[Pacemaker] lenny + clvm + pacemaker/openais...
Dejan Muhamedagic
dejanmm at fastmail.fm
Thu May 28 17:43:49 UTC 2009
Hi,
On Thu, May 28, 2009 at 05:19:34PM +0000, Alain St-Denis wrote:
>
> Andrew Beekhof wrote:
> > You might want to check out Martin's packages.
> > If I understood correctly, he's built the version of clvm used by SUSE
> > (which we know works) against 0.80.5
> >
> > Look for his email with the subject "lvm2-clvm RPMs in opensuse.org
> > package repo?"
>
> Thanks!
>
> I installed Martin's packages. Here's what I have:
>
> pacemaker-openais 1.0.3+svn20090522-2~bpo50+1
> clvm-openais 2.02.44-4~bpo50+1
> libopenais-legacy-2 0.80.5+svn20090522-2~bpo50+1
> openais-legacy 0.80.5+svn20090522-2~bpo50+1
> heartbeat-common 2.99.2+sles11r9-3~bpo50+1
> libheartbeat2 2.99.2+sles11r9-3~bpo50+1
>
> Now, soon after I start clvmd, aisexec dies with a segv (in
> openais_conn_private_data_get). On my 3 nodes test cluster, I start openais
> on all nodes, then I start clvmd on one of the nodes. Not long after, aisexec
> dies on the other nodes. Here are the last messages logged by aisexec:
>
> May 28 16:19:04.924914 [TOTEM] entering GATHER state from 11.
> May 28 16:19:05.079052 [TOTEM] Saving state aru 20 high seq received 20
> May 28 16:19:05.079094 [TOTEM] Storing new sequence id for ring 298
> May 28 16:19:05.079155 [TOTEM] entering COMMIT state.
> May 28 16:19:05.079500 [TOTEM] entering RECOVERY state.
> May 28 16:19:05.079558 [TOTEM] position [0] member 142.135.16.107:
> May 28 16:19:05.079571 [TOTEM] previous ring seq 660 rep 142.135.16.107
> May 28 16:19:05.079578 [TOTEM] aru a high delivered a received flag 1
> May 28 16:19:05.079587 [TOTEM] position [1] member 142.135.16.109:
> May 28 16:19:05.079594 [TOTEM] previous ring seq 660 rep 142.135.16.109
> May 28 16:19:05.079612 [TOTEM] aru 20 high delivered 20 received flag 1
> May 28 16:19:05.079627 [TOTEM] Did not need to originate any messages in
> recovery.
> May 28 16:19:05.080669 [CLM ] CLM CONFIGURATION CHANGE
> May 28 16:19:05.080711 [CLM ] New Configuration:
> May 28 16:19:05.080724 [CLM ] r(0) ip(142.135.16.109)
> May 28 16:19:05.080733 [CLM ] Members Left:
> May 28 16:19:05.080774 [CLM ] Members Joined:
> May 28 16:19:05.080790 [crm ] notice: pcmk_peer_update: Transitional
> membership event on ring 664: memb=1, new=0, lost=0
> May 28 16:19:05.080805 [crm ] info: pcmk_peer_update: memb: lab09 1829799822
> May 28 16:19:05.080843 [CLM ] CLM CONFIGURATION CHANGE
> May 28 16:19:05.080855 [CLM ] New Configuration:
> May 28 16:19:05.080865 [CLM ] r(0) ip(142.135.16.107)
> May 28 16:19:05.080901 [CLM ] r(0) ip(142.135.16.109)
> May 28 16:19:05.080914 [CLM ] Members Left:
> May 28 16:19:05.080923 [CLM ] Members Joined:
> May 28 16:19:05.080938 [CLM ] r(0) ip(142.135.16.107)
> May 28 16:19:05.080972 [crm ] notice: pcmk_peer_update: Stable membership
> event on ring 664: memb=2, new=1, lost=0
> May 28 16:19:05.080985 [MAIN ] info: update_member: Node 1796245390/lab07 is
> now: member
> May 28 16:19:05.081001 [crm ] info: pcmk_peer_update: NEW: lab07 1796245390
> May 28 16:19:05.081036 [crm ] info: pcmk_peer_update: MEMB: lab07 1796245390
> May 28 16:19:05.081044 [crm ] info: pcmk_peer_update: MEMB: lab09 1829799822
> May 28 16:19:05.081063 [crm ] info: send_member_notification: Sending
> membership update 664 to 2 children
> May 28 16:19:05.081118 [SYNC ] This node is within the primary component and
> will provide service.
> May 28 16:19:05.081144 [TOTEM] entering OPERATIONAL state.
> May 28 16:19:05.082382 [MAIN ] info: update_member: 0x7f1188002510 Node
> 1796245390 (lab07) born on: 664
> May 28 16:19:05.082416 [crm ] info: send_member_notification: Sending
> membership update 664 to 2 children
> May 28 16:19:05.082757 [CLM ] got nodejoin message 142.135.16.107
> May 28 16:19:05.082832 [CLM ] got nodejoin message 142.135.16.109
> May 28 16:19:05.087292 [CPG ] got joinlist message from node 1829799822
>
> Then it crashes. Martin (or anybody), have you seen this? I attached my
> openais.conf file. Maybe I'm doing something stupid in there?
You should file a bugzilla for openais. Please use hb_report, it
is going to get all the relevant stuff including the stack traces
(I hope that there was a core dumped).
Thanks,
Dejan
> Alain
>
> --
> Alain St-Denis
> Supercomputing, Systems and Storage / Superinformatique, syst?mes et stockage,
> High Performance Computing Support / Soutien aux calculs en haute performance
> Chief Information Officer Branch / Direction G?n?rale du dirigeant principal
> de l'information
> Environment Canada / Environnement Canada
> Tel: +1 514 421 4697
> # Please read the openais.conf.5 manual page
>
> aisexec {
> # Run as root - this is necessary to be able to manage resources with Pacemaker
> user: root
> group: root
> }
>
> service {
> # Load the Pacemaker Cluster Resource Manager
> name: pacemaker
> ver: 0
> }
>
> totem {
> version: 2
>
> # How long before declaring a token lost (ms)
> token: 10000
>
> # How many token retransmits before forming a new configuration
> token_retransmits_before_loss_const: 20
>
> # How long to wait for join messages in the membership protocol (ms)
> join: 60
>
> # How long to wait for consensus to be achieved before starting a new round of membership configuration (ms)
> consensus: 4800
>
> # Turn off the virtual synchrony filter
> vsftype: none
>
> # Number of messages that may be sent by one processor on receipt of the token
> max_messages: 20
>
> # Limit generated nodeids to 31-bits (positive signed integers)
> clear_node_high_bit: yes
>
> # Disable encryption
> secauth: off
>
> # How many threads to use for encryption/decryption
> threads: 0
>
> # Optionally assign a fixed node id (integer)
> # nodeid: 1234
>
> interface {
> ringnumber: 0
>
> # The following values need to be set based on your environment
> bindnetaddr: 142.135.16.0
> mcastaddr: 226.94.1.1
> mcastport: 5405
> }
> }
>
> logging {
> debug: on
> fileline: off
> to_syslog: yes
> to_stderr: yes
> syslog_facility: daemon
> timestamp: on
> }
>
> amf {
> mode: disabled
> }
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
More information about the Pacemaker
mailing list