On Thu, Feb 28, 2008 at 6:33 PM, Schmidt, Florian
<[EMAIL PROTECTED]> wrote:
> I got it working by erasing the 2 "-INFINITY"-rules for (drbd slave and 
> stopped) and added one with "INFINITY" for drbd master..this works
>
>  I cannot explain why but maybe one of you can ;)

Unfortunately there is a bug in the PE thats causing those particular
constraints to be mis-interpreted.
But luckily they're redundant and can be removed in favor of a +INFINITY one.

>
>  Regards
>  Florian
>
>
>  > -----Ursprüngliche Nachricht-----
>  > Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
>  > ha.org] Im Auftrag von Schmidt, Florian
>  > Gesendet: Donnerstag, 28. Februar 2008 17:22
>  > An: General Linux-HA mailing list
>  > Betreff: [Linux-HA] Multistate-Resource Constraint Problem
>
>
> >
>  > Hi again,
>  >
>  >
>  > I do not get my filesystem started, i think it is a problem with the
>  > constraints, but I don't find the problem.
>  >
>  > These are the constraints:
>  >
>  > <rsc_colocation id="Run_Group1_never_on_DRBD_stopped" from="Group1"
>  > to="DRBD" to_role="stopped" score="-INFINITY"/>
>  > <rsc_colocation id="Run_Group1_never_on_DRBD_slave" from="Group1"
>  > to="DRBD" to_role="slave" score="-INFINITY"/>
>  > <rsc_order id="Start_Group1_after_DRBD" from="Group1" to="DRBD"
>  > action="start" to_action="promote"/>
>  > <rsc_colocation id="Run_Cluster_IP_only_on_DRBD" from="Cluster_IP"
>  > to="DRBD" to_role="master" score="INFINITY"/>
>  >
>  > I copied the first 3 lines from the Linux-HA-page and modified them for
>  > my resource-names. I understand that they deny running Group1 on slave
>  > or stopped DRBD. So it implies that Group1 can only run on Master-DRBDs.
>  >
>  > But in the Log, there are 2 warnings:
>  >
>  > pengine[6452]: 2008/02/28_17:07:44 info: determine_online_status: Node
>  > noderz is online
>  > pengine[6452]: 2008/02/28_17:07:44 ERROR: unpack_rsc_colocation: No
>  > resource (con=Run_SAPROUTER_only_on_DRBD, rsc=SAPROUTER)
>  > pengine[6452]: 2008/02/28_17:07:44 notice: clone_print: Master/Slave
>  > Set: DRBD
>  > pengine[6452]: 2008/02/28_17:07:45 notice: native_print:     drbd_r0:0
>  > (heartbeat::ocf:drbd):  Started noderz
>  > pengine[6452]: 2008/02/28_17:07:45 notice: native_print:     drbd_r0:1
>  > (heartbeat::ocf:drbd):  Stopped
>  > pengine[6452]: 2008/02/28_17:07:45 notice: group_print: Resource Group:
>  > Group1
>  > pengine[6452]: 2008/02/28_17:07:45 notice: native_print:     Filesystem
>  > (heartbeat::ocf:Filesystem):    Stopped
>  > pengine[6452]: 2008/02/28_17:07:45 notice: native_print: Cluster_IP
>  > (heartbeat::ocf:IPaddr):        Stopped
>  > pengine[6452]: 2008/02/28_17:07:45 WARN: native_color: Resource
>  > drbd_r0:1 cannot run anywhere
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_promotion_order: Merging
>  > weights for DRBD
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_color: Promoting
>  > drbd_r0:0
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_color: DRBD: Promoted 1
>  > instances of a possible 1 to master
>  > pengine[6452]: 2008/02/28_17:07:45 notice: DemoteRsc: noderz    Demote
>  > drbd_r0:0
>  > pengine[6452]: 2008/02/28_17:07:45 notice: NoRoleChange: Leave resource
>  > drbd_r0:0       (noderz)
>  > pengine[6452]: 2008/02/28_17:07:45 notice: PromoteRsc: noderz   Promote
>  > drbd_r0:0
>  > pengine[6452]: 2008/02/28_17:07:45 notice: DemoteRsc: noderz    Demote
>  > drbd_r0:0
>  > pengine[6452]: 2008/02/28_17:07:45 notice: NoRoleChange: Leave resource
>  > drbd_r0:0       (noderz)
>  > pengine[6452]: 2008/02/28_17:07:45 notice: PromoteRsc: noderz   Promote
>  > drbd_r0:0
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_color: DRBD: Promoted 1
>  > instances of a possible 1 to master
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_color: DRBD: Promoted 1
>  > instances of a possible 1 to master
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_color: DRBD: Promoted 1
>  > instances of a possible 1 to master
>  > pengine[6452]: 2008/02/28_17:07:45 WARN: native_color: Resource
>  > Filesystem cannot run anywhere
>  > pengine[6452]: 2008/02/28_17:07:45 info: master_color: DRBD: Promoted 1
>  > instances of a possible 1 to master
>  > pengine[6452]: 2008/02/28_17:07:45 notice: StartRsc:  noderz    Start
>  > Cluster_IP
>  > pengine[6452]: 2008/02/28_17:07:46 notice: RecurringOp: noderz
>  > Cluster_IP_monitor_30000
>  > crmd[6444]: 2008/02/28_17:07:46 info: do_state_transition: State
>  > transition S_POLICY_ENGINE -> S_TRANSITION_ENGINE [
>  > input=I_PE_SUCCESS
>  > cause=C_IPC_MESSAGE origin=route_message ]
>  > tengine[6451]: 2008/02/28_17:07:46 info: unpack_graph: Unpacked
>  > transition 1: 11 actions in 11 synapses
>  > pengine[6452]: 2008/02/28_17:07:46 WARN: process_pe_message: Transition
>  > 1: WARNINGs found during PE processing. PEngine Input stored in:
>  > /var/lib/heartbeat/pengine/pe-warn-1150.bz2
>  > tengine[6451]: 2008/02/28_17:07:46 info: te_pseudo_action: Pseudo action
>  > 22 fired and confirmed
>  > pengine[6452]: 2008/02/28_17:07:46 info: process_pe_message:
>  > Configuration ERRORs found during PE processing.  Please run "crm_verify
>  > -L" to identify issues.
>  > tengine[6451]: 2008/02/28_17:07:46 info: send_rsc_command: Initiating
>  > action 36: Cluster_IP_start_0 on noderz
>  > tengine[6451]: 2008/02/28_17:07:46 info: send_rsc_command: Initiating
>  > action 50: drbd_r0:0_pre_notify_promote_0 on noderz
>  >
>  >
>  > I understand, that Resource drbd_r0:1 cannot run anywhere, because my
>  > second node is powered off. But why can Filesystem not run anywhere?
>  >
>  >
>  > Thanks
>  > Florian
>  > _______________________________________________
>  > Linux-HA mailing list
>  > [email protected]
>  > http://lists.linux-ha.org/mailman/listinfo/linux-ha
>  > See also: http://linux-ha.org/ReportingProblems
>  _______________________________________________
>  Linux-HA mailing list
>  [email protected]
>  http://lists.linux-ha.org/mailman/listinfo/linux-ha
>  See also: http://linux-ha.org/ReportingProblems
>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to