0

I have a Linux system (Fedora 32, kernel 5.7-rc2) booting from NFS. After exactly one hour it locks up and doesn't respond to anything. It's perfectly reproducible and journalctl does not show anything, either because nothing shows up or because it does not catch the error quickly enough before everything gets locked.

Mount options are: rw,nolock,noacl,vers=4

Export options are: rw,async,no_root_squash,fsid=3

Using NFS3 does not help either. I also disabled chrony service because I suspected it running some stuff hourly but this also was not successful Any ideas how to debug this or what might be the cause?

0

1 Answer 1

0

Okay, it was not some obscure stuff like solar storms, Thanos, rainbow unicorns ... But the lease time of the dhcp server.

At least turning it to 10 min lead to the same problem in, surprise, 10 minutes. Just testing right now what happens when I extend it beyond hour ...

btw, did someone check the power plug?

Edit: After setting the lease time to infinite and the client now running already for 63 minutes, I can confirm, that it was definitely the lease time. Case closed :)

2
  • The question i think you should be asking is why the DHCP lease affected the system at all. (They generally get renewed to the same IP address so that shouldn't be changing.) Apr 26, 2020 at 13:45
  • 1
    That's a good question indeed and I'll look into that. Maybe the client lost it's IP which of course would sucks when using NFS? Anyways, it's been running for 5h now. I had a similar setup before and can't remember specifying the lease time but I'm gonna look it up. New system, new bugs ... Apr 26, 2020 at 17:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .