<div dir="ltr"><div><div><div>Niels,<br><br></div>Allesandro's configuration does not have shard enabled. So it has definitely not got anything to do with shard not supporting seek fop.<br></div><br></div><div>Copy-pasting volume-info output from the first mail:<br><br>Volume Name: datastore2<br>
Type: Replicate<br>
Volume ID: c95ebb5f-6e04-4f09-91b9-<wbr>bbbe63d83aea<br>
Status: Started<br>
Snapshot Count: 0<br>
Number of Bricks: 1 x (2 + 1) = 3<br>
Transport-type: tcp<br>
Bricks:<br>
Brick1: srvpve2g:/data/brick2/brick<br>
Brick2: srvpve3g:/data/brick2/brick<br>
Brick3: srvpve1g:/data/brick2/brick (arbiter)<br>
Options Reconfigured:<br>
nfs.disable: on<br>
performance.readdir-ahead: on<br>
transport.address-family: inet<br><br></div><div><br></div>-Krutika<br><br></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, May 9, 2017 at 7:40 PM, Niels de Vos <span dir="ltr"><<a href="mailto:ndevos@redhat.com" target="_blank">ndevos@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">...<br>
<span class="">> > client from<br>
> > srvpve2-162483-2017/05/08-10:<wbr>01:06:189720-datastore2-<wbr>client-0-0-0<br>
> > (version: 3.8.11)<br>
> > [2017-05-08 10:01:06.237433] E [MSGID: 113107] [posix.c:1079:posix_seek]<br>
> > 0-datastore2-posix: seek failed on fd 18 length 42957209600 [No such<br>
> > device or address]<br>
<br>
</span>The SEEK procedure translates to lseek() in the posix xlator. This can<br>
return with "No suck device or address" (ENXIO) in only one case:<br>
<br>
ENXIO whence is SEEK_DATA or SEEK_HOLE, and the file offset is<br>
beyond the end of the file.<br>
<br>
This means that an lseek() was executed where the current offset of the<br>
filedescriptor was higher than the size of the file. I'm not sure how<br>
that could happen... Sharding prevents using SEEK at all atm.<br>
<br>
...<br>
<span class="">> > The strange part is that I cannot seem to find any other error.<br>
> > If I restart the VM everything works as expected (it stopped at ~9.51<br>
> > UTC and was started at ~10.01 UTC) .<br>
> ><br>
> > This is not the first time that this happened, and I do not see any<br>
> > problems with networking or the hosts.<br>
> ><br>
> > Gluster version is 3.8.11<br>
> > this is the incriminated volume (though it happened on a different one too)<br>
> ><br>
> > Volume Name: datastore2<br>
> > Type: Replicate<br>
> > Volume ID: c95ebb5f-6e04-4f09-91b9-<wbr>bbbe63d83aea<br>
> > Status: Started<br>
> > Snapshot Count: 0<br>
> > Number of Bricks: 1 x (2 + 1) = 3<br>
> > Transport-type: tcp<br>
> > Bricks:<br>
> > Brick1: srvpve2g:/data/brick2/brick<br>
> > Brick2: srvpve3g:/data/brick2/brick<br>
> > Brick3: srvpve1g:/data/brick2/brick (arbiter)<br>
> > Options Reconfigured:<br>
> > nfs.disable: on<br>
> > performance.readdir-ahead: on<br>
> > transport.address-family: inet<br>
> ><br>
> > Any hint on how to dig more deeply into the reason would be greatly<br>
> > appreciated.<br>
<br>
</span>Probably the problem is with SEEK support in the arbiter functionality.<br>
Just like with a READ or a WRITE on the arbiter brick, SEEK can only<br>
succeed on bricks where the files with content are located. It does not<br>
look like arbiter handles SEEK, so the offset in lseek() will likely be<br>
higher than the size of the file on the brick (empty, 0 size file). I<br>
don't know how the replication xlator responds on an error return from<br>
SEEK on one of the bricks, but I doubt it likes it.<br>
<br>
We have <a href="https://bugzilla.redhat.com/show_bug.cgi?id=1301647" rel="noreferrer" target="_blank">https://bugzilla.redhat.com/<wbr>show_bug.cgi?id=1301647</a> to support<br>
SEEK for sharding. I suggest you open a bug for getting SEEK in the<br>
arbiter xlator as well.<br>
<br>
HTH,<br>
Niels<br>
</blockquote></div><br></div>