On Mon, 2010-02-08 at 16:10 +0900, [email protected] wrote: > Hi All, > > I built a cluster by the next combination. > cib.xml does not arrange it. > > * RHEL5.4(x86) on VMWare Esxi > * corosync 1.2.0 > * Pacemaker 1.0.6 and Pacemaker 1.0.7 > > However, only in a single node, Pacemaker does not become the stable > state(S_IDLE state). > In addition, as for this problem, interface does not occur in the case of > single. > (rrp_mode occurs in the case of active or passive.) > > Bug, this problem does not happen when I used corosync1.1.2. > > Is this a bug? > Is this problem already reported? > Hideo, I read through the thread.
I have not heard this problem reported before but I don't have vmware esxi. It is difficult to assess the situation without corosync logs. Do you have iptables on in your "host" or the os running your VMs? The host's iptables will block children virtual machines from communicating and cause corosync to appear faulty. try turning iptables off completely. Regards -steve Regards -steve > I show the link of the exchanges of the email with Mr. Andrew next. > > Please watch the link of this place in detail. > link : http://www.gossamer-threads.com/lists/linuxha/pacemaker/61103 > > Best Regards, > Hideo Yamauchi. > > _______________________________________________ > Openais mailing list > [email protected] > https://lists.linux-foundation.org/mailman/listinfo/openais _______________________________________________ Openais mailing list [email protected] https://lists.linux-foundation.org/mailman/listinfo/openais
