<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <br>
    <div class="moz-cite-prefix">On 07/21/2017 02:55 PM, yayo (j) wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAGK=3kx=XufQ3_bbmF1+R3g+gqCa-nTUf05KNbA00eBcHM92+g@mail.gmail.com">
      <div dir="ltr">
        <div class="gmail_extra">
          <div class="gmail_quote">2017-07-20 14:48 GMT+02:00
            Ravishankar N <span dir="ltr">&lt;<a
                href="mailto:ravishankar@redhat.com" target="_blank"
                moz-do-not-send="true">ravishankar@redhat.com</a>&gt;</span>:<br>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div text="#000000" bgcolor="#FFFFFF"><span class="">
                  <p><br>
                  </p>
                </span> But it does  say something. All these gfids of
                completed heals in the log below are the for the ones
                that you have given the getfattr output of. So what is
                likely happening is there is an intermittent connection
                problem between your mount and the brick process,
                leading to pending heals again after the heal gets
                completed, which is why the numbers are varying each
                time. You would need to check why that is the case.<br>
                Hope this helps,<br>
                Ravi
                <div>
                  <div class="h5"><br>
                    <br>
                    <blockquote type="cite">
                      <div dir="ltr">
                        <div class="gmail_extra">
                          <div class="gmail_quote">
                            <div><br>
                            </div>
                          </div>
                        </div>
                        <blockquote style="margin:0px 0px 0px
                          40px;border:none;padding:0px">
                          <div class="gmail_extra">
                            <div class="gmail_quote">
                              <div>
                                <div><i>[2017-07-20 09:58:46.573079] I
                                    [MSGID: 108026]
                                    [afr-self-heal-common.c:1254:a<wbr>fr_log_selfheal]
                                    0-engine-replicate-0: Completed data
                                    selfheal on
                                    e6dfd556-340b-4b76-b47b-7b6f5b<wbr>d74327.
                                    sources=[0] 1  sinks=2</i></div>
                              </div>
                            </div>
                          </div>
                          <div class="gmail_extra">
                            <div class="gmail_quote">
                              <div>
                                <div><i>[2017-07-20 09:59:22.995003] I
                                    [MSGID: 108026]
                                    [afr-self-heal-metadata.c:51:_<wbr>_afr_selfheal_metadata_do]
                                    0-engine-replicate-0: performing
                                    metadata selfheal on
                                    f05b9742-2771-484a-85fc-5b6974<wbr>bcef81</i></div>
                              </div>
                            </div>
                          </div>
                          <div class="gmail_extra">
                            <div class="gmail_quote">
                              <div>
                                <div><i>[2017-07-20 09:59:22.999372] I
                                    [MSGID: 108026]
                                    [afr-self-heal-common.c:1254:a<wbr>fr_log_selfheal]
                                    0-engine-replicate-0: Completed
                                    metadata selfheal on
                                    f05b9742-2771-484a-85fc-5b6974<wbr>bcef81.
                                    sources=[0] 1  sinks=2</i></div>
                              </div>
                            </div>
                          </div>
                        </blockquote>
                      </div>
                    </blockquote>
                  </div>
                </div>
              </div>
            </blockquote>
          </div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
          <div class="gmail_signature" data-smartmail="gmail_signature">Hi,</div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
          <div class="gmail_signature" data-smartmail="gmail_signature">But
            we ha1e 2 gluster volume on the same network and the other
            one (the "Data" gluster) don't have any problems. Why you
            think there is a network problem?</div>
        </div>
      </div>
    </blockquote>
    <br>
    Because pending self-heals come into the picture when I/O from the
    clients (mounts) do not succeed on some bricks. They are mostly due
    to <br>
    (a) the client losing connection to some bricks (likely),<br>
    (b) the I/O failing on the bricks themselves (unlikely).<br>
    <br>
    If most of the i/o is also going to the 3rd brick (since you say the
    files are already present on all bricks and I/O is successful) ,
    then it is likely to be (a).<br>
    <br>
    <blockquote type="cite"
cite="mid:CAGK=3kx=XufQ3_bbmF1+R3g+gqCa-nTUf05KNbA00eBcHM92+g@mail.gmail.com">
      <div dir="ltr">
        <div class="gmail_extra">
          <div class="gmail_signature" data-smartmail="gmail_signature"> 
            How to check this on a gluster infrastructure?</div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
        </div>
      </div>
    </blockquote>
    In the fuse mount logs for the engine volume, check if there are any
    messages for brick disconnects. Something along the lines of
    "disconnected from volname-client-x".<br>
    Just guessing here, but maybe even the 'data' volume did experience
    disconnects and self-heals later but you did not observe it when you
    ran heal info. See the glustershd log or mount log for for self-heal
    completion messages on <i> 0-data-replicate-0 </i>also.<br>
    <br>
    Regards,<br>
    Ravi<br>
    <blockquote type="cite"
cite="mid:CAGK=3kx=XufQ3_bbmF1+R3g+gqCa-nTUf05KNbA00eBcHM92+g@mail.gmail.com">
      <div dir="ltr">
        <div class="gmail_extra">
          <div class="gmail_signature" data-smartmail="gmail_signature">Thank
            you</div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
          <div class="gmail_signature" data-smartmail="gmail_signature"><br>
          </div>
        </div>
      </div>
    </blockquote>
    <br>
  </body>
</html>