Re: [Linux-HA] pacemaker 1.0.2 memory leak

On Wed, Mar 11, 2009 at 8:27 PM, Andrew Beekhof wrote: > On Wed, Mar 11, 2009 at 18:17, Pavel Georgiev wrote: >> I've noticed that pacemaker`s /usr/lib/heartbeat/cib leaks ~200kb >> every time a resource is migrated. I`ve setup a resource to fail ~ 2 >> minutes after

[Linux-HA] pacemaker 1.0.2 memory leak

I've noticed that pacemaker`s /usr/lib/heartbeat/cib leaks ~200kb every time a resource is migrated. I`ve setup a resource to fail ~ 2 minutes after it is started and the cib proc quickly grows in size. I`ve upgraded pacemaker to 1.0.2-11.1, which is the latest centos rpm, but the problem persists.

Re: [Linux-HA] Resource does not come online after failure-timeout has expired

On Mon, Mar 9, 2009 at 10:44 PM, Andreas Kurz wrote: > Have a look at the output of "/usr/lib/heartbeat/crmd metadata" and search > for "cluster_recheck_interval" ... and add it to your crm_config. > > I think Andrew already added a patch to pacemaker so the > cluster_recheck_interval is enabled b

[Linux-HA] Resource does not come online after failure-timeout has expired

I`m running heartbeat 2.99 + pacemaker 1.0. I have migration-threshold=2 and failure-timeout=600s. When I simulate 2 resource failures per server, the resource is migrated to the next node (as expected). When all servers eligible for running the resource reach migration-threshold failures (2), the

Re: [Linux-HA] Heartbeat quorum question

anyone knows a good solution here? I`m basically triyng to take quorum decisions based only of three of the nodes of a four node cluster. Thanks. On Tue, Feb 17, 2009 at 9:29 AM, Andrew Beekhof wrote: > On Tue, Feb 17, 2009 at 02:14, Pavel Georgiev wrote: > > I have a setup with

Re: [Linux-HA] Heartbeat score calculation

Thanks, I`ll upgrade. Any input on issue (2) and (3) On Wed, Feb 18, 2009 at 9:15 PM, Serge Dubrouski wrote: > 2.1.3 has a well known bug that was fixed in later releases. fail > count doesn't get increased when resource fails. You have to upgrade. > > On Wed, Feb 18, 2009 a

[Linux-HA] Heartbeat score calculation

I`m using heartbeat 2.1.3, default Centos 5 rpm. I`m running 3 nodes with a single resource (LSB RA) which has an equal score on each server. I`m having two "issues" (which actually might be feature) 1) If I stop the resource between two heartbeat "monitor" intervals, it detects it is down and rest

[Linux-HA] Heartbeat quorum question

I have a setup with 4 nodes, 3 of them may run a resource in active/passive mode and that resource should never run on the 4th node. I also have a second resource which may run on either one of the nodes (again active/passive). Since I can run the first resource on only 3 of the nodes, can I make

Re: [Linux-HA] Heartbeat notifications on resource take over

Thanks, that was helpfull. On Monday 23 June 2008 13:40:15 Dejan Muhamedagic wrote: > Hi, > > On Fri, Jun 20, 2008 at 08:44:16PM +0300, Pavel Georgiev wrote: > > Hi. > > > > I`m trying get some custom scripts triggered whenever an appliance > > acquires/rele

Re: [Linux-HA] Heartbeat notifications on resource take over

it into a group with your primary resources. On Fri, Jun 20, 2008 at 11:44 AM, Pavel Georgiev <[EMAIL PROTECTED]> wrote: Hi. I`m trying get some custom scripts triggered whenever an appliance acquires/releases a resource. What is the best practice for doing this? I saw HA executes s

[Linux-HA] Heartbeat notifications on resource take over

Hi. I`m trying get some custom scripts triggered whenever an appliance acquires/releases a resource. What is the best practice for doing this? I saw HA executes scripts in /etc/ha.d/resource.d/ for the specified resources, so that looks the way to go for, I`m expecting that /etc/ha.d/resource.