[Gluster-users] Rebuilding a failed cluster

[Richard Betel]

On 29/11/2023 22:58 Diego Zuccato wrote:
>If the old volume is still online, it's better if you copy from its FUSE
mount point to the new one.

The old servers are offline and cannot be rebooted, but the bricks
themselves are running fine. The whole point of the exercise is to bring
the volumes back online *somehow* using new servers.

> Much depends on the original volume layout. For replica volumes you'll
find multiple copies of the same file on different bricks. And sometimes
> 0-byte files that are placeholders of renamed files: do not overwrite a
good file with its empty version!

It's a dispersed 3-brick volume, but I'm trying to treat the old bricks as
read only as much as possible.

>But since it's a temporary "backup", there's no need to use another
Gluster volume as the destination: just use a USB drive directly
> connected to the old nodes (one at a time) or to a machine that can still
FUSE mount the old volume. Once you have a backup, write-protect
> it and experiment freely :)

The problem is the *disperse*. with replicas, you end up with multiple
copies of the whole file. Unless my bricks are hosed too and I'm
misunderstanding things, the files I have are not normal complete files
because of the disperse volume.
If I had used simple replication back in the day, I'd be able to just make
a backup of a brick and have all my files, but past me saw disperse as a
form of RAID for gluster, and now I have 3 disks (aka bricks) that are
largely unreadable until I can manage to
convince a new cluster to properly mount them...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20231130/3f40e60a/attachment.html>