On 05/24/2010 02:51 PM, Vladislav Bogdanov wrote: > Hi all, > > Sorry for being out of "References", just subscribed. > > >> On Fri, 2010-05-21 at 16:19 +0200, Alain.Moulle wrote: >> >>> Hi >>> >>> These new releases of corosync do not start successfully on RHEL5 : >>> corosync-1.2.2-1.1.el5 >>> corosynclib-1.2.2-1.1.el5 >>> I 've joined the messages trace. >>> >>> whereas on same machines, these old ones works fine : >>> corosync-1.2.1-1.el5 >>> corosynclib-1.2.1-1.el5 >>> >>> I've reinstalled these old ones and it works fine again. >>> And ... I can't test furthermore with the new releases before around 10 >>> days. >>> >>> Regards >>> Alain >>> >>> >> Building from sources on rhel5, corosync starts properly. I didn't give >> pacemaker a go. >> >> could you provide more information: >> 1) where did you download the corosync rpms >> 2) Which version of RHEL are you running >> >> Then I can look into reproducing >> > I confirm that both 1.2.2 and 1.2.3 segfault on CentOS 5.5 when > pacemaker is enabled (this is critical, corosync alone starts just fine). > Tried with both clusterlabs 1.2.2-1.1 RPM and home-brew 1.2.3 RPM. > > Segfault is originated from exec/logsys.c:760, in strlen(rec->buffer) > > Can't post gdb output, console buffer is lost yet due to urgent downgrade. > >
Reproducible? can you run corosync-fplay and send the list the output. Please send your conf file. Thanks -steve > Best, > Vladislav > _______________________________________________ > 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
