ere is a a rsa stonith agent but you probably also can use IPMI (most
servers support this)
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker
Project Home: http://www.clust
I would prefer to configure fencing based on the blade-chassis functionality
but I do not know if there is a stonith agent for you're specific chassis...
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.c
in a single cluster.
However, you should set the appropriate restrictions.
It is likely that some resources should not run on host x or some combination
of resources should not run on the same machine.
Regards,
Robert van Leeuwen
___
Pacemaker mailing lis
documentation and examples can be found here ;-)
http://www.clusterlabs.org/doc/en-US/Pacemaker/1.0/html/Pacemaker_Explained/
Regards,
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/p
ore"
You should also configure fencing devices like an APC powerswitch (preffered
way) or e.g. an IPMI device (almost all servers support this)
Regards,
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://os
ckiness it will migrate the
resource.
The stickiness off the cluster suite is set as follows:
crm_attribute --type rsc_defaults --attr-name resource-stickiness --attr-value
50
Regards,
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker@oss.clust
er can not recover itself but it would
have recovered if all nodes were still running.
I could see these features being very useful in non-ha environments like
testing/development but most off MY customers cannot use this feature or would
rather pay the power bill...
Just my 2 cents,
Robert
imit limits the number of cluster actions.
Nice to know but in this case everything points to the same SAN / pool with
LUN's.
So limiting by-node is not really helpful in this case.
Thanks,
Robert van Leeuwen
___
Pacemaker mailing list: Pacem
migrate a resource at a rate of 1
resource every 2 minutes e.g.?
batch-limit will stop multiple resource migration processing in parallel but
can you set the interval between batch-limit?
Thanks,
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker
sb
primitive amazonip lsb:/usr/local/bin/amazonip
order stunnel-after-haproxy inf: haproxy stunnel
order amazonip-after-stunnel inf: stunnel amazonip
commit
Regarding you're custom scripts I would create a OCF or LSB compliant script as
in the example above.
You can place these in
r to edit the IPaddr2 resource script so it does not wait for
the arp commands.
At you're own risk of course ;-)
Kind regards,
Robert van Leeuwen
___
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/li
would be a node at a third DC but you still run into problems
with the fencing devices.
I doubt you can remotely power-off the non-responding DC :-)
So a split brain situation is likely to happen sometime.
So for 100% data integrity I think it is best to let the cluster freeze
itself...
Best Regar
it also uses a TCP/IP connection to fence...
If that network is up you could also just add cluster communication over this
network ;-)
So in my opinion it does not add much value over node-fencing...
Best Regards,
Robert van Leeuwen
___
Pacemaker
> In a simple case, we want to prevent two nodes using a shared file-system
> without coordination. If they loose contact with each other and can't
> coordinate their access, one must die. (Even if you make it hard for them to
> loose contact by providing multiple communication channels, you h
On Tue, 30 Mar 2010 13:34:43 +0800, Yan Gao wrote:
> Hi,
>
> On 03/29/10 18:06, Robert wrote:
>> Hello,
>>
>> I am using the Madkiss repo for Debian Lenny. The management client
>> seems to be missing though. I tried to build the package
>> Pacemaker-Python-GUI-0f1490eaa8d8 but it fails:
>>
>> S
15 matches
Mail list logo
