|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
From: IT Resource Center (support_feedback_at_us-support-mail.external.hp.com)
Date: Sun Aug 11 2002 - 09:18:22 CDT
HP Support Information Digests
===============================================================================
o IT Resource Center World Wide Web Service
---------------------------------------------------
If you subscribed through the IT Resource Center and would
like to be REMOVED from this mailing list, access the
IT Resource Center on the World Wide Web at:
http://www.itresourcecenter.hp.com/
Login using your IT Resource Center User ID and Password.
Then select Support Information Digests (located under
Maintenance and Support). You may then unsubscribe from the
appropriate digest.
To download a patch referenced below, access the
IT Resource Center on the World Wide Web at:
http://www.itresourcecenter.hp.com/
Login using your IT Resource Center User ID and Password.
Then select Individual Patches (under Maintenance and Support)
to access the patch. You may also download a patch via anonymous
ftp(1) from ftp.itrc.hp.com.
===============================================================================
Digest Name: weekly HP-UX series 800 10.X patch digest
Created: Sun Aug 11 3:05:02 PDT 2002
Table of Contents:
Document ID Title
--------------- -----------
PHCO_27422 s700_800 10.20 cumulative cron/at/crontab patch
PHCO_27133 s700_800 10.20 lpspool subsystem cumulative patch
PHSS_27638 s700_800 10.20 OV NNM6.2 http server fix
PHKL_26745 s800 10.26 audit cumulative patch
The documents are listed below.
-------------------------------------------------------------------------------
Document ID: PHCO_27422
Date Loaded: 20020808
Title: s700_800 10.20 cumulative cron/at/crontab patch
Patch Name: PHCO_27422
Patch Description: s700_800 10.20 cumulative cron/at/crontab patch
Creation Date: 02/08/06
Post Date: 02/08/08
Hardware Platforms - OS Releases:
s700: 10.20
s800: 10.20
Products: N/A
Filesets:
OS-Core.CMDS-MIN OS-Core.CMIN-ENG-A-MAN
Automatic Reboot?: No
Status: General Release
Critical: No
Path Name: /hp-ux_patches/s700_800/10.X/PHCO_27422
Symptoms:
PHCO_27422:
1.at(1) aborts with "rc=2" error when processing multibyte
characters if 'atjobs' script contain expressions to set
the locale to a multibyte locale.
2.at/cron is not able to create large files.
3.For jobs scheduled for 31st of a month, cron starts the
jobs multiple times on the last day of the month if the
month has 30 days.
4.atjobs are appending unexpected characters to the
environment variables with multibyte characters.
PHCO_22768:
1.crontab(1) incorrect functionality.
2.When DST shift is due before the scheduled time, at(1)
does not schedule the jobs correctly.
3.at(1) incorrect argument parsing.
4.cron(1M) does not schedule jobs during daylight saving
shift.
5.In trusted mode, crontab(1) does not create correct
audit record.
6.at(1) reports "bad date specification" when specifying
four digits time .
PHCO_19985:
1. at now + [n] months ignores leap year.
2. crontab(1) -e creates crontab job files
with incorrect permissions.
3. at(1) manpage needs to be modified.
4. Patches for cron(1M) should stop cron and
start the new cron.
5. PHCO_17552 does not have libc dependencies specified.
6. at(1) assumes the given time to be AM if the am_pm
string in the specified locale is NULL.
PHCO_17552:
at(1) does not accept input of "00" or "0" for year for
all locales.
PHCO_16864:
1. On trusted systems, cron(1) has slower performance due
to calling getspwent() when getspwnam() should have
been used.
2. Cron from patch PHCO_15679 does not run at(1) and
batch(1) jobs.
3. at(1) schedules jobs incorrectly relative to
28 Feb of non-leap years.
4. at(1) does not recognize year 00 (2000). When year
is input as 00, at(1) exits with the error message -
"bad date specification"
PHCO_15679:
On trusted systems, /etc/passwd is inherited as stdin
by processes spawned by cron.
PHCO_15234:
Year displayed in two digits.
PHCO_14179:
1. at non able to schedule job for Feb 29 2000 correctly.
2. at creates an invalid file in /var/spool/cron when
given a non-existant job file with the -f option.'
PHCO_10737:
Cron used to dump core and die when jobs submitted
with at -m were killed.
PHCO_10123:
The following symptoms correspond to the descriptions
in the defect descriptions field for PHCO_10123:
1. option -t now deal with year 2000 input correctly.
2. year limit has been increased from 2030 to 2037.
PHCO_9348:
1. Cron does not mail stdout and stderr to user on a
trusted system.
2. '.' not present in the default PATH
3. Cron has a hard-coded limit of MAXRUN (40) child
processes which can be running simultaneously. Now
the limit has been increased to 100.
PHCO_8277:
at(1) records incorrect audit id on trusted systems
PHCO_7836:
at(1) cannot schedule job using weekdays in the last
week of the month.
Defect Description:
PHCO_27422:
1. at(1) creates 'atjobs' scripts for executing batched
commands at a later time. But at(1) causes execution
error, when 'atjobs' scripts contain expressions to set
the LC_CTYPE to a multibyte locale. For example: if
the atjob script has the following expressions:
setenv LANG ja_JP.SJIS
setenv AAA xxxx
at -t 03261500 << EOF./test1.sh
EOF
Where "xxxx" is Japanese character and it is SJIS
Code '0x93609197'.
'atjobs' script is created in /var/spool/cron/atjobs
after above command. But when the above job is scheduled,
at(1) aborts with "rc=2" error. The log file
/var/adm/cron/log will contain the above error.
Resolution:
/usr/bin/sh does not handle the change in LC_CTYPE in the
same session. This behavior of sh-posix(1) is as per the
requirements of POSIX standards.The man page documentation
of sh-posix(1) also states this behavior. sh-posix(1) man
page states the following:
"Changing the value of LC_CTYPE after the shell has
started does not affect the lexical processing of shell
commands in the current shell execution environment or
its subshells. See environ(5)."
As the above job involves changing the LC_CTYPE,
at(1) used to process the data in the previous set
locales resulting in error.
at(1) is modified to take care of these environment
Variables. Now, When scheduling 'atjobs' involving
multibyte locales, at(1) inserts a new shell session
in the 'atjobs' file, if it encounters any of LC_ALL,
LC_CTYPE and LANG settings.
2. at/cron jobs are not able to create large files even
when the user had the required permissions.
This was happening because at(1) was not checking
whether user had permission to create largefiles.
