<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>&lt;BUMP&gt; so I can get some sort of resolution on the issue
      (i.e. is it hardware, Gluster etc)<br>
    </p>
    <p>I guess what I really need to know is <br>
    </p>
    <p>1) Node 2 complains that it cant reach node 1 and node 3.  If
      this was an OS/Hardware networking issue and not internal to
      Gluster , then why didn't node1 and node3 have error message
      complaining about not reaching node2 <br>
    </p>
    <div class="moz-forward-container">2) how significant is it that the
      node was running 6.5 while node 1 and node 2 were running 6.4</div>
    <div class="moz-forward-container"><br>
    </div>
    <div class="moz-forward-container">-wk<br>
    </div>
    <div class="moz-forward-container"><br>
      -------- Forwarded Message --------
      <table class="moz-email-headers-table" cellspacing="0"
        cellpadding="0" border="0">
        <tbody>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">Subject:
            </th>
            <td>VM freeze issue on simple gluster setup.</td>
          </tr>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">Date: </th>
            <td>Thu, 5 Dec 2019 16:23:35 -0800</td>
          </tr>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">From: </th>
            <td>WK <a class="moz-txt-link-rfc2396E" href="mailto:wkmail@bneit.com">&lt;wkmail@bneit.com&gt;</a></td>
          </tr>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">To: </th>
            <td>Gluster Users <a class="moz-txt-link-rfc2396E" href="mailto:gluster-users@gluster.org">&lt;gluster-users@gluster.org&gt;</a></td>
          </tr>
        </tbody>
      </table>
      <br>
      <br>
      I have a replica2+arbiter setup that is used for VMs.<br>
      <br>
      ip #.1 is the arb<br>
      <br>
      ip #.2 and #.3 are the kvm hosts.<br>
      <br>
      Two Volumes are involved and its gluster 6.5/Ubuntu 18.4/fuse The
      Gluster networking uses a  two ethernet card teamd/round-robin
      setup which *should* have stayed up if one of the ports had
      failed.<br>
      <br>
      I just had a number of VMs go Read-Only due to the below
      communication failure at 22:00 but only on kvm host  #2<br>
      <br>
      VMs on the same gluster volumes on kvm host 3 were unaffected.<br>
      <br>
      The logs on host #2 show the following:<br>
      <br>
      [2019-12-05 22:00:43.739804] C
      [rpc-clnt-ping.c:155:rpc_clnt_ping_timer_expired]
      0-GL1image-client-2: server 10.255.1.1:49153 has not responded in
      the last 21 seconds, disconnecting.<br>
      [2019-12-05 22:00:43.757095] C
      [rpc-clnt-ping.c:155:rpc_clnt_ping_timer_expired]
      0-GL1image-client-1: server 10.255.1.3:49152 has not responded in
      the last 21 seconds, disconnecting.<br>
      [2019-12-05 22:00:43.757191] I [MSGID: 114018]
      [client.c:2323:client_rpc_notify] 0-GL1image-client-2:
      disconnected from GL1image-client-2. Client process will keep
      trying to connect to glusterd until brick's port is available<br>
      [2019-12-05 22:00:43.757246] I [MSGID: 114018]
      [client.c:2323:client_rpc_notify] 0-GL1image-client-1:
      disconnected from GL1image-client-1. Client process will keep
      trying to connect to glusterd until brick's port is available<br>
      [2019-12-05 22:00:43.757266] W [MSGID: 108001]
      [afr-common.c:5608:afr_notify] 0-GL1image-replicate-0:
      Client-quorum is not met<br>
      [2019-12-05 22:00:43.790639] E
      [rpc-clnt.c:346:saved_frames_unwind] (--&gt;
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(_gf_log_callingfn+0x139)[0x7f030d045f59]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xcbb0)[0x7f030cdf0bb0]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xccce)[0x7f030cdf0cce]
      (--&gt;
/usr/lib/x86_64-linux-gnu/libgfrpc.so.0(rpc_clnt_connection_cleanup+0x95)[0x7f030cdf1c45]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xe890)[0x7f030cdf2890]
      ))))) 0-GL1image-client-2: forced unwinding frame type(GlusterFS
