[Gluster-devel] CPU usage in AFR
vikas at gluster.com
Mon Dec 28 11:13:31 UTC 2009
Gordan Bobic wrote:
> I thought there were always at least 2 lock servers, no? And they
> rotate around depending on which servers are up (when the current lock
> server dies, it fails over to the next one). Is that not the case?
> Also if the servers leave and re-join, does the lock server
> functionality rotate back or does it remain on the failed-over server?
No, by default there is only one lock server. You can increase it using the
options I mentioned in the previous mail.
If servers re-join, the lock servers do rotate back, since the definition
of "first n subvolumes that are up" changes.
More information about the Gluster-devel