[Pacemaker] NFS server shutdown issue

Fri Nov 20 21:58:59 UTC 2009

On Fri, Nov 20, 2009 at 1:29 PM, Andrew Beekhof <andrew at beekhof.net> wrote:

> On Fri, Nov 20, 2009 at 7:12 PM, Judd Tracy <judd at thetracys.net> wrote:
> > It is a weird issue.  Redhat kills the nfs daemons with a kill -2 in
> their
> > init script.  But when pacemaker starts the nfs daemons the only way to
> kill
> > the daemons is a kill -9.  I don't have a clue what pacemaker is doing
> that
> > would cause this behavior.
>
> Pacemaker isn't doing anything, it just calls the scripts you tell it.
>

I know, I have stepped through the script and have not found anything out of
the ordinary.  But I still have the problem that when pacemaker starts the
nfs daemons they cannot be killed without using -9.  So something is going
on that I cannot figure out.  I was just hoping that someone else has seen
this issue before and figured out what was going on.

>
> >
> > As far as the mysql order constraint since the file system is in the
> group
> > mysql won't it be brought after the drbd master is promoted?  And don't
> the
> > members of the group get started in order?  I am a newbie so please
> correct
> > me if I am wrong.
>
> Oh, I may have missed the group part.
>
> >
> > Judd
> >
> > On Fri, Nov 20, 2009 at 2:29 AM, Andrew Beekhof <andrew at beekhof.net>
> wrote:
> >>
> >> On Thu, Nov 19, 2009 at 10:24 PM, Judd Tracy <judd at thetracys.net>
> wrote:
> >> > I am trying to setup a drbd/nfs server in pacemaker on RHEL5 and am
> >> > experiencing some wierd issues the the server is started using
> >> > pacemaker.
> >> > Pacemaker starts the daemons just fine, but when it tries to shutdown
> it
> >> > cannot.  It calls /etc/init.d/nfs to shutdown the daemons, but they do
> >> > not
> >> > respond.
> >>
> >> I guess the scripts have a problem then.
> >> Or perhaps you need some more ordering constraints so that the
> >> services talking to nfs are shut down first.
> >>
> >> > When I start the daemons mysql using the same script I am able to
> >> > kill them using the init.d script.  The logs show success trying to
> >> > shutdown
> >> > the daemons even though they do not.  I was wondering if anyone else
> has
> >> > seen this issue?
> >>
> >> btw. Shouldn't
> >>   order mysql_after_drbd_mysql inf: ms_drbd_mysql:promote mysql:start
> >> be between mysql and the filesystem?
> >>
> >> >
> >> >
> >> > Judd
> >> >
> >> > Configuration:
> >> >
> >> > node filer1
> >> > node filer2
> >> > node mysql1
> >> > node mysql2
> >> > primitive drbd_mysql ocf:linbit:drbd \
> >> >         params drbd_resource="mysql" \
> >> >         op monitor interval="15s"
> >> > primitive fs1_drbd ocf:linbit:drbd \
> >> >         params drbd_resource="fs1"
> >> > primitive fs1_lvm ocf:heartbeat:LVM \
> >> >         params volgrpname="data_vg"
> >> > primitive fs1_nfs ocf:heartbeat:nfsserver \
> >> >         params nfs_init_script="/etc/init.d/nfs"
> >> > nfs_notify_cmd="/sbin/rpc.statd" nfs_shared_infodir="/var/lib/nfs/"
> >> > nfs_ip="fs1"
> >> > primitive fs1_nfs_recovery_fs ocf:heartbeat:Filesystem \
> >> >         params device="/dev/data_vg/v4recovery"
> >> > directory="/var/lib/nfs/v4recovery/" fstype="ext3"
> >> > primitive mysql_daemon ocf:heartbeat:mysql \
> >> >         params binary="/usr/bin/mysqld_safe"
> >> > pid="/var/run/mysqld/mysqld.pid"
> >> > primitive mysql_fs ocf:heartbeat:Filesystem \
> >> >         params device="/dev/drbd/by-res/mysql"
> >> > directory="/var/lib/mysql"
> >> > fstype="ext3"
> >> > primitive mysql_ip ocf:heartbeat:IPaddr2 \
> >> >         params ip="mysql" nic="eth0:0"
> >> > group fs1 fs1_lvm fs1_nfs_recovery_fs fs1_nfs \
> >> >         params target_role="stopped"
> >> > group mysql mysql_fs mysql_ip mysql_daemon
> >> > ms ms_drbd_mysql drbd_mysql \
> >> >         meta master-max="1" master-node-max="1" clone-max="2"
> >> > clone-node-max="1" notify="true" globally-unique="false"
> >> > ms ms_fs1_drbd fs1_drbd \
> >> >         meta master-max="1" master-node-max="1" clone-max="2"
> >> > clone-node-max="1" notify="true" globally-unique="false"
> >> > location drbd_mysql_on_mysql1 ms_drbd_mysql 100000: mysql1
> >> > location drbd_mysql_on_mysql2 ms_drbd_mysql 100000: mysql2
> >> > location fs1_drbd_on_filer1 ms_fs1_drbd 100000: filer1
> >> > location fs1_drbd_on_filer2 ms_fs1_drbd 100000: filer2
> >> > location fs1_on_filer1 fs1 100000: filer1
> >> > location fs1_on_filer2 fs1 100000: filer2
> >> > location mysql_on_mysql1 mysql 100000: mysql1
> >> > location mysql_on_mysql2 mysql 100000: mysql2
> >> > colocation fs1_on_fs1_drbd inf: fs1 ms_fs1_drbd:Master
> >> > colocation mysql_on_drbd_mysql inf: mysql ms_drbd_mysql:Master
> >> > order fs1_after_fs1_drbd inf: ms_fs1_drbd:promote fs1:start
> >> > order mysql_after_drbd_mysql inf: ms_drbd_mysql:promote mysql:start
> >> > property $id="cib-bootstrap-options" \
> >> >         dc-version="1.0.5-462f1569a43740667daf7b0f6b521742e9eb8fa7" \
> >> >         cluster-infrastructure="openais" \
> >> >         expected-quorum-votes="4" \
> >> >         symmetric-cluster="false" \
> >> >         stonith-enabled="false" \
> >> >         last-lrm-refresh="1258664854"
> >> >
> >> > _______________________________________________
> >> > Pacemaker mailing list
> >> > Pacemaker at oss.clusterlabs.org
> >> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >> >
> >> >
> >>
> >> _______________________________________________
> >> Pacemaker mailing list
> >> Pacemaker at oss.clusterlabs.org
> >> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >
> >
> > _______________________________________________
> > Pacemaker mailing list
> > Pacemaker at oss.clusterlabs.org
> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >
> >
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091120/a8721e89/attachment-0002.htm>