      4.x v1) op(FXATTROP(34)) called at 2019-12-05 22:00:19.736456
      (xid=0x825bffb)<br>
      [2019-12-05 22:00:43.790655] W [MSGID: 114031]
      [client-rpc-fops_v2.c:1614:client4_0_fxattrop_cbk]
      0-GL1image-client-2: remote operation failed<br>
      [2019-12-05 22:00:43.790686] E
      [rpc-clnt.c:346:saved_frames_unwind] (--&gt;
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(_gf_log_callingfn+0x139)[0x7f030d045f59]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xcbb0)[0x7f030cdf0bb0]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xccce)[0x7f030cdf0cce]
      (--&gt;
/usr/lib/x86_64-linux-gnu/libgfrpc.so.0(rpc_clnt_connection_cleanup+0x95)[0x7f030cdf1c45]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xe890)[0x7f030cdf2890]
      ))))) 0-GL1image-client-1: forced unwinding frame type(GlusterFS
      4.x v1) op(FXATTROP(34)) called at 2019-12-05 22:00:19.736428
      (xid=0x89fee01)<br>
      [2019-12-05 22:00:43.790703] W [MSGID: 114031]
      [client-rpc-fops_v2.c:1614:client4_0_fxattrop_cbk]
      0-GL1image-client-1: remote operation failed<br>
      [2019-12-05 22:00:43.790774] E [MSGID: 114031]
      [client-rpc-fops_v2.c:1393:client4_0_finodelk_cbk]
      0-GL1image-client-1: remote operation failed [Transport endpoint
      is not connected]<br>
      [2019-12-05 22:00:43.790777] E
      [rpc-clnt.c:346:saved_frames_unwind] (--&gt;
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(_gf_log_callingfn+0x139)[0x7f030d045f59]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xcbb0)[0x7f030cdf0bb0]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xccce)[0x7f030cdf0cce]
      (--&gt;
/usr/lib/x86_64-linux-gnu/libgfrpc.so.0(rpc_clnt_connection_cleanup+0x95)[0x7f030cdf1c45]
      (--&gt;
      /usr/lib/x86_64-linux-gnu/libgfrpc.so.0(+0xe890)[0x7f030cdf2890]
      ))))) 0-GL1image-client-2: forced unwinding frame type(GlusterFS
      4.x v1) op(FXATTROP(34)) called at 2019-12-05 22:00:19.736542
      (xid=0x825bffc)<br>
      [2019-12-05 22:00:43.790794] W [MSGID: 114029]
      [client-rpc-fops_v2.c:4873:client4_0_finodelk]
      0-GL1image-client-1: failed to send the fop<br>
      [2019-12-05 22:00:43.790806] W [MSGID: 114031]
      [client-rpc-fops_v2.c:1614:client4_0_fxattrop_cbk]
      0-GL1image-client-2: remote operation failed<br>
      [2019-12-05 22:00:43.790825] E [MSGID: 114031]
      [client-rpc-fops_v2.c:1393:client4_0_finodelk_cbk]
      0-GL1image-client-2: remote operation failed [Transport endpoint
      is not connected]<br>
      [2019-12-05 22:00:43.790842] W [MSGID: 114029]
      [client-rpc-fops_v2.c:4873:client4_0_finodelk]
      0-GL1image-client-2: failed to send the fop<br>
      <br>
      the fop/transport not connected errors just repeat for another 50
      lines or so until I hit 22:00:46 seconds at which point the
      Volumes appear to be fine (though the VMs were still read-only
      until I rebooted.<br>
      <br>
      [2019-12-05 22:00:46.987242] W [fuse-bridge.c:2827:fuse_readv_cbk]
      0-glusterfs-fuse: 91701328: READ =&gt; -1
      gfid=d883b7c4-97f5-4f12-9373-7987cfc7dee4 fd=0x7f02f005b708
      (Transport endpoint is not connected)<br>
      [2019-12-05 22:00:47.029947] W [fuse-bridge.c:2827:fuse_readv_cbk]
      0-glusterfs-fuse: 91701329: READ =&gt; -1
      gfid=d883b7c4-97f5-4f12-9373-7987cfc7dee4 fd=0x7f02f005b708
      (Transport endpoint is not connected)<br>
      [2019-12-05 22:00:49.901075] W [fuse-bridge.c:2827:fuse_readv_cbk]
      0-glusterfs-fuse: 91701330: READ =&gt; -1
      gfid=c342dba6-a2a2-49a8-be3f-cd320e90c956 fd=0x7f02f002bee8
      (Transport endpoint is not connected)<br>
      [2019-12-05 22:00:49.923525] W [fuse-bridge.c:2827:fuse_readv_cbk]
      0-glusterfs-fuse: 91701331: READ =&gt; -1
      gfid=c342dba6-a2a2-49a8-be3f-cd320e90c956 fd=0x7f02f002bee8
      (Transport endpoint is not connected)<br>
      [2019-12-05 22:00:49.970219] W [fuse-bridge.c:2827:fuse_readv_cbk]
      0-glusterfs-fuse: 91701332: READ =&gt; -1
      gfid=fcec6b7a-ad23-4449-aa09-107e113877a1 fd=0x7f02f008dd58
      (Transport endpoint is not connected)<br>
