<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=us-ascii"><meta name=Generator content="Microsoft Word 14 (filtered medium)"><style></style></head><body lang=EN-US link=blue vlink=purple><div class=WordSection1>> Stonith = fencing, so the same thing<o:p></o:p><o:p> </o:p>But the name is cooler! “Shoot the other node in the head”, that’s awesome.<o:p></o:p><o:p> </o:p><o:p> </o:p><o:p> </o:p><o:p> </o:p>Jamison Maxwell Sr. Systems Administrator HD Supply - Facilities Maintenance<o:p></o:p><o:p> </o:p>From: linux-cluster-bounces@redhat.com [mailto:linux-cluster-bounces@redhat.com] On Behalf Of emmanuel segura Sent: Thursday, February 28, 2013 8:36 AM To: linux clustering Subject: Re: [Linux-cluster] Cannot connect to rgmanager<o:p></o:p><o:p> </o:p>Stonith = fencing, so the same thing For use corosync with udp look here <a href="http://www.thatsgeeky.com/2011/12/installing-corosync-on-ec2/">http://www.thatsgeeky.com/2011/12/installing-corosync-on-ec2/</a><o:p></o:p><div>2013/2/27 Maxwell, Jamison [HDS] <<a href="mailto:JMaxwell@pbp1.com" target="_blank">JMaxwell@pbp1.com</a>><o:p></o:p><div><div>Figured it out! I wasn’t allowing multicast on the vlan interface on my switch. I never even thought about this as an issue because I’m not routing, the nodes are right next to each other. However, tcpdump showed that I was sending tons of multicast, but not receiving any. As soon as I enabled multicast on the vlan interface everything came up.<o:p></o:p> <o:p></o:p>Also, in frustration I dropped cman, rgmanager, modclusterd in favor of corosync + pacemaker, but I believe I would have been as successful with either option. I think I prefer stonith to fencing, though.<o:p></o:p> <o:p></o:p>Lastly, is there a way to use unicast? I realize that multicast would be greatly preferable in 3+ node clusters, but in this two node it would be easier to use unicast.<o:p></o:p><div> <o:p></o:p> <o:p></o:p> <o:p></o:p> <o:p></o:p>Jamison Maxwell Sr. Systems Administrator HD Supply - Facilities Maintenance<o:p></o:p> <o:p></o:p></div>From: <a href="mailto:linux-cluster-bounces@redhat.com" target="_blank">linux-cluster-bounces@redhat.com</a> [mailto:<a href="mailto:linux-cluster-bounces@redhat.com" target="_blank">linux-cluster-bounces@redhat.com</a>] On Behalf Of emmanuel segura Sent: Sunday, February 24, 2013 1:32 PM<o:p></o:p><div><div> To: linux clustering Subject: Re: [Linux-cluster] Cannot connect to rgmanager<o:p></o:p></div></div><div><div> <o:p></o:p>Hello Sorry For my late reply, try to configure your fence devices and after start the rgmanager service<o:p></o:p><div>2013/2/19 Maxwell, Jamison [HDS] <<a href="mailto:JMaxwell@pbp1.com" target="_blank">JMaxwell@pbp1.com</a>><o:p></o:p><div><div>Yes, rgmanager, cman, ricci, and modclustered are started and start automatically in run levels three through five…<o:p></o:p> <o:p></o:p> <o:p></o:p> <o:p></o:p> <o:p></o:p><div>Jamison Maxwell Sr. Systems Administrator<o:p></o:p></div>HD Supply - Facilities Maintenance<o:p></o:p> <o:p></o:p>From: <a href="mailto:linux-cluster-bounces@redhat.com" target="_blank">linux-cluster-bounces@redhat.com</a> [mailto:<a href="mailto:linux-cluster-bounces@redhat.com" target="_blank">linux-cluster-bounces@redhat.com</a>] On Behalf Of emmanuel segura Sent: Tuesday, February 19, 2013 11:03 AM To: linux clustering Subject: Re: [Linux-cluster] Cannot connect to rgmanager<o:p></o:p><div><div> <o:p></o:p>did you started rgmanager?<o:p></o:p><div>2013/2/19 Maxwell, Jamison [HDS] <<a href="mailto:JMaxwell@pbp1.com" target="_blank">JMaxwell@pbp1.com</a>><o:p></o:p><div><div>I am attempting to create a two node cluster where the only resource required would be a shared IP address, however, after a couple of attempts I continue to fail. I have followed a guide located at <a href="http://www.openlogic.com/wazi/bid/188071/" target="_blank">http://www.openlogic.com/wazi/bid/188071/</a> . Everything appears to work fine until I get to the point where I actually add the IP address resource, both cluster nodes appear as online and quorate and the configuration validates, but will not enable the new resource. Below I am including any information that I think may be relevant, but feel free to ask for more.