----- Original Message -----
> From: "Andrew Beekhof" <andrew at beekhof.net>
> To: "The Pacemaker cluster resource manager" <pacemaker at oss.clusterlabs.org>
> Sent: Tuesday, May 15, 2012 7:59:47 PM
> Subject: Re: [Pacemaker] how to enable verbose logging for failed
>
> On Tue, May 15, 2012 at 9:27 PM, Igor Zinovik
> <zinovik.igor at gmail.com> wrote:
> > 2012/5/14 Andrew Beekhof <andrew at beekhof.net>:
> >> On Sat, May 12, 2012 at 11:41 PM, Igor Zinovik
> >> <zinovik.igor at gmail.com> wrote:
> >>> Hello.
> >>>
A little late to the party but...
Where are you getting the slapd RA from? And/or what version of it?
What OS?
I had a small bit of trouble when I first tried it - I since got a small
patch applied to the RA and it runs nicely in my cluster.
Jake
> >>> How can i increase verbosity level for resource that fails to
> >>> start:
> >>> # egrep -ei '(warn|error)' /var/log/pacemaker.log
> >>> May 12 17:25:12 ldap2 lrmd: [6279]: WARN: Core dumps could be
> >>> lost if
> >>> multiple dumps occur.
> >>> May 12 17:25:12 ldap2 lrmd: [6279]: WARN: Consider setting
> >>> non-default
> >>> value in /proc/sys/kernel/core_pattern (or equivalent) for
> >>> maximum
> >>> supportability
> >>> May 12 17:25:12 ldap2 lrmd: [6279]: WARN: Consider setting
> >>> /proc/sys/kernel/core_uses_pid (or equivalent) to 1 for maximum
> >>> supportability
> >>> May 12 17:25:13 ldap2 corosync[6270]: [pcmk ] ERROR:
> >>> pcmk_wait_dispatch: Child process mgmtd exited (pid=6283, rc=100)
> >>> May 12 17:28:02 ldap2 crmd: [6282]: info: process_lrm_event: LRM
> >>> operation slapd_mirrormode_monitor_10000 (call=5, rc=1,
> >>> cib-update=14,
> >>> confirmed=false) unknown error
> >>> May 12 17:28:04 ldap2 crmd: [6282]: info: process_lrm_event: LRM
> >>> operation slapd_mirrormode_monitor_10000 (call=8, rc=1,
> >>> cib-update=17,
> >>> confirmed=false) unknown error
> >>> May 12 17:28:05 ldap2 crmd: [6282]: info: process_lrm_event: LRM
> >>> operation slapd_mirrormode_monitor_10000 (call=11, rc=1,
> >>> cib-update=20, confirmed=false) unknown error
> >>>
> >>> In slapd log i do not see any messages which shows the culprit of
> >>> failure.
> >>
> >> Is it producing any? Did you grep for "slapd" too?
> >
> > Unfortunately nothing.
>
> The only additional thing Pacemaker can log is output from the RA.
> Does it produce any?
> Is slapd even being called? If so, you'll have to look at the slapd
> config to see how to increase it's verbosity.
>
> >
> > My config file is correct:
> > ldap2:~# /usr/lib/openldap/slapd -Tt
> > config file testing succeeded
> >
> > I can successfully start slapd by hands:
> > ldap2:~ # service ldap start
> > redirecting to systemctl
> > ldap2:~ # service ldap status
> > redirecting to systemctl
> > ldap.service - LSB: OpenLDAP Server (slapd)
> > Loaded: loaded (/etc/init.d/ldap)
> > Active: active (running) since Tue, 15 May 2012 15:19:51
> > +0400; 3s ago
> > Process: 2278 ExecStop=/etc/init.d/ldap stop (code=exited,
> > status=0/SUCCESS)
> > Process: 2762 ExecStart=/etc/init.d/ldap start
> > (code=exited,
> > status=0/SUCCESS)
> > CGroup: name=systemd:/system/ldap.service
> > 2852 /usr/lib/openldap/slapd -h ldap:/// ldaps:///
> > ldapi:/// -f /etc/openldap/slapd.conf -u ldap -g lda...
