[Pacemaker] Problem with rrp_mode

[Dan Frincu]

Hi,

Andrew Beekhof wrote:
> I dont think rrp is well tested by upstream.
> You might want to ask on the corosync ML to be sure.
>
> On Wed, Jan 19, 2011 at 3:43 PM, Michael Schwartzkopff
> <misch at clusterbau.com> wrote:
>   
>> Hi,
>>
>> I have two network cards and configured corosync-1.2.7 with
>> rrp_mode: active
>>
>> at first corosync-cfg -s tells me
>> Printing ring status.
>> Local node ID 1210452490
>> RING ID 0
>>        id      = 10.10.38.72
>>        status  = ring 0 active with no faults
>> RING ID 1
>>        id      = 10.10.40.115
>>        status  = ring 1 active with no faults
>>
>> after a very short time I see the following:
>>
>> RING ID 1
>>        id      = 10.10.40.115
>>        status  = Incrementing problem counter for seqid 1352 iface
>> 10.10.40.115 to [3 of 10]
>>
>> and finally:
>> RING ID 1
>>        id      = 10.10.40.115
>>        status  = Marking seqid 1390 ringid 1 interface 10.10.40.115 FAULTY -
>> adminisrtative intervention required.
>>
>> Anybody being successful at all using rrp_mode with corosync?
>>     
I use rrp_mode: active with two network bonds (4 network cards) and what 
I can tell you is that when I perform tests by shutting down the switch 
ports for one of the bonds, it starts to increment the counter then 
marks the ring as faulty, just as you mentioned, which is normal because 
I just shut the ports.

If you're encountering this without having network connectivity 
problems, then yes, it's an issue, otherwise, when the connection per 
ring is lost, that's the normal output.

Also when the network connectivity is restored the ring doesn't restore 
it's previous condition automatically, it's a feature to be implemented 
in the corosync 2.y.z branch (a.k.a. Weaver's Needle). For now after 
restoring the network connectivity to get the ring back you either do a 
corosync-cfgtool -r or have some script monitor and do it for you.

HTH,
Dan
>> Greetings.
>>
>> --
>> Dr. Michael Schwartzkopff
>> Guardinistr. 63
>> 81375 München
>>
>> Tel: (0163) 172 50 98
>>
>> _______________________________________________
>> 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>>
>>
>>     
>
> _______________________________________________
> 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>   

-- 
Dan FRINCU
Systems Engineer
CCNA, RHCE
Streamwide Romania

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.clusterlabs.org/pipermail/pacemaker/attachments/20110119/c6740252/attachment.htm>