Re: [Pacemaker] failover problem with pacemaker & drbd

I agree with Gerry, why would you disallow automatic failover, that is the complete opposite idea to having a cluster. In any case, I think you could put in some rule constraint tying the resources to run in only one node with a value of infinity. If you wanted to move the resource, you would

[Pacemaker] non-standby node scores

hi, i have a node that is placed in standby mode. when i use showscores.sh, i can see that resources have a score of -100 (-INFINITY) for this host. is it possible to get a "if this node wasn't in standby" score for those resources? i'm asking because i fear that when i bring the node online

Re: [Pacemaker] problem when i reboot a node

Is the rebooted node host2? Then it makes sense because location score is INFINITE in host2. hj On Thu, Aug 20, 2009 at 10:14 AM, Gerry kernan wrote: > Hi > > > > I have a 2 node cluster using openais & pacemaker, > > > > If I reboot the standby node when the node comes back up all resources >

[Pacemaker] problem when i reboot a node

Hi I have a 2 node cluster using openais & pacemaker, If I reboot the standby node when the node comes back up all resources migrate automatically to that node from the node they where running on. Is this normal behavior or how do I stop this happening. [r...@host1 ~]# crm configure show n

Re: [Pacemaker] Problem with drbd 2 primary + OCFS2 + server

Hi, On Wed, Aug 19, 2009 at 07:51:58PM +0200, cu...@sonet.internetdsl.pl wrote: > hj lee pisze: >> I found two issues in your configuration. >> >> 1. Still the last location constraint does not make sense. The server-www >> is a group, not multistate master/slave, so role=master is wrong. This >

[Pacemaker] showscores.sh patch (was: Re: issue regarding command line parsing)

hi, as a follow up to the command line parsing "issue", i created a small patch to make showscores.sh working again. i think it should be ok to apply this patch regardless of the state of my report as using "ptest -L -s" should be valid anyways :) thanks, raoul -- __

[Pacemaker] issue regarding command line parsing

hi, as already mentioned on irc, i encounter some issues when passing mixed upper- and lowercase arguments to pacemaker binaries. e.g. > ptest -Ls > ptest: option `-Ls' is ambiguous > ... > # crm_resource -WQr nfs-kernel-server|head > crm_resource: option `-WQr' is ambiguous > ... if i seperat

Re: [Pacemaker] A question about the official support of Corosync.(and about Quorum control)

Hi Andrew, > I expect my testing to start October 1st, so not long after that. > I'll still fix any bugs people find before then, I just don't plan be > using it much personally until I start at RedHat. Thank you for comment. I look forward to release and use it. And I report the problem when

Re: [Pacemaker] pacemaker-mgmt packages in CentOS 5 repo (again)

On Thu, Aug 20, 2009 at 11:34 AM, Lars Marowsky-Bree wrote: > On 2009-08-19T11:39:24, Florian Haas wrote: > >> And one more: >> >> file /etc/ha.d/shellfuncs from install of heartbeat-3.0.0-32.1.x86_64 >> conflicts with file from package resource-agents-1.0-29.1.x86_64 >> >> Sorry for not rolling a

Re: [Pacemaker] A question about the official support of Corosync.(and about Quorum control)

On Thu, Aug 20, 2009 at 4:14 AM, wrote: > Hi, > > We confirmed some behavior of Pacemaker on Corosync1.00. > When it combined Corosync1.00, it seemed to move almost well. > > When is the combination with Corosync1.00 going to be supported formally? I expect my testing to start October 1st, so not

Re: [Pacemaker] pacemaker-mgmt packages in CentOS 5 repo (again)

On 2009-08-19T11:39:24, Florian Haas wrote: > And one more: > > file /etc/ha.d/shellfuncs from install of heartbeat-3.0.0-32.1.x86_64 > conflicts with file from package resource-agents-1.0-29.1.x86_64 > > Sorry for not rolling all of this into one email as I should have. heartbeat hasn't been