Just like said Lon, this idea is very easy and perfect for me ! No need to handle the cluster suite in this case :-) Because I wanna know if a node has stopped working, I'll do : on each node, a script watch a file on the gfs every minute, if cannot find it, stop httpd and send me a email. Because httpd will be down, the server will be remove from the servers pool. Simple and should work. Best of both work. Thanks Matthew and Lon for the responses !!! Patton, Matthew F, CTR, OSD-PA&E wrote: > > Classification: UNCLASSIFIED > > this doesnt solve your problem but why not just leave all httpd > processes running on all the boxes and put some smarts in the > loadbalancer such that if a page fetch comes up empty/invalid to mark > that server as offline. So if GFS goes wondering off, cluster services > doesnt have to figure if http is valid or not. Obviously cluster > services needs to figure out that GFS went screwy... > > ------------------------------------------------------------------------ > > -- > Linux-cluster mailing list > Linux-cluster@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/linux-cluster -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
