On Mon, Oct 5, 2009 at 5:06 PM, Fausto Lombardi <faulo...@gmail.com> wrote:
> Do anyone know this option?
I just spoke to upstream and apparently there isn't an option yet.
But what you can do is add
echo 1 > /sys/kernel/config/dlm/cluster/protocol
to the start action in the "controld" resource agent.
At some point I'll need to add a parameter to turn this behavior on.
>
> In my mind I would use the rrp feature to stop the drbd in case of
> network/resource/node failure.
> (With the rrp activated if I unplug the replication link the node with
> role DC creates a constraint in pacemaker ("fence-peer
> /usr/lib64/drbd/crm-fence.sh" in drbd.conf) through the second network
> link and the drbd resource goes to secondary state in the other node).
>
> How could I do without use a stonith hardware device?
>
>
>
>
>
> 2009/10/1 Fausto Lombardi <faulo...@gmail.com>:
>> Hi,
>> I have removed the second rrp_interface from openais.conf and now all is
>> OK!!!!
>>
>> What is this function?
>> Thanks!
>>
>>
>>
>> 2009/10/1 Andrew Beekhof <and...@beekhof.net>
>>>
>>> Apparently the dlm needs a special option in order to support multiple
>>> interfaces.
>>> (I don't recall what it is off the top of my head).
>>>
>>> The short-term solution is to remove the second interface from
>>> openais.conf until I get a chance to look up what it is.
>>>
>>> On Wed, Sep 30, 2009 at 9:45 AM, Fausto Lombardi <faulo...@gmail.com>
>>> wrote:
>>> > Hi of all,
>>> > I'm trying to setup a dual primary DRDB with Pacemaker and OCFS2.
>>> >
>>> > I have followed the 'Clusters from Scratch - Apache on Fedora11' pdf but
>>> > when mount the ocfs2 partition I'm receiving this error message:
>>> >
>>> > Sep 30 00:03:17 fedora-kvm1 kernel: dlm: TCP protocol can't handle
>>> > multi-homed hosts, try SCTP
>>> > Sep 30 00:03:17 fedora-kvm1 kernel: dlm: cannot start dlm lowcomms -22
>>> > Sep 30 00:03:17 fedora-kvm1 kernel: (7101,1):ocfs2_dlm_init:2828 ERROR:
>>> > status = -22
>>> > Sep 30 00:03:17 fedora-kvm1 kernel: (7101,1):ocfs2_mount_volume:1741
>>> > ERROR:
>>> > status = -22
>>> > Sep 30 00:03:17 fedora-kvm1 kernel: ocfs2: Unmounting device (147,1) on
>>> > (node 0)
>>> >
>>> > DLM and O2CB are started.
>>> >
>>> > Why?
>>> >
>>> > Furthermore in this manual it is suggested to modify the OCF sript o2cb
>>> > to
>>> > improve the compatibility with fedora 11.
>>> >
>>> > sed -i.gres s/start_daemon//g /usr/lib/ocf/resource.d/ocfs2/o2cb
>>> > sed -i.gres s/pidofproc/pidof/g /usr/lib/ocf/resource.d/ocfs2/o2cb
>>> >
>>> > What are the incompatibility with this release?
>>> >
>>> > Thanks.
>>> >
>>> > _______________________________________________
>>> > Pacemaker mailing list
>>> > Pacemaker@oss.clusterlabs.org
>>> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>>> >
>>> >
>>>
>>> _______________________________________________
>>> Pacemaker mailing list
>>> Pacemaker@oss.clusterlabs.org
>>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>>
>>
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker@oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
_______________________________________________
Pacemaker mailing list
Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker
