<div><div dir="auto">The disks were deployed with ceph-deploy / ceph-volume using the default style (lvm) and not simple-mode.</div></div><div dir="auto"><br></div><div dir="auto">The disks were provisioned as a whole, no resizing. I never touched the disks after deployment.</div><div dir="auto"><br></div><div dir="auto">It is very strange that this first happened after the update, never met such an error before.</div><div dir="auto"><br></div><div dir="auto">I found a BUG in the tracker, that also shows such an error with count 0. That was closed with „can’t reproduce“ (don’t have the link ready). For me this seems like the data itself is fine and I just hit a bad transaction in the replay (which maybe caused the crash in the first place).</div><div dir="auto"><br></div><div dir="auto">I need one of three disks back. Object corruption would not be a problem (regarding drop of a journal), as this cluster hosts backups which will fail validation and regenerate. Just marking the OSD lost does not seem to be an option.</div><div dir="auto"><br></div><div dir="auto">Is there some sort of fsck for BlueFS?</div><div dir="auto"><br></div><div dir="auto">Kevin</div><div dir="auto"><br></div><div><br><div class="gmail_quote"><div dir="ltr">Igor Fedotov <<a href="mailto:ifedotov@suse.de">ifedotov@suse.de</a>> schrieb am Mi. 3. Okt. 2018 um 13:01:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
  
    
  
  <div text="#000000" bgcolor="#FFFFFF">
    <p>I've seen somewhat similar behavior in a log from Sergey Malinin
      in another thread ("mimic: 3/4 OSDs crashed...")</p>
    <p>He claimed it happened after LVM volume expansion. Isn't this the
      case for you? <br>
    </p>
    <p>Am I right that you use LVM volumes?</p></div><div text="#000000" bgcolor="#FFFFFF">
    <br>
    <div class="m_2445726827436418132moz-cite-prefix">On 10/3/2018 11:22 AM, Kevin Olbrich
      wrote:<br>
    </div>
    </div><div text="#000000" bgcolor="#FFFFFF"><blockquote type="cite"></blockquote></div><div text="#000000" bgcolor="#FFFFFF"><blockquote type="cite">
      
      <div dir="ltr">Small addition: the failing disks are in the same
        host.
        <div>This is a two-host, failure-domain OSD cluster.<br>
          <br>
          <br>
          <div class="gmail_quote">
            <div dir="ltr">Am Mi., 3. Okt. 2018 um 10:13 Uhr schrieb
              Kevin Olbrich <<a href="mailto:ko@sv01.de" target="_blank">ko@sv01.de</a>>:<br>
            </div>
            <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div dir="ltr">
                <div dir="ltr">Hi!<br>
                  <br>
                  Yesterday one of our (non-priority) clusters failed
                  when 3 OSDs went down (EC 8+2) together.<br>
                  <b>This is strange as we did an upgrade from 13.2.1 to
                    13.2.2 one or two hours before.</b><br>
                  They failed exactly at the same moment, rendering the
                  cluster unusable (CephFS).</div>
                <div dir="ltr">We are using CentOS 7 with latest updates
                  and ceph repo. No cache SSDs, no external journal /
                  wal / db.<br>
                  <br>
                  <b><font size="4">OSD 29 (no disk failure in dmesg):</font></b><br>
                  2018-10-03 09:47:15.074 7fb8835ce1c0  0 set uid:gid to
                  167:167 (ceph:ceph)<br>
                  2018-10-03 09:47:15.074 7fb8835ce1c0  0 ceph version
                  13.2.2 (02899bfda814146b021136e9d8e80eba494e1126)
                  mimic (stable), process ceph-osd, pid 20899<br>
                  2018-10-03 09:47:15.074 7fb8835ce1c0  0 pidfile_write:
                  ignore empty --pid-file<br>
                  2018-10-03 09:47:15.100 7fb8835ce1c0  0 load: jerasure
                  load: lrc load: isa <br>
                  2018-10-03 09:47:15.100 7fb8835ce1c0  1 bdev create
                  path /var/lib/ceph/osd/ceph-29/block type kernel<br>
                  2018-10-03 09:47:15.100 7fb8835ce1c0  1
                  bdev(0x561250a20000 /var/lib/ceph/osd/ceph-29/block)
                  open path /var/lib/ceph/osd/ceph-29/block<br>
                  2018-10-03 09:47:15.100 7fb8835ce1c0  1
                  bdev(0x561250a20000 /var/lib/ceph/osd/ceph-29/block)
                  open size 1000198897664 (0xe8e0800000, 932 GiB)
                  block_size 4096 (4 KiB) rotational<br>
