[Gluster-users] Impact of low optical power on GlusterFS?

Pieter Baele pieter.baele at gmail.com
Fri Nov 25 13:49:13 UTC 2016


We've some important volumes on replica 2 sets (Red Hat Storage). Today we
had an incident in which we couldn't mount one of the volumes anymore, the
other volumes (less used) no problems.

More or less on the same moment we also received a warning / alarm from our
networking team that one of the optical/fibre 10G uplinks on those servers
is having low power issues (< -10dB).

Which impact can this have on Gluster?

in the logs; chronologically:
- a lot of selfheals: dht_log_new_layout_for_dir_selfheal
- afther this; both servers from a client:  has not responded in the last
42 seconds, disconnecting.
- saved_frames_unwind
- remote operation failures

I can upload a lot of files to RH support, but AFAIK the best expertise is
available here :-)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20161125/a8707e3c/attachment.html>

More information about the Gluster-users mailing list