[ClusterLabs] Antw: split brain

ciprian niculescu cnicules at gmail.com
Fri Apr 24 13:46:27 UTC 2015


Hi,
as i have 2 nodes, can't i use other than mcast? maybe it fixes things?
i tryed the member {} in the interface {} but no succes.
the logs yes, they tell lot of things, and nothing i can see it would
say the problem. I can give them to you.
Ciprian

On Fri, Apr 24, 2015 at 4:17 PM, Ulrich Windl
<Ulrich.Windl at rz.uni-regensburg.de> wrote:
>>>> ciprian niculescu <cnicules at gmail.com> schrieb am 24.04.2015 um 09:58 in
> Nachricht
> <CAEMue=NRq=uaRfPzQkbU7MitCgZZJr-HYHjHCGuuXR3wMeCGtA at mail.gmail.com>:
>> Hello,
>>
>> I did a 2 node corosync+pacemaker configuration, all worked well.
>> Debian 7, all the packages from the standard repository.
>> We had a power failure on both nodes and now they don't see each other.
>> On node 1, 2 is offline. On node 2, 1 is offline.
>
> We hat that in two variants (with SLES11):
> 1) Both nodes were discussing who's going to be DC for a very long time. I think that bug is fixed meanwhile.
> 2) Both nodes could talk to each other at TCP level, but not at corosync level, and each node thought it's DC, trying to play with the resources. That one is still under investigation, but it wasn't the latest patch level of the cluster software
>
>> The network between them is working is can ping, recheck the configs
>> they seem ok.
>>
>> I did a little search on the net but didn't found the information. Is
>> there a way to force resync?
>>
>> the corosync.conf is the folowing:
>> totem {
>>         version: 2
>>         secauth: off
>>         threads: 0
>>
>>         interface {
>>                 ringnumber: 0
>>                 bindnetaddr: 192.168.2.0
>>                 mcastaddr: 226.94.1.1
>>                 mcastport: 5405
>>                 ttl: 1
>>         }
>> }
>>
>> amf {
>>         mode: disabled
>> }
>>
>> service {
>>         # Load the Pacemaker Cluster Resource Manager
>>         ver:       0
>>         name:      pacemaker
>> }
>>
>> aisexec {
>>         user:   root
>>         group:  root
>> }
>>
>> logging {
>>         fileline: off
>>         to_stderr: no
>>         to_logfile: yes
>>         logfile: /var/log/corosync/corosync.log
>>         to_syslog: yes
>>         syslog_facility: daemon
>>         debug: on
>>         timestamp: on
>>         logger_subsys {
>>                 subsys: AMF
>>                 debug: on
>>                 tags: enter|leave|trace1|trace2|trace3|trace4|trace6
>>         }
>> }
>
> What do the logs say? Usually a lot...
>
> Ulrich
>
>
>
> _______________________________________________
> Users mailing list: Users at clusterlabs.org
> http://clusterlabs.org/mailman/listinfo/users
>
> 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 Users mailing list