Re: remove-brick failure on distributed with 5.6

["Taste-Of-IT" <kontakt@xxxxxxxxxxxxxx>]

Hi,

i had similar problem. In my case the rebalance didnt finish because of not enough free space to migrate the space to other nodes. Reason was 1% Reservation Options which is setup by default in distributed, but i set it to 0%, which was ignored by gluster

Greatings 

Taste


Am 25.05.2019 07:00:13, schrieb Nithya Balachandran:

Hi Brandon,

Please send the following:

1. the gluster volume info

2. Information about which brick was removed

3. The rebalance log file for all nodes hosting removed bricks.

Regards,

Nithya

On Fri, 24 May 2019 at 19:33, Ravishankar N <ravishankar@xxxxxxxxxx> wrote:

Adding a few DHT folks for some possible suggestions.

-Ravi

On 23/05/19 11:15 PM, brandon@xxxxxxxxxxxxx wrote:

Does anyone know what should be done on a glusterfs v5.6 "gluster volume remove-brick" operation that fails?  I'm trying to remove 1 of 8 distributed smaller nodes for replacement with larger node.

 

The "gluster volume remove-brick ... status" command reports status failed and failures = "3"

 

cat /var/log/glusterfs/volbackups-rebalance.log

...

[2019-05-23 16:43:37.442283] I [MSGID: 109028] [dht-rebalance.c:5070:gf_defrag_status_get] 0-volbackups-dht: Rebalance is failed. Time taken is 545.00 secs

 

All servers are confirmed in good communications and updated and freshly rebooted and retried the remove-brick few times with fail each time

 

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users