[Pacemaker] GFS resource configuration

[Andrew Beekhof]

On Sat, Feb 26, 2011 at 12:48 PM, Jonathan Gowar <jon at whiteheat.org.uk> wrote:
> Hi,
>
>  I'm not able to configure the GFS resource, as detailed in Cluster From
> Scratch (Beekhof), so I'm running in debug mode.
>
> root at squeeze:~# sh -x /usr/lib/ocf/resource.d/pacemaker/controld start ;
> echo $?
> + . /usr/lib/ocf/resource.d/heartbeat/.ocf-shellfuncs
> + unset LC_ALL
> + export LC_ALL
> + unset LANGUAGE
> + export LANGUAGE
> + basename /usr/lib/ocf/resource.d/pacemaker/controld
> + __SCRIPT_NAME=controld
> + [ -z /usr/lib/ocf ]
> + : /usr/lib/ocf/resource.d/heartbeat
> + . /usr/lib/ocf/resource.d/heartbeat/.ocf-binaries
> +
> PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/sbin:/bin:/usr/sbin:/usr/bin
> +
> PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/ucb
> + export PATH
> + : mawk
> + : /bin/grep -E
> + : -a
> + : mail
> + : /bin/ping
> + : /bin/bash
> + : /usr/bin/test
> + : /usr/bin/test
> + : basename
> + : blockdev
> + : cat
> + : fsck
> + : fuser
> + : getent
> + : grep
> + : ifconfig
> + : iptables
> + : ip
> + : mdadm
> + : modprobe
> + : mount
> + : msgfmt
> + : netstat
> + : perl
> + : python
> + : raidstart
> + : raidstop
> + : route
> + : umount
> + : reboot
> + : poweroff
> + : wget
> + : whoami
> + : strings
> + : scp
> + : ssh
> + : swig
> + : gzip
> + : tar
> + : md5
> + : drbdadm
> + : drbdsetup
> + . /usr/lib/ocf/resource.d/heartbeat/.ocf-returncodes
> + OCF_SUCCESS=0
> + OCF_ERR_GENERIC=1
> + OCF_ERR_ARGS=2
> + OCF_ERR_UNIMPLEMENTED=3
> + OCF_ERR_PERM=4
> + OCF_ERR_INSTALLED=5
> + OCF_ERR_CONFIGURED=6
> + OCF_NOT_RUNNING=7
> + OCF_RUNNING_MASTER=8
> + OCF_FAILED_MASTER=9
> + . /usr/lib/ocf/resource.d/heartbeat/.ocf-directories
> + prefix=/usr
> + exec_prefix=/usr
> + : /etc/init.d
> + : /etc/ha.d
> + : /etc/ha.d/rc.d
> + : /etc/ha.d/conf
> + : /etc/ha.d/ha.cf
> + : /var/lib/heartbeat
> + : /var/run/heartbeat/rsctmp
> + : /var/lib/heartbeat/fifo
> + : /usr/lib/heartbeat
> + : /usr/sbin
> + : %Y/%m/%d_%T
> + : /dev/null
> + : /etc/ha.d/resource.d
> + : /usr/share/doc/heartbeat
> + : controld
> + : controld[8408]
> + : /var/run/
> + : /var/lock/subsys/
> + : 0
> + __ocf_set_defaults start
> + __OCF_ACTION=start
> + unset LANG
> + LC_ALL=C
> + export LC_ALL
> + [ -z  ]
> + : 0
> + [ ! -d /usr/lib/ocf ]
> + [ -z  ]
> + : controld
> + [ -z  ]
> + : We are being invoked as an init script.
> + : Fill in some things with reasonable values.
> + : default
> + return 0
> + : false
> + : /sys/kernel/config
> + : false
> + : -q 0
> + : gfs_controld.pcmk
> + controld_validate
> + check_binary gfs_controld.pcmk
> + have_binary gfs_controld.pcmk
> + sed -e s/ -.*//
> + echo gfs_controld.pcmk
> + bin=gfs_controld.pcmk
> + which gfs_controld.pcmk
> + [ -x /usr/sbin/gfs_controld.pcmk ]
> + return 0
> + :
> + [ -d /var/run/cluster ]
> + return 0
> + controld_start
> + controld_monitor
> + killall -0 gfs_controld.pcmk
> gfs_controld.pcmk: no process found
> + rc=1
> + return 7
> + rc=7
> + [ ! -e /sys/kernel/config ]
> + grep type configfs
> + mount
> + [ 0 != 0 ]
> + [ ! -e /sys/kernel/config/dlm ]
> + gfs_controld.pcmk -q 0
> unknown option: q

Looks pretty relevant.
My guess is that you have an old version of the controld agent.
Try:
   http://hg.clusterlabs.org/pacemaker/1.1/raw-file/tip/extra/resources/controld

> + sleep 1
> + controld_monitor
> + killall -0 gfs_controld.pcmk
> gfs_controld.pcmk: no process found
> + rc=1
> + return 7
> + rc=7
> + exit 7
> 7
> root at squeeze:~#
>
> I've installed these 'gfs' packages:-
>
> ii  gfs-pcmk                           3.0.12-2                    Red Hat
> cluster suite - GFS pacemaker module
> ii  gfs2-tools                         3.0.12-2                    Red Hat
> cluster suite - global file system 2 tools
>
> Not sure what else I can do to debug this problem, or how to resolve it.
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs:
> http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>