Event Monitor Notification - only a vague idea

[cnb]

Jason,

Sorry I forgot to include to check the system activity log and not just the infolog activity:

# cstm
cstm> ls

Rgds,

[jasonK_1]

cnb,

cstm>ial
^-- (InfoActLog) is currently disabled. --

cstm>ls
Thu Feb 19 17:27:42 2009: Aborting daemon process (cclogd) with process
identifier (1541) as support tool system is shutting
down.

Thu Feb 19 17:27:42 2009: Aborting daemon process (diaglogd) with process
identifier (1542) as support tool system is shutting
down.

Thu Feb 19 17:27:42 2009: Aborting daemon process (memlogd) with process
identifier (1543) as support tool system is shutting
down.

Thu Feb 19 17:27:42 2009: Aborting daemon process (psmctd) with process
identifier (1544) as support tool system is shutting
down.

Thu Feb 19 17:35:58 2009: Diagmond daemon started.

Thu Feb 19 17:37:08 2009: Launching daemon executable
(/usr/sbin/stm/uut/bin/sys/cclogd).

Thu Feb 19 17:37:08 2009: Launching daemon executable
(/usr/sbin/stm/uut/bin/sys/diaglogd).

Thu Feb 19 17:37:08 2009: Launching daemon executable
(/usr/sbin/stm/uut/bin/sys/memlogd).

Thu Feb 19 17:37:08 2009: Launching daemon executable
(/usr/sbin/stm/uut/bin/sys/psmctd).

Fri Feb 20 09:35:11 2009: An IPC message of 0 length was read.

Possible Causes/Recommended Action:

The sender of the IPC message connected but did
not send any data or exited before sending any
data.
Check the support tool system activity log and
the tool activity logs for more information on
the sending process.

Fri Feb 20 09:35:11 2009: Attempt to read the IPC header from a port failed.
The port is on system name (hnrgnw) at IP address
(205.135.30.74) and has system port (1508).

Fri Feb 20 09:35:11 2009: Message received from system port had an invalid
header. Message will be ignored.

Fri Feb 20 10:25:44 2009: An IPC message of 0 length was read.

Possible Causes/Recommended Action:

The sender of the IPC message connected but did
not send any data or exited before sending any
data.
Check the support tool system activity log and
the tool activity logs for more information on
the sending process.

Fri Feb 20 10:25:44 2009: Attempt to read the IPC header from a port failed.
The port is on system name (hnrgnw) at IP address
(205.135.30.74) and has system port (1508).

Fri Feb 20 10:25:44 2009: Message received from system port had an invalid
header. Message will be ignored.

[cnb]

Jason,

Try this:

echo "sel dev all;info;wait;ial"|/usr/sbin/cstm

-or-
Try the report again manually

#cstm
cstm> sel all
cstm> info
cstm> il
quit
done
cstm> ial
quit
save
file-name
cstm> quit



hth,

[jasonK_1]

cnb,

OK, I got something out of that command. Please see the attachment.

[cnb]

Jason,

A few things are not right and they may or may not be related to the system firmware being way out of date. You system is at 41.39 and critcial fixes were published after that version. The latest is 44.28 for your system.

For 11.00:

http://www13.itrc.hp.com/service/patch/patchDetail.do?BC=main|patchDetail{PHSS_31800,{hpux:11.00,}}|&patchid=PHSS_32704&sel={hpux:11.00,}

a) According to the diag tool, one of your processors isn't responding properly. This is interesting because part of log report has both processors active and configured. So I'm not sure why it's failing to report properly, PDC?

+++++
from infolog report for proc at 166:

Processor PIM Information: PIM contains NO data

+++++

-- Information Tool Activity Log for CPU on path 166 --

Log creation time:
Tue Jun 2 14:44:53 2009

Tue Jun 2 14:44:53 2009: Information tool (cpu) starting on path (166).

Tue Jun 2 14:44:53 2009: Failure Summary:

The returned status from PDC_SYSTEM_INFO call:
(-4).

Possible Causes/Recommended Action:

The PDC_SYSTEMINFO call returned status indicating
that selected processor is not present.

This is an unexpected status. No recommended
action available.

+++++

However this issue was supposedly fixed with PDC 41.39:

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&taskId=120&prodSeriesId=377380&prodTypeId=15351&objectID=c00906010

I have a feeling it didn't and you may want to have HP update all of your system firmware to the latest supported versions.


b) For some reason the /dev/diag/diagx files are either missing or the tool can't read them. You'll need to resolve this or recreate the files with 'insf -e'. See insf man pages:

Tue Jun 2 14:44:52 2009

Tue Jun 2 14:44:52 2009: Information tool (pci) starting on path (0/7/0/0).

Tue Jun 2 14:44:52 2009: The device file /dev/diag/diag1 is not present on the
system. Communications with diag1 diagnostic pseudo
driver is not possible without this file.

Possible Causes/Recommended Action:

Please use "insf -e " from the /dev directory to
create the /dev/diag/diag1 device file.

Tue Jun 2 14:44:52 2009: Failed to obtain device_id, revision_id,
device_status, vendor_id and class_code for the card.

An error or warning occurred while obtaining the
information.


You might want to check /etc/rc.log and verify everything started up correctly the last time it booted. If you find errors then investigate.

IMHO I would get the system firmware updated to match the installed software requirements first. SInce the diag tool isn't reporting the adapter type I can't tell what h/w is supported for your config. You can check here for some adapter minimum PDC versions (although 11.00 isn't listed you can get an idea of where your system should be at):

http://docs.hp.com/en/SFWM1/ch04.html#ftn.dc3

The diagnostic subsystem is not functioning properly and you might not have any accurate warnings until something really breaks.


hth,


BTW:
http://forums13.itrc.hp.com/service/forums/helptips.do?#33

;-)

[jasonK_1]

cnb,

I'll do what you said

Thanks