Op 22-8-2012 13:08, Jan Friesse schreef:
Maurits van de Lande napsal(a):
Hello,
I hope this is the right mailing list for this question.
Problem:
On a cman cluster with two redundant rings running on Centos6.3, Ring
1 is marked as faulty every few seconds and then recovers again.
Ring 0:
Two 10Gbe adapters bonded in mode 1 (bond0)
Ring 1:
An 1Gbe adapter connected to a dedicated " secondary ring network"
The problem look similar as problem 3 in this thread
http://www.spinics.net/lists/corosync/msg01637.html
I believe it's same problem. Patch is not yet included in RH 6.3.z (so
also not in COS 6.3). If you are able to, you can try to recompile
package by yourself with patch (in upstream git it's
17e15322b7500247c539166441fb2e78cc95d02b).
I have compiled the packages. Is it correct that the following packages
are build with 'make rpm'?
After selecting the commit I have rebooted the build system ones after
installing updates.
corosync-1.4.3-1.4.17e15.el6.x86_64.rpm
corosync-debuginfo-1.4.3-1.4.17e15.el6.x86_64.rpm
corosynclib-1.4.3-1.4.17e15.el6.x86_64.rpm
corosynclib-devel-1.4.3-1.4.17e15.el6.x86_64.rpm
I'll install and test it tomorrow.
Thanks,
Maurits
Regards,
Honza
_______________________________________________
discuss mailing list
discuss@xxxxxxxxxxxx
http://lists.corosync.org/mailman/listinfo/discuss
