<html xmlns="http://www.w3.org/1999/xhtml" xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office"><head><!--[if gte mso 9]><xml><o:OfficeDocumentSettings><o:AllowPNG/><o:PixelsPerInch>96</o:PixelsPerInch></o:OfficeDocumentSettings></xml><![endif]--></head><body>
As I mentioned, I’ve never had such a case.<div>Give it a try in a test environment and if it works - go ahead.</div><div><br></div><div>Best Regards,</div><div>Strahil Nikolov <br><br><p class="yahoo-quoted-begin" style="font-size: 15px; color: #715FFA; padding-top: 15px; margin-top: 0">On Wednesday, August 16, 2023, 1:21 PM, Sebastian Neustein <sebastian.neustein@arc-aachen.de> wrote:</p><blockquote class="iosymail"><div id="yiv2370176778"><div>
    <div class="yiv2370176778moz-cite-prefix">Strahil Nikolov:<br clear="none">
    </div>
    <blockquote type="cite">
      I’ve never had such situation and I don’t recall someone sharing
      something similar.</blockquote>
    <br clear="none">
    <br clear="none">
    That's strange, it is really easy to reproduce. This is from a fresh
    test environment:<br clear="none">
    <br clear="none">
    summary:<br clear="none">
    - There is one snapshot present. <br clear="none">
    - On one node glusterd is stopped. <br clear="none">
    - During the stop, one snapshot is deleted. <br clear="none">
    - The node is brought up again<br clear="none">
    - On that node there is an orphaned snapshot<br clear="none">
    <br clear="none">
    <br clear="none">
    detailed version:<br clear="none">
    # on node 1:<br clear="none">
    root@gl1:~# cat /etc/debian_version<br clear="none">
    11.7<br clear="none">
    <br clear="none">
    root@gl1:~# gluster --version<br clear="none">
    glusterfs 10.4<br clear="none">
    <br clear="none">
    root@gl1:~# gluster volume info<br clear="none">
    Volume Name: glvol_samba<br clear="none">
    Type: Replicate<br clear="none">
    Volume ID: 91cb059e-10e4-4439-92ea-001065652749<br clear="none">
    Status: Started<br clear="none">
    Snapshot Count: 1<br clear="none">
    Number of Bricks: 1 x 3 = 3<br clear="none">
    Transport-type: tcp<br clear="none">
    Bricks:<br clear="none">
    Brick1: gl1:/data/glusterfs/glvol_samba/brick0/brick<br clear="none">
    Brick2: gl2:/data/glusterfs/glvol_samba/brick0/brick<br clear="none">
    Brick3: gl3:/data/glusterfs/glvol_samba/brick0/brick<br clear="none">
    Options Reconfigured:<br clear="none">
    cluster.granular-entry-heal: on<br clear="none">
    storage.fips-mode-rchecksum: on<br clear="none">
    transport.address-family: inet<br clear="none">
    nfs.disable: on<br clear="none">
    performance.client-io-threads: off<br clear="none">
    features.barrier: disable<br clear="none">
    <br clear="none">
    root@gl1:~# gluster snapshot list<br clear="none">
    snaps_GMT-2023.08.15-13.05.28<br clear="none">
    <br clear="none">
    <br clear="none">
    <br clear="none">
    # on node 3:<br clear="none">
    root@gl3:~# systemctl stop glusterd.service<br clear="none">
    <br clear="none">
    <br clear="none">
    <br clear="none">
    # on node 1:<br clear="none">
    root@gl1:~# gluster snapshot deactivate
    snaps_GMT-2023.08.15-13.05.28<br clear="none">
    Deactivating snap will make its data inaccessible. Do you want to
    continue? (y/n) y<br clear="none">
    Snapshot deactivate: snaps_GMT-2023.08.15-13.05.28: Snap deactivated
    successfully<br clear="none">
    <br clear="none">
    root@gl1:~# gluster snapshot delete snaps_GMT-2023.08.15-13.05.28<br clear="none">
    Deleting snap will erase all the information about the snap. Do you
    still want to continue? (y/n) y<br clear="none">
    snapshot delete: snaps_GMT-2023.08.15-13.05.28: snap removed
    successfully<br clear="none">
    <br clear="none">
    root@gl1:~# gluster snapshot list<br clear="none">
    No snapshots present<br clear="none">
    <br clear="none">
    <br clear="none">
    <br clear="none">
    # on node 3:<br clear="none">
    root@gl3:~# systemctl start glusterd.service<br clear="none">
    <br clear="none">
    root@gl3:~# gluster snapshot list<br clear="none">
    snaps_GMT-2023.08.15-13.05.28<br clear="none">
    <br clear="none">
    root@gl3:~# gluster snapshot deactivate
    snaps_GMT-2023.08.15-13.05.28<br clear="none">
    Deactivating snap will make its data inaccessible. Do you want to
    continue? (y/n) y<br clear="none">
    snapshot deactivate: failed: Pre Validation failed on gl1.ad.arc.de.