                  2018-10-03 09:47:15.101 7fb8835ce1c0  1
                  bluestore(/var/lib/ceph/osd/ceph-29) _set_cache_sizes
                  kv_min_ratio 1 > kv_ratio 0.5<br>
                  2018-10-03 09:47:15.101 7fb8835ce1c0  1
                  bluestore(/var/lib/ceph/osd/ceph-29) _set_cache_sizes
                  cache_size 536870912 meta 0 kv 1 data 0<br>
                  2018-10-03 09:47:15.101 7fb8835ce1c0  1
                  bdev(0x561250a20000 /var/lib/ceph/osd/ceph-29/block)
                  close<br>
                  2018-10-03 09:47:15.358 7fb8835ce1c0  1
                  bluestore(/var/lib/ceph/osd/ceph-29) _mount path
                  /var/lib/ceph/osd/ceph-29<br>
                  2018-10-03 09:47:15.358 7fb8835ce1c0  1 bdev create
                  path /var/lib/ceph/osd/ceph-29/block type kernel<br>
                  2018-10-03 09:47:15.358 7fb8835ce1c0  1
                  bdev(0x561250a20000 /var/lib/ceph/osd/ceph-29/block)
                  open path /var/lib/ceph/osd/ceph-29/block<br>
                  2018-10-03 09:47:15.359 7fb8835ce1c0  1
                  bdev(0x561250a20000 /var/lib/ceph/osd/ceph-29/block)
                  open size 1000198897664 (0xe8e0800000, 932 GiB)
                  block_size 4096 (4 KiB) rotational<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1
                  bluestore(/var/lib/ceph/osd/ceph-29) _set_cache_sizes
                  kv_min_ratio 1 > kv_ratio 0.5<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1
                  bluestore(/var/lib/ceph/osd/ceph-29) _set_cache_sizes
                  cache_size 536870912 meta 0 kv 1 data 0<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1 bdev create
                  path /var/lib/ceph/osd/ceph-29/block type kernel<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1
                  bdev(0x561250a20a80 /var/lib/ceph/osd/ceph-29/block)
                  open path /var/lib/ceph/osd/ceph-29/block<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1
                  bdev(0x561250a20a80 /var/lib/ceph/osd/ceph-29/block)
                  open size 1000198897664 (0xe8e0800000, 932 GiB)
                  block_size 4096 (4 KiB) rotational<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1 bluefs
                  add_block_device bdev 1 path
                  /var/lib/ceph/osd/ceph-29/block size 932 GiB<br>
                  2018-10-03 09:47:15.360 7fb8835ce1c0  1 bluefs mount<br>
                  2018-10-03 09:47:15.538 7fb8835ce1c0 -1 bluefs _replay
                  file with link count 0: file(ino 519 size 0x31e2f42
                  mtime 2018-10-02 12:24:22.632397 bdev 1 allocated
                  3200000 extents
[1:0x7008200000+100000,1:0x7009000000+100000,1:0x7009100000+100000,1:0x7009200000+100000,1:0x7009300000+100000,1:0x7009400000+100000,1:0x7009500000+100000,1:0x7009600000+100000,1:0x7009700000+100000,1:0x7009800000+100000,1:0x7009900000+100000,1:0x7009a00000+100000,1:0x7009b00000+100000,1:0x7009c00000+100000,1:0x7009d00000+100000,1:0x7009e00000+100000,1:0x7009f00000+100000,1:0x700a000000+100000,1:0x700a100000+100000,1:0x700a200000+100000,1:0x700a300000+100000,1:0x700a400000+100000,1:0x700a500000+100000,1:0x700a600000+100000,1:0x700a700000+100000,1:0x700a800000+100000,1:0x700a900000+100000,1:0x700aa00000+100000,1:0x700ab00000+100000,1:0x700ac00000+100000,1:0x700ad00000+100000,1:0x700ae00000+100000,1:0x700af00000+100000,1:0x700b000000+100000,1:0x700b100000+100000,1:0x700b200000+100000,1:0x700b300000+100000,1:0x700b400000+100000,1:0x700b500000+100000,1:0x700b600000+100000,1:0x700b700000+100000,1:0x700b800000+100000,1:0x700b900000+100000,1:0x700ba00000+100000,1:0x700bb00000+100000,1:0x700bc00000+100000,1:0x700bd00000+100000,1:0x700be00000+100000,1:0x700bf00000+100000,1:0x700c000000+100000])<br>
                  2018-10-03 09:47:15.538 7fb8835ce1c0 -1 bluefs mount
                  failed to replay log: (5) Input/output error<br>
                  2018-10-03 09:47:15.538 7fb8835ce1c0  1 stupidalloc
                  0x0x561250b8d030 shutdown<br>
                  2018-10-03 09:47:15.538 7fb8835ce1c0 -1
                  bluestore(/var/lib/ceph/osd/ceph-29) _open_db failed
                  bluefs mount: (5) Input/output error<br>
                  2018-10-03 09:47:15.538 7fb8835ce1c0  1
                  bdev(0x561250a20a80 /var/lib/ceph/osd/ceph-29/block)
                  close<br>
