Oracle DB. was hanging, no error messages in alert_LOG , but sys log say something ...

[Shahbaz_1]

Hi Friends,
Please help to,

Typical situation, when users were compleining, that db was not responding, I checked, was unable to login to Oracle ,
but the alert_LOG didn't show any error messages.
The system log, was showing some errors, around 4:33pm was, the starting point of system hanging.

How to diagnos the problem? Could it be os or network problem,

Please find the attached file, the content of alert_LOG and system log file, during the time, system hanging.

Please ...

Thanks & Regards
Syed

[Ron Kinner]

What does it say if you do:


/opt/resmon/b
in/resdata -R 148176898 -r /system/events/core_hw/core_hw -n 148176903 -a

(not sure if that is one command or two)

Ron

[Jeff Schussele]

Hi Syed,

You need to execute the following commands:

/opt/resmon/b
in/resdata -R 148176898 -r /system/events/core_hw/core_hw -n 148176903 -a

/opt/resmon/b
in/resdata -R 148176898 -r /system/events/core_hw/core_hw -n 148176904 -a

Since it's pointing to core_hw it's possible you have a HW problem.

HTH,
Jeff

[Shahbaz_1]

Hi Ron & Jeff
Thank lot for the response.

I run the following, but,
/opt/resmon/bin/resdata -R 148176898 -r /system/events/core_hw/core_hw -n 148176903 -a

There is no active monitor request for the resource /system/events/core_hw/core_
hw associated with the request id 148176898.

/opt/resmon/bin/resdata -R 148176898 -r /system/events/core_hw/core_hw -n 148176904 -a
There is no active monitor request for the resource /system/events/core_hw/core_
hw associated with the request id 148176898.

So where should I look at this error?

Is there any way to have thorugh check of hardware?

So Please !!!
Thanks & Regards
Syed

[Jeff Schussele]

You need to look in the log - located in:
/etc/opt/resmon/log

Should be the client.log

Rgds,
Jeff

[Jeff Schussele]

BUT...it could be a diff one - check all .logs there & correlate date/time.

HTH,
Jeff

[T G Manikandan]

Hello,
Please look into the STM logs.
Your syslog file does not show any point to the error except it tells that it is a core hardware problem(event from the EMS)
check the logs and browse the below link for the errors.

http://docs.hp.com/hpux/onlinedocs/diag/ems/dm_core_hw.htm

Thanks

[Shahbaz_1]

Hi Jeff,
I didn't find much info.in /etc/opt/resmon/log

Going back again, If in our absense (dba), if users compleined, that they were un-able to login to oracle-system.
First definitely we'll look at alert_LOG,
then sys_LOG.
What else we should do, at the O/S or Network level, to diagnos the problem?

Could it be possible that, it was a database problem, but nothing recorded in alert_LOG by oracle?
Please ....

Thanks & Regards
Syed

[Shahbaz_1]

Hi Manikandan,
Thanks lot, the HP-Doc site refered by you is great.
Can you pls tell more about STM logs?

Regards
Syed

[Ruediger Noack]

Poosibly it was a connection problem to the database. Have a look at
$ORACLE_HOME/network/log/listener.log

Ruediger