Re: [Pacemaker] Multi-site support in pacemaker (tokens, deadman, CTR)

2011-05-16 Thread Andrew Beekhof
On Sun, May 15, 2011 at 11:12 PM, Lars Marowsky-Bree wrote: > On 2011-05-13T08:54:19, Andrew Beekhof wrote: > >> > No. Because starting/acquiring the resources would still depend on a >> > "local" fence of the other side. This is protected against by our >> > regular dependencies already. >> > >>

Re: [Pacemaker] Multi-site support in pacemaker (tokens, deadman, CTR)

2011-05-16 Thread Lars Marowsky-Bree
On 2011-05-16T09:55:13, Andrew Beekhof wrote: > >> Couldn't both sides start shooting each other until one looses the token? > > > > Not more than before, though I'd expect a side that freshly rebooted to > > come up w/o any token. > > > > In general, the possibility for a fencing deathmatch is n

Re: [Pacemaker] Active/Active firewall cluster, with load balancing

2011-05-16 Thread Michael Schwartzkopff
> Hi there! > > I'm looking for some light regarding active/active firewall clustering with > load balancing. > > I heard some references to ldirectord and LVS, but in the first's webpage ( > http://horms.net/projects/ldirectord/) there aren't any documentation, and > in the LVS site (http://www.

[Pacemaker] Error while performing crm attribute operation. and cl_status commands

2011-05-16 Thread Rakesh K
Hi All I had configured Mysql as multi-state resource in HA I had configured HA cluster consisting of two nodes and one Mysql multistate resource crm_attribute --type rsc_location --node mysql2 --get-value score non-option ARGV-elements: score scope=rsc_location value=(null) Error performi

Re: [Pacemaker] [pacemaker] Error while executing crm_attribute command from command line

2011-05-16 Thread Rakesh K
Andrew Beekhof writes: > > # CIB_file=./cib.xml crm_attribute -N mysql3 -n master-Mysql:0 -l > forever -v 3 -V > crm_attribute[23703]: 2011/05/10_08:42:59 info: Invoked: crm_attribute > -N mysql3 -n master-Mysql:0 -l forever -v 3 -V > crm_attribute[23703]: 2011/05/10_08:42:59 ERROR: Type ID does

[Pacemaker] Preventing auto-fail-back

2011-05-16 Thread Daniel Bozeman
For the life of me, I cannot prevent auto-failback from occurring in a master-slave setup I have in virtual machines. I have a very simple configuration: node $id="4fe75075-333c-4614-8a8a-87149c7c9fbb" ha2 \ attributes standby="off" node $id="70718968-41b5-4aee-ace1-431b5b65fd52" ha1 \

Re: [Pacemaker] Always reload/restart a resource

2011-05-16 Thread Michael Liebl
Am Freitag, den 13. Mai 2011 schrubte Vladislav: > > I'm looking for a possibility to always reload or restart a resource > > independent of the action (start/stop). > > I always thought that smbd reloads its config automatically whenever > it changes and is valid. Doesn't it? I am not sure if

Re: [Pacemaker] Always reload/restart a resource

2011-05-16 Thread Michael Liebl
Am Freitag, den 13. Mai 2011 schrubte Michael: > Hello List! > > I'm looking for a possibility to always reload or restart a resource > independent of the action (start/stop). Just to clarify, I am looking for something like the old heartbeat 1 behaviour. I had for example in my haresources "sys