Patch Name: PHCO_16864 Patch Description: s700_800 10.20 Year 2000 cumulative cron/at/crontab patch Creation Date: 98/11/06 Post Date: 99/01/22 Warn: 99/02/12 - This Critical Warning has been issued by HP. The /sbin/cron process that is running immediately after the installation or removal of this patch may cause certain jobs initiated by cron(1M) to hang. This is known to affect cron jobs using the fbackup(1M) command. The problem will only occur when the patch is installed or removed in a swinstall(1M) or swremove(1M) session that does *not* cause the system to be rebooted. During the installation and removal of this patch, cron(1M) is started by the postinstall and postremove scripts. This invocation of cron(1M) will not allow certain types of commands, like fbackup(1M), to complete successfully. Once cron(1M) is restarted via a system reboot or manual invocation, the problem will not be experienced. To avoid this problem, it is recommended that cron(1M) be restarted immediately following the installation and removal of the patch. This can be accomplished either via a system reboot or by killing and restarting cron(1M). Please note that it will be necessary to use the SIGKILL signal (signal 9) to kill cron: 1. kill -9 2. /sbin/init.d/cron start Hardware Platforms - OS Releases: s700: 10.20 s800: 10.20 Products: N/A Filesets: OS-Core.CMDS-MIN Automatic Reboot?: No Status: General Superseded With Warnings Critical: No Path Name: /hp-ux_patches/s700_800/10.X/PHCO_16864 Symptoms: PHCO_16864: 1.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 diplayed 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 symtoms 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_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 diplayed 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: 1653274167 5003425777 5003426254 1653269746 1653266239 1653241117 4701383794 1653215038 4701347153 4701334763 4701328179 4701393728 Patch Files: /usr/sbin/cron /usr/bin/at /usr/bin/crontab what(1) Output: /usr/sbin/cron: PATCH_10_20 PHCO_10737 $Revision: 78.2.1.15 $ PATCH_10_20: hpux_rel.o cron.o funcs.o 98/11/05 /usr/bin/at: PATCH_10_20 PHCO_10737 $Revision: 78.2.1.15 $ $Revision: 76.2.1.3 $ $Revision: 76.2.1.1.1.4 $ PATCH_10_20: hpux_rel.o at.o pdt.o ptime.o pdate.o f uncs.o permit.o 98/11/05 /usr/bin/crontab: PATCH_10_20 PHCO_10737 $Revision: 78.2.1.15 $ PATCH_10_20: hpux_rel.o crontab.o permit.o funcs.o 9 8/11/05 cksum(1) Output: 2964472964 40960 /usr/sbin/cron 9209656 45056 /usr/bin/at 4233445326 24576 /usr/bin/crontab Patch Conflicts: None Patch Dependencies: None Hardware Dependencies: None Other Dependencies: None Supersedes: PHCO_7836 PHCO_8277 PHCO_9348 PHCO_10123 PHCO_10737 PHCO_14179 PHCO_15234 PHCO_15679 Equivalent Patches: PHCO_16942: s700: 11.00 s800: 11.00 PHCO_16863: s700: 10.10 s800: 10.10 PHCO_16953: s700: 10.01 s800: 10.01 Patch Package Size: 170 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_16864 5a. For a standalone system, run swinstall to install the patch: swinstall -x autoreboot=true -x match_target=true \ -s /tmp/PHCO_16864.depot 5b. For a homogeneous NFS Diskless cluster run swcluster on the server to install the patch on the server and the clients: swcluster -i -b This will invoke swcluster in the interactive mode and force all clients to be shut down. WARNING: All cluster clients must be shut down prior to the patch installation. Installing the patch while the clients are booted is unsupported and can lead to serious problems. The swcluster command will invoke an swinstall session in which you must specify: alternate root path - default is /export/shared_root/OS_700 source depot path - /tmp/PHCO_16864.depot To complete the installation, select the patch by choosing "Actions -> Match What Target Has" and then "Actions -> Install" from the Menubar. 5c. For a heterogeneous NFS Diskless cluster: - run swinstall on the server as in step 5a to install the patch on the cluster server. - run swcluster on the server as in step 5b to install the patch on the cluster clients. By default swinstall will archive the original software in /var/adm/sw/patch/PHCO_16864. 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_16864.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_16864.depot of=/dev/rmt/0m bs=2k Special Installation Instructions: The /sbin/cron process that is running immediately after the installation or removal of this patch may cause certain jobs initiated by cron(1M) to hang. This is known to affect cron jobs using the fbackup(1M) command. The problem will only occur when the patch is installed or removed in a swinstall(1M) or swremove(1M) session that does *not* cause the system to be rebooted. During the installation and removal of this patch, cron(1M) is started by the postinstall and postremove scripts. This invocation of cron(1M) will not allow certain types of commands, like fbackup(1M), to complete successfully. Once cron(1M) is restarted via a system reboot or manual invocation, the problem will not be experienced. To avoid this problem, it is recommended that cron(1M) be restarted immediately following the installation and removal of the patch. This can be accomplished either via a system reboot or by killing and restarting cron(1M). Please note that it will be necessary to use the SIGKILL signal (signal 9) to kill cron: 1. kill -9 2. /sbin/init.d/cron start