<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">On 07/04/2018 09:20 PM, Anh Vo wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAPUsBJ1RRwvmpfb_60WT-yD6prBkQyCs3fRnFJXsyxSqDANsXQ@mail.gmail.com">
<div dir="ltr">I forgot to mention we're using 3.12.10</div>
<div class="gmail_extra"><br>
<div class="gmail_quote">On Wed, Jul 4, 2018 at 8:45 AM, Anh Vo
<span dir="ltr"><<a href="mailto:vtqanh@gmail.com"
target="_blank" moz-do-not-send="true">vtqanh@gmail.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr">If I run "sudo gluster volume heal gv0
split-brain latest-mtime /" I get the following:
<div>
<div><br>
</div>
<div>Lookup failed on /:Invalid argument.</div>
<div>Volume heal failed.</div>
</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
<br>
Can you share the glfsheal-<volname>.log on the node where you
ran this failed command?<br>
<blockquote type="cite"
cite="mid:CAPUsBJ1RRwvmpfb_60WT-yD6prBkQyCs3fRnFJXsyxSqDANsXQ@mail.gmail.com">
<div class="gmail_extra">
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr">
<div>
<div><br>
</div>
</div>
<div>node2 was not connected at that time, because if we
connect it to the system after a few minutes gluster
will become almost unusable and we have many jobs
failing. This morning I reconnected it and ran heal info
and we have about 30000 entries to heal (15K from
gfs-vm000 and 15k from gfs-vm001, 80% are all gfid, 20%
have file names). It's not feasible for us to check the
individual gfid so we kinda rely on gluster self heal to
handle those gfid. The "/" is a concern because it
prevents us from mounting nfs. We do need to mount nfs
for some of our management because gluster fuse mount is
much slower compared to nfs when it comes to recursive
operations like 'du' </div>
<div><br>
</div>
<div>Do you have any suggestion for healing the metadata
on '/' ?</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
You can manually delete the afr xattrs on node 3 as a workaround:<br>
<tt>setfattr -x trusted.afr.gv0-client-0 gluster/brick/brick0</tt><tt><br>
</tt><tt>setfattr -x trusted.afr.gv0-client-1 gluster/brick/brick0</tt><br>
<br>
This should remove the split-brain on root.<br>
<br>
HTH,<br>
Ravi<br>
<blockquote type="cite"
cite="mid:CAPUsBJ1RRwvmpfb_60WT-yD6prBkQyCs3fRnFJXsyxSqDANsXQ@mail.gmail.com">
<div class="gmail_extra">
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr">
<div><br>
</div>
<div>Thanks</div>
<span class="HOEnZb"><font color="#888888">
<div>Anh</div>
</font></span></div>
<div class="HOEnZb">
<div class="h5">
<div class="gmail_extra"><br>
<div class="gmail_quote">On Tue, Jul 3, 2018 at 8:02
PM, Ravishankar N <span dir="ltr"><<a
href="mailto:ravishankar@redhat.com"
target="_blank" moz-do-not-send="true">ravishankar@redhat.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<p>Hi,</p>
<p>What version of gluster are you using?</p>
<p>1. The afr xattrs on '/' indicate a meta-data
split-brain. You can resolve it using one of
the policies listed in <a
class="m_-2457352565565741707m_3342788517930548362moz-txt-link-freetext"
href="https://docs.gluster.org/en/latest/Troubleshooting/resolving-splitbrain/"
target="_blank" moz-do-not-send="true">https://docs.gluster.org/en/la<wbr>test/Troubleshooting/resolving<wbr>-splitbrain/</a></p>
<p>For example, "<code>gluster volume heal gv0
split-brain latest-mtime / "<br>
</code></p>
2. Is the file corresponding to the other gfid
(81289110-867b-42ff-ba3b-1373a<wbr>187032b)
present in all bricks? What do the getfattr
outputs for this file indicate?<br>
<br>
3. As for the discrepancy in output of heal
info, is node2 connected to the other nodes?
