Event Monitor error messages

[Miquel_2]

Hello,

Event monitor sends e-mails like this every 20 minutes

From root@itaca.mgc.es Sat Nov 5 00:07:44 MET 2005
Received: (from root@localhost)
by itaca.mgc.es (8.11.1/8.11.1) id jA4N7iU09582;
Sat, 5 Nov 2005 00:07:44 +0100 (MET)
Date: Sat, 5 Nov 2005 00:07:44 +0100 (MET)
Message-Id: <200511042307.jA4N7iU09582@itaca.mgc.es>
To: root@itaca.mgc.es
From: root@itaca.mgc.es
Subject: itaca.mgc.es: Event Monitor Restart Notification

>------------ Event Monitoring Service Event Notification ------------<

Notification Time: Sat Nov 5 00:07:44 2005

itaca.mgc.es sent Event Monitor Restart information:

Title: dm_ses_enclosure
Command: /usr/sbin/stm/uut/bin/tools/monitor/dm_ses_enclosure
Vendor: Hewlett-Packard Company
Version: B.01.00
Monitor PID: 9542

Resources currently monitored:
/storage/events/enclosures/ses_enclosure/0_2_1_0.15.0
/storage/events/enclosures/ses_enclosure/0_3_1_0.15.0

>---------- End Event Monitoring Service Event Notification ----------<



Then i look at /etc/opt/resmon/log/api.log and seems to something fails but i don't know exactly what.

Can someone help me ?
Thanks.


-------------------Start Event--------------------
User event occurred at Sat Nov 5 00:06:41.843411 2005
Process ID: 9477 (/usr/sbin/stm/uut/bin/tools/.../dm_stape) Log Level: Error
Tool is exiting due to receipt of an unexpected signal (11).

SIGSEGV (11) signal indicates a segmentation violation.

Possible Causes/Recommended Action:

Internal Application error. Tool attempted to reference an invalid address.
Usually a NULL or bad pointer.
-------------------End Event----------------------

-------------------Start Event--------------------
User event occurred at Sat Nov 5 00:07:42.233744 2005
Process ID: 9542 (/usr/sbin/stm/uut/.../dm_ses_enclosure) Log Level: Error
perform_polling: internal failure to initialize the device
at address 0/2/1/0.15.0.
-------------------End Event----------------------

-------------------Start Event--------------------
User event occurred at Sat Nov 5 00:07:42.500250 2005
Process ID: 9542 (/usr/sbin/stm/uut/.../dm_ses_enclosure) Log Level: Error
perform_polling: internal failure to initialize the device
at address 0/3/1/0.15.0.
-------------------End Event----------------------

[Mel Burslan]

did you have a tape drive on this server at one time and it was removed recently, maybe one with an autochanger ?

[Matti_Kurkela]

Looks like the event monitor software is having some internal trouble. Check your Event Monitor software version and patch level. Search the patch database for known bugs.

Of course, in the background there might be an actual hardware fault that triggers this bug somehow. Are there any other error messages? Recent tombstone files at /var/tombstones? Check the kernel message buffer (the "dmesg" command) and /var/adm/syslog/syslog.log.

You could also use the STM diagnostics (cstm/mstm/xstm) to check out the devices at hardware paths 0/2/1/0.15.0 and 0/3/1/0.15.0.

[Miquel_2]

Sorry, someone power down the tape!!

Thanks!

[Andrew Merritt_2]

Hi Miguel,
That could explain the messages from dm_stape, but not the ones from dm_ses_enclosure.

Did turning the tape back on stop the messages saying that dm_ses_enclosure was being restarted?

What version of OnlineDiags are you running (run 'cstm' and see what it reports)?

Andrew

[Andrew Merritt_2 · Accepted solution]

Some more information regarding dm_ses_enclosure. This problem has been seen before in a couple of cases where there was bad hardware configuration.

This happened when either:

a) the device was set to SAF-TE mode instead of SES mode. See below for more details.

b) there are duplicate worldwide names in the two controllers. This problem is fixed in current versions of the OnlineDiags (HWE0303 or later), so is less likely.

See http://www.docs.hp.com/en/diag/stm/stm_upd.htm#table for the list of OnlineDiags versions.

Configuration - Setting BCC DIP Switches:

- Two DIP switch banks on BCC A (left slot from rear) determine configuration
- If BCC A is removed, DIP switches on BCC B determine configuration
- The switch settings on BCC A must match the switch settings on BCC B
- According to the manual, the BCC's are shipped with all switches in the
"1" (OFF/ENABLED) position

NOTE: If the DS2300 is being connected to an HP-UX system, the 'SES/SAF-TE' switch should be set to 'SES' (0) mode. If the disk system is being connected to a Netserver, the switch should be set to 'SAF-TE' (1) mode

Andrew

[Miquel_2]

Thanks Andrew, i changed the DIP switch and it's all ok!

No more messages received since the change.

Thanks,
Miquel.