<o:p></o:p> <o:p></o:p>===========================================<o:p></o:p>[root@ hostname]# cat /etc/cluster/cluster.conf<o:p></o:p><?xml version="1.0"?><o:p></o:p><cluster config_version="4" name="hacluster"><o:p></o:p> <cman expected_votes="1" two_node="1"/><o:p></o:p> <clusternodes><o:p></o:p> <clusternode name="actual name of node" nodeid="1" votes="1"><o:p></o:p> <fence><o:p></o:p> <method name="single"/><o:p></o:p> </fence><o:p></o:p> </clusternode><o:p></o:p> <clusternode name=" actual name of node " nodeid="2" votes="1"><o:p></o:p> <fence><o:p></o:p> <method name="single"/><o:p></o:p> </fence><o:p></o:p> </clusternode><o:p></o:p> </clusternodes><o:p></o:p> <fencedevices/><o:p></o:p> <rm><o:p></o:p> <failoverdomains/><o:p></o:p> <resources/><o:p></o:p> <service autostart="1" exclusive="0" name="IP" recovery="relocate"><o:p></o:p> <ip address="actual shared IP address" monitor_link="on" sleeptime="10"/><o:p></o:p> </service><o:p></o:p> </rm><o:p></o:p></cluster><o:p></o:p>===========================================<o:p></o:p> <o:p></o:p>[root@ hostname]# clusvcadm -e IP<o:p></o:p>Local machine trying to enable service:IP...Could not connect to resource group manager<o:p></o:p>===========================================<o:p></o:p> <o:p></o:p>[root@ hostname]# strace clusvcadm -e IP<o:p></o:p>…<o:p></o:p>connect(5, {sa_family=AF_FILE, path="/var/run/cluster/<a href="http://rgmanager.sk" target="_blank">rgmanager.sk</a>"}, 110) = -1 ENOENT (No such file or directory)<o:p></o:p>close(5) = 0<o:p></o:p>write(1, "Could not connect to resource gr"..., 44Could not connect to resource group manager<o:p></o:p>) = 44<o:p></o:p>exit_group(1) = ?<o:p></o:p>===========================================<o:p></o:p> <o:p></o:p>I would most like to call your attention to the line “write(1, " connect(5, {sa_family=AF_FILE, path="/var/run/cluster/<a href="http://rgmanager.sk" target="_blank">rgmanager.sk</a>"}, 110) = -1 ENOENT (No such file or directory)”. There was also a someone who mailed this list with what appears to be the same problem, however, no issue is present in the conversation. The topic is located here: <a href="http://www.redhat.com/archives/linux-cluster/2012-August/msg00156.html" target="_blank">http://www.redhat.com/archives/linux-cluster/2012-August/msg00156.html</a> . <o:p></o:p> <o:p></o:p>This is version 6.3, no iptables and no selinux until I can get this working. I greatly appreciate any assistance that can be offered.<o:p></o:p> <o:p></o:p> <o:p></o:p>Jamison Maxwell Sr. Systems Administrator HD Suppy - Facilities Maintenance<o:p></o:p> <o:p></o:p></div></div> -- Linux-cluster mailing list <a href="mailto:Linux-cluster@redhat.com" target="_blank">Linux-cluster@redhat.com</a> <a href="https://www.redhat.com/mailman/listinfo/linux-cluster" target="_blank">https://www.redhat.com/mailman/listinfo/linux-cluster</a><o:p></o:p></div> -- esta es mi vida e me la vivo hasta que dios quiera <o:p></o:p></div></div></div></div> -- Linux-cluster mailing list <a href="mailto:Linux-cluster@redhat.com" target="_blank">Linux-cluster@redhat.com</a> <a href="https://www.redhat.com/mailman/listinfo/linux-cluster" target="_blank">https://www.redhat.com/mailman/listinfo/linux-cluster</a><o:p></o:p></div> -- esta es mi vida e me la vivo hasta que dios quiera <o:p></o:p></div></div></div></div> -- Linux-cluster mailing list <a href="mailto:Linux-cluster@redhat.com">Linux-cluster@redhat.com</a> <a href="https://www.redhat.com/mailman/listinfo/linux-cluster" target="_blank">https://www.redhat.com/mailman/listinfo/linux-cluster</a><o:p></o:p></div> -- esta es mi vida e me la vivo hasta que dios quiera <o:p></o:p></div></body></html>