[Pacemaker] crmsh resource update issue

Dejan Muhamedagic dejanmm at fastmail.fm
Thu Sep 27 17:03:26 UTC 2012


On Thu, Sep 27, 2012 at 05:02:24PM +0200, Dejan Muhamedagic wrote:
> Hi Vladimir,
> 
> On Thu, Sep 27, 2012 at 10:22:43AM +0300, Vladislav Bogdanov wrote:
> > Hi Dejan, list,
> > 
> > It looks like shell 1.2.0 (I use
> > b58a3398bf11621fe7811380f00245dac52d34c6 with patch you sent recently)
> > incorrectly replaces the whole cib with just resources section, so all
> > node state sections are dropped.
> 
> Node state: do you mean the <status> section? That's bad. I'll
> see what happened.

This should fix it:

http://hg.savannah.gnu.org/hgweb/crmsh/rev/86b59c2c2b49

Please let me know if it did.

Cheers,

Dejan

> > Logs and code analysis show that crm calls 'cibadmin -p -R'
> > (commit-_doc()) instead of 'cibadmin -p -R -o resources' (or uses
> > resources section as source instead of the whole cib).
> > 
> > Pacemaker is 1.1.8.
> > Both 'crm configure edit' and 'crm configure load update' suffer from this.
> 
> Both are doing basically the same thing.
> 
> Thanks,
> 
> Dejan
> 
> > Relevant log lines are (one attribute was replaced):
> > (many) Sep 27 06:55:28 vd01-a cibadmin[5239]:   notice: crm_log_args:
> > Invoked: cibadmin -Ql -o resources
> > Sep 27 06:55:28 vd01-a cibadmin[5240]:   notice: crm_log_args: Invoked:
> > cibadmin -Ql
> > Sep 27 06:55:29 vd01-a crm_verify[5241]:   notice: crm_log_args:
> > Invoked: crm_verify -V -p
> > Sep 27 06:55:29 vd01-a lrmadmin: [5249]: WARN: lrm_signon: can not
> > initiate connection
> > Sep 27 06:55:29 vd01-a lrmadmin: [5249]: WARN: Can't connect to lrmd!
> > Sep 27 06:55:29 vd01-a stonith: [5250]: ERROR: Invalid device type:
> > 'fence_xvm'
> > Sep 27 06:55:29 vd01-a cib[5302]:     info: crm_update_callsites:
> > Enabling callsites based on priority=6, files=(null), functions=(null),
> > formats=(null), tags=(
> > Sep 27 06:55:29 vd01-a cibadmin[5303]:   notice: crm_log_args: Invoked:
> > cibadmin -p -R
> > Sep 27 06:55:29 vd01-a crmd[1734]:   notice: do_state_transition: State
> > transition S_IDLE -> S_POLICY_ENGINE [ input=I_PE_CALC
> > cause=C_FSA_INTERNAL origin=abort
> > Sep 27 06:55:29 vd01-a rsyslogd-2177: imuxsock lost 183 messages from
> > pid 1729 due to rate-limiting
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: log_cib_diff: cib:diff:
> > Diff: --- 0.1337.84
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: log_cib_diff: cib:diff:
> > Diff: +++ 0.1339.1
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --
> > <nvpair value="pentium"
> > id="bubble-test01.vds-ok.com-vm-instance_attributes-cpu_model" />
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --     <node_state
> > id="1107559690" uname="vd01-c" in_ccm="true" crmd="online" join="member"
> > expected="memb
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --
> > <transient_attributes id="1107559690" >
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --
> > <instance_attributes id="status-1107559690" >
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --
> > <nvpair id="status-1107559690-probe_complete" name="probe_complete"
> > value="true" />
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --
> > <nvpair id="status-1107559690-last-failure-libvirtd"
> > name="last-failure-libvirtd" value="1348
> > Sep 27 06:55:29 vd01-a cib[1729]:   notice: cib:diff: --
> > </instance_attributes>
> > ...
> > Sep 27 06:55:29 vd01-a rsyslogd-2177: imuxsock begins to drop messages
> > from pid 1729 due to rate-limiting
> > 
> > Vladislav
> > 
> > _______________________________________________
> > 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




More information about the Pacemaker mailing list