On Wed, 2009-05-20 at 14:59 +0100, Chrissie Caulfield wrote: > > So, I can fix it I think with poll(0), but if an admin kills qdiskd with > > SIGKILL (or any other fatal signal), restarting qdiskd will prevent > > correct vote registration (though as I have found out, polling still > > works great). > > When qdiskd restarts, if you get EBUSY from _register then you could > deregister and reregister with the new information. Well... cman_poll_quorum_device(x, 0), then unregister, then register. > There's an argument here for a cman API call to change the number of > votes associated with the quorum disk though ... what do you think ? Either one works for me. I'm happy to fix up qdiskd, and I don't think it's unreasonable to require clean de-register / re-register to adjust vote count. -- Lon -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
