bdf command hanging

[Steven E. Protter]

Shalom,

Long term, I would recommend udpating youre Operating Environment, which will update the network and i/o subsystems among many other issues.

Short term, you will need to trim that syslog file and reboot.

One of your mount points is stale.

grep nfs /etc/fstab

Please post. You also might want to look at /etc/mnttab for mount points. This behavior mimics the way a system operates when a remote NFS mount has gone stale.

SEP

[Arunabha Banerjee]

This is a known  issue with the older mpt driver and it can be fixed by installing the newer version. If you clear the syslog file and when you restart the daemon you will get the same alert and syslog will grow very fast. You need to update SCSI Ultra320 A6961-60011 driver to get rid with this issue.

You can find latest HP-UX Ultra320 SCSI driver@ https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=scsiU320-00

The scsiU320-00 bundle delivers the mpt driver, which supports the following adapters:
•HP A7173A PCI-X Dual Channel Ultra320 SCSI Host Bus Adapter

•HP Ultra320 SCSI core I/O controllers

•HP AB290A PCI-X Ultra320 SCSI and Gigabit Ethernet Combination Card

Latest version as follows:

Product #:  scsiU320-00 

Version:  11.11.0911/11.23.0809/11.31.0909 

Software specification: HP-UX 11.11 / HP-UX 11.23 / HP-UX 11.31 

Can you please give me the following output. 

# uptime

# uname -a

# model

# cat /etc/mnttab

# autopath display all (If securepath installed)

# swlist -l product | grep -i scsiU320

For syslog you can null the file and then restart syslogd daemon

# >/var/adm/syslog/syslog.log

# /sbin/init.d/syslogd stop

# /sbin/init.d/syslogd start

[Laurent Menase]

the 2 known problems which can make that results are:

mtp like in prevous reply and Symantec 1439223

https://sort.symantec.com/public/documents/sfha/5.1sp1/hp-ux/productguides/html/sfrac_notes/ch01s09s...

1439223 The lmxpollport function times out with an incorrect timeout value on systems that have been running for more than 410 days. The issue was caused by the lbolt variable, which resets after 410 days.
 

So if you have Veritas Storage Foundation for Oracle RAC check the version. (it is solved in 5.1sp1 apparently, but fix exist for previous versions too.

so like the others

# >/var/adm/syslog/syslog.log

# /sbin/init.d/syslogd stop

# /sbin/init.d/syslogd start

reboot

update mpt driver, update VSF if necessary,

[Mark Read · Accepted solution]

The typical cause of the bdf hanging is that one or more of the volumes has lost connectivity with the server.  This could be a possible SAN volume and SAN connectivity issue, or a volume that gets unmounted for a snapshot backup and is not remounted properly afterwards.

Sometimes if the logical volumes are not mounted properly the bdf will hang.  I have seen cases which after a reboot the volumes mount normally and the bdf works as expected.

The 'bdf -l' command will list only locally configured volumes (not NFS or remote shares). This would only rule out if the cause of the hanging bdf is related to a remote mounted volume.

[-_]

Dear all,

Thanks a lot for all your valuable suggestions. Cause of this issue is volume groups are disconnected from storage due to storage hardware problem so that volume groups are unavilable and syslog is full.

Issue resolved from storage end by replace the hardware part. so iam able to access my volume groups, execute all commands bdf,bdf -l df etc., syslog file cleared .

Regards,

B.satish reddy