Page 1 of 1
Linux Servers with NFS mount
Posted: Thu Apr 06, 2017 8:14 am
by TBT
We have several Linux Servers mounting a common share. If the share (NAS device) goes off-line, it appears all Host and Service checks for said Linux servers fail with time-out or no response errors. Can anyone explain this behavior?
Re: Linux Servers with NFS mount
Posted: Thu Apr 06, 2017 11:49 am
by mcapra
How are you monitoring these machines? If it's via NRPE, is the NRPE or xinetd binary located on the NFS mount?
Re: Linux Servers with NFS mount
Posted: Thu Apr 06, 2017 12:38 pm
by TBT
Guess that would be helpful too. We're using SNMP.
Re: Linux Servers with NFS mount
Posted: Thu Apr 06, 2017 4:06 pm
by mcapra
Are you able to send over a system profile? Or at least some samples of the check commands being run. The snmpd.conf from one of the machines (sanitized of course) may also be helpful.
Re: Linux Servers with NFS mount
Posted: Fri Apr 07, 2017 7:27 am
by TBT
mcapra wrote:Are you able to send over a system profile? Or at least some samples of the check commands being run. The snmpd.conf from one of the machines (sanitized of course) may also be helpful.
We're using the Linux SNMP wizard to add a check, along with a mount point (on /). Have the mount go off-line and you should be able to replicate this issue.
Additional notes: This typically happens when the mount point is unavailable due to time-out (gone stale, not unmounted purposely). Also, as mentioned the mount point is /dir not the conventional /mnt/dir.
Re: Linux Servers with NFS mount
Posted: Fri Apr 07, 2017 1:17 pm
by tgriep
If all of the SNMP checks to a remote host fails when the share on the remote host is down, sounds like the SNMP Daemon on the remote server stopped responding to the Nagios server.
Can you login to one of the remote systems that had the issue, check the /var/log/messages file to see if the SNMP daemon crashed or logged an error?
Re: Linux Servers with NFS mount
Posted: Fri Apr 07, 2017 2:24 pm
by TBT
tgriep wrote:If all of the SNMP checks to a remote host fails when the share on the remote host is down, sounds like the SNMP Daemon on the remote server stopped responding to the Nagios server.
Can you login to one of the remote systems that had the issue, check the /var/log/messages file to see if the SNMP daemon crashed or logged an error?
By remote system you mean one of the servers that has the NFS mount, correct? Nothing of interest in /var/log/messages
Re: Linux Servers with NFS mount
Posted: Fri Apr 07, 2017 2:35 pm
by scottwilkerson
Out of curiosity, can you still run snmpwalk or snmpget commands against this host when it is in this state?
e.g.
Code: Select all
snmpwalk -v 2c -c public your.host.com
Re: Linux Servers with NFS mount
Posted: Fri Apr 07, 2017 3:28 pm
by TBT
scottwilkerson wrote:Out of curiosity, can you still run snmpwalk or snmpget commands against this host when it is in this state?
e.g.
Code: Select all
snmpwalk -v 2c -c public your.host.com
We'll have to recreate the issue and report back.
Re: Linux Servers with NFS mount
Posted: Fri Apr 07, 2017 3:50 pm
by cdienger
Thanks. Please update us once you do.