[ClusterLabs] to heal gfid which is not there -- IGNORE wrong list

[lejeczek]

On 04/02/2024 12:57, lejeczek via Users wrote:
> hi guys.
>
> So, I've manged to make my volume go haywire, here:
>
> -> $ gluster volume heal VMAIL info
> Brick 10.1.1.100:/devs/00.GLUSTERs/VMAIL
> Status: Connected
> Number of entries: 0
>
> Brick 10.1.1.101:/devs/00.GLUSTERs/VMAIL
> <gfid:452e0075-8b69-4c78-a443-213103a7fc37>/dovecot-uidlist
> Status: Connected
> Number of entries: 1
>
> Brick 10.1.1.99:/devs/00.GLUSTERs/VMAIL-arbiter
> Status: Connected
> Number of entries: 0
>
> That _gfid_ nor _dovecot-uidlist_ do not exist on any of 
> the bricks, but I've noticed that all bricks have:
> .glusterfs-anonymous-inode-462a1850-a31a-4a17-934d-26f3996dc9b8
> which is an empty dir.
>
> That volume I fear won't heal on its own and might need 
> some intervention - would you know how/where to start?
>
> many thanks, L.
>
> _______________________________________________
> Manage your subscription:
> https://lists.clusterlabs.org/mailman/listinfo/users
>
> ClusterLabs home:https://www.clusterlabs.org/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20240204/fe6f239f/attachment.htm>