[Linux-cluster] problems with gfs and openais cluster.

[Steven Dake]

On Sat, 2009-08-15 at 02:11 -0400, Enrique Sanchez wrote:
> folks,
>     I have little experience with RH and even less with RH clusters,
> but kind of know my way around, I am concerned about an RH cluster
> where I am getting nonstop errors like these...
> 
> Aug 15 05:45:42 XXXXX   openais[5283]: [MAIN ] Received message has
> invalid digest... ignoring.
> Aug 15 05:45:42 XXXXX   openais[5283]: [MAIN ] Invalid packet data
> 
> 
> We are using RedHat's GFS and I believe the error should be fixed
> following this post, however  I have a copule of questions.....
> 
> http://www.mail-archive.com/linux-cluster@redhat.com/msg06172.html
> 
> a) here is my /etc/ais/openais.conf file (same config for both hosts
> in the cluster)...
> ======================================================
> totem {
>         version: 2
>         secauth: off
>         threads: 0
>         interface {
>                 ringnumber: 0
>                 bindnetaddr: 192.168.2.0
>                 mcastaddr: 226.94.1.1
>                 mcastport: 5405
>         }
> }
> logging {
>         debug: off
>         timestamp: on
> }
> amf {
>         mode: disabled
> }
> ===================================================
> but here is a list of my IP addresses:
> 
> ============================
> ifconfig -a | grep inet | grep Mask
>           inet addr:10.205.226.91  Bcast:10.205.239.255  Mask:255.255.240.0
>           inet addr:10.205.39.26  Bcast:10.205.39.255  Mask:255.255.255.0
>           inet addr:127.0.0.1  Mask:255.0.0.0
>           inet addr:192.168.122.1  Bcast:192.168.122.255
> Mask:255.255.255.0 ((virbr0))
> ============================
> first of all, I believe the  bindnetaddr parameter should have been
> set to 192.168.122.0 but I am not sure.....
> 
> next, there is a 2nd cluster with the exact same configuration but
> different IP addresses except the debugging setup...
> 
> here it is...
> ========================================
> totem {
>         version: 2
>         secauth: off
>         threads: 0
>         interface {
>                 ringnumber: 0
>                 bindnetaddr: 192.168.2.0
>                 mcastaddr: 226.94.1.1
>                 mcastport: 5405
>         }
> }
> logging {
>         debug: on
>         logfile: /var/log/openais.log
>         timestamp: on
> }
> 
> amf {
>         mode: disabled
> }
> ===================================================
> and here are the IP addresses found in the server:
> ============================================
> #ifconfig -a | grep inet | grep Mask
>           inet addr:10.205.26.38  Bcast:10.205.26.255  Mask:255.255.255.0
>           inet addr:10.205.39.28  Bcast:10.205.39.255  Mask:255.255.255.0
>           inet addr:127.0.0.1  Mask:255.0.0.0
>           inet addr:192.168.122.1  Bcast:192.168.122.255
> Mask:255.255.255.0 (virbr0)
> ============================================
> 
> 
> given that 10.205.26.38 and 10.205.226.91 are on different VLANs I
> doubt that is the problem, but it could be in the adapters using the
> 10.205.39.x IP address range since they are in the same VLAN and given
> the interface binding is incorrect the cluster picked up an interface
> and clashed with the existing server.... is that correct?
> 
> 
> 2nd question: I see openais service is running when chkconfig --list
> openais ; shows it should be down ....
> 
> chkconfig --list openais
> openais         0:off   1:off   2:off   3:off   4:off   5:off   6:off
> 
> who's starting it???
> 
> thank you,
> enrique sanchez.
> 
> 
> 

The openais.conf file is not used when starting with the cman service.
Instead cman uses its own configuration file
in /etc/cluster/cluster.conf.  openais should not be chkconfig'ed on
unless running without cman.

One reason you could have the error is that you have multiple clusters
on the same multicast address as specified in /etc/cluster/cluster.conf.

Regards
-steve

> 
>