    Snapshot (snaps_GMT-2023.08.15-13.05.28) does not exist.<br clear="none">
    Pre Validation failed on gl2. Snapshot
    (snaps_GMT-2023.08.15-13.05.28) does not exist.<br clear="none">
    Snapshot command failed<br clear="none">
    <br clear="none">
    root@gl3:~# lvs -a<br clear="none">
      LV                                 VG        Attr       LSize 
    Pool      Origin    Data%  Meta%  Move Log Cpy%Sync Convert<br clear="none">
      669cbc14fa7542acafb2995666284583_0 vg_brick0 Vwi-aotz-- 15,00g
    tp_brick0 lv_brick0 0,08<br clear="none">
      lv_brick0                          vg_brick0 Vwi-aotz-- 15,00g
    tp_brick0           0,08<br clear="none">
      [lvol0_pmspare]                    vg_brick0 ewi------- 20,00m<br clear="none">
      tp_brick0                          vg_brick0 twi-aotz--
    18,00g                     0,12   10,57<br clear="none">
      [tp_brick0_tdata]                  vg_brick0 Twi-ao---- 18,00g<br clear="none">
      [tp_brick0_tmeta]                  vg_brick0 ewi-ao---- 20,00m<br clear="none">
    <br clear="none">
    <br clear="none">
    <br clear="none">
    <br clear="none">
    Would it be dangerous to just delete following items on node 3 while
    gluster is down:<br clear="none">
    - the orphaned directories in /var/lib/glusterd/snaps/<br clear="none">
    - the orphaned lvm, here 669cbc14fa7542acafb2995666284583_0 <br clear="none">
    <br clear="none">
    Or is there a self-heal command?<br clear="none">
    <br clear="none">
    Regards<br clear="none">
    Sebastian<br clear="none">
    <br clear="none">
    <div class="yiv2370176778moz-cite-prefix">Am 10.08.2023 um 20:33 schrieb Strahil
      Nikolov:<br clear="none">
    </div>
    <blockquote type="cite">
      </blockquote></div><div id="yiv2370176778yqt07030" class="yiv2370176778yqt4182361365"><div>
      I’ve never had such situation and I don’t recall someone sharing
      something similar.
