Unsolved
This post is more than 5 years old
9 Posts
0
4654
NFS4 bad sequence-id errors
I have an FS7610 setup with some unix style containers that are being mounted via NFS4 to a number of Linux (CentOS 6.6 - 6.8) client machines. Looking through my kernel messages, I'm seeing a bunch of sporadic NFS errors.
This is on a closed network, so I can't just copy and paste my logs here, but essentially the error is:
kernel: NFS: v4 server returned a bad sequence-id error!
I found a redhat article (here) regarding this error when connecting to a NetApp. The article states that this is pretty much a generic error and to get a real handle on this, I need to run tcpdump to grab some packets. I'm not an expert on the inner workings of NFS, so the packet inspection likely won't help me much.
I'm hoping someone else here has seen this or something similar and can provide some help.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
December 6th, 2016 12:00
Hello,
Can you repost the link with just copy/paste? I can't access it in your posting.
What version of FluidOS are you running and what version of EQL firmware?
It's not a known problem that I know of. I'm checking with some other folks just in case.
Is it actually causing a problem?
Regards,
Don
amartin177
9 Posts
0
December 6th, 2016 12:00
Sure thing - the link to the RedHat article is: https://access.redhat.com/solutions/1328073 (you'll need a redhat support account to view the contents).
We are running v4.0.300098 of FluidFS. The EQL firmware version is 8.1.1. It should be noted that this is a multi-member group, but they are all running 8.1.1
While I wouldn't say it is causing issues directly, we have been noticing some strange behavior with regards to our NFS mounts. I'd like to make these messages go away if possible so I can narrow down some of the other issues that I'm seeing.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
December 6th, 2016 13:00
Hello,
Have you opened a support case? That would be the best option, since digging into the logs, and possible wiresharks traces will be needed I suspect.
Regards,
Don
amartin177
9 Posts
0
December 6th, 2016 15:00
I haven't opened a support case as of yet. I wanted to see if anyone in the community had seen this before. If no one else chimes in, I'll certainly go that route.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
December 7th, 2016 05:00
Hello,
That error is addressed in the V5 version of FluidFS, however, that version is not available for the FS76xx family of products.
I'm not sure it's related to your issue. I would suggest opening a support case so they can review the diags, and the network.
Regards,
Don
amartin177
9 Posts
0
December 8th, 2016 06:00
Do you happen to have the recommended options for mounting NFS shares? I have isolated the behavior to a number of machines that seem to be using the recommended options from the FS7610 install guide.
example from the manual:
mount -o rw,bg,hard,nointr,tcp,vers=3,timeo=2,retrans=10,rsize=32768,
wsize=32768 client_access_vip:hostname:/container_name /exported_directory /local_directory