That… Am 04.05.2012 um 20:57 schrieb Ben Pfaff:
> On Fri, May 04, 2012 at 08:42:25PM +0200, Oliver Francke wrote: >> Does one have any experiences with workload for ca. 150 VMs, what >> das ovs-vswitchd say to this, with at least one rule for every VM? > > It should be fine. We regularly use 10000 rules or more. … sounds cool. We just have our nodes running with 20-50% of ( one) CPU-usage. And another story, with 98% showing lines ala: May 4 21:06:57 fcmsnode10 ovs-vswitchd: 3827527|poll_loop|WARN|Dropped 635 log messages in last 1 seconds (most recently, 1 seconds ago) due to excessive rate May 4 21:06:57 fcmsnode10 ovs-vswitchd: 3827528|poll_loop|WARN|wakeup due to [POLLIN] on fd 16 (unknown anon_inode:[eventpoll]) at lib/dpif-linux.c:1191 (98% CPU usage) May 4 21:06:57 fcmsnode10 ovs-vswitchd: 3827529|poll_loop|WARN|wakeup due to [POLLIN] on fd 16 (unknown anon_inode:[eventpoll]) at lib/dpif-linux.c:1191 (98% CPU usage) --- 8-< --- in the syslog with 1.4.0 ever so often. Dunno yet, what's going on… But, another story. Thnx for now, Oliver. > _______________________________________________ > discuss mailing list > [email protected] > http://openvswitch.org/mailman/listinfo/discuss _______________________________________________ discuss mailing list [email protected] http://openvswitch.org/mailman/listinfo/discuss
