EMS restarts every 15-20 mins.

[RAC_1]

HP-UX 11.0

# OnlineDiag B.11.00.15.08 HPUX 11.0 Support Tools Bundle
OnlineDiag.Sup-Tool-Mgr B.11.00.15.08 Support Tools Manager for HPUX systems
OnlineDiag.EMS-KRMonitor A.11.00.02 EMS Kernel Resource Monitor
OnlineDiag.EMS-Core A.03.20 EMS Core Product
OnlineDiag.EMS-Config A.03.20 EMS Config
OnlineDiag.Predictive C.11.00.16 HP Predictive Support
OnlineDiag.Contrib-Tools B.11.00.15.08 Contributed Tools
OnlineDiag.LIF-LOAD B.11.00.15 HP LIF LOAD Tools


The EMS restarts every 15-20 mins on it's own. the syslog message is as follows.

Jul 2 10:49:18 xxxx EMS [8507]: ----- EMS Monitor Restart ----- Title: dm_core_hw Command: /usr/sbin/stm/uut/bin/tools/monitor/dm_core_hw Vendor: Hewlett-Packard Company Version: B.01.00 To obtain a list of currently monitored resources, execute the following: /opt/resmon/bin/resdata -M 1998375097

The command /opt/resmon/bin/resdata -M 1998375097 gives --
/system/events/core_hw/core_hw

I have tried doing following.

Restart EMS/diagnostics
1. /sbin/init.d/ems stop/start
/etc/opt/resmon/lbin/monconfig
Disable monitoring/enable it.
2. /sbin/init.d/diagnostics stop/start
3. Reset GSP.

I am not concerned about message, I am concerned about why it is restarting every 15-20 mins. It is filling up syslog.log
Upgrade won't be a good thing to do at the moment.

Anil

[Scot Bean]

Don't run /sbin/init.d/ems stop/start. These scripts actually just configure EMS.
EMS runs from /etc/inittab.

Things you can try:

- These diags are really old, Dec2000 or so. Install the latest 11.00 diags (Mar04).

- Latest patch for these old diags is PHSS_25843, do you have this installed? There is a defect fixed in the patch stream that has this decsription, sounds like your problem:

"3. dm_core_hw monitor
EMS monitor dm_core_hw will exit at each polling cycle,
which is once every 15 minutes. It will then be restarted,
and a "restart" event message will be logged in the event
log /var/opt/resmon/log/event.log and send the same message
to all targets associated with this monitor."


Note: Diags and patch installs for this should not require a reboot, just install & go.

[RAC_1]

Thank you Scot.

"3. dm_core_hw monitor
EMS monitor dm_core_hw will exit at each polling cycle,
which is once every 15 minutes. It will then be restarted,
and a "restart" event message will be logged in the event
log /var/opt/resmon/log/event.log and send the same message
to all targets associated with this monitor."

>>>Would it be able to control this polling interval?? Can we avoid those messages??

Anil

[RAC_1]

No takers for this???


Anil

[Scot Bean]

Edit /var/stm/config/tools/monitor/dm_core_hw.cfg.

You will see POLL_INTERVAL variable.

[RAC_1]

Scot, thank you again.

As you explained, the restart is caused because of the polling interval of 15 mins.
After that the EMS (whole) is restarted. Right???

Anil

[Scot Bean · Accepted solution]

The restart is ultimately caused by a bug in the code. You should really patch or upgrade diags.

If you increase the config file parameter, you will not fix the problem, but you can make it less frequent.

[RAC_1]

Thsnk scot. It really helped.

Anil

[Andrew Merritt_2]

Just to enforce what has already been said, the problem is caused by a bug in a very old version of the OnlineDiags (for which there does appear to be a patch).

Changing the poll interval just means that the monitor will restart less frequently, but won't help since it means the monitor won't actually be doing anything useful; it's starting, waiting for the configured poll interval then dying when it tries to do the polling; it won't be reporting any events.

You should upgrade to a much more recent version of OnlineDiags; the version you have will not be supported by HP.

[RAC_1]

Closed