      <div><span style=""><br clear="none">
        </span></div>
      <div><span style="">Most probably
          it’s easier to remove the node from the TSP and re-add it.</span></div>
      <div>
        <div>Of course , test the case in VMs just to validate that it’s
          possible to add a mode to a cluster with snapshots.</div>
        <div><br clear="none">
        </div>
        <div>I have a vague feeling that you will need to delete all
          snapshots.</div>
        <div>
          <div><br clear="none">
          </div>
          <div>Best Regards,</div>
          <div>Strahil Nikolov </div>
          <div><br clear="none">
            <p style="font-size:15px;color:#715FFA;padding-top:15px;margin-top:0;" class="yiv2370176778yahoo-quoted-begin">On
              Thursday, August 10, 2023, 4:36 AM, Sebastian Neustein
              <a rel="nofollow noopener noreferrer" shape="rect" ymailto="mailto:sebastian.neustein@arc-aachen.de" target="_blank" href="mailto:sebastian.neustein@arc-aachen.de" class="yiv2370176778moz-txt-link-rfc2396E"><sebastian.neustein@arc-aachen.de></a> wrote:</p>
            <blockquote class="yiv2370176778iosymail">
              <div id="yiv2370176778">
                <div>
                  <div class="yiv2370176778c-message_kit__blocks yiv2370176778c-message_kit__blocks--rich_text">
                    <div class="yiv2370176778c-message__message_blocks yiv2370176778c-message__message_blocks--rich_text">
                      <div class="yiv2370176778p-block_kit_renderer">
                        <div class="yiv2370176778p-block_kit_renderer__block_wrapper yiv2370176778p-block_kit_renderer__block_wrapper--first">
                          <div class="yiv2370176778p-rich_text_block">
                            <div class="yiv2370176778p-rich_text_section">Hi<br clear="none">
                              <br clear="none">
                              Due to an outage of one node, after
                              bringing it up again, the node has some
                              orphaned snapshosts, which are already
                              deleted on the other nodes. <br clear="none">
                              <br clear="none">
                              <span class="yiv2370176778c-mrkdwn__br"></span>How
                              can I delete these orphaned snapshots?
                              Trying the normal way produceses these
                              errors:<br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">[2023-08-08
                                19:34:03.667109 +0000] E [MSGID: 106115]
[glusterd-mgmt.c:118:gd_mgmt_v3_collate_errors] 0-management: Pre
                                Validation failed on B742. Please check
                                log file for details.</code><br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">[2023-08-08
                                19:34:03.667184 +0000] E [MSGID: 106115]
[glusterd-mgmt.c:118:gd_mgmt_v3_collate_errors] 0-management: Pre
                                Validation failed on B741. Please check
                                log file for details.</code><br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">[2023-08-08
                                19:34:03.667210 +0000] E [MSGID: 106121]
[glusterd-mgmt.c:1083:glusterd_mgmt_v3_pre_validate] 0-management: Pre
                                Validation failed on peers</code><br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">[2023-08-08
                                19:34:03.667236 +0000] E [MSGID: 106121]
[glusterd-mgmt.c:2875:glusterd_mgmt_v3_initiate_snap_phases]
                                0-management: Pre Validation Failed</code><span class="yiv2370176778c-mrkdwn__br"></span><br clear="none">
                              <br clear="none">
                              Even worse: I followed <a rel="nofollow noopener noreferrer" shape="rect" target="_blank" href="https://access.redhat.com/documentation/en-us/red_hat_gluster_storage/3.5/html/administration_guide/troubleshooting_snapshots" class="yiv2370176778c-link">read hat gluster
                                snapshot trouble guide</a> and deleted
                              one of those directories defining a
                              snapshot. Now I receive this on the cli:<br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">run-gluster-snaps-e4dcd4166538414c849fa91b0b3934d7-brick6-brick[297342]:
                                [2023-08-09 08:59:41.107243 +0000] M
                                [MSGID: 113075]