Steps to reproduce the defect :
$ ulimit
unlimited
$ /usr/bin/at now <<EOF
ulimit >t1
EOF
job 1023756162.a at Tue Jun 11 06:12:42 2002
$ cat t1
4194303 (expect output here is "unlimited")
Resolution:
at(1) has been modified so that the atjobs can
create large files for the users having the required
permission.
3. A job scheduled for 31st of a month starts looping
on the last day of the month incase the month does
not have 31 days (Looping also happens for jobs
scheduled for 29th and 30th of Feb incase of non-leap
years and for jobs scheduled for 30th of Feb incase
of leap years).
This problem can be reproduced by the following
entry in the crontab filer:
58 12 31 * * /usr/bin/date > /tmp/afile123 2>&1
cron log (file is /var/adm/cron/log) shows:
< vts 4426 c Tue Apr 30 12:58:21 IST 2002
< vts 4425 c Tue Apr 30 12:58:21 IST 2002
> CMD: /usr/bin/date > /tmp/afile123 2>&1
> vts 4433 c Tue Apr 30 12:58:21 IST 2002
< vts 4433 c Tue Apr 30 12:58:21 IST 2002
< vts 4430 c Tue Apr 30 12:58:21 IST 2002
< vts 4427 c Tue Apr 30 12:58:21 IST 2002
> CMD: /usr/bin/date > /tmp/afile123 2>&1
> vts 4435 c Tue Apr 30 12:58:21 IST 2002
> CMD: /usr/bin/date > /tmp/afile123 2>&1
> vts 4436 c Tue Apr 30 12:58:21 IST 2002
> CMD: /usr/bin/date > /tmp/afile123 2>&1
> vts 4437 c Tue Apr 30 12:58:21 IST 2002
...
...
This erroneous behavior is due to an unnecessary day
comparison introduced by the DST fix.
Resolution:
Now the day comparison made in cron has been commented
out to fix the problem.
4. at(1) command is appending backslash for multibyte
characters while writing the environment variables to
the atjob file.
This problem can be reproduced by the following
commands :
$ LANG=japanese ; export LANG
$ NIHONGO=`/bin/echo '\0223\0372\0226{\0214\0352'`
$ export NIHONGO
$ echo $NIHONGO | od -c
0000000 223 372 226 { 214 352 \n
0000007
$ at now + 1 minute
echo $NIHONGO | od -c
warning: commands will be executed using /bin/sh
job 832329266.a at ...
octal dump of at-job file :
0000400 e x p o r t N
0000420 I H O N G O ; N I H O N G O =
0000440 \ 223 \ 372 \ 226 \ { \ 214 352 \n
The above output shows that a backslash is
appended for every byte of multibyte character.
This was happening because at(1) was not checking
for the multibyte characters while writing into
atjob file.
Resolution:
Now an additional check for multibyte characters
has been added before writing to atjob file. In
case of a multibyte character, all the bytes are
written to the atjob file without appending
backslash.
PHCO_22768:
1. crontab(1) incorrect functionality.
Resolution:
Code has been modified to take care of this issue.
2. When DST shift is due, there will be a non-existing
time in a day. When jobs are scheduled for such
non-existing times through commands like -
at now + <offset> minutes, at fails with message
"too late".
Resolution:
Now the code has been modified to take care of
jobs scheduled for non-existing times.
3. at(1) incorrect argument parsing.
Resolution:
The code has been modified to take care of this.
4.cron(1M) executes the job 2 days later than the
scheduled date, if a shift in daylight happens between
the current date and the scheduled date.
Resolution:
Now the code has been modified to take care of
jobs scheduled for non-existing times.
5. In trusted mode, crontab(1) does not create audit
records because required data is not generated by
the underlying library call .
Resolution:
Appropriate calls to get audit ids are used to
resolve the problem.
6.at(1) shows bad date specification when specifying
four digits time.
Resolution:
Now the code has been modified to accept the
four digits time.
PHCO_19985:
1. Setting the system date such that the day is any day
after 28th and executing the at(1) command to schedule
the job for the month of February in a leap year,
schedules the jobs incorrectly. The following
shows the incorrect behaviour of at(1).
$ date 1229084495
date: do you really want to run time backwards?[yes/no]yes
Fri Dec 29 08:44:00 EST 1995
$ at now + 2 month
warning: commands will be executed using /usr/bin/sh
job 825687848.a at Fri Mar 1 08:44:08 1996
at(1) should have scheduled the job for Feb 29.
Instead it is scheduling the job for Mar 1.
2. crontab when used with -e creates the job files
with read permission for all.
3. The manpage of at(1) needs to be more clear
regarding the ambiguity of dates.
4. The cron(1M) patches do not stop the old cron daemon
and start the new cron daemon. The starting and
stopping of the daemon does not correctly work in the
swinstall scripts since the signal mask of the cron
daemon gets modified through SD scripts.
5. The cron/at patches should have the libc
dependencies specified. Without the libc patch
PHCO_8108 installed, the at(1) command will fail
with an unresolved symbol error for 'snprintf'.
6. For certain locales like fr_FR.iso88591, where am_pm
string is defined as null ("";""), at(1) treats the
specified time to be AM. The following illustrates
the problem.
$ export LANG=fr_FR.iso88591
$ echo date | at 07:00 demain # OK
job 879487201.a at Vendredi 14 november 1997 07:00:00
$ echo date | at 20:00 # OK
job 879447600.a at Jeudi 13 november 1997 20:00:00
$ echo date | at 20:00 demain # BUG
Hours Field Is Invalid
Resolution:
1. Checking for the leap year before scheduling the
job, solves the problem.
2. The crontab command was relying on a macro for
permissions on the job file. This macro was incorrect.
This has been set right now. Now, crontab(1) creates
job files with read permission only for the owner.
3. Necessary changes are made to the at(1) manpage to
solve this problem.
4. Resetting the signal handlers in cron to their
respective defaults solves the problem.
5. Specifying the libc patch, PHCO_8108, in the patch
dependencies field solves the problem.
6. Checking for the value of am_pm string for the locale
solves the problem.
PHCO_17552:
at(1) treats year input of 00 or 0 as incorrect when
locale is set to one which defines D_T_FMT (date-time)
string to have year in fields other than the third.
For eg. japanese locale defines the date string to
be of format yy/mm/dd. With locale set to japanese,
an attempt to schedule a job using at(1) for a date
00/12/20 or 0/12/20 fails.
