[Pacemaker] HA KVM over DRBD primary/secondary configuration

Rasto Levrinc rasto.levrinc at gmail.com
Fri Apr 19 10:38:56 UTC 2013


On Fri, Apr 19, 2013 at 11:15 AM, Alexandr A. Alexandrov
<shurrman at gmail.com> wrote:
> Hi!
>
> I do not need CMAN config to be created, since I do not use CMAN...

It's possible, you would have to on RHEL6 system.

> I tried LCMC, it seems to recognize cluster/drbd settings.
> However, when I was finishing looking at it, I got corosync dead! :-(

Did Corosync actually sefgault? Looks like Corosync problem, LCMC doesn't
interact with it directly.

Rasto

>
>
> Apr 19 12:45:30 kennedy corosync[20271]:   [QB    ] IPC credentials
> authenticated (20271-19215-31)
> Apr 19 12:45:30 kennedy cib[17082]:    error: pcmk_cpg_dispatch: Connection
> to the CPG API failed: 2
> Apr 19 12:45:30 kennedy cib[17082]:    error: cib_ais_destroy: Corosync
> connection lost!  Exiting.
> Apr 19 12:45:30 kennedy crmd[16172]:    error: pcmk_cpg_dispatch: Connection
> to the CPG API failed: 2
> Apr 19 12:45:30 kennedy cib[17082]:    error: cib_ipc_closed: Client crm_mon
> not found in the hashtable
> Apr 19 12:45:30 kennedy crmd[16172]:    error: crm_ipc_read: Connection to
> cib_shm failed
> Apr 19 12:45:30 kennedy crmd[16172]:    error: mainloop_gio_callback:
> Connection to cib_shm[0x6503b0] closed (I/O condition=17)
> Apr 19 12:45:30 kennedy crmd[16172]:    error: crmd_cib_connection_destroy:
> Connection to the CIB terminated...
> Apr 19 12:45:31 kennedy pacemakerd[17080]:    error: cfg_connection_destroy:
> Connection destroyed
> Apr 19 12:45:31 kennedy pacemakerd[17080]:    error: cpg_connection_destroy:
> Connection destroyed
> Apr 19 12:45:31 kennedy pacemakerd[17080]:   notice: pcmk_shutdown_worker:
> Shuting down Pacemaker
> Apr 19 12:45:31 kennedy pacemakerd[17080]:   notice: stop_child: Stopping
> crmd: Sent -15 to process 16172
> Apr 19 12:45:31 kennedy stonith-ng[17083]:    error: pcmk_cpg_dispatch:
> Connection to the CPG API failed: 2
> Apr 19 12:45:31 kennedy stonith-ng[17083]:    error:
> stonith_peer_ais_destroy: AIS connection terminated
> Apr 19 12:45:31 kennedy crmd[16172]:    error: do_log: FSA: Input I_ERROR
> from crmd_cib_connection_destroy() received in state S_NOT_DC
> Apr 19 12:45:31 kennedy crmd[16172]:   notice: do_state_transition: State
> transition S_NOT_DC -> S_RECOVERY [ input=I_ERROR cause=C_FSA_INTERNAL
> origin=crmd_
> cib_connection_destroy ]
> Apr 19 12:45:31 kennedy crmd[16172]:    error: do_recover: Action A_RECOVER
> (0000000001000000) not supported
> Apr 19 12:45:31 kennedy crmd[16172]:    error: do_log: FSA: Input
> I_TERMINATE from do_recover() received in state S_RECOVERY
> Apr 19 12:45:31 kennedy attrd[17085]:    error: pcmk_cpg_dispatch:
> Connection to the CPG API failed: 2
> Apr 19 12:45:31 kennedy attrd[17085]:     crit: attrd_ais_destroy: Lost
> connection to Corosync service!
> Apr 19 12:45:31 kennedy attrd[17085]:   notice: main: Exiting...
> Apr 19 12:45:31 kennedy attrd[17085]:   notice: main: Disconnecting client
> 0x61ec10, pid=16172...
> Apr 19 12:45:31 kennedy attrd[17085]:    error:
> attrd_cib_connection_destroy: Connection to the CIB terminated...
> Apr 19 12:45:31 kennedy pacemakerd[17080]:    error: pcmk_child_exit: Child
> process attrd exited (pid=17085, rc=1)
> Apr 19 12:45:31 kennedy pacemakerd[17080]:    error: send_cpg_message:
> Sending message via cpg FAILED: (rc=9) Bad handle
> Apr 19 12:45:31 kennedy pacemakerd[17080]:    error: send_cpg_message:
> Sending message via cpg FAILED: (rc=9) Bad handle
>
>
> 19.04.2013 11:44, Rasto Levrinc пишет:
>
> On Fri, Apr 19, 2013 at 9:11 AM, Alexandr A. Alexandrov
> <shurrman at gmail.com> wrote:
>
> Hi Rasto,
>
> Note that on RHEL 6/CentOS 6, you should run the Pacemaker through CMAN and
> not a Corosync plugin
>
> I wonder if that's still true, but better be safe than sorry.
>
>
> Not glad to hear that... We are using  Pacemaker+Corosync everywhere (SuSe,
> CentOS, OracleLinux servers).
> Is there any way to use LCMC in this setup?
>
> there's no problem using LCMC in this setup. It will not help to create the
> cman config, though. That's a feature with very low priority for me at the
> moment.
>
> Rasto
>
> 19.04.2013 09:30, Rasto Levrinc пишет:
>
> We used the amazing LCMC tool for the hosts, drbd and pacemaker and Corosync
> setup
> http://lcmc.sourceforge.net/
>
> LCMC does lots of the setup automatic - a huge timesaver.
>
> I'm glad to hear that.
>
> _______________________________________________
> 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
>



-- 
Dipl.-Ing. Rastislav Levrinc
rasto.levrinc at gmail.com
Linux Cluster Management Console
http://lcmc.sf.net/




More information about the Pacemaker mailing list