dmesg -> got error .

[Idham]

Today suddenly in my dmesg encounter this error log....

DIAGNOSTIC SYSTEM WARNING:
The diagnostic logging facility has started receiving excessive
errors from the I/O subsystem. I/O error entries will be lost
until the cause of the excessive I/O logging is corrected.
If the diaglogd daemon is not active, use the Daemon Startup command
in stm to start it.
If the diaglogd daemon is active, use the logtool utility in stm
to determine which I/O subsystem is logging excessive errors.

[Suraj K Sankari]

Hi,

Did you check your /var/adm/syslog/syslog.log ?

Is there any error over there at syslog.log

Suraj

[Idham]

got same error....in /var/adm/syslog/syslog.log


Nov 26 21:58:17 camrac01 uvservd[17350]: stdout = $AgentGUID=2E7F2CD5A9FBD5EA93C554102099D980;AgentStatus=1;
Nov 26 22:07:56 camrac01 vmunix: DIAGNOSTIC SYSTEM WARNING:
Nov 26 22:07:56 camrac01 vmunix: The diagnostic logging facility has started receiving excessive
Nov 26 22:07:56 camrac01 vmunix: errors from the I/O subsystem. I/O error entries will be lost
Nov 26 22:07:56 camrac01 vmunix: until the cause of the excessive I/O logging is corrected.
Nov 26 22:07:56 camrac01 vmunix: If the diaglogd daemon is not active, use the Daemon Startup command
Nov 26 22:07:56 camrac01 vmunix: in stm to start it.
Nov 26 22:07:56 camrac01 vmunix: If the diaglogd daemon is active, use the logtool utility in stm
Nov 26 22:07:56 camrac01 vmunix: to determine which I/O subsystem is logging excessive errors.

[Wim Rombauts]

You have a storage problem.
Look at "ioscan -fnC disk" and check for lines with "NO_HW" in it.
Alterbatively, run "vgsdisplay" (without options) and see what warnings/errors it returns about missing physical volumes.

[Idham]

Win,

Thanx for the reply..

but i already check on both command earlier. No such error/NO_HW or unclaimed disk. Everything looks OK.

[Jeeshan]

this will also occur in your HBA.

check the HBA card(s).

[Idham]

Dear ahsan

Thanks for the idea..
BTW, please refer below...looks like no problem

Class I H/W Path Driver S/W State H/W Type Description
=================================================================
fc 0 0/3/2/0 td CLAIMED INTERFACE HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td0
fc 1 0/4/2/0 td CLAIMED INTERFACE HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td1

[Jeeshan]

your FC card is seems ok.

if its SAN/FC based, it could be a problem with the SAN, FC card or cables.

may be latest patch set may solve this.

you may apply latest hwenabler path bundles.

[Wim Rombauts]

The only time I get erros like that is when I loose one side of our mirrored LV's completely. The error keeps repeating until the mirror-side is back alive.

I am not an stm-specialist. I opened stm myself to look at this logtool.

Maybe you just should do what the error messages says : run stm (having an X-Display works more comfortable) and query your hardware from there to see what device is giving you these IO errors. STM is a default software as part of on-line diagnostics.

[Jozef_Novak]

Generally speaking, this can happen anytime there is some sort of problem with the system peripherals and sudden burst of errors is being logged by the STM. The best way to find the cause is to run the STM and check your system for possible malfunctions. (e.g. via logtool)

Jozef

[Idham]

Thanks for the info....

But can anyone assist me on how to run the stm properly....
Maybe a steps by steps instruction will do....

p/s:do I need to run on X11 or just CLI?
will if affect my server performace or oracle db?

[~sesh]

There are 3 ways to run STM:

1. Menu Mode (mstm)
2. X Windows environment (xstm)
3. Command Line (cstm)

In STM you have 3 options: Information Tool, Log Tool & Diagnose Tool.

Which way you run STM depends on how comfortable you are and whether X window system has been setup or not. If you have X windows, then you could run xstm to run the tool in graphical mode.

Log Tool would be more useful in your situation. STM would require a little system resource so your server resources have to considered before running it while the server is in production.