On Mon, 2009-03-23 at 11:07 +0100, Fabio M. Di Nitto wrote: > On Sun, 2009-03-22 at 10:05 -0700, Steven Dake wrote: > > This patch is a forward port of the patch previously posted to rid > > ourselves of the race condition introduced when using totem token > > received callbacks. > > > > This adds two new apis to allow the scheduling of work for a later time > > (or destroying that request). > > Hi Steven, > > I am having problems with this patch. Tested on a 2 and 3 nodes cluster. > > - firing cpgbench will make one of the corosync processes spinning 100% > of the CPU > - corosync doesn't complete init properly on big endian (tested on ppc)
^^ the bigendian problem is not a problem. It was a fault on the test environment. > - shutdown blocks a lot of time > > so I am sorry but this is not good to go in. > > Fabio > > > _______________________________________________ > 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