Resolution:
at(1) assumed that the year field can be only in the
third field of the date string. This assumption has been
removed. Also, the function which returned error if the
date string had 0 or 00 in the first or second fields
has been changed not to check and return error for such
a condition. Now, at(1) accepts input of 0 or 00 for year
in any field of the date string, as specified by D_T_FMT
string for the locale.
PHCO_16864:
1. On trusted systems, cron(1) called getspwent() which
performs a linear search in the passwd database, instead
of getspwnam() which performs a keyed search.
2. After installing PHCO_15679, jobs submitted from
at(1) are never actually run, although they are
removed from the /var/spool/cron/atjobs queue.
3. Jobs scheduled after 28 Feb are scheduled
erratically by at(1). The following shows the
incorrect behaviour of at(1).
# date 0227112099
Sat Feb 27 11:20:00 GMT 1999
# at now + 2 day
warning: commands will be executed using /usr/bin/sh
job 920287223.a at Mon Mar 1 11:20:23 1999
# at now + 3 day
warning: commands will be executed using /usr/bin/sh
job 920287243.a at Mon Mar 1 11:20:43 1999
In the above example, job is scheduled for March 1st
for both the inputs "now + 2 day" and "now + 3 day".
For the input "now + 3 day", at(1) should have
scheduled the job for March 2nd 1999.
The source code has been changed to fix the defect.
After the fix, at(1) correctly schedules jobs
relative to Feb, 28th. In the above example, for
input "now + 3 day", at(1) schedules the job for
March 2nd.
4. at(1) treats year input of 00 as incorrect. Following
shows how at(1) behaves with 00 as input year.
# at 12:00 1 Jan 00
bad date specification
The command has been changed to accept 00 as year
2000.
After the fix, at(1) treats year input of 00 as
year 2000 and does not show any error messages.
PHCO_15679:
On trusted systems, /etc/passwd is inherited as stdin
by processes spawned by cron.
PHCO_15234:
Year displayed in two digits.
PHCO_14179:
1. Job submitted for Feb 29, 2000 was actually scheduled
for Mar 1, 2000.
2. At fails to clean up the file created under
/var/spool/cron when invoked with -f option and
PHCO_10737:
Cron used to dump core and die when jobs submitted
with at -m were killed.
PHCO_10123:
The following symptoms correspond to the descriptions
in the defect descriptions field for PHCO_10123:
1. option -t now deal with year 2000 input correctly.
2. year limit has been increased from 2030 to 2037.
PHCO_9348:
1. Cron does not mail stdout and stderr to user on a
trusted system.
2. '.' not present in the default PATH
3. Cron has a hard-coded limit of MAXRUN (40) child
processes which can be running simultaneously. Now
the limit has been increased to 100.
PHCO_8277:
at(1) records incorrect audit id on trusted systems
PHCO_7836:
at(1) cannot schedule job using weekdays in the last
week of the month.
non-existant job file.
SR:
8606189207 8606179733 8606130624 1653170621 8606165710
8606125591 8606157697 8606158467 8606140830 4701411173
8606106557 1653312975 5003459313 1653293407 5003451278
5003457150 1653238113 4701416891 1653274167 5003425777
5003426254 1653269746 1653266239 4701393728 1653241117
4701383794 1653215038 4701347153 4701334763 4701328179
Patch Files:
/usr/lib/nls/msg/C/at.cat
/usr/lib/nls/msg/C/cron.cat
/usr/lib/nls/msg/C/crontab.cat
/usr/sbin/cron
/usr/bin/at
/usr/bin/crontab
/usr/share/man/man1.Z/at.1
/usr/share/man/man1.Z/batch.1
what(1) Output:
/usr/lib/nls/msg/C/at.cat:
None
/usr/lib/nls/msg/C/cron.cat:
None
/usr/lib/nls/msg/C/crontab.cat:
None
/usr/sbin/cron:
PATCH_10_20 $Revision: 78.2.1.38 $
PATCH_10_20: hpux_rel.o cron.o funcs.o 02/08/06
/usr/bin/at:
PATCH_10_20 $Revision: 78.2.1.38 $
$Revision: 76.2.1.12 $
$Revision: 76.2.1.1.1.7 $
PATCH_10_20: hpux_rel.o at.o pdt.o ptime.o pdate.o f
uncs.o permit.o 02/08/06
/usr/bin/crontab:
PATCH_10_20 $Revision: 78.2.1.38 $
PATCH_10_20: hpux_rel.o crontab.o permit.o funcs.o 0
2/08/06
/usr/share/man/man1.Z/at.1:
None
/usr/share/man/man1.Z/batch.1:
None
cksum(1) Output:
1511457102 3277 /usr/lib/nls/msg/C/at.cat
4002162402 2532 /usr/lib/nls/msg/C/cron.cat
2252532415 1121 /usr/lib/nls/msg/C/crontab.cat
807075404 40960 /usr/sbin/cron
3807016117 45056 /usr/bin/at
3237755393 24576 /usr/bin/crontab
2908365674 9017 /usr/share/man/man1.Z/at.1
2908365674 9017 /usr/share/man/man1.Z/batch.1
Patch Conflicts: None
Patch Dependencies:
s700: 10.20: PHCO_8108
s800: 10.20: PHCO_8108
Hardware Dependencies: None
Other Dependencies: None
Supersedes:
PHCO_7836 PHCO_8277 PHCO_9348 PHCO_10123 PHCO_10737 PHCO_14179
PHCO_15234 PHCO_15679 PHCO_16864 PHCO_17552 PHCO_19985 PHCO_22768
Equivalent Patches:
PHCO_27141:
s700: 11.00
s800: 11.00
PHCO_26562:
s700: 11.11
s800: 11.11
Patch Package Size: 190 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 PHCO_27422
5a. For a standalone system, run swinstall to install the
patch:
swinstall -x autoreboot=true -x match_target=true \
-s /tmp/PHCO_27422.depot
By default swinstall will archive the original software in
/var/adm/sw/patch/PHCO_27422. If you do not wish to retain a
copy of the original software, you can create an empty file
named /var/adm/sw/patch/PATCH_NOSAVE.
WARNING: If this file exists when a patch is installed, the
patch cannot be deinstalled. Please be careful
when using this feature.
It is recommended that you move the PHCO_27422.text file to
/var/adm/sw/patch for future reference.
