Schlick Rupert
2021-Apr-21 16:28 UTC
[Gluster-users] Brick mount points are "bent" after restoring a snapshot
Dear List! We are using (user serviceable) snapshots and recently, we needed to restore a snapshot for a whole volume. The effect is now, that gluster uses a brick path to the mount of the lvm snapshot. Brick xyz:/run/gluster/snaps/d44afa00d24e4a249de440dc13bfe42c/brick1/gfs_home_brick1 The original brick mount point (/data/glusterfs_home/gfs_home_brick1) is stuck on the original logical volume, which is now completely out of date, because further work is going to the other logical volume (it seems). In the meantime, a third brick has been added to the (replicated) volume, having the "correct" path. Is there a preferred/easy/clean way to resolve this, without breaking anything? Best Rupert -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20210421/1f5712ec/attachment.html>
Strahil Nikolov
2021-Apr-22 03:18 UTC
[Gluster-users] Brick mount points are "bent" after restoring a snapshot
I have used remove-brick+add-brick on a replica volume. Yet, for large bricks
that could be a pain due to the ammount of data that has to be synced.
Best Regards,Strahil Nikolov
On Wed, Apr 21, 2021 at 19:47, Schlick Rupert<Rupert.Schlick at
ait.ac.at> wrote: <!--#yiv3523213834 _filtered {} _filtered
{}#yiv3523213834 #yiv3523213834 p.yiv3523213834MsoNormal, #yiv3523213834
li.yiv3523213834MsoNormal, #yiv3523213834 div.yiv3523213834MsoNormal
{margin:0cm;margin-bottom:.0001pt;font-size:11.0pt;font-family:"Calibri",
sans-serif;}#yiv3523213834 a:link, #yiv3523213834 span.yiv3523213834MsoHyperlink
{color:#0563C1;text-decoration:underline;}#yiv3523213834 a:visited,
#yiv3523213834 span.yiv3523213834MsoHyperlinkFollowed
{color:#954F72;text-decoration:underline;}#yiv3523213834
span.yiv3523213834E-MailFormatvorlage17 {font-family:"Calibri",
sans-serif;color:windowtext;}#yiv3523213834 .yiv3523213834MsoChpDefault
{font-family:"Calibri", sans-serif;} _filtered {}#yiv3523213834
div.yiv3523213834WordSection1 {}-->
Dear List!
?
We are using (user serviceable) snapshots and recently, we needed to restore a
snapshot for a whole volume.
The effect is now, that gluster uses a brick path to the mount of the lvm
snapshot.
?
Brick
xyz:/run/gluster/snaps/d44afa00d24e4a249de440dc13bfe42c/brick1/gfs_home_brick1
?
The original brick mount point (/data/glusterfs_home/gfs_home_brick1) is stuck
on the original logical volume, which is now completely out of date, because
further work is going to the other logical volume (it seems).
In the meantime, a third brick has been added to the (replicated) volume, having
the ?correct? path.
?
Is there a preferred/easy/clean way to resolve this, without breaking anything?
?
Best
Rupert
________
Community Meeting Calendar:
Schedule -
Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC
Bridge: https://meet.google.com/cpu-eiue-hvk
Gluster-users mailing list
Gluster-users at gluster.org
https://lists.gluster.org/mailman/listinfo/gluster-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.gluster.org/pipermail/gluster-users/attachments/20210422/1790a57d/attachment.html>