[Gluster-users] Fwd: gluster heal does not work

Serkan Çoban cobanserkan at gmail.com
Mon Mar 28 13:00:11 UTC 2016


What I understand in below situation is; while fio was writing I
shutdown four servers (IPs ending with 186,187,188,189), so files that
are written to those servers become different from others.But why
there is no entry in server with IP 186? Am I doing something wrong?
Can you explain me your interpretations?



---------- Forwarded message ----------
From: Serkan Çoban <cobanserkan at gmail.com>
Date: Mon, Mar 28, 2016 at 3:15 PM
Subject: gluster heal does not work
To: Gluster Users <gluster-users at gluster.org>


Hi,
I am doing some tests on 78x(16+4) dist-disp volume, gluster 3.7.9. I
rebooted four servers at the same time while fio was writing on the
volume. After servers came online there are still entries in "gluster
volume heal v0 info" output like below:

Brick 1.1.1.185:/bricks/25
/1.1.1.135.t7
/1.1.1.143.t2
/1.1.1.131.t9
/1.1.1.144.t4
/1.1.1.142.t4
/1.1.1.113.t9
/1.1.1.144.t15
/1.1.1.113.t5
/1.1.1.104.t19
/1.1.1.110.t11
/1.1.1.155.t14
/1.1.1.110.t6
/1.1.1.146.t2
/1.1.1.152.t20
/1.1.1.151.t4
/1.1.1.151.t16
/1.1.1.137.t17
Number of entries: 17

Brick 1.1.1.186:/bricks/25
Number of entries: 0

Brick 1.1.1.187:/bricks/25
<gfid:e57ff441-c4d0-48df-b86e-20fd37308d3e>
/1.1.1.131.t9
Number of entries: 2

Brick 1.1.1.188:/bricks/25
<gfid:e57ff441-c4d0-48df-b86e-20fd37308d3e>
/1.1.1.131.t9
Number of entries: 2

Brick 1.1.1.189:/bricks/25
<gfid:e57ff441-c4d0-48df-b86e-20fd37308d3e>
<gfid:a67bd0ca-9f09-4863-9268-874abc18da5b>
Number of entries: 2

Brick 1.1.1.190:/bricks/25
/1.1.1.135.t7
/1.1.1.143.t2
/1.1.1.131.t9
/1.1.1.144.t4
/1.1.1.142.t4
/1.1.1.113.t9
/1.1.1.144.t15
/1.1.1.113.t5
/1.1.1.104.t19
/1.1.1.110.t11
/1.1.1.110.t6
/1.1.1.146.t2
/1.1.1.155.t14
/1.1.1.152.t20
/1.1.1.151.t4
/1.1.1.151.t16
/1.1.1.137.t17
Number of entries: 17

I tried to do heal full but it did not changed anything. How can I
resolve this issue?


More information about the Gluster-users mailing list