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHCO_27422.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions: None
-----End of Document ID: PHCO_27422------------------------------------------
Document ID: PHCO_27133
Date Loaded: 20020808
Title: s700_800 10.20 lpspool subsystem cumulative patch
Patch Name: PHCO_27133
Patch Description: s700_800 10.20 lpspool subsystem cumulative patch
Creation Date: 02/08/05
Post Date: 02/08/08
Hardware Platforms - OS Releases:
s700: 10.20
s800: 10.20
Products: N/A
Filesets:
PrinterMgmt.LP-SPOOL
Automatic Reboot?: No
Status: General Release
Critical: No
Path Name: /hp-ux_patches/s700_800/10.X/PHCO_27133
Symptoms:
PHCO_27133:
1. lpsched(1m) was not killing its children when lpshut(1m)
was run.
2. lpadmin(1m) dumps core in certain conditions.
3. rlpdaemon(1m) is not behaving properly in certain
conditions.
4. lp spooler commands dumps core in certain conditions.
PHCO_25109:
1. rlpdaemon(1M) is mis-behaving in certain conditions.
2. rlpdaemon(1M) is not functioning as expected in certain
cases.
PHCO_24699:
1. rlpdaemon is not functioning as expected in certain
cases.
2. Multiple lpscheds can be running simultaneously.
3. lp was sending a corrupted mail message.
PHCO_22364:
1. If a printing request is cancelled, the other requests
in the printer queue are not getting printed until
another request is given to the queue.
2. rlp may hang indefinitely when the remote machine goes
down and comes up.
3. The remote request with the user name having one or more
spaces, is not getting processed.
4. rlpdaemon, lpstat, lp, cancel, disable, enable, rlpstat
and rcancel have memory overflow.
5. When a request is transferred from a non BSD (Berkeley
Software Distribution) printer to a remote BSD printer,
the remote BSD printer gets disabled without printing
the job. In case of BSD to remote non-BSD printer, the
request is printed but the residual datafile remains on
the system.
6. If rlpdaemon writes the messages (for eg: when
lp or lpalt are invoked with -w option for remote
printers) without stripping the control characters,
some of these control characters can make the
user's terminal behave abnormally. Some times the user
terminal may be closed.
7. Data does not get printed when a remote request
is given with a very long title.
8. When large number of requests (around 120-130)
are given, sometimes the spooler stops printing requests
in between.
9. lpadmin does not configure a remote printer properly
if the length of printer name is exactly 14 characters.
PHCO_18779:
lpstat gives "Status received is corrupt
message" with remote printers on Sun.
PHCO_17713:
1. cancel(1) does not work with printer queue
length==14 chars
2. datafiles not removed after an lpalt from
PHCO_13133 onwards.
3. cancel(1) returns exit value of 0 even if
failed w/ permission denied.
4. rlpdaemon lockfiles remain with FQDNs after
patch PHCO_12894.
PHCO_16724:
kill -9 on interface script and other lp
processes causes a corruption in the pstatus
file entries.
PHCO_16294:
lpsched deadlocks on remote printers when
remote system is down.
PHCO_16106:
lp(1) does not link data files to the SPOOL
directory as the man page states.
PHCO_14431:
1. LP spooler needs to support P405 and HP5000D640
model printers.
2. The value of LANG env variable is improperly
passed to the model script
PHCO_14029:
1. If you submit more than 52 files lp does not
remove those above 52.
2. lpsched -v writes strange strings in log file
if LANG is set.
3. lpsched -v intermittently stops by SIGSEGV in
catgets if LANG is set.
PHCO_13133:
1)incomplete lpstat output when -a/-p option used
with printer name and class
2)After PHCO_11462, moving request with lpalt
leaves data file in req dir
PHCO_12894:
1. Jobs submitted to a LP remote printer that points
to a "disabled" remote HPDPS printer are discarded.
2. When more than one jobs, whose job sequence numbers
are the same, are simultaneously submitted from the same
host to the same remote printer through RFC1179 protocol,
rlpdaemon(1M) fails to spool the jobs on the remote
system due to data/control file name collision.
3. All jobs from any HPDPS printers are spooled under the
/usr/spool/lp/request/PDRequest by rlpdaemon(1M). This
single PDRequest directory increases the chance of file
name collision due to simultaneous job submission.
PHCO_11462:
1. Cancelling a lp job with empty control file
clears the whole queue.
2. Printer gets disabled on cancelling a printing
job.
PHCO_10891:
1. "lpstat -i -o<printer-name>" doesn't inhibit the
reporting of the remote status if a HPDPS printer name
is specified. This problem happens only when the HPDPS
patch, PHCO_10608, is installed on the system. Because
of this, CDE print manager doesn't work correctly for
HPDPS printers.
PHCO_10749:
1. LP-spool hang or qstatus corrupt when /var full
log says:Bogus request
2. Parent lpsched hangs when lpalt is used to move
a request.
3. Model scripts does not handle SIGTERM any more.
4. lpsched introduced 5 sec delay between prints at
10.20
PHCO_8488:
1. lpsched hangs under FIFO full condition
2. Printers go idle even when requests are present
in the queue
3. Jobs not printed when remote host(IBM) immediately
saps connection.
4. If a BSD system spooler sends a '\001' to rlp(1)
after the control file has been sent, rlp(1) simply
ignores the '\001' and removes the entry.
5. lp, cancel, and lpstat hang when forwarding jobs
or requests to HPDPS.
6. Implicit requirement that a rlpdeamon client's
remote printer name must be the same as its
local pseudo name.
Defect Description:
PHCO_27133:
1. For every print request, lpsched forks a child
which handles that request. lpsched was not killing
its child processes when lpshut was run, printing
jobs were not stopped.
Resolution:
The problem was in handling the signals. Signal
handlers for the signal SIGTERM were not properly
set. The changes are made such that signal handlers
are set at proper places.
2. lpadmin dumps core and does not display proper
messages.
Resolution:
Changes are done in the code such that core dump
does not occur and messages are printed properly.
3. rlpdaemon (1m) is not behaving properly in certain
conditions.
Resolution:
Code is modified such that rlpdaemon (1m) behaves
properly.
4. lp spooler commands dumps core in certain conditions
and does not display proper messages.
Resolution:
Changes are done in the code such that core dump
does not occur and messages are printed properly.
PHCO_25109:
1. rlpdaemon(1M) is mis-behaving in certain conditions.
Resolution:
Code is modified so that rlpdaemon(1M) would behave
correctly.
2. rlpdaemon(1M) is not functioning as expected in
certain cases because of improper file handling.