                  2018-10-03 09:47:15.616 7fb8835ce1c0  1
                  bdev(0x561250a20000 /var/lib/ceph/osd/ceph-29/block)
                  close<br>
                  2018-10-03 09:47:15.870 7fb8835ce1c0 -1 osd.29 0
                  OSD:init: unable to mount object store<br>
                  2018-10-03 09:47:15.870 7fb8835ce1c0 -1  ** ERROR: osd
                  init failed: (5) Input/output error
                  <div><br>
                  </div>
                  <div><b><font size="4">OSD 42:</font></b></div>
                  <div>disk is found by lvm, tmpfs is created but
                    service immediately dies on start without log...</div>
                  <div>This might be failed.</div>
                  <div><br>
                  </div>
                  <div><b><font size="4">OSD 47 (same as above, seems
                        not be died, no dmesg trace):</font></b></div>
                  <div>
                    <div>2018-10-03 10:02:25.221 7f4d54b611c0  0 set
                      uid:gid to 167:167 (ceph:ceph)</div>
                    <div>2018-10-03 10:02:25.221 7f4d54b611c0  0 ceph
                      version 13.2.2
                      (02899bfda814146b021136e9d8e80eba494e1126) mimic
                      (stable), process ceph-osd, pid 8993</div>
                    <div>2018-10-03 10:02:25.221 7f4d54b611c0  0
                      pidfile_write: ignore empty --pid-file</div>
                    <div>2018-10-03 10:02:25.247 7f4d54b611c0  0 load:
                      jerasure load: lrc load: isa </div>
                    <div>2018-10-03 10:02:25.248 7f4d54b611c0  1 bdev
                      create path /var/lib/ceph/osd/ceph-46/block type
                      kernel</div>
                    <div>2018-10-03 10:02:25.248 7f4d54b611c0  1
                      bdev(0x564072f96000
                      /var/lib/ceph/osd/ceph-46/block) open path
                      /var/lib/ceph/osd/ceph-46/block</div>
                    <div>2018-10-03 10:02:25.248 7f4d54b611c0  1
                      bdev(0x564072f96000
                      /var/lib/ceph/osd/ceph-46/block) open size
                      1000198897664 (0xe8e0800000, 932 GiB) block_size
                      4096 (4 KiB) rotational</div>
                    <div>2018-10-03 10:02:25.249 7f4d54b611c0  1
                      bluestore(/var/lib/ceph/osd/ceph-46)
                      _set_cache_sizes kv_min_ratio 1 > kv_ratio 0.5</div>
                    <div>2018-10-03 10:02:25.249 7f4d54b611c0  1
                      bluestore(/var/lib/ceph/osd/ceph-46)
                      _set_cache_sizes cache_size 536870912 meta 0 kv 1
                      data 0</div>
                    <div>2018-10-03 10:02:25.249 7f4d54b611c0  1
                      bdev(0x564072f96000
                      /var/lib/ceph/osd/ceph-46/block) close</div>
                    <div>2018-10-03 10:02:25.503 7f4d54b611c0  1
                      bluestore(/var/lib/ceph/osd/ceph-46) _mount path
                      /var/lib/ceph/osd/ceph-46</div>
                    <div>2018-10-03 10:02:25.504 7f4d54b611c0  1 bdev
                      create path /var/lib/ceph/osd/ceph-46/block type
                      kernel</div>
                    <div>2018-10-03 10:02:25.504 7f4d54b611c0  1
                      bdev(0x564072f96000
                      /var/lib/ceph/osd/ceph-46/block) open path
                      /var/lib/ceph/osd/ceph-46/block</div>
                    <div>2018-10-03 10:02:25.504 7f4d54b611c0  1
                      bdev(0x564072f96000
                      /var/lib/ceph/osd/ceph-46/block) open size
                      1000198897664 (0xe8e0800000, 932 GiB) block_size
                      4096 (4 KiB) rotational</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1
                      bluestore(/var/lib/ceph/osd/ceph-46)
                      _set_cache_sizes kv_min_ratio 1 > kv_ratio 0.5</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1
                      bluestore(/var/lib/ceph/osd/ceph-46)
                      _set_cache_sizes cache_size 536870912 meta 0 kv 1
                      data 0</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1 bdev
                      create path /var/lib/ceph/osd/ceph-46/block type
                      kernel</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1
                      bdev(0x564072f96a80
