System Hangs.. HDD error??

[baluchen]

Dear Admin

We have HP ClassD9000 Server with 11.0 OS. Past few weeks it has been started to hang without any error on console as well as syslog. I could able to ping but no login prompt for telnet session. Finally I need to hard boot. I want to fix this problem. When I look one of its HDD from STM (Information menu)it shows

Write errors â 2
Non-medium errors â 6

Is this error comes under normal circumstance?

Could this problem due to because of this? If so, is there any utility in HP-UX to check hard disk without loosing data? Please advice incase if an

[Robert-Jan Goossens · Accepted solution]

Hi,

Looks like one of your disk is dying.

Try a read test.

# dd if=/dev/rdsk/cxtydz of=/dev/null bs=1024k

This will take a while, if toy get any errors call HP and replace your disk.

Hope this helps,
Robert-Jan

[SANTOSH S. MHASKAR]

Hi Balu,

I also faced same problem on same System
(HP9000 D-Class). The problem came at leastonce in a week . What I did is I changed the Internal SCSI connection for the HDDs. Now the system is OK.

-Santosh

[SANTOSH S. MHASKAR]

Also check OLDsyslog for errors if any.

-Santosh

[PVR]

CHeck each Hard disk with dd command as suggested by robert.

Check for the SCSI terminations. If it is not proper, system can behave like this.

[SureshKumar_2]

Hi Balu

Could u confirm the following things..

*) Error is always keep on displaying, or when u take some backup is that happening???
*) Could u get the time stamp of the error mesg.
*) When u access some particular FS is this problem is coming up??
*) could u give /var/adm/mess*
*) and also syslog mesg...

so that we could able to help u further...!!

suresh

[baluchen]

Thanks for your replies

1.Usually this machine does not connected with monitor. We are accessing the terminal through console hardware, which will display console output to our PC terminal through telnet screen.

2. no I canâ t get time stamp from STM

3. No. I canâ t figure out when the system going to hang.

4. File doesnâ t exist

5. Attached with this post (var/adm/syslog/OLDsyslog

I will do as Robert & santhosh

[baluchen]

Dear Robert

As you suggest i ran read test. It retuns with same number of in and out counts. it didn't end with any error.. Can you suggest what could be other problem?

Balu

[KapilRaj]

You say , when the system is hung u can still puing it but no telnet . Are you able to login via console and see if inetd is running ?.

It looks like inetd issue or it is running out of memory but a rogue process

Regds,

Kaps

[Sanjiv Sharma_1]

Hello,

It seems that the inetd daemon dies and hence you are not able to telnet to the server.

Check that you have installed the latest patch bundles:
http://www1.itrc.hp.com/service/patch/locateBundles.do

hth.

[baluchen]

Dear Kapil,

Actually this server does not connected with keyboard and mouse even monitor. We have a separate console device, which connects serial ports of multiple HP servers with mapping different port number. We use the port number to connect particular HP server (by establishing telnet session between Client to Console device)

I couldnâ t find any errors on my Console device. Even I left for an hour after established console connectivity.nothing displayed except login screen.

Any utility to check hardware problem incase if an

[baluchen]

Hi Admin,

On Friday I connected a monitor on this particular server. When I came to office on Monday the system was hung. But there was no error log on console. Simply login screen appears on monitor. When I enter user ID it doesnâ t ask password.

But still I can ping. How do I monitor incase of memory full? Else could be hardware problem? How do I diagnose?

Thanks in advance
Bal

[baluchen]

LCD Display Shows RUNF02F

Balu

[Naveej.K.A]

Hi balu,

I have had a similar problem with one of my D220s and this turned out to be the problem with the root hard disk. The system used to hand wouldn't respond to any thing from the console. Even a ping was not working.

Eventhough your scenario is a little different, take a serious diagnosis of the hard disk (which STM is reporting errors) if it is a root hard disk. Try the dd command two to three times and see if any error comes.

With best wishes
Naveej

[SureshKumar_2]

Hi

Run F02F -> sytems is in normal load,

Check for swap/root volumes any bad blocks. There could be background process, to re-trying some process.

Mostly it could be Disk problem, Check with HP, and get it done replaced.

*) And also confirm, without any Database/application running it on the server, do u find probem?? If that is the case bad sector may be in swap/root volumes. confirm these, by dd.

*) If u suspect u can very well check thru'
#vmstat 1 50

*) When u say after typing it has not asked for password, have u got after some time (after 10/20 mts), then the possible cause will be Bad block reallocation.

Hope these inputs will Help you

suresh

[baluchen]

Thanks dear all,

As Robert jan suggested I did read test repeatedly. I got the same results. I did the same to all volumes.
For inetd issueâ ¦ I just prepared one script, which will send report at the interval of every 10min about inetd daemon. Let me update if there is any problem in inetd daemon.

Can you please clarify this doubt?
I am getting the following mesg when booting this particular machine. Is this error is normal. Or do I have to do anything?
â CPU 0
WARNING: Memory configuration is not optimized for performance. Referto the System Installation or Memory Installation manuals for memory configuration guidelines.â

Is there any other utility to check hard disk bad sectors without loosing data?

Thanks for spending your valuable time with m

[baluchen]

i just put a simple script in crontab to check inetd daemon at the interval of every 10min. this script will make entry in a particular file about the status. But when the system starts to hang, the cron job also stopped at the same time. There after there were no entries made in that file.

Does cron related with inetd.?? Any clue???

Thanks in Advance

Balu

[baluchen]

This is the right time to wish all of you.

Suddenly I noticed there was â POWERFAILEDâ error in log file. This error belongs to root disk. So I replaced new disk and then ignite backup were restored.

Now my machine is all right. Thanks for your participation