Resolution:
Code is modified to handle the files properly.
PHCO_24699:
1. rlpdaemon is not functioning as expected in certain
cases because of improper memory handling.
Resolution:
A new condition is added in the code to handle the
problem.
2. In lpsched startup, there is a large window for a race
condition that allows multiple lpscheds to start.
Resolution:
The fix is that ~lp/SCHEDLOCK file is opened and it is
locking the file descriptor. If one lpsched is already
running,the second/any next lpsched will not run as
there is already lock being aquired by another process
(which is already running)
3. A newline character was missing in lpsched mail text ,
due to which the mail mesage was getting corrupted when
read through /usr/bin/mail.
Resolution:
The newline character '/n' has been added in the end
of mail message for lp -m , so that messagees don't
get corrupted with lp -m.
PHCO_22364:
1. When a printing request is cancelled, there is no
instruction for lpsched to continue processing other
existing requests in the queue.
Resolution:
When the printing process is killed, an instruction
is given to the scheduler to continue with the other
requests in the spooler.
2. rlp sends a request to rlpdaemon and waits for the
acknowledgement in read call on the socket. read()
is waiting indefinitely even when the remote
machine goes down without closing the socket.
Resolution:
SO_KEEPALIVE option is set for the socket so that
if there is no response for 2 hours, the read() call
returns with an error instead of waiting indefinitely.
3. The remote request with the username having spaces
is not getting processed because of the incorrect
parsing of arguments in rlpdaemon.
Resolution:
It is made sure that the parsing of arguments is
done correctly and the request is processed.
4. Memory fault in the commands is because of the
incorrect handling of buffer for error message.
Resolution:
The buffer size is increased for the error message.
If the argument is longer than the allocated buffer
size, it is truncated before printing the error
message.
5. When the requests are transfered from non-BSD printer
to remote BSD printer, datafilename has been renamed
incorrectly and the request remains unprocessed. When
the request is transfered from BSD printer to remote
non-BSD printer, the request gets processed but the
datafile, which is named incorrectly, remains in the
request directory.
Resolution:
The fix is to add more conditions so that
the datafile is renamed appropriately while
the request is being transfered from BSD to
non-BSD printer and vise-versa.
6. rlpdaemon does not filter control characters.
Because of this, some control characters may
turn the user's terminal insane.
Resolution:
Control characters are replaced with blanks in
rlpdaemon.
7. Data is not getting printed due to control file
corruption in the remote machine. This is because
of the buffer overflow for the array which stores
title in rlpdaemon.
Resolution:
The array size which contains the title on the
remote printer is increased to hold the title
length specified in the RFC1179.
8. Spooler stops processing requests in between
because, the file which contains the request
information (outputq) is not getting compressed
when a large number of requests are given. This
occurs because of the missing return statement
in a function.
Resolution:
The return statement is added in the appropriate
function.
PHCO_18779:
According to RFC-1179, the messages from the
remote printer daemon could be anything, but
the starting and ending of these messages should
adhere to the correct protocol. Portions where
unrequired checks are being made to the messages
resulting in the "Status corrupt" messages have
to be removed.
Resolution:
Unrequired checking in rlpstat(1) has been removed
to avoid the "Status corrupt" message.
Any message got from the remote printer daemon is
printed as is got.
PHCO_17713:
1. Since the sequence numbers of BSD(3 digit) and
for non BSD (4 digits) are not in proper format,
removal of data files results in a failure after
an lpalt operation from PHCO_13133 onwards.
2. As the array size for the printer name was not
uniform, cancel(1) of requests greater than
14 characters was resulting in a failure.
3. cancel(1) does not set and pass on the correct
exit values. As a result of this, the exit value
remains "0" even after the command fails.
4. rlpdaemon lockfiles remain with FQDNs after patch
PHCO_12894
Resolution:
1. The formatting of the seqno while copying
them onto a buffer has been corrected to
take care of BSD and non BSD sequence
numbers.
2. The value of DESTMAX denoting the Max
chracters allowed for a printer name has
been made uniform to accept 14 character
names.
3. The exit value in cancel(1) is now set and
passed on properly to ensure the correct
value is got in case of the command failure.
4. Check for the lock before initialising the
lckfile string.
PHCO_16724:
During spooler startup, the write offset in the
pstatus file was not updated properly.
PHCO_16294:
lpsched used to hang intermittently for remote
requests when remote printer/system was down.
PHCO_16106:
lp(1) was not linking files to the SPOOL directory
as the complete path of the directory was not
specified.
PHCO_14431:
1. The model scripts corresponding to P405, HP5000D640
were not there.
2. Since the LANG variable was incorrectly passed to the
control file, the "Invalid options" error message was
appearing on the banner.
PHCO_14029:
1. Spooler was only deleting the data files whose
names start with "d".
2. with LANG set, lpsched was improperly handling
repeated calls to catgets().
3. Due to sharing of same fd's of catalogue message
files between parent and child processes in lpsched,
a SIGSEGV used to cause coredumps intermittently.
PHCO_13133:
1) lpstat was not giving the correct output for
certain sequences of printer/class lists due to
incorrect referencing of both pstatus and qstatus
files.
2) lpsched was not clearing the data files present
in the spooler directory due to descripencies in
the names of these files, caused by the lpalt.
PHCO_12894:
1. Jobs submitted to a LP remote printer that points to
a "disabled" remote HPDPS printer are discarded.
2. When more than one jobs, whose job sequence numbers are
the same, are simultaneously submitted from the same
host to the same remote printer through RFC1179
protocol, rlpdaemon(1M) fails to spool the jobs on the
remote system due to data/control file name collision.
3. All jobs from any HPDPS printers are spooled under the
/usr/spool/lp/request/PDRequest by rlpdaemon(1M). This
single PDRequest directory increases the chance of file
name collision due to simultaneous job submission.
PHCO_11462:
1. Cancelling a lp job with empty control file
clears the whole queue.
2. Printer gets disabled on cancelling a printing
job.
PHCO_10891:
1. "lpstat -i -o<printer-name>" doesn't inhibit the
reporting of the remote status if a HPDPS printer name
is specified. This problem happens only when the HPDPS
patch, PHCO_10608, is installed on the system. Because
of this, CDE print manager doesn't work correctly for
HPDPS printers.
PHCO_10749:
1. LP-spool hang or qstatus corrupt when /var full
log says:Bogus request
2. Parent lpsched hangs when lpalt is used to move
a request.
