On Tue, 2009-10-13 at 11:02 +0200, Andrew Beekhof wrote: > ACK - with my reworked pcmk_shutdown() function the stack is now able > to shut down. > If you could apply this and r2520 to flatiron I'd be very grateful. > > -- Andrew > Andrew
this patch still needa a bit of work around individual service unload (corosync-cfgtool -u cpg -v 0 for example). I'll sort that out soon. Regards -steve > On Mon, Oct 12, 2009 at 10:58 PM, Steven Dake <[email protected]> wrote: > > Currently exec_exit_fn doesn't allow the possibility to continue work at > > a later time. Service engines can deadlock if they block (pacemaker > > specifically). To avoid this situation, we allow exec_exit_fn to return > > -1 when it has work it needs to continue. When it is done, it returns 0 > > and will be unloaded. > > > > Still a work in progress (ie individual service engine unload doesn't > > work from corosync-cfgtool). > > > > Regards > > -steve > > > > _______________________________________________ > > 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
