input / output error on /etc/pve

[paulwarren]

Hi Guys,

Just a quick question, I'm getting input / output errors on /etc/pve aka /dev/fuse.

This server is a single server not a node / cluster or anything. If i did /etc/init.d/pve-cluster restart what would happen, baring in mind this is a live server with live vm's that I can not take down. Also would this fix the input / output errors?


Thanks
Paul

 

[dietmar]

Also would this fix the input / output errors?

Maybe, but I do not know the reason for the IO errors. Maybe the root partition is full? Any hints in /var/log/syslog?

 

[paulwarren]

Sorry, I should have said

df -h
/dev/fuse 30M 24K 30M 1% /etc/pve

so using 24k of 30m partition

The only partition close to being full is /var/lib/vz at 88% the
reason behind this is we run a backup of a qemu at 1am, and it can't write the lock file to run the back up. on looking into this I found that /etc/pve can't be written to and we get input / output errors if I try to write to /etc/pve/ If i try to write to /etc/ everything is fine. So looks to me like a issue with the partition.

I've looked through the syslog and dmesg but can't find anything to suggest why this is a issue. I was thinking of maybe unmount / mount the /dev/fuse partition but didn't want to just do it in case it took servers down due to /etc/pve not being there any more.

 

[dietmar]

What is the output of

# df -h /

 

[paulwarren]

root@prox2:~# df -h /
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 10G 1.3G 8.3G 13% /

 

[paulwarren]

any one have any ideas on this, as this server is not backing up vm's cause of it

 

[dietmar]

Already tried to restart the pve-cluster service? Try:

# service pve-cluster restart
# service pvedaemon restart
# service pveproxy restart

 

[paulwarren]

Hi,

Will this take down any services? this is a live server running live websites. I did not want to do the above and the vm's go down.

Thanks
Paul

 

[dietmar]

Will this take down any services? /QUOTE]

Should not stop any VMs.

 

[paulwarren]

Thanks for this, i've now run the first 2, cluster and daemon, both have fixed the issue. proxy didn't exist as a server.

Many thanks for your help and merry xmas and happy new year.