3. Model scripts does not handle SIGTERM any more.
4. lpsched introduced 5 sec delay between prints at
10.20
PHCO_8488:
1. lpsched hangs under FIFO full condition
2. Printers go idle even when requests are present
in the queue
3. Jobs not printed when remote host(IBM) immediately
snaps connection.
4. If a BSD system spooler sends a '\001' to rlp(1)
after the control file has been sent, rlp(1) simply
ignores the '\001' and removes the entry.
5. lp, cancel, and lpstat hang when forwarding jobs
or requests to HPDPS.
6. Implicit requirement that a rlpdeamon client's
remote printer name must be the same as its
local pseudo name.
SR:
8606226693 8606227217 8606233628 8606262488 8606221482
8606215197 8606201567 8606197229 8606197042 8606170526
8606154599 8606106120 8606125159 8606152584 8606155796
8606155795 8606110628 8606127627 8606103553 8606104672
1653298620 4701387050 4701383828 1653245696 1653248195
1653248153 1653235002 1653232694 1653233296 5003347351
4701356097 4701350769 5003359760 1653198119 1653203109
1653208504 5003330688 1653172023 4701323964 5003318188
4701334128 1653261974 4701392043 5003428094 5003439273
4701397182 5003455048
Patch Files:
/usr/lib/lp/model/HPP405
/usr/lib/lp/model/PCL4
/usr/lib/lp/model/PCL5
/usr/lib/lp/model/colorlaserjet
/usr/lib/lp/model/deskjet1600CM
/usr/lib/lp/model/hp5000c30
/usr/lib/lp/model/hpC1208a
/usr/lib/lp/model/hpd640
/usr/lib/lp/model/laserjet
/usr/lib/lp/model/laserjet4
/usr/lib/lp/model/laserjet4Si
/usr/lib/lp/model/laserjet4v
/usr/lib/lp/model/laserjet5Si
/usr/lib/lp/model/laserjetIIISi
/usr/lib/lp/model/paintjet
/usr/lib/lp/model/paintjetXL300
/usr/lib/lp/model/rmodel
/usr/bin/cancel
/usr/bin/disable
/usr/bin/enable
/usr/bin/lpstat
/usr/sbin/accept
/usr/sbin/lpfence
/usr/sbin/lpmove
/usr/sbin/lpshut
/usr/sbin/reject
/usr/bin/lp
/usr/bin/lpalt
/usr/lbin/rwrite
/usr/sbin/lpadmin
/usr/sbin/lpsched
/usr/sbin/rcancel
/usr/sbin/rlpdaemon
/usr/sbin/rlpstat
/usr/sbin/rlp
/usr/lib/nls/msg/C/lp.cat
what(1) Output:
/usr/lib/lp/model/HPP405:
None
/usr/lib/lp/model/PCL4:
None
/usr/lib/lp/model/PCL5:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/colorlaserjet:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/deskjet1600CM:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/hp5000c30:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/hpC1208a:
None
/usr/lib/lp/model/hpd640:
$Header: hpd640,v 82.1 98/02/24 15:51:13 ssa Exp $
HP-UX 10.x
/usr/lib/lp/model/laserjet:
None
/usr/lib/lp/model/laserjet4:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/laserjet4Si:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/laserjet4v:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/laserjet5Si:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/laserjetIIISi:
$Header: PCL5,v 76.2.1.4 98/04/27 19:20:27 ssa Exp $
/usr/lib/lp/model/paintjet:
None
/usr/lib/lp/model/paintjetXL300:
None
/usr/lib/lp/model/rmodel:
$Header: rmodel,v 72.3.1.2 97/04/28 19:49:14 ssa Exp
$
/usr/bin/cancel:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o cancel.o 02/08/05
/usr/bin/disable:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o disable.o 02/08/05
/usr/bin/enable:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o enable.o 02/08/05
/usr/bin/lpstat:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpstat.o 02/08/05
/usr/sbin/accept:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o accept.o 02/08/05
/usr/sbin/lpfence:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpfence.o 02/08/05
/usr/sbin/lpmove:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpmove.o 02/08/05
/usr/sbin/lpshut:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpshut.o 02/08/05
/usr/sbin/reject:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o reject.o 02/08/05
/usr/bin/lp:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lp.o 02/08/05
/usr/bin/lpalt:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpalt.o 02/08/05
/usr/lbin/rwrite:
PATCH_10_20: rwrite.o 02/08/05
/usr/sbin/lpadmin:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpadmin.o 02/08/05
/usr/sbin/lpsched:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o lpsched.o 02/08/05
/usr/sbin/rcancel:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o rcancel.o 02/08/05
/usr/sbin/rlpdaemon:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o rlpdaemon.o 02/08/05
/usr/sbin/rlpstat:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o rlpstat.o 02/08/05
/usr/sbin/rlp:
$Revision: 78.29.1.112 $
PATCH_10_20: hpux_rel.o rlp.o 02/08/05
/usr/lib/nls/msg/C/lp.cat:
None
cksum(1) Output:
483369663 35251 /usr/lib/lp/model/HPP405
3092667237 23718 /usr/lib/lp/model/PCL4
2678772530 51661 /usr/lib/lp/model/PCL5
2678772530 51661 /usr/lib/lp/model/colorlaserjet
2678772530 51661 /usr/lib/lp/model/deskjet1600CM
2678772530 51661 /usr/lib/lp/model/hp5000c30
4158767097 39056 /usr/lib/lp/model/hpC1208a
3155102849 71817 /usr/lib/lp/model/hpd640
3092667237 23718 /usr/lib/lp/model/laserjet
2678772530 51661 /usr/lib/lp/model/laserjet4
2678772530 51661 /usr/lib/lp/model/laserjet4Si
2678772530 51661 /usr/lib/lp/model/laserjet4v
2678772530 51661 /usr/lib/lp/model/laserjet5Si
2678772530 51661 /usr/lib/lp/model/laserjetIIISi
1439765765 6373 /usr/lib/lp/model/paintjet
1811378279 46204 /usr/lib/lp/model/paintjetXL300
332041795 5287 /usr/lib/lp/model/rmodel
666854031 40960 /usr/bin/cancel
3047459008 24576 /usr/bin/disable
3654941588 16384 /usr/bin/enable
2944843974 49152 /usr/bin/lpstat
3011229638 16384 /usr/sbin/accept
3255587211 20480 /usr/sbin/lpfence
1967896288 28672 /usr/sbin/lpmove
1922815281 16384 /usr/sbin/lpshut
2300559328 16384 /usr/sbin/reject
519186272 45056 /usr/bin/lp
2623393167 36864 /usr/bin/lpalt
1662401969 16384 /usr/lbin/rwrite
4133804509 45056 /usr/sbin/lpadmin
1046128378 53248 /usr/sbin/lpsched
2905584379 32768 /usr/sbin/rcancel
3873444901 73728 /usr/sbin/rlpdaemon
1917446736 36864 /usr/sbin/rlpstat
1331170932 24576 /usr/sbin/rlp
3800179760 16137 /usr/lib/nls/msg/C/lp.cat
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes:
PHCO_8488 PHCO_10749 PHCO_10891 PHCO_11462 PHCO_12894 PHCO_13133
PHCO_14029 PHCO_14431 PHCO_16106 PHCO_16294 PHCO_16724 PHCO_17713
PHCO_18779 PHCO_22364 PHCO_24699 PHCO_25109
Equivalent Patches:
PHCO_27132:
s700: 11.00
s800: 11.00
PHCO_27020:
s700: 11.11
s800: 11.11
Patch Package Size: 950 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 PHCO_27133
5a. For a standalone system, run swinstall to install the
patch:
swinstall -x autoreboot=true -x match_target=true \
-s /tmp/PHCO_27133.depot
By default swinstall will archive the original software in
/var/adm/sw/patch/PHCO_27133. If you do not wish to retain a
copy of the original software, you can create an empty file
named /var/adm/sw/patch/PATCH_NOSAVE.