                                [posix-helpers.c:2161:posix_health_check_thread_proc]
0-e4dcd4166538414c849fa91b0b3934d7-posix: health-check failed, going
                                down</code><br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">run-gluster-snaps-e4dcd4166538414c849fa91b0b3934d7-brick6-brick[297342]:
                                [2023-08-09 08:59:41.107243 +0000] M
                                [MSGID: 113075]
                                [posix-helpers.c:2161:posix_health_check_thread_proc]
0-e4dcd4166538414c849fa91b0b3934d7-posix: health-check failed, going
                                down</code><br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">run-gluster-snaps-e4dcd4166538414c849fa91b0b3934d7-brick6-brick[297342]:
                                [2023-08-09 08:59:41.107292 +0000] M
                                [MSGID: 113075]
                                [posix-helpers.c:2179:posix_health_check_thread_proc]
0-e4dcd4166538414c849fa91b0b3934d7-posix: still alive! -> SIGTERM</code><br clear="none">
                              <code class="yiv2370176778c-mrkdwn__code">run-gluster-snaps-e4dcd4166538414c849fa91b0b3934d7-brick6-brick[297342]:
                                [2023-08-09 08:59:41.107292 +0000] M
                                [MSGID: 113075]
                                [posix-helpers.c:2179:posix_health_check_thread_proc]
0-e4dcd4166538414c849fa91b0b3934d7-posix: still alive! -> SIGTERM</code><span class="yiv2370176778c-mrkdwn__br"></span><br clear="none">
                              <br clear="none">
                              What are my options? <br clear="none">
                              - is there an easy way to remove all those
                              snapshots?<br clear="none">
                              - or would it be easier to remove and
                              rejoin the node to the gluster cluster?<br clear="none">
                              <br clear="none">
                              Thank you for any help!<br clear="none">
                              <br clear="none">
                              Seb<br clear="none">
                              <span dir="ltr" class="yiv2370176778c-message__edited_label"></span></div>
                          </div>
                        </div>
                      </div>
                    </div>
                  </div>
                  <pre class="yiv2370176778moz-signature">-- 
Sebastian Neustein

Airport Research Center GmbH
Bismarckstraße 61
52066 Aachen
Germany

Phone: +49 241 16843-23
Fax: +49 241 16843-19
e-mail: <a rel="nofollow noopener noreferrer" shape="rect" ymailto="mailto:sebastian.neustein@arc-aachen.de" target="_blank" href="mailto:sebastian.neustein@arc-aachen.de" class="yiv2370176778moz-txt-link-abbreviated yiv2370176778moz-txt-link-freetext">sebastian.neustein@arc-aachen.de</a>
Website: <a rel="nofollow noopener noreferrer" shape="rect" target="_blank" href="http://www.airport-consultants.com" class="yiv2370176778moz-txt-link-freetext yiv2370176778moz-txt-link-freetext">http://www.airport-consultants.com</a>

Register Court: Amtsgericht Aachen HRB 7313
Ust-Id-No.: DE196450052

Managing Director:
Dipl.-Ing. Tom Alexander Heuer</pre>
                </div>
              </div>
              ________<br clear="none">
              <br clear="none">
              <br clear="none">
              <br clear="none">
              Community Meeting Calendar:<br clear="none">
              <br clear="none">
              Schedule -<br clear="none">
              Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC<br clear="none">
              Bridge: <a rel="nofollow noopener noreferrer" shape="rect" target="_blank" href="https://meet.google.com/cpu-eiue-hvk" class="yiv2370176778moz-txt-link-freetext">https://meet.google.com/cpu-eiue-hvk</a><br clear="none">
              Gluster-users mailing list<br clear="none">
              <a rel="nofollow noopener noreferrer" shape="rect" ymailto="mailto:Gluster-users@gluster.org" target="_blank" href="mailto:Gluster-users@gluster.org" class="yiv2370176778moz-txt-link-freetext">Gluster-users@gluster.org</a><br clear="none">
              <a rel="nofollow noopener noreferrer" shape="rect" target="_blank" href="https://lists.gluster.org/mailman/listinfo/gluster-users" class="yiv2370176778moz-txt-link-freetext">https://lists.gluster.org/mailman/listinfo/gluster-users</a><br clear="none">
            </blockquote>
          </div>
        </div>
      </div>
    
    <br clear="none">
  </div></div></div><blockquote></blockquote></blockquote></div>
</body></html>