[Gluster-users] storage.health-check-interval

[Eivind Sarto]

I am new to gluster and I am evaluating using it for a project.  One of the
things I am looking at is various failure scenarios and how gluster deals
with them.

As far as I can tell, gluster is supposed to recover from a hung filesystem
brick.  I see that storage.health-check-interval is set to 30 sec by
default and that this is supposed to detect a hung filesystem and recover.

My attempt to test this was to create a 1x2 replicated volume with XFS
bricks, and then "hang" one of the bricks by executing xfs_freeze on it.
But, if I attempt to modify the volume, the client hangs forever.

Would this method be appropriate to simulate a hung filesystem?
Or is there something else I need to enable in gluster-3.7.4

-eivind
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150918/bdcf51f1/attachment.html>