Re: [Pacemaker] Cluster tools

2013-03-15 Thread matonb
Maxwell, Jamison [HDS] JMaxwell@... writes: Pssh complained about wanting python-2.5, but I wasn't going to downgrade from 2.6, so I skipped the dependency. I've tested several areas within the shell and all appears well... crmsh (with pssh) from the SUSE ha repo works just fine, the

Re: [Pacemaker] Cluster tools

2013-03-15 Thread matonb
Andrew Beekhof andrew@... writes: I believe this should make everyone happy: https://github.com/beekhof/pacemaker/commit/3c9275e Are you able to test it? I cloned the repo and used the pacemaker.spec.in file. I did have to tweak the spec file a bit to get it to build: %global

Re: [Pacemaker] crm resource cleanup / reprobe

2013-03-15 Thread Dejan Muhamedagic
Hi, On Thu, Mar 14, 2013 at 02:55:52PM +, Hinds, Luke (NSN - GB/Farnborough) wrote: Hi, What are the key differences between 'crm resource cleanup' and 'crm resource reprobe'? The first removes all status records of the resource. One consequence is that CRM orders a probe on all nodes

Re: [Pacemaker] mysql/drbd on wheezy active/passive setup issues

2013-03-15 Thread Dejan Muhamedagic
Hi, On Thu, Mar 14, 2013 at 12:21:04PM -0400, christopher barry wrote: Hi, I get this when I run ocf-tester: ocf-tester -n p_mysql /usr/lib/ocf/resource.d/heartbeat/mysql Beginning tests for /usr/lib/ocf/resource.d/heartbeat/mysql... /usr/sbin/ocf-tester: 214: /usr/sbin/ocf-tester:

Re: [Pacemaker] Cluster tools

2013-03-15 Thread Dejan Muhamedagic
Hi, On Thu, Mar 14, 2013 at 03:03:21PM -0400, Maxwell, Jamison [HDS] wrote: I was actually able to get the crm shell back onto a system that I had upgraded unawares of the removal of the tool. How bad an idea is that? I downloaded RPMs for crmsh-1.2.5-55 and pssh-1.4.3 and installed them.

Re: [Pacemaker] problem on moving resource

2013-03-15 Thread Dejan Muhamedagic
Hi, On Thu, Mar 14, 2013 at 05:35:54PM -0300, Charles Mean wrote: Hey Guys, I am running heartbeat 3.0.5 and pacemaker 1.1.7 with two hosts(host01, host02). The thing is, when I try to define host01 as the preferred by the following command: crm configure location prefer-host01

Re: [Pacemaker] resource modification and resource agent update

2013-03-15 Thread Dejan Muhamedagic
Hi, On Thu, Mar 14, 2013 at 11:48:16PM -0400, Matthew O'Connor wrote: Hi!! Two quick questions. 1. I have a resource that many other resources depend on. I need to modify this base resource, but if I modify this resource while it is online, will pacemaker restart it to effect the changes?

Re: [Pacemaker] resource modification and resource agent update

2013-03-15 Thread Leon Fauster
Am 15.03.2013 um 10:49 schrieb Dejan Muhamedagic deja...@fastmail.fm: Hi, On Thu, Mar 14, 2013 at 11:48:16PM -0400, Matthew O'Connor wrote: Hi!! Two quick questions. 1. I have a resource that many other resources depend on. I need to modify this base resource, but if I modify this

Re: [Pacemaker] resource modification and resource agent update

2013-03-15 Thread Vladislav Bogdanov
15.03.2013 13:52, Leon Fauster wrote: Am 15.03.2013 um 10:49 schrieb Dejan Muhamedagic deja...@fastmail.fm: Hi, On Thu, Mar 14, 2013 at 11:48:16PM -0400, Matthew O'Connor wrote: Hi!! Two quick questions. 1. I have a resource that many other resources depend on. I need to modify this base

Re: [Pacemaker] Cluster tools

2013-03-15 Thread Maxwell, Jamison [HDS]
Correct me if I am wrong, but I believe crm_mon and the other crm_* commands are provided by pacemaker-cli, not by crmsh. (At least in the RHEL repos.) The reason I wanted crmsh back so badly was because I had entered and documented several aliases to make managing the cluster more convenient

Re: [Pacemaker] Cluster tools

2013-03-15 Thread Lars Marowsky-Bree
On 2013-03-15T10:05:30, Maxwell, Jamison [HDS] jmaxw...@pbp1.com wrote: Is the HA community abandoning crmsh and why? The HA community is NOT abandoning crmsh. crmsh remains supported by the community, *and* on SLE HA. We are and will continue to develop it. RHT has made a decision not to

Re: [Pacemaker] resource modification and resource agent update

2013-03-15 Thread Dejan Muhamedagic
On Fri, Mar 15, 2013 at 04:16:52PM +0300, Vladislav Bogdanov wrote: 15.03.2013 13:52, Leon Fauster wrote: Am 15.03.2013 um 10:49 schrieb Dejan Muhamedagic deja...@fastmail.fm: Hi, On Thu, Mar 14, 2013 at 11:48:16PM -0400, Matthew O'Connor wrote: Hi!! Two quick questions. 1. I have

Re: [Pacemaker] Cluster tools

2013-03-15 Thread Maxwell, Jamison [HDS]
Thanks for the explanation! Jamison Maxwell Sr. Systems Administrator HD Supply - Facilities Maintenance -Original Message- From: Lars Marowsky-Bree [mailto:l...@suse.com] Sent: Friday, March 15, 2013 10:35 AM To: The Pacemaker cluster resource manager Subject: Re: [Pacemaker]

Re: [Pacemaker] problem on moving resource

2013-03-15 Thread Dejan Muhamedagic
On Fri, Mar 15, 2013 at 10:46:48AM +0100, Dejan Muhamedagic wrote: Hi, On Thu, Mar 14, 2013 at 05:35:54PM -0300, Charles Mean wrote: Hey Guys, I am running heartbeat 3.0.5 and pacemaker 1.1.7 with two hosts(host01, host02). The thing is, when I try to define host01 as the preferred

Re: [Pacemaker] resource modification and resource agent update

2013-03-15 Thread Vladislav Bogdanov
15.03.2013 18:38, Dejan Muhamedagic wrote: [ ... ] putting the resource in unmanaged mode, thought? No, monitors are still run in unmanaged mode. Global maintainance-mode=true disables them. Right. Probably the simplest option. Or just temporary delete 'op monitor' lines from affected

[Pacemaker] Monitor interval vs. timeout

2013-03-15 Thread Shawn Paul Smith
I have a few questions about monitors. * Is the timeout counted from each monitor attempt individually, or from the oldest attempt, or from the newest? * Does it make more sense to set the interval shorter than the timeout, or longer? * Would setting timeout and

[Pacemaker] Nagios in pacemaker

2013-03-15 Thread kaushal jani
hi all, i installed pacemaker 1.8.7 in two node two resource. res_nagios_1lsb:nagiosres_httpd_1 lsb:httpd resource move is ok but problem is when resource moved to primary/secondary node the nagios check_ping give me high response time with critical status,which is runing nice