WARNING: If this file exists when a patch is installed, the
patch cannot be deinstalled. Please be careful
when using this feature.
It is recommended that you move the PHCO_27133.text file to
/var/adm/sw/patch for future reference.
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHCO_27133.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions:
Run /usr/sbin/lpshut before the installation of this patch
and /usr/sbin/lpsched after installation of the patch.
-----End of Document ID: PHCO_27133------------------------------------------
Document ID: PHSS_27638
Date Loaded: 20020805
Title: s700_800 10.20 OV NNM6.2 http server fix
Patch Name: PHSS_27638
Patch Description: s700_800 10.20 OV NNM6.2 http server fix
Creation Date: 02/08/01
Post Date: 02/08/05
Hardware Platforms - OS Releases:
s700: 10.20
s800: 10.20
Products:
OpenView Network Node Manager 6.2
Filesets:
OVPlatform.OVWWW-SRV,B.06.20.00
Automatic Reboot?: No
Status: General Release
Critical: No
Path Name: /hp-ux_patches/s700_800/10.X/PHSS_27638
Symptoms:
PHSS_27638:
http server process cannot handle bad HTTP request types
correctly.
Defect Description:
PHSS_27638:
http server process was not handling the bad requests
properly.
Resolution:
Code has been added for handling the bad requests properly.
SR:
8606270796
Patch Files:
/opt/OV/httpd/bin/ab
/opt/OV/httpd/bin/apachectl
/opt/OV/httpd/bin/apxs
/opt/OV/httpd/bin/dbmmanage
/opt/OV/httpd/bin/htdigest
/opt/OV/httpd/bin/htpasswd
/opt/OV/httpd/bin/httpd
/opt/OV/httpd/bin/logresolve
/opt/OV/httpd/bin/rotatelogs
what(1) Output:
/opt/OV/httpd/bin/ab:
None
/opt/OV/httpd/bin/apachectl:
None
/opt/OV/httpd/bin/apxs:
None
/opt/OV/httpd/bin/dbmmanage:
None
/opt/OV/httpd/bin/htdigest:
None
/opt/OV/httpd/bin/htpasswd:
None
/opt/OV/httpd/bin/httpd:
patched for CAN-2002-0392 [CERT VU#944335]
/opt/OV/httpd/bin/logresolve:
None
/opt/OV/httpd/bin/rotatelogs:
None
cksum(1) Output:
3027199269 24677 /opt/OV/httpd/bin/ab
1446960109 6579 /opt/OV/httpd/bin/apachectl
3998113415 18538 /opt/OV/httpd/bin/apxs
3168729259 6428 /opt/OV/httpd/bin/dbmmanage
1433302304 24671 /opt/OV/httpd/bin/htdigest
232185699 20561 /opt/OV/httpd/bin/htpasswd
3724501601 436633 /opt/OV/httpd/bin/httpd
1178741435 20541 /opt/OV/httpd/bin/logresolve
3748759209 20537 /opt/OV/httpd/bin/rotatelogs
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes: None
Equivalent Patches:
PHSS_27639:
s700: 11.00 11.11
s800: 11.00 11.11
PSOV_03199:
sparcSOL: 2.6 2.7 2.8
Patch Package Size: 640 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 PHSS_27638
5a. For a standalone system, run swinstall to install the
patch:
swinstall -x autoreboot=true -x match_target=true \
-s /tmp/PHSS_27638.depot
By default swinstall will archive the original software in
/var/adm/sw/patch/PHSS_27638. If you do not wish to retain a
copy of the original software, you can create an empty file
named /var/adm/sw/patch/PATCH_NOSAVE.
WARNING: If this file exists when a patch is installed, the
patch cannot be deinstalled. Please be careful
when using this feature.
It is recommended that you move the PHSS_27638.text file to
/var/adm/sw/patch for future reference.
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHSS_27638.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions:
========================================================
BEFORE LOADING THIS PATCH...
If there are any ovw sessions running on either the
Management Station or on any Management Consoles, these
sessions need to be closed. The "ovstop" command
should then be executed.
WARNING : the patch will not install if any
Network Node Manager processes are
detected.
BACKUP ALL OpenView DIRECTORIES TO TAPE. This
pre-caution ensures that if you have problems after
installing the patch, you will be able to restore your
current state from tape.
NOTE : ovstart is NOT executed after the patch is loaded
You will need to manually run ovstart.
Also, in order to keep a history of the patches
installed on your system the post-install script
will attempt to copy this <patch_name>.text file
to /usr/OV/ReleaseNotes/patches
(or /opt/OV/ReleaseNotes/patches). Please make
sure this file is in /tmp before installation.
