Open bobwhitelock opened 7 years ago
From @vlj91 on May 17, 2016 9:56
1.5 build on OpenStack. After purging a depot, the following appears when running df on compute hosts:
df
[alces@node00(hpc1) ~]$ df -h df: ‘/opt/gridware/depots/8153b7b7’: Stale file handle Filesystem Size Used Avail Use% Mounted on /dev/vda1 40G 3.8G 37G 10% / devtmpfs 1.9G 0 1.9G 0% /dev tmpfs 1.9G 0 1.9G 0% /dev/shm tmpfs 1.9G 17M 1.9G 1% /run tmpfs 1.9G 0 1.9G 0% /sys/fs/cgroup 10.75.0.3:/home 40G 4.0G 37G 10% /home 10.75.0.3:/opt/gridware/depots/f705bd34 40G 4.0G 37G 10% /opt/gridware/depots/f705bd34 10.75.0.3:/opt/gridware/etc 40G 4.0G 37G 10% /opt/gridware/etc 10.75.0.3:/opt/gridware/depots/f2e8534d 40G 4.0G 37G 10% /opt/gridware/depots/f2e8534d tmpfs 380M 0 380M 0% /run/user/1000
Copied from original issue: alces-software/clusterware#152
From @mjtko on August 1, 2016 14:58
I think this happens because we don’t yet trigger/handle any kind of "unmount on purge" event.
We should look into doing that to handle this condition properly.
From @vlj91 on May 17, 2016 9:56
1.5 build on OpenStack. After purging a depot, the following appears when running
df
on compute hosts:Copied from original issue: alces-software/clusterware#152