<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Folks,<br>
<br>
I have having trouble starting my DRBD+OCFS2 filesystem. It seems to
be a timing thing, with the filesystem trying to come up before DRBD
has gotten the second node of the cluster into Primary mode. I find
this in the logs:<br>
<blockquote>Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) FATAL: Module scsi_hostadapter not
found.<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) blockdev: <br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) cannot open /dev/drbd/by-res/share<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) : <br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) Wrong medium type<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) mount.ocfs2<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) : <br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) I/O error on channel<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) <br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) while opening device /dev/drbd1<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: RA output:
(p_fs_share:1:start:stderr) <br>
Dec 4 15:50:05 aztestc4 Filesystem[1631]: ERROR: Couldn't mount
filesystem /dev/drbd/by-res/share on /share<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: WARN: Managed
p_fs_share:1:start process 1631 exited with return code 1.<br>
Dec 4 15:50:05 aztestc4 lrmd: [1177]: info: operation start[15]
on p_fs_share:1 for client 1180: pid 1631 exited with return code
1<br>
Dec 4 15:50:05 aztestc4 crmd: [1180]: debug:
create_operation_update: do_update_resource: Updating resouce
p_fs_share:1 after complete start op (interval=0)<br>
Dec 4 15:50:05 aztestc4 crmd: [1180]: info: process_lrm_event:
LRM operation p_fs_share:1_start_0 (call=15, rc=1, cib-update=18,
confirmed=true) unknown error<br>
</blockquote>
If I simply wait a little while (maybe a minute, maybe less) and
then "crm resource cleanup cl_fs_share", the filesystem starts
properly on both nodes. Here are the pertinent parts of my
configuration:<br>
<blockquote>primitive p_drbd_share ocf:linbit:drbd \<br>
params drbd_resource="share" \<br>
op monitor interval="15s" role="Master" timeout="20s" \<br>
op monitor interval="20s" role="Slave" timeout="20s" \<br>
op start interval="0" timeout="240s" \<br>
op stop interval="0" timeout="100s"<br>
primitive p_fs_share ocf:heartbeat:Filesystem \<br>
params device="/dev/drbd/by-res/share" directory="/share"
fstype="ocfs2" options="rw,noatime" \<br>
op start interval="0" timeout="60" \<br>
op stop interval="0" timeout="60" \<br>
op monitor interval="20" timeout="40"<br>
primitive p_o2cb ocf:pacemaker:o2cb \<br>
params stack="cman" \<br>
op start interval="0" timeout="90" \<br>
op stop interval="0" timeout="100" \<br>
op monitor interval="10" timeout="20"<br>
ms ms_drbd_share p_drbd_share \<br>
meta master-max="2" notify="true" interleave="true"
clone-max="2" is-managed="true" target-role="Started"<br>
clone cl_fs_share p_fs_share \<br>
meta interleave="true" notify="true" globally-unique="false"
target-role="Started"<br>
clone cl_o2cb p_o2cb \<br>
meta interleave="true" globally-unique="false"<br>
order o_ocfs2 inf: ms_drbd_share:promote cl_o2cb<br>
order o_share inf: cl_o2cb cl_fs_share<br>
</blockquote>
Should I increase the timeout value in <br>
<blockquote>primitive p_fs_share ocf:heartbeat:Filesystem \<br>
... \<br>
op start interval="0" timeout="60"<br>
</blockquote>
to take care of this? I am dubious because I think cl_o2cb is
starting, which allows cl_fs_share to start, before ms_drbd_share is
done promote-ing.<br>
<br>
Thanks,<br>
-- Art Z.<br>
<br>
<div class="moz-signature">-- <br>
<p style="font-style: italic;">
Art Zemon, President<br>
<a href="http://www.hens-teeth.net/">Hen's Teeth Network</a> for
reliable web hosting and programming<br>
(866)HENS-NET / (636)447-3030 ext. 200 / <a class="moz-txt-link-abbreviated" href="http://www.hens-teeth.net">www.hens-teeth.net</a></p>
</div>
</body>
</html>