On Tue, Jul 29, 2014 at 03:01:03PM -0700, Alex Wang wrote: > Hey Ben, > > Sorry for the confusion, > > On Tue, Jul 29, 2014 at 2:52 PM, Ben Pfaff <b...@nicira.com> wrote: > > > On Tue, Jul 29, 2014 at 11:30:12AM -0700, Alex Wang wrote: > > > I think we should not check 'ofconn's for OFCONN_SERVICE... This > > > change broke several tests: > > > > > > *Please send `tests/testsuite.log' and all information you think might > > > help:* > > > > > > * To: <b...@openvswitch.org <b...@openvswitch.org>>* > > > > > > * Subject: [openvswitch 2.3.0] testsuite: 737 739 747 817 failed* > > > > I'm a little confused now. > > > > I'm not sure why you addressed this to Simon. I don't think he has any > > recent commits. > > > > > 1. The test failure is caused by this RFC patch, > > 2. connmgr_wants_packet_in_on_miss() was added by Simon. I think this > function should only check for controller connections. And want to > confirm > it with Simon. > > 3. unit test 737 corresponds to this: (which is also added by Simon) > > 527ae97e (Simon Horman 2014-03-13 15:52:54 +0900 1565) > 527ae97e (Simon Horman 2014-03-13 15:52:54 +0900 1566) > AT_SETUP([ofproto-dpif - table-miss flow (OpenFlow 1.0)]) > 527ae97e (Simon Horman 2014-03-13 15:52:54 +0900 1567) > OVS_VSWITCHD_START([dnl > 527ae97e (Simon Horman 2014-03-13 15:52:54 +0900 1568) add-port br0 > p1 -- set Interface p1 type=dummy > 527ae97e (Simon Horman 2014-03-13 15:52:54 +0900 1569) ]) > > > I think that's why I sent email asking him.
I don't wish to add to any confusion but I will none the less make a comment on this change which you may choose to ignore if you like. The motivation for adding connmgr_wants_packet_in_on_miss() is to help implement per-OpenFlow version behaviour of packet_in messages. So if monitors should have per-OpenFlow version behaviour then I am not sure your patch is correct. My recollection is that this behaviour of monitors seemed useful for the purpose of testing the per-OpenFlow version behaviour of packet in messages. But it was a while ago so my recollection may not be accurate. > > From the [openvswitch 2.3.0] it looks like you ran the tests against > > branch-2.3. I don't see the same failures locally when I run the tests > > on branch-2.3 (commit 9067b54) or master (commit 57fa816). Are these > > intermittent failures or do you see them every time? Do they appear > > only after the two recent additional commits that you pushed? > > > > > 4. I saw them every time. Those failures are caused only by this RFC > patch (unrelated to the two recent commits I just pused) > > 5. The cause of failure is that connmgr_wants_packet_in_on_miss() now > checks only the 'ofconn's of type OFCONN_PRIMARY. so, the 'ofconn' > for 'ofctl monitor' is ignored, and we always drop the pkt_ins. > > Thanks, > Alex Wang, _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
