Flaky NFS mount can bring down the server?

[Aaron Bliss]

If I'm understanding what happened, you may want to try using the soft and
timeout mount options when mounting the share:

http://uw713doc.sco.com/en/FS_manager/nfsD.nfsopt.html

Aaron

-----Original Message-----
From: redhat-list-bounces at redhat.com [mailto:redhat-list-bounces at redhat.com]
On Behalf Of S M
Sent: Monday, March 23, 2009 2:11 PM
To: redhat-list at redhat.com
Subject: Flaky NFS mount can bring down the server?

Hi All,
Just observed a very strange behaviour on our RHEL 4.5 x86_64 server
(database server), where a box which exposed a NFS mount point to the
database server was not available, causing the database server to become
completely unresponsive, and had to be power cycled.  The logs are as below:

Mar 22 20:59:18 crmdb kernel: nfs: server x.x.x.x not responding, still
trying

Mar 22 21:01:50 crmdb kernel: nfs: server x.x.x.x not responding, still
trying

Mar 22 21:03:50 crmdb last message repeated 2 times

Mar 22 21:05:50 crmdb last message repeated 2 times

Mar 22 21:07:50 crmdb last message repeated 2 times

Mar 22 21:09:50 crmdb last message repeated 2 times

Mar 22 21:11:50 crmdb last message repeated 2 times

Mar 22 22:54:26 crmdb sshd(pam_unix)[26930]: session closed for user ssadmin

Mar 22 22:54:26 crmdb sshd(pam_unix)[26755]: session closed for user ssadmin

Mar 22 22:55:59 crmdb sshd(pam_unix)[6416]: session opened for user ssadmin
by (uid=0)

Mar 22 22:57:21 crmdb sshd(pam_unix)[6513]: session opened for user ssadmin
by (uid=0)

*Mar 22 23:22:09 crmdb kernel: nfs: server x.x.x.x OK*

*Mar 22 23:23:50 crmdb kernel: nfs: server x.x.x.x not responding, still
trying*

*Mar 22 23:24:09 crmdb last message repeated 11 times*

Mar 22 23:40:46 crmdb sshd(pam_unix)[7452]: session opened for user ssadmin
by (uid=0)

Mar 23 07:43:52 crmdb syslogd 1.4.1: restart.

Mar 23 07:43:52 crmdb syslog: syslogd startup succeeded

Mar 23 07:43:52 crmdb kernel: klogd 1.4.1, log source = /proc/kmsg started.

The question is:
1. Should the unavailability of a NFS mount bring down the system?
2. What can I do to prevent this - the NFS mount is a box that may go down
off and on as it is a VTL that may be taken down for maintainence etc.
3. Are there any settings/parameters that we can use to prevent this
behaviour?

Thanks in advance,

Sanjeev
-- 
redhat-list mailing list
unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list