[Gluster-users] gluster volume rebalance making things more unbalanced

[Joe Julian]

When a file should be moved based on its dht hash mapping but the target that it should be moved to has less free space than the origin, the rebalance command does not move the file and leaves the dht pointer in place. When you use "force", you override that behavior and always move each file regardless of free space.

In theory, eventually when the rebalance is finished you should end up with utilization mostly balanced but as the rebalance is processing you may end up in the state you show.

On August 27, 2020 7:32:16 AM PDT, Pat Haley <phaley at mit.edu> wrote:
>
>Hi,
>
>We have distributed gluster volume spread across 4 bricks. Yesterday I 
>noticed that the remaining space was uneven (about 2.7TB, 1.7TB, 1TB, 
>1TB) so I issued the following rebalance command
>
>  * |gluster volume rebalance <VOLNAME> start force|
>
>Today I see that instead, things have gotten even more unbalanced (64G 
>853G 6.2T 20K).  I'm killing the rebalance now.  What should I do to 
>make sure that I get a successful rebalance?
>
>Thanks
>
>Pat||
>
>-- 
>
>-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
>Pat Haley                          Email:  phaley at mit.edu
>Center for Ocean Engineering       Phone:  (617) 253-6824
>Dept. of Mechanical Engineering    Fax:    (617) 253-8125
>MIT, Room 5-213                    http://web.mit.edu/phaley/www/
>77 Massachusetts Avenue
>Cambridge, MA  02139-4301

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20200827/cd5a8fbc/attachment.html>