[Gluster-users] slow lstat on 3.12 disperse volume

[Ingard Mevåg]

Hi

We've recently done some testing with a 3.12 disperse cluster. The
performance of filesystem stat calls was terrible, taking multiple seconds.
We dumped client side stats to see what was going on and noticed gluster
STAT was the culprit. tpcdump shows a STAT call being sent and replied to
quite fast, but still the client hangs for multiple seconds before
returning.

After downgrading client side to 3.10 (servers still being 3.12), stat
calls return instantaneous. After dumping client side stats when running
the 3.10 client we noticed that the linux lstat gets translated to a STATFS
instead of a STAT.

Has something changed to cause this behaviour? Is there some additional
tuning that needs to be done when running 3.12?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180216/a23f15ac/attachment.html>