> >
> > When i create resource and commit configuration, i see following in
> > pacemaker log:
> > ldap2:~# egrep -e 'May 15 15:14' /var/log/pacemaker.log
> > May 15 15:14:47 ldap2 crmd: [13576]: info: update_dc: Unset DC
> > ldap1
> > May 15 15:14:47 ldap2 crmd: [13576]: info: do_state_transition:
> > State
> > transition S_NOT_DC -> S_PENDING [ input=I_PENDING
> > cause=C_FSA_INTERNAL origin=do_election_count_vote ]
> > May 15 15:14:47 ldap2 crmd: [13576]: info: update_dc: Set DC to
> > ldap1 (3.0.5)
> > May 15 15:14:47 ldap2 crmd: [13576]: info: do_state_transition:
> > State
> > transition S_PENDING -> S_NOT_DC [ input=I_NOT_DC
> > cause=C_HA_MESSAGE
> > origin=do_cl_join_finalize_respond ]
> >
> > ldap2:~# crm_mon -1 -o -r
> > ============
> > Last updated: Tue May 15 15:24:05 2012
> > Last change: Tue May 15 15:14:47 2012 by root via cibadmin on ldap2
> > Stack: openais
> > Current DC: ldap1 - partition with quorum
> > Version: 1.1.6-9971ebba4494012a93c03b40a2c58ec0eb60f50c
> > 2 Nodes configured, 2 expected votes
> > 2 Resources configured.
> > ============
> >
> > Online: [ ldap2 ldap1 ]
> >
> > Full list of resources:
> >
> > ldap_virtual_ip (ocf::heartbeat:IPaddr2): Started
> > ldap1
> > slapd_mirrormode (ocf::heartbeat:slapd): Stopped
> >
> > Operations:
> > * Node ldap1:
> > ldap_virtual_ip: migration-threshold=1000000
> > + (33) start: rc=0 (ok)
> > + (34) monitor: interval=5000ms rc=0 (ok)
> > slapd_mirrormode:0: migration-threshold=1000000
> > fail-count=1000000
> > + (7) start: rc=1 (unknown error)
> > slapd_mirrormode: migration-threshold=3 fail-count=3
> > + (31) monitor: interval=10000ms rc=1 (unknown error)
> > + (32) stop: rc=0 (ok)
> > * Node ldap2:
> > slapd_mirrormode: migration-threshold=3 fail-count=3
> > + (11) monitor: interval=10000ms rc=1 (unknown error)
> > + (12) stop: rc=0 (ok)
> >
> > Failed actions:
> > slapd_mirrormode:0_start_0 (node=ldap1, call=7, rc=1,
> > status=complete): unknown error
> > slapd_mirrormode_monitor_10000 (node=ldap1, call=31, rc=1,
> > status=complete): unknown error
> > slapd_mirrormode_monitor_10000 (node=ldap2, call=11, rc=1,
> > status=complete): unknown error
> >
> > I tried following change to slapd agent:
> > --- slapd.orig 2012-05-15 15:25:19.625554295 +0400
> > +++ slapd 2012-05-15 15:25:27.330815814 +0400
> > @@ -38,7 +38,7 @@
> > # Initialization:
> >
> > : ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/resource.d/heartbeat}
> > -. ${OCF_FUNCTIONS_DIR}/.ocf-shellfuncs
> > +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs
> >
> > : ${OCF_RESKEY_slapd="/usr/sbin/slapd"}
> > : ${OCF_RESKEY_ldapsearch="ldapsearch"}
> >
> > because in other agents that come with pacemaker there was no dot
> > before `ocf-shellfuncs', e.g.:
> > ldap2:~# grep ocf-shellfuncs
> > /usr/lib/ocf/resource.d/heartbeat/{nginx,mysql,pgsql}
> > /usr/lib/ocf/resource.d/heartbeat/nginx:.
> > ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs
> > /usr/lib/ocf/resource.d/heartbeat/mysql:.
> > ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs
> > /usr/lib/ocf/resource.d/heartbeat/pgsql:.
> > ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs
> >
> > _______________________________________________
> > 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://bugs.clusterlabs.org
>
> _______________________________________________
> 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://bugs.clusterlabs.org
>
>