Did you pull the latest code, or that commit (May 9th)? I couldn't build packages from that May 9th snapshot. I was getting a failed test: -------- 561: ofproto - flow monitoring pause and resume FAILED (ofproto.at:1721 ) ofproto-dpif -------- which might be related, and a configure error. -------- make: *** [build-arch-stamp] Error 1 -------- FWIW I got the same failed test and error with the latest snapshot too.
BUT... building without the test seemed to work, and here's what I see: It looks like it's working much better now, with one exception: When a higher priority entry is installed, the counters from the low-priority one are not being transferred or reset (which is great), but I think that the counters *from the last time the lower-priority one was updated* (<=1 second) are being added to the newly installed higher-priority one. Does this make sense? If that hypothesis is true (I have some data to back it), can anything be done about that?
_______________________________________________ discuss mailing list discuss@openvswitch.org http://openvswitch.org/mailman/listinfo/discuss
