No problem. 6 of the tests fail because we failed to revalidate the flows fast enough.
e.g the pktact priority tests add flow in_port=1, match M actions=output:2 with priority 250 sends a packet and verifies its egress port is 2 add flow in_port=1, match M actions=output:3 with priority 1250 sends a packet and verifies its egress port is 3 (which fails) The flows gets eventually revalidated, so I do not see this as an issue. Also note that if both flows were added together and then we send a packet, it would use the second flow as expected. This is the reason the sleep 1 sec was fixing the issue. To identify all the tests failing for the same reason I commented out the poll_block() at the end of revalidate loop so flows get revalidated as soon as we add a new flow. That eliminates 6 of the above tests. That leaves me with FAIL: pktact.DirectLlcPackets FAIL: flow_expire.FlowExpire FAIL: detailed_contr_sw_messages.DeleteEmerFlow FAIL: flow_stats.SingleFlowStats FAIL: counters.LookupMatchedCount to debug. Thanks. On Wed, Dec 3, 2014 at 5:44 PM, Ben Pfaff <b...@nicira.com> wrote: > Thanks a lot for picking this up! It always makes me unhappy to fail > test cases without knowing why, but I don't have time to look at the > details this week. > > It's very interesting that it's timing-related; that's a good hint > already. > > On Wed, Dec 03, 2014 at 11:25:11AM -0800, Madhu Challa wrote: > > Its interesting to see that some of these tests pass when run under pdb > or > > if I put a sleep(1) between verify flow and install of new flow. That > tells > > me it has something to do with the timing. Here are the tests that fail. > > > > pktact.WildcardPriorityWithDelete ... FAIL > > pktact.WildcardPriority ... FAIL > > pktact.FloodMinusPort ... FAIL > > pktact.SingleWildcardMatchPriority ... FAIL > > pktact.DirectLlcPackets ... FAIL > > flow_expire.FlowExpire ... FAIL > > detailed_contr_sw_messages.ModifyAction ... FAIL > > detailed_contr_sw_messages.Outport2 ... FAIL > > detailed_contr_sw_messages.DeleteEmerFlow ... FAIL > > detailed_contr_sw_messages.StrictModifyAction ... FAIL > > flow_stats.SingleFlowStats ... FAIL > > > > The pktact ones seem to be fine when I run under pdb which kind of makes > it > > difficult to debug. > > > > I will debug further and update later this week. > > > > Thanks. > > > > On Wed, Nov 26, 2014 at 2:29 PM, Ben Pfaff <b...@nicira.com> wrote: > > > > > I received an off-list report that OVS 2.3.0 fails a number of OFtest > > > test cases. I don't know whether they are important failures or not. > > > Anyone have some time to run a "make check-oftest" against the latest > > > version of OFtest and do a little investigation? Much appreciated. > > > > > > Thanks, > > > > > > Ben. > > > _______________________________________________ > > > dev mailing list > > > dev@openvswitch.org > > > http://openvswitch.org/mailman/listinfo/dev > > > > _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
