Hi,
I am using Pacemaker 1.0.6. I noticed when I use crm_mon --failcounts when I
have a resource in the failed state I get the migration threshold and the
fail-count
Migration summary:
* Node qpr2:
default_route: migration-threshold=5 fail-count=6 last-failure='Wed Jan 27
14:47:49 2010'
*
Am Mittwoch, 27. Januar 2010 18:10:48 schrieb Rasto Levrinc:
> On Wed, January 27, 2010 5:41 pm, Michael Schwartzkopff wrote:
> > Am Mittwoch, 27. Januar 2010 16:13:09 schrieb Martin Gerhard Loschwitz:
> >> Packages available for Lenny on amd64 and i386 from the usual source:
> >>
> >>
> >> deb htt
On Wed, January 27, 2010 5:41 pm, Michael Schwartzkopff wrote:
> Am Mittwoch, 27. Januar 2010 16:13:09 schrieb Martin Gerhard Loschwitz:
>> Packages available for Lenny on amd64 and i386 from the usual source:
>>
>>
>> deb http://people.debian.org/~madkiss/ha lenny main deb-src
>> http://people.d
Am Mittwoch, 27. Januar 2010 17:56:38 schrieb Raoul Bhatia [IPAX]:
> On 01/27/2010 05:41 PM, Michael Schwartzkopff wrote:
> > Where do I get your old packages?
>
> # apt-cache policy pacemaker
> pacemaker:
> Installed: (none)
> Candidate: 1.0.7+hg20100127-0test1~bpo50+1
> Version table:
>
On 01/27/2010 05:41 PM, Michael Schwartzkopff wrote:
> Where do I get your old packages?
# apt-cache policy pacemaker
pacemaker:
Installed: (none)
Candidate: 1.0.7+hg20100127-0test1~bpo50+1
Version table:
1.0.7+hg20100127-0test1~bpo50+1 0
50 http://people.debian.org lenny/main
Am Mittwoch, 27. Januar 2010 16:13:09 schrieb Martin Gerhard Loschwitz:
> Ladies and Gentleman,
>
> it's a great pleasure for me to announce the availability of up-to-date
> packages of the standard cluster stack components once more, including all
> the recently released release-candidates from He
Am Mittwoch, 27. Januar 2010 16:13:09 schrieb Martin Gerhard Loschwitz:
> Ladies and Gentleman,
>
> it's a great pleasure for me to announce the availability of up-to-date
> packages of the standard cluster stack components once more, including all
> the recently released release-candidates from He
Nice work!
On Wed, Jan 27, 2010 at 4:13 PM, Martin Gerhard Loschwitz
wrote:
> Ladies and Gentleman,
>
> it's a great pleasure for me to announce the availability of up-to-date
> packages
> of the standard cluster stack components once more, including all the recently
> released release-candidate
Ladies and Gentleman,
it's a great pleasure for me to announce the availability of up-to-date packages
of the standard cluster stack components once more, including all the recently
released release-candidates from Heartbeat, Cluster-Glue and Cluster-Agents as
well as Pacemaker 1.0.7.
Packages a
>> seems like it is a bit more complicated then I initially though. Now I
>> tried to set the timeout longer so that there will not be any warning.
>> However, the second group is not created neither:
>> # crm
>> crm(live)# configure
>> crm(live)configure# group udom udom-drbd-udom0 udom-drbd-udom1
On 01/27/2010 11:26 AM, Oliver Ladner wrote:
> Hello,
>
> Can someone please tell me the difference between the resource agents
> ocf:heartbeat:drbd and ocf:linbit:drbd?
>
> I'm using the legacy heartbeat:drbd agent at the moment, as drbd on SLES 11
> is only at 8.2.7. Problem is that the two-n
On Tue, Jan 26, 2010 at 6:33 PM, Koch, Sebastian
wrote:
> Hi,
>
>
>
> i am kind of new to pacemaker. I am trying to configure a active/passive
> pacemaker/drbd/mysql cluster. Preferably all resources should run on node1
> if available, if not they should migrate to node2. I already read all the
>
Hello,
Can someone please tell me the difference between the resource agents
ocf:heartbeat:drbd and ocf:linbit:drbd?
I'm using the legacy heartbeat:drbd agent at the moment, as drbd on SLES 11 is
only at 8.2.7. Problem is that the two-node cluster only moves the drbd
resource when the active n
Done.
Though perhaps you could amend
http://www.clusterlabs.org/wiki/Debian_Lenny_HowTo if there were
additional steps it didnt cover.
On Tue, Jan 26, 2010 at 9:52 AM, Koch, Sebastian
wrote:
> Hi,
>
>
>
> i would like to contribute to your great project. I needed to figure out
> some strange err
Hi,
On Tue, Jan 26, 2010 at 09:38:56PM +0100, Lars Ellenberg wrote:
> On Tue, Jan 26, 2010 at 04:47:12PM +, Maros Timko wrote:
> > OK,
> >
> > seems like it is a bit more complicated then I initially though. Now I
> > tried to set the timeout longer so that there will not be any warning.
> >
Hello,
found the problem myself, if anyone wonders what the problem was:
having the timeout value for the ping resource set to 5 seconds and the number
of pings per host set to "default - 4", the probe needs at least 4 seconds per
node to complete, so it got always killed by the lrm :-)
Thanks
16 matches
Mail list logo
