14:05:07 #startmeeting OF plugin meeting 14:05:07 Meeting started Tue Sep 16 14:05:07 2014 UTC. The chair is abhijitkumbhare. Information about MeetBot at http://ci.openstack.org/meetbot.html. 14:05:07 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:05:07 The meeting name has been set to 'of_plugin_meeting' 14:05:41 Hi kamal_rameshan vishnoianil oflibMichal 14:05:55 Hi all 14:05:57 hi 14:06:08 oflibMichal - is michal_rehak around? 14:06:24 yes he is 14:06:30 he will be online in a sec 14:06:41 OK 14:07:14 oflibMichal: sent an email with updated info on the NPE, fwiw 14:07:41 There is a central page now for the bugs for all the projects - the OF plugin bugs are at: https://wiki.opendaylight.org/view/OpenDaylight_Bugs#openflowplugin 14:08:03 Or actually from that link: https://bugs.opendaylight.org/buglist.cgi?bug_severity=blocker&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&bug_severity=trivial&bug_status=UNCONFIRMED&bug_status=CONFIRMED&bug_status=IN_PROGRESS&bug_status=WAITING_FOR_REVIEW&columnlist=product%2Ccomponent%2Cassigned_to%2Cbug_status%2Cresolution%2Cshort_desc%2Cchangeddate%2Ccf_target_milestone&f1=cf_target_milest 14:08:03 one&list_id=16708&n1=1&o1=substring&product=openflowplugin&query_format=advanced&resolution=---&v1=Lithium 14:08:48 OK best to use the link https://wiki.opendaylight.org/view/OpenDaylight_Bugs#openflowplugin and navigate to the Helium bugs query 14:08:49 tbachman: ok, i will check it 14:08:56 oflibMichal: thx! 14:09:49 There are a few new bugs sent by Peter Gubka 14:10:00 The first one is: https://bugs.opendaylight.org/show_bug.cgi?id=1765 14:10:59 Actually before we embark on those - I think we should also look at which ones are a must fix for Helium - as there is only 1.5 weeks left 14:11:49 So while looking at bugs individually - lets also decide whether we can defer it to Lithium 14:12:12 So https://bugs.opendaylight.org/show_bug.cgi?id=1765 14:12:15 tbachman: Action [_action=NxActionOutputRegNodesNodeTableFlowApplyActionsCase [_nxOutputReg=NxOutputReg [_maxLen=65535, _src=Src [_ofsNbits=31, 14:12:26 yeah — that ofsNbits, right? 14:12:36 java.lang.NullPointerException: nullat org.opendaylight.openflowjava.nx.codec.action.RegLoadCodec.serialize(RegLoadCodec.java:32) ~[na:na] 14:12:37 is that correct? 14:12:57 I think that’s what we’re trying to set it to (31) 14:13:18 it looks like you want to serialize Outputreg action ... but it goes into regload action codec 14:13:54 What is Peter trying to say on this bug? 14:14:10 OK bad HTTP request 14:14:46 I think its breaking at the restconf layer 14:14:47 Is it actually a bad request? 14:14:47 looks like update is not working 14:14:58 i.e. bad xml 14:15:09 xml looks ok 14:15:25 what about the "strict" part? 14:15:53 "Schema node "strict" was not found in module" 14:15:58 It might be caused by the POST method - I believe that rest specification prescribes to use PUT for update 14:16:52 OK - michal_rehak shall we ask Peter to try out with PUT? 14:17:09 hi is already gone for today 14:17:21 I will try this xml 14:17:23 OK - I will put that in the bug 14:17:26 OK 14:17:56 Next one: https://bugs.opendaylight.org/show_bug.cgi?id=1840 14:18:27 oflibMichal: so, to make sure I understand — the flow that I dumped looks bad? 14:19:46 This one is probably not causing issues - right? things have been stored different in operational than config 14:20:06 abhijitkumbhare: I can take 1840 14:20:10 OK 14:20:31 abhijitkumbhare: done 14:20:43 Great 14:21:06 Next one: https://bugs.opendaylight.org/show_bug.cgi?id=1846 14:22:25 abhijitkumbhare: I can take that one as well 14:22:30 some flows in the test provider not working - f26 and f27 14:22:32 OK 14:22:40 thx mbobak 14:22:54 abhijitkumbhare: u r welcome ;-) 14:23:15 michal_rehak - next one is https://bugs.opendaylight.org/show_bug.cgi?id=1759 14:24:06 This is one of the VTN bugs 14:24:25 this is partially caused by bug in AD-SAL 14:24:33 OK 14:25:22 are u advocating him to directly use MD-SAL for this case? from your