<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div><div>On 19.03.2012, at 20:26, Florian Haas wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div>On Mon, Mar 19, 2012 at 8:14 PM, Mathias Nestler<br><<a href="mailto:mathias.nestler@barzahlen.de">mathias.nestler@barzahlen.de</a>> wrote:<br><blockquote type="cite">Hi everyone,<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">I am trying to setup an active/passive (2 nodes) Linux-HA cluster with corosync and pacemaker to hold a PostgreSQL-Database up and running. It works via DRBD and a service-ip. If node1 fails, node2 should take over. The same if PG runs on node2 and it fails. Everything works fine except the STONITH thing.<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Between the nodes is an dedicated HA-connection (10.10.10.X), so I have the following interface configuration:<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">eth0 eth1 host<br></blockquote><blockquote type="cite">10.10.10.251 172.10.10.1 node1<br></blockquote><blockquote type="cite">10.10.10.252 172.10.10.2 node2<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Stonith is enabled and I am testing with a ssh-agent to kill nodes.<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">crm configure property stonith-enabled=true<br></blockquote><blockquote type="cite">crm configure property stonith-action=poweroff<br></blockquote><blockquote type="cite">crm configure rsc_defaults resource-stickiness=100<br></blockquote><blockquote type="cite">crm configure property no-quorum-policy=ignore<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">crm configure primitive stonith_postgres stonith:external/ssh \<br></blockquote><blockquote type="cite"> params hostlist="node1 node2"<br></blockquote><blockquote type="cite">crm configure clone fencing_postgres stonith_postgres<br></blockquote><br>You're missing location constraints, and doing this with 2 primitives<br>rather than 1 clone is usually cleaner. The example below is for<br>external/libvirt rather than external/ssh, but you ought to be able to<br>apply the concept anyhow:<br><br><a href="http://www.hastexo.com/resources/hints-and-kinks/fencing-virtual-cluster-nodes">http://www.hastexo.com/resources/hints-and-kinks/fencing-virtual-cluster-nodes</a><br><br></div></blockquote><div><br></div><div>As is understood the cluster decides which node has to be stonith'ed. Besides this, I already tried the following configuration:</div><div><br></div><div><div>crm configure primitive stonith1_postgres stonith:ssh \</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>params hostlist="node1"</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>op monitor interval="25" timeout="10"</div><div>crm configure primitive stonith2_postgres stonith:ssh \</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>params hostlist="node2"</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>op monitor interval="25" timeout="10"</div><div>crm configure location stonith1_not_on_node1 stonith1_postgres \</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>-inf: node1</div><div>crm configure location stonith2_not_on_node2 stonith2_postgres \</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>-inf: node2</div><div><br></div><div>The result is the same :/</div></div><div><br></div><br><blockquote type="cite"><div>Hope this helps.<br>Cheers,<br>Florian<br><br></div></blockquote><br></div><div>Best</div><div>Mathias</div><br></body></html>