========================================================
-----End of Document ID: PHSS_27638------------------------------------------
Document ID: PHKL_26745
Date Loaded: 20020805
Title: s800 10.26 audit cumulative patch
Patch Name: PHKL_26745
Patch Description: s800 10.26 audit cumulative patch
Creation Date: 02/06/14
Post Date: 02/08/05
Hardware Platforms - OS Releases:
s800: 10.26
Products: N/A
Filesets:
BLS.BLS-CORE OS-Core.CORE-KRN BLS.BLS-ENG-A-MAN
Automatic Reboot?: Yes
Status: General Release
Critical:
Yes
PHKL_26745: ABORT
Path Name: /hp-ux_patches/s800/10.X/PHKL_26745
Symptoms:
PHKL_26745:
Audit daemon does not behave as expected when audit
record size is greater than 64K.
PHKL_18794:
While system call auditing is off, the changed process
attributes e.g. loginuid, uid and privileges etc. are
not reflected correctly in the subsequent command
generated audit records.
PHKL_17900:
System may reach deadlock condition in auditing code
PHKL_17891:
10.26 audits only the first 358 system calls.
Defect Description:
PHKL_26745:
Audit daemon does not behave as expected when audit
record size is greater than 64K.
Resolution:
When the record size is greater than 64K audit driver will
return an error(EINVAL). Also the audit driver generates
an audit event of type ET_AUDIT. This is to notify the
error occurance during audit.
PHKL_18794:
When system call auditing is off, kernel is not
generating audit records to reflect the new state of the
process before genearting new audit records.
Resolution
Changed kernel to generate the missing audit record to
reflect the changed state of the process.
PHKL_17900:
After allocating a big chunk of audit buffers, the audit
subsystem may try to write another audit record before
releasing the buffers. The second request may put the
process in context to sleep to honour the space constraints
enforced by the audit subsystem. The process will
be deadlocked. Subsequently, all other processes will
also be put to sleep waiting for audit buffer space.
Resolution:
Prevent the deadlock by overriding the space constraints
for the second audit request.
PHKL_17891:
The current 10.26 code audits only the first 358 system
calls. The base 10.20 supports a total of 453 system calls
including large filesystem calls. The 10.26 code has been
enhanced to audit the large filesystem calls which happens
to be from 359 to 371. The system calls from 372 to 453
are not supported in 10.26.
Resolution:
Update audit_table and auditmsg.h to support the large
filesystem calls.
SR:
0000000000
Patch Files:
/usr/conf/lib/libsec.a(audit_dev.o)
/usr/conf/lib/libsec.a(sec_audit.o)
/usr/include/sys/audit.h
/usr/include/sys/auditmsg.h
/etc/conf/h/audit.h
/usr/conf/h/auditmsg.h
/usr/share/man/man3.Z/authaudit.3
what(1) Output:
/usr/conf/lib/libsec.a(sec_audit.o):
99/06/08 kern/sec/sec_audit.c, hpux, hpux_10.26, ic5
dc Revision 1.7 PATCH_10.26 (PHKL_18794)
99/03/09 kern/sec/audit_table, hpux, hpux_10.26, ic5
dc Revision 1.3 PATCH_10.26 (PHKL_17891)
/usr/conf/lib/libsec.a(audit_dev.o):
02/04/16 kern/sec/audit_dev.c, hpux, hpux_10.26, ic5
gs Revision 1.8 PATCH_10.26 (PHKL_26745)
/usr/include/sys/audit.h:
$Revision: 1.5 kern/h/audit.h, hpux, hpux_10.26, ic5
bn $ $Date: 99/03/10 14:05:14 $ Hewlett-Pack
ard Co. */
99/03/10 kern/h/audit.h, hpux, hpux_10.26, ic5bn Rev
ision 1.5 PATCH_10.26 (PHKL_17900) */
/usr/include/sys/auditmsg.h:
$Revision: 1.3 kern/h/auditmsg.h, hpux, hpux_10.26,
ic5bn $ $Date: 99/03/09 16:08:54 $ Hewlett-P
ackard Co. */
$Revision: Hewlett-Packard ISSL 1.6 kern/h/auditmsg.
h, sysaudit, vvos_davis, davis7 $ $Date: 96/
12/05 07:42:32 $ */
99/03/09 kern/h/auditmsg.h, hpux, hpux_10.26, ic5bn
Revision 1.3 PATCH_10.26 (PHKL_17891) */
/etc/conf/h/audit.h:
$Revision: 1.5 kern/h/audit.h, hpux, hpux_10.26, ic5
bn $ $Date: 99/03/10 14:05:14 $ Hewlett-Pack
ard Co. */
99/03/10 kern/h/audit.h, hpux, hpux_10.26, ic5bn Rev
ision 1.5 PATCH_10.26 (PHKL_17900) */
/usr/conf/h/auditmsg.h:
$Revision: 1.3 kern/h/auditmsg.h, hpux, hpux_10.26,
ic5bn $ $Date: 99/03/09 16:08:54 $ Hewlett-P
ackard Co. */
$Revision: Hewlett-Packard ISSL 1.6 kern/h/auditmsg.
h, sysaudit, vvos_davis, davis7 $ $Date: 96/
12/05 07:42:32 $ */
99/03/09 kern/h/auditmsg.h, hpux, hpux_10.26, ic5bn
Revision 1.3 PATCH_10.26 (PHKL_17891) */
/usr/share/man/man3.Z/authaudit.3:
None
cksum(1) Output:
818332283 22388 /usr/conf/lib/libsec.a(sec_audit.o)
1866385554 16612 /usr/conf/lib/libsec.a(audit_dev.o)
3832503544 55723 /usr/include/sys/audit.h
3972675672 25519 /usr/include/sys/auditmsg.h
3832503544 55723 /etc/conf/h/audit.h
3972675672 25519 /usr/conf/h/auditmsg.h
2658750108 4181 /usr/share/man/man3.Z/authaudit.3
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes:
PHKL_17891 PHKL_17900 PHKL_18794
Equivalent Patches:
PHKL_26735:
s700: 10.26
Patch Package Size: 270 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 PHKL_26745
5a. For a standalone system, run swinstall to install the
patch:
swinstall -x autoreboot=true -x match_target=true \
-s /tmp/PHKL_26745.depot
By default swinstall will archive the original software in
/var/adm/sw/patch/PHKL_26745. If you do not wish to retain a
copy of the original software, you can create an empty file
named /var/adm/sw/patch/PATCH_NOSAVE.
WARNING: If this file exists when a patch is installed, the
patch cannot be deinstalled. Please be careful
when using this feature.
It is recommended that you move the PHKL_26745.text file to
/var/adm/sw/patch for future reference.
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHKL_26745.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions: None
-----End of Document ID: PHKL_26745------------------------------------------
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]