comment looks like it 14:27:33 so michal_rehak will keep working on this 1759 - right? we can move to the next one 14:29:28 I have asked Luis an update for https://bugs.opendaylight.org/show_bug.cgi?id=1468 14:29:39 but Ed told us not to change AD-SAL 14:29:55 so this is ready, just waiting for merge 14:29:59 OK - michal_rehak 14:31:08 but will VTN hit the issue or are they avoiding it by using MD-SAL? 14:32:11 for the new plugin is made default VTN is using AD-SAL "front-end" then adapters and md-sal 14:32:45 we fixed adapters so that TOS value from AD-SAL is handled as DSCP (subset of TOS) 14:34:51 so will you be pushing the code change? 14:35:04 already pushed 14:35:16 OK - great 14:35:24 just needs to be merged? 14:35:55 https://git.opendaylight.org/gerrit/#/c/11132/ 14:35:55 https://git.opendaylight.org/gerrit/11114 14:35:57 yes 14:36:01 one is in controller 14:37:04 OK - can you add the info to the bug? 14:38:18 So folks - the following to me looks like a candidate for Lithium: https://bugs.opendaylight.org/show_bug.cgi?id=1725 . What do you folks think? 14:39:13 Its a protocol spec conformance issue - good to fix - but probably Lithium timeframe looks better 14:39:28 any thoughts or objections? 14:40:33 OK - moving it to Lithium 14:43:46 kamal_rameshan - do you have the status on https://bugs.opendaylight.org/show_bug.cgi?id=1509 ? 14:44:30 i need to investigate on this. Was reported by Learning Switch project 14:44:37 OK - we had put this to Lithium - will change the deadline 14:44:45 ok 14:45:20 How about https://bugs.opendaylight.org/show_bug.cgi?id=1540 kamal_rameshan ? 14:46:29 well this one i have not seen it happening. Will talk to moiz today and get it closed, unless he objects 14:46:53 OK - kamal_rameshan . mbobak - any thoughts on https://bugs.opendaylight.org/show_bug.cgi?id=1764 14:47:32 mbobak: Bug 1764 is pushing of default config for OF switch should be moved into separate module 14:48:00 Do you think we should do this at the late stage or in Lithium? 14:49:58 abhijitkumbhare: I suppose Lithium it is, there are some things in controller that need to be done before OF switch config can be separated in openflowplugin 14:51:17 How about https://bugs.opendaylight.org/show_bug.cgi?id=1587 ? 14:51:50 abhijitkumbhare: that's still waiting for 1484 to be completed 14:52:04 Yes - saw that comment 14:53:24 mbobak : https://bugs.opendaylight.org/show_bug.cgi?id=1667 14:53:46 abhijitkumbhare: that just needs to be merged 14:53:50 You have the patch there - you want michal_rehak to review it? 14:54:16 abhijitkumbhare: oh yes, that woul help much 14:55:07 so michal_rehak - can you review https://git.opendaylight.org/gerrit/#/c/10943/ :) ? 14:56:05 rebasing 14:56:40 oflibMichal tbachman - I will just assign https://bugs.opendaylight.org/show_bug.cgi?id=1907 to oflibMichal. After initial investigation - we can figure out - is that OK? 14:56:49 this will turn on statistics exposed via jmx in integration/distributions/base 14:57:00 abhijitkumbhare: that works for me :) 14:57:15 abhijitkumbhare: this will be a bug in openflowplugin translation or in controller 14:57:54 looks like last outputReg action is being serialized by regLoadCodec 14:57:58 which is not good 14:58:06 i believe it is not ofjava bug 14:58:29 can you add the info to the bugs 14:58:35 sure 14:58:38 I mean to the bug 14:59:04 we are out of time - we can continue/finish tomorrow 14:59:09 are any of your guys attending the design summit? 14:59:39 the topic that folks are asking on in the simultaneous release meeting is the testing projects are doing 15:00:28 kamal_rameshan - I am attending the first day (till around just before 4 pm) 15:01:03 michal_rehak & oflibMichal - are you folks attending? 15:01:28 oh ok. thx abhijitkumbhare 15:01:54 I will try 15:01:58 I am travelling that evening - so will miss the second day 15:02:01 I have meeting in 28 mins 15:02:24 talking about the design forum in santa clara michal_rehak 15:03:03 OK - will do end meeting 15:03:10 #endmeeting