[Linux-cluster] Ricci doesn't work

[Chip Burke]

This gives me the same behavior. On each node I ran:


#rm -f /var/lib/ricci/certs/clients/*

#service ricci restart

Then from one node I simply incremented the version in cluster.conf and ran

# cman_tool version -r
You have not authenticated to the ricci daemon on xanadunode1
Password: 
You have not authenticated to the ricci daemon on xanadunode2
Password: 
You have not authenticated to the ricci daemon on xanadunode3
Password: 
The connection to xanadunode1 died unexpectedly
The connection to xanadunode2 died unexpectedly
The connection to xanadunode3 died unexpectedly
cman_tool: ccs_sync failed.
If you have distributed the config file yourself, try re-running with -S



Let me know if there is any other info I can give you.

Thanks again!

________________________________________
Chip Burke







On 9/5/12 9:34 AM, "Jan Pokorný" <jpokorny at redhat.com> wrote:

>On 20/08/12 14:51 +0000, Chip Burke wrote:
>> Thanks for sticking with me on this.
>
>Sorry for delay.
>
>The traceback from your initial email, which ended with:
>> TypeError: No object (name: translator) has been registered for this
>> thread
>
>I overlooked this one, but independently hit it too [1].
>
>But this is orthogonal to the main issue you are stating
>and which I am having troubles to figure out.
>
>Could you please try following as a possible workaround?
>At the particular node (or all within the cluster), try running
>rm -f /var/lib/ricci/certs/clients/* and restarting ricci.
>As a consequence, you may be prompted for the password to authenticate
>against ricci instance(s) and/or re-add the cluster in luci even if
>this was already done before.
>
>[1] https://bugzilla.redhat.com/show_bug.cgi?id=853151
>
>Thanks,
>Jan
>
>--
>Linux-cluster mailing list
>Linux-cluster at redhat.com
>https://www.redhat.com/mailman/listinfo/linux-cluster