Does heal info still print the details of all 3
bricks when you run it on node2 ?<br>
-Ravi
<div>
<div class="m_-2457352565565741707h5"><br>
<br>
<div
class="m_-2457352565565741707m_3342788517930548362moz-cite-prefix">On
07/04/2018 01:47 AM, Anh Vo wrote:<br>
</div>
<blockquote type="cite">Actually we just
discovered that the heal info command was
returning different things when executed
on the different nodes of our 3-replica
setup.
<div>When we execute it on node2 we did
not see the split brain reported "/" but
if I execute it on node0 and node1 I am
seeing:</div>
<div><br>
<div>x@gfs-vm001:~$ sudo gluster volume
heal gv0 info | tee heal-info</div>
<div>Brick
gfs-vm000:/gluster/brick/brick<wbr>0</div>
<div><gfid:81289110-867b-42ff-ba3b-<wbr>1373a187032b></div>
<div>/ - Is in split-brain</div>
<div><br>
</div>
<div>Status: Connected</div>
<div>Number of entries: 2</div>
<div><br>
</div>
<div>Brick
gfs-vm001:/gluster/brick/brick<wbr>0</div>
<div>/ - Is in split-brain</div>
<div><br>
</div>
<div><gfid:81289110-867b-42ff-ba3b-<wbr>1373a187032b></div>
<div>Status: Connected</div>
<div>Number of entries: 2</div>
<div><br>
</div>
<div>Brick
gfs-vm002:/gluster/brick/brick<wbr>0</div>
<div>/ - Is in split-brain</div>
<div><br>
</div>
<div>Status: Connected</div>
<div>Number of entries: 1</div>
</div>
<div><br>
</div>
<div><br>
</div>
<div>I ran getfattr -d -m . -e hex
/gluster/brick/brick0 on all three nodes
and I am seeing node2 has slightly
different attr:</div>
<div>node0:</div>
<div>
<div>sudo getfattr -d -m . -e hex
/gluster/brick/brick0</div>
<div>getfattr: Removing leading '/' from
absolute path names</div>
<div># file: gluster/brick/brick0</div>
<div>trusted.afr.gv0-client-2=0x000<wbr>000000000000100000000</div>
<div>trusted.gfid=0x000000000000000<wbr>00000000000000001</div>
<div>trusted.glusterfs.dht=0x000000<wbr>010000000000000000ffffffff</div>
<div>trusted.glusterfs.volume-id=0x<wbr>7fa3aac372d543f987ed0c66b77f02<wbr>e2</div>
</div>
<div><br>
</div>
<div>node1:</div>
<div>
<div>sudo getfattr -d -m . -e hex
/gluster/brick/brick0</div>
<div>getfattr: Removing leading '/' from
absolute path names</div>
<div># file: gluster/brick/brick0</div>
<div>trusted.afr.gv0-client-2=0x000<wbr>000000000000100000000</div>
<div>trusted.gfid=0x000000000000000<wbr>00000000000000001</div>
<div>trusted.glusterfs.dht=0x000000<wbr>010000000000000000ffffffff</div>
<div>trusted.glusterfs.volume-id=0x<wbr>7fa3aac372d543f987ed0c66b77f02<wbr>e2</div>
</div>
<div><br>
</div>
<div>node2:</div>
<div>
<div>sudo getfattr -d -m . -e hex
/gluster/brick/brick0</div>
<div>getfattr: Removing leading '/' from
absolute path names</div>
<div># file: gluster/brick/brick0</div>
<div>trusted.afr.dirty=0x0000000000<wbr>00000000000000</div>
<div>trusted.afr.gv0-client-0=0x000<wbr>000000000000200000000</div>
<div>trusted.afr.gv0-client-1=0x000<wbr>000000000000200000000</div>
<div>trusted.afr.gv0-client-2=0x000<wbr>000000000000000000000</div>
<div>trusted.gfid=0x000000000000000<wbr>00000000000000001</div>
<div>trusted.glusterfs.dht=0x000000<wbr>010000000000000000ffffffff</div>
<div>trusted.glusterfs.volume-id=0x<wbr>7fa3aac372d543f987ed0c66b77f02<wbr>e2</div>
</div>
<div><br>
</div>
<div>Where do I go from here? Thanks</div>
</blockquote>
<br>
</div>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</blockquote>
<br>
</body>
</html>