
<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div>Memory defect?<br><br>Stefan<div><br></div><div>Excuse my typo s<span style="font-size: 13pt;">ent from my mobile phone.</span></div></div><div><br>Am 14.09.2014 um 15:41 schrieb Alexandre DERUMIER <<a href="mailto:aderumier@odiso.com">aderumier@odiso.com</a>>:<br><br></div><blockquote type="cite"><div><blockquote type="cite"><blockquote type="cite"><span>I am curios - you have done that on all nodes, or only on the failing 2 nodes?</span><br></blockquote></blockquote><span></span><br><span>Yes, I need to do it on all nodes.</span><br><span></span><br><span></span><br><span></span><br><span>I have done more invesgations, and now I can reproduce the problem 100%</span><br><span></span><br><span>The problem seem to come from a specific node: kvm11</span><br><span></span><br><span>When I start cman on this node,</span><br><span></span><br><span>I have :</span><br><span>pmxcfs[31484]: [status] notice: cpg_send_message retry XX</span><br><span></span><br><span>on all other nodes</span><br><span></span><br><span>Same hardware than other nodes, I need to check the network layer.</span><br><span></span><br><span></span><br><span>On the faulty node, I see also some pmxcfs segfaults in dmesg</span><br><span></span><br><span>[976776.602200] pmxcfs[3130]: segfault at 7ff1dcadef08 ip 00007ff1dcadef08 sp 00007fffd89cfe68 error 15</span><br><span>[977517.260211] pmxcfs[4947]: segfault at 1956b00 ip 0000000001956b00 sp 00007ffff3b109e8 error 15</span><br><span>[980494.722550] pmxcfs[15205]: segfault at 7f712457ef08 ip 00007f712457ef08 sp 00007fff4a916668 error 15</span><br><span></span><br><span></span><br><span></span><br><span>----- Mail original ----- </span><br><span></span><br><span>De: "Dietmar Maurer" <<a href="mailto:dietmar@proxmox.com">dietmar@proxmox.com</a>> </span><br><span>À: "Alexandre DERUMIER" <<a href="mailto:aderumier@odiso.com">aderumier@odiso.com</a>> </span><br><span>Cc: <a href="mailto:pve-devel@pve.proxmox.com">pve-devel@pve.proxmox.com</a> </span><br><span>Envoyé: Dimanche 14 Septembre 2014 12:53:45 </span><br><span>Objet: RE: [pve-devel] corosync problems - need help </span><br><span></span><br><blockquote type="cite"><span>Ok,I finally solved, </span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>kill -9 dlm_controld </span><br></blockquote><blockquote type="cite"><span>kill -9 corosync -f </span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>and service cman start </span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Now all is working fine again. </span><br></blockquote><span></span><br><span>I am curios - you have done that on all nodes, or only on the failing 2 nodes? </span><br><span>_______________________________________________</span><br><span>pve-devel mailing list</span><br><span><a href="mailto:pve-devel@pve.proxmox.com">pve-devel@pve.proxmox.com</a></span><br><span><a href="http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel">http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel</a></span><br></div></blockquote></body></html>