                      /var/lib/ceph/osd/ceph-46/block) open path
                      /var/lib/ceph/osd/ceph-46/block</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1
                      bdev(0x564072f96a80
                      /var/lib/ceph/osd/ceph-46/block) open size
                      1000198897664 (0xe8e0800000, 932 GiB) block_size
                      4096 (4 KiB) rotational</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1 bluefs
                      add_block_device bdev 1 path
                      /var/lib/ceph/osd/ceph-46/block size 932 GiB</div>
                    <div>2018-10-03 10:02:25.505 7f4d54b611c0  1 bluefs
                      mount</div>
                    <div>2018-10-03 10:02:25.620 7f4d54b611c0 -1 bluefs
                      _replay file with link count 0: file(ino 450 size
                      0x169964c mtime 2018-10-02 12:24:22.602432 bdev 1
                      allocated 1700000 extents
[1:0x6fd9500000+100000,1:0x6fd9600000+100000,1:0x6fd9700000+100000,1:0x6fd9800000+100000,1:0x6fd9900000+100000,1:0x6fd9a00000+100000,1:0x6fd9b00000+100000,1:0x6fd9c00000+100000,1:0x6fd9d00000+100000,1:0x6fd9e00000+100000,1:0x6fd9f00000+100000,1:0x6fda000000+100000,1:0x6fda100000+100000,1:0x6fda200000+100000,1:0x6fda300000+100000,1:0x6fda400000+100000,1:0x6fda500000+100000,1:0x6fda600000+100000,1:0x6fda700000+100000,1:0x6fda800000+100000,1:0x6fda900000+100000,1:0x6fdaa00000+100000,1:0x6fdab00000+100000])</div>
                    <div>2018-10-03 10:02:25.620 7f4d54b611c0 -1 bluefs
                      mount failed to replay log: (5) Input/output error</div>
                    <div>2018-10-03 10:02:25.620 7f4d54b611c0  1
                      stupidalloc 0x0x564073102fc0 shutdown</div>
                    <div>2018-10-03 10:02:25.620 7f4d54b611c0 -1
                      bluestore(/var/lib/ceph/osd/ceph-46) _open_db
                      failed bluefs mount: (5) Input/output error</div>
                    <div>2018-10-03 10:02:25.620 7f4d54b611c0  1
                      bdev(0x564072f96a80
                      /var/lib/ceph/osd/ceph-46/block) close</div>
                    <div>2018-10-03 10:02:25.763 7f4d54b611c0  1
                      bdev(0x564072f96000
                      /var/lib/ceph/osd/ceph-46/block) close</div>
                    <div>2018-10-03 10:02:26.010 7f4d54b611c0 -1 osd.46
                      0 OSD:init: unable to mount object store</div>
                    <div>2018-10-03 10:02:26.010 7f4d54b611c0 -1  **
                      ERROR: osd init failed: (5) Input/output error</div>
                  </div>
                  <div><br>
                  </div>
                  <div>We had failing disks in this cluster before but
                    that was easily recovered by out + rebalance.</div>
                  <div>For me, it seems like one disk died (there was
                    large I/O on the cluster when this happened) and
                    took two additional disks with it.</div>
                  <div>It is very strange that this happened about two
                    hours after the upgrade + reboot.</div>
                  <div><br>
                  </div>
                  <div><b>Any recommendations?</b></div>
                  <div><b>I have 8 PGs down, the remeining are active
                      and recovery / rebalance.</b></div>
                  <div><br>
                  </div>
                  <div>Kind regards</div>
                  <div>Kevin</div>
                </div>
              </div>
            </blockquote>
          </div>
        </div>
      </div>
      <br>
      <fieldset class="m_2445726827436418132mimeAttachmentHeader"></fieldset>
      <br>
      </blockquote></div><div text="#000000" bgcolor="#FFFFFF"><blockquote type="cite"><pre>_______________________________________________
ceph-users mailing list
<a class="m_2445726827436418132moz-txt-link-abbreviated" href="mailto:ceph-users@lists.ceph.com" target="_blank">ceph-users@lists.ceph.com</a>
<a class="m_2445726827436418132moz-txt-link-freetext" href="http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com" target="_blank">http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com</a>
</pre>
    </blockquote>
    <br>
  </div>

</blockquote></div></div>-- <br><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><br>Mit freundlichen Grüßen / best regards,<br>Kevin Olbrich.</div></div></div></div>