Redhat 6.6 and weird NFS issues (tl;dr, try NFS v3 rather than default v4.x)

So I recently built a new VM with RHEL 6.6 for a production Domino server. I mounted the data and log directories to a NetApp filer volume as I’d done in the past with RHEL 5.x, and went about my way.

A couple of days later I got some alerts about the server crashing. I ran a top and found that RPCIOD was using the entirety of my CPU’s, not only that but the filer was maxing and according to vSphere my VM was using 2gps network traffic.

The only way out of this mess was to hard reboot the box, as trying to kill these processes or unmounting didn’t work.

After some investigation and looking at the logs I came across these two bug reports – both closely match the problems I was having and it seems that kernel 2.6.32-504.16.2.el6 may have issues with NFSv4.

https://access.redhat.com/solutions/1220203 & https://access.redhat.com/solutions/1485703

After changing fstab to specifically use NFSv3 rather than v4 the problem hasn’t reoccurred yet.

I could reproduce the error on a different VM, with a different mapped volume on the same filer.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s