Error in syslog

[Mohammad Tariq Ayub]

Hi,

I am getting the following message in my syslog. what should i do.

DIAGNOSTIC SYSTEM WARNING:
Jun 3 16:16:57 billing vmunix: If the diaglogd daemon is not active, use the Daemon Startup command
Jun 3 16:16:57 billing vmunix: in stm to start it.
Jun 3 16:16:57 billing vmunix: If the diaglogd daemon is active, use the logtool utility in stm
Jun 3 16:16:57 billing vmunix: to determine which I/O subsystem is logging excessive errors.
Jun 3 16:16:57 billing vmunix: DIAGNOSTIC SYSTEM WARNING:
Jun 3 16:16:57 billing vmunix: The diagnostic logging facility is no longer receiving excessive
Jun 3 16:16:57 billing vmunix: errors from the I/O subsystem. 1 I/O error entries were lost.
Jun 3 16:16:57 billing vmunix: The diagnostic logging facility has started receiving excessive
Jun 3 16:16:57 billing vmunix: errors from the I/O subsystem. I/O error entries will be lost
Jun 3 16:16:57 billing vmunix: until the cause of the excessive I/O logging is corrected.

[Eugeny Brychkov]

Hello,
first of all start diaglogd:
check /etc/rc.config.d/diagnostic config file to contain string "DIAGNOSTICS=1", and then run "/sbin/init.d/diagnostic start" and check with 'ps -ef' for running diag daemons.
The you can see extended messages in syslog, and examine evens using logtool in STM (support tools manager - cstm, mstm, xstm)
Eugeny

[Rajesh G. Ghone]

Hi,

You can check the error message through STM,but before that you make sure that you have started Diagnostics which will be available in /etc/rc.config.d
DIAGNOSTIC=1

Regards,
Rajesh G.

[Mohammad Tariq Ayub]

Hi,

From logtool i got a device 1/0/8/0/2.2.22.0.0.1.7 is creating the problem. But how can i correct that problem

[Bill McNAMARA_1]

well it's the lun of something or other,

attach an ioscan -fnk in your next post. It may be an unavailable device.

[Mohammad Tariq Ayub]

Hi,

Here is the ioscan -fnk output

disk 41 1/0/8/0/0.2.22.0.0.1.7 sdisk CLAIMED DEVICE HP A6189A
/dev/dsk/c10t1d7 /dev/rdsk/c10t1d7

actually this lun should be connected bu other node not with this node.

[Bill McNAMARA_1]

the fact that it shows up in ioscan is normal even if it's shared hardware and not in use on this node.
Do you see any problems on the other node with the same lun?

[Mohammad Tariq Ayub]

Hi,

It is in cluster solution. The other node is ok. This problem occured when we tried to run nickel. It tried to access a lun that is supposed to be in other package. thats why from this server it is noit getting that path.

[Zeev Schultz]

Hm,va7400,check zoning / binds to hosts.

[Bill McNAMARA_1]

Yea, at this point I'd post your question in the storage forum!.

What was nickel doing at the time? diskinfo? ioscan? vgdisplays? What version of nickel? open a call with nickel too! http://www.grc.hp.com/docs/nickel

Later,
Bill

[Mohammad Tariq Ayub]

Hi,

At that time it was executing lvlnboot -v and trying to collect information about a lun that was not this systems volumegroup.

[Eugeny Brychkov]

Zeev is absolutely correct. This is VA7400 device, controlleris attached to 1GB brocade port #6 and LUN #15. First of all check configuration: host behavior table. This host should be recognized by VA as s hpux host, and VA should have either default host port behavior = HPUX or it should be present in host behavior table as HPUX host.
Check
armdsp -a
for VA overall state report
armlog -e
for VA logs to see anything unusual
Eugeny