      [2019-12-05 22:00:50.023932] W [fuse-bridge.c:2827:fuse_readv_cbk]
      0-glusterfs-fuse: 91701333: READ =&gt; -1
      gfid=fcec6b7a-ad23-4449-aa09-107e113877a1 fd=0x7f02f008dd58
      (Transport endpoint is not connected)<br>
      [2019-12-05 22:00:54.807833] I [rpc-clnt.c:2028:rpc_clnt_reconfig]
      0-GL1image-client-2: changing port to 49153 (from 0)<br>
      [2019-12-05 22:00:54.808043] I [rpc-clnt.c:2028:rpc_clnt_reconfig]
      0-GL1image-client-1: changing port to 49152 (from 0)<br>
      [2019-12-05 22:00:46.115076] E [MSGID: 133014]
      [shard.c:1799:shard_common_stat_cbk] 0-GL1image-shard: stat
      failed: 7a5959d6-75fc-411d-8831-57a744776ed3 [Transport endpoint
      is not connected]<br>
      [2019-12-05 22:00:54.820394] I [MSGID: 114046]
      [client-handshake.c:1106:client_setvolume_cbk]
      0-GL1image-client-1: Connected to GL1image-client-1, attached to
      remote volume '/GLUSTER/GL1image'.<br>
      [2019-12-05 22:00:54.820447] I [MSGID: 114042]
      [client-handshake.c:930:client_post_handshake]
      0-GL1image-client-1: 10 fds open - Delaying child_up until they
      are re-opened<br>
      [2019-12-05 22:00:54.820549] I [MSGID: 114046]
      [client-handshake.c:1106:client_setvolume_cbk]
      0-GL1image-client-2: Connected to GL1image-client-2, attached to
      remote volume '/GLUSTER/GL1image'.<br>
      [2019-12-05 22:00:54.820568] I [MSGID: 114042]
      [client-handshake.c:930:client_post_handshake]
      0-GL1image-client-2: 10 fds open - Delaying child_up until they
      are re-opened<br>
      [2019-12-05 22:00:54.821381] I [MSGID: 114041]
      [client-handshake.c:318:client_child_up_reopen_done]
      0-GL1image-client-1: last fd open'd/lock-self-heal'd - notifying
      CHILD-UP<br>
      [2019-12-05 22:00:54.821406] I [MSGID: 108002]
      [afr-common.c:5602:afr_notify] 0-GL1image-replicate-0:
      Client-quorum is met<br>
      [2019-12-05 22:00:54.821446] I [MSGID: 114041]
      [client-handshake.c:318:client_child_up_reopen_done]
      0-GL1image-client-2: last fd open'd/lock-self-heal'd - notifying
      CHILD-UP<br>
      <br>
      What is odd is that the gluster logs on the #3 and #1 show
      absolutely ZERO gluster errors around that time nor do I show any
      Network/teamd errors on any of the  3 nodes (including the problem
      node #2)<br>
      <br>
      I've checked dmesg/syslog and every other log file on the box.<br>
      <br>
      According to a staff member, we had this same kvm host have the
      same problem about 3 weeks ago, it was written up as a fluke
      possible due to excess disk I/O, since we have been using gluster
      for years and rarely have seen issues, especially with very basic
      gluster usage.<br>
      <br>
      In this case those VMs weren't overly busy and now we have a
      repeat problem.<br>
      <br>
      So I am wondering where else I can look to diagnose the problem or
      should I abandon the hardware/setup?<br>
      <br>
      I assume its a networking issue and not on gluster, but I am
      confused why gluster nodes #1 and #3 didn't complain about not
      seeing #2? If the networking did drop out should they have
      noticed?<br>
      <br>
      There also doesn't appear to be any visible hard disk issues
      (smartd is running)<br>
      <br>
      Side Note: I have reset the tcp-timeout back to 42 seconds and
      will look at upgrading to 6.6. I also see that the ARB and the
      unaffected Gluster node were running Gluster 6.4 (I don't know why
      #2 is on 6.5 but I am checking on that as well, we turn off
      auto-upgrade)<br>
      <br>
      Maybe the mismatched versions are the culprit?<br>
      <br>
      Also, we have a large of these replica 2+1 gluster setups running
      gluster version from 5.x up and none of the others have had this
      issue<br>
      <br>
      Any advise would be appreciated.<br>
      <br>
      Sincerely,<br>
      <br>
      Wk<br>
      <br>
      <br>
      <br>
      <br>
      <br>
    </div>
  </body>
</html>