a reboot), use the following command:
echo 'async_read_avoidance_enabled?W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
To patch the in-memory kernel (does not survive a reboot),
use the following commands:
echo 'async_read_avoidance_enabled/W 0d1' |
adb -w -k /stand/vmunix /dev/kmem
A kernel regeneration will require these steps to be done
again to enable the fix.
To resolve the process hang/deadlock due to
unkillable processes defect (JAGab70789/JAGad28779),
PHKL_23406, PHKL_23407, PHKL_23408, and PHKL_23409
must also be installed along with this patch. All
five of these patches may be installed independently.
If all five patches are not installed, the fix will
not be enabled.
If NFS/TCP is configured, it is recommended to have
the HP-UX 11.00 Extension Pack 9905 from May 1999,
or later, installed. Any revision of the XSWGR1100
patch bundle equal to, or greater than, B.11.00.45
will satisfy this recommendation.
For NFS/TCP, these minimum values are recommended
for the kernel parameters in /stand/system:
nkthread=2048
max_thread_proc=256
ncallout=2064
Note that the default value of ncallout is
nkthread + 16.
|
Versión para imprimir
