I kept looking at logs and could not find anything that is useful. I ordered some servers as a testbed. I will try to reproduce the problem with that equipment.
Regards,
Ludwig
On Thu, Jun 22, 2017 at 11:55 AM, Atin Mukherjee <amukherj@xxxxxxxxxx> wrote:
On Tue, Jun 20, 2017 at 7:32 PM, Ludwig Gamache <ludwig@xxxxxxxxxxxxx> wrote:All,Over the week-end, one of my volume became unavailable. All clients could not access their mount points. On some of the clients, I had user processes that we using these mount points. So, I could not do a umount/mount without killing these processes.I also noticed that when I restarted the volume, the port changed on the server. So, clients that were still using the previous TCP/port could not reconnect.If a volume is restarted and a new set of ports are assigned to the bricks clients will eventually get to know about these new ports from glusterd at reconnect attempt and should be able to successfully connect over automatically. If that hasn't happened for your case you might have hit an issue for which you'd need to either open a bug through bugzilla with all the logs attached from client and servers or attach over the email.
I was wondering if there is a way to tell to the client that a specific volume as a new port to connect to?Regards,Ludwig--Ludwig Gamache
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users
Ludwig Gamache
IT Director - Element AI
4200 St-Laurent, suite 1200
514-704-0564
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-users
