Re: [Pacemaker] Configuring 3rd Node as Quorum Node in 2 Node Cluster

2012-01-10 Thread Florian Haas
On Wed, Jan 11, 2012 at 1:44 AM, Andrew Beekhof wrote: > On Wed, Jan 11, 2012 at 3:30 AM, Andrew Martin wrote: >> 3. Limit the DRBD, nfs, and smbd resources to only node1 and node2 by adding >> a location rule for the g_nfs group (which includes p_fs_drbd0 >> p_lsb_nfsserver p_exportfs_drbd0 p_ip

Re: [Pacemaker] Configuring 3rd Node as Quorum Node in 2 Node Cluster

2012-01-10 Thread Andrew Beekhof
On Wed, Jan 11, 2012 at 3:30 AM, Andrew Martin wrote: > Hello, > > I am creating a 2 node active/passive cluster for a fileserver (NFS and > Samba) using DRBD + Heartbeat + Pacemaker on Ubuntu 10.04 following this > guide: http://www.linbit.com/fileadmin/tech-guides/ha-nfs.pdf. I'd like to > enabl

Re: [Pacemaker] need cluster-wide variables

2012-01-10 Thread Arnold Krille
On Tuesday 10 January 2012 20:08:50 Dejan Muhamedagic wrote: > On Thu, Jan 05, 2012 at 04:59:13AM +, shashi wrote: > > So we have two probable options: > > 1. An work around to achieve this tri-state efficiently without changing > > pacemaker internals. > > 2. Modify pacemaker to add this tri-s

Re: [Pacemaker] need cluster-wide variables

2012-01-10 Thread Dejan Muhamedagic
Hi, On Thu, Jan 05, 2012 at 04:59:13AM +, shashi wrote: > > Its not something we had planned, this is the first time anyone has asked. > > I'd be happy to help point someone on your side in the right > > direction, but if I have to do all of it, you might be waiting a > > while. > > > > Can y

[Pacemaker] Configuring 3rd Node as Quorum Node in 2 Node Cluster

2012-01-10 Thread Andrew Martin
Hello, I am creating a 2 node active/passive cluster for a fileserver (NFS and Samba) using DRBD + Heartbeat + Pacemaker on Ubuntu 10.04 following this guide: http://www.linbit.com/fileadmin/tech-guides/ha-nfs.pdf . I'd like to enable quorum and STONITH to ensure data integrity. Thus I want

Re: [Pacemaker] Cannot Create Primitive in CRM Shell

2012-01-10 Thread Andrew Martin
Hi Jake, No, unfortunately I did not record the status of libglib2.0-0 before installing. However, I do have the list of packages that were installed when I did "apt-get install drbd8-utils heartbeat pacemaker" from the PPA and it is not listed among them: binutils build-essential cluster-g

Re: [Pacemaker] Invoking "crm node standby" command on active command leads to stopping of resource on both Active and standby node

2012-01-10 Thread Andreas Kurz
Hello, On 01/10/2012 09:47 AM, neha chatrath wrote: > Hello, > I am using a cluster with following configuration: > > [root@MCG1 neha]# crm configure show > node $id="0686a4d1-c9de-4334-8d33-1a9f6f0755dd" ggns2mexsatsdp22 > node $id="76246d46-f0e4-4ba8-9179-d60aa7c697c8" ggns2mexsatsdp23 > node

[Pacemaker] Invoking "crm node standby" command on active command leads to stopping of resource on both Active and standby node

2012-01-10 Thread neha chatrath
Hello, I am using a cluster with following configuration: [root@MCG1 neha]# crm configure show node $id="0686a4d1-c9de-4334-8d33-1a9f6f0755dd" ggns2mexsatsdp22 node $id="76246d46-f0e4-4ba8-9179-d60aa7c697c8" ggns2mexsatsdp23 node $id="9d59c9e6-24e0-4684-94ab-c07af7e7a2f0" mcg1 \ attribute

Re: [Pacemaker] [ANNOUNCE] The Booth Cluster Ticket Manager - part of multi-site support in pacemaker

2012-01-10 Thread 李泰勲
Hi Jiaju Thanks for your reply. I confirmed source code. And I have some question about booth. This is my simple test note 1.I make installed booth with Pacemaker1.1 in RHEL6.1 2.I have two-site(siteA,B) cluster and one-arbitrator 3.Each site that is consists of one node have dummy resource 4.b