14:01:18 #startmeeting OFplugin Daily 14:01:18 Meeting started Mon Aug 11 14:01:18 2014 UTC. The chair is edwarnicke. Information about MeetBot at http://ci.openstack.org/meetbot.html. 14:01:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:18 The meeting name has been set to 'ofplugin_daily' 14:01:24 #topic Rollcall 14:01:28 #info edwarnicke 14:01:34 #info oflibMichal 14:01:34 OK folks.. please #info in 14:01:54 michal_rehak: Madhu_ jmedved are you guys here? 14:02:20 #info michal_rehak 14:02:31 morning 14:05:49 abhijitkumbhare: Error: Can't start another meeting, one is in progress. Use #endmeeting first. 14:05:56 #info abhijitkumbhare 14:06:13 Sorry folks - a little late 14:07:19 #link https://wiki.opendaylight.org/view/OpenDaylight_OpenFlow_Plugin:Meeting_Logs 14:07:29 #undo 14:07:52 #info The meeting logs are available at: 14:07:57 #link https://wiki.opendaylight.org/view/OpenDaylight_OpenFlow_Plugin:Meeting_Logs 14:08:08 #chair abhijitkumbhare 14:08:08 Current chairs: abhijitkumbhare edwarnicke 14:08:16 abhijitkumbhare: It's all yours :) 14:08:21 Thanks edwarnicke 14:08:23 Run us through :) 14:08:41 Just give me a moment 14:09:33 Shall we go over https://trello.com/b/TtnyEJZr/helium-release ? 14:09:49 and then 2) performance 14:09:58 3) documentation 14:10:04 Any other topics? 14:10:13 Actually bug list 14:10:36 So: 14:10:50 #info Agenda 14:10:59 #info 1) Bug list 14:11:12 #info 2) Trello 14:11:24 #info 3) performance 14:11:35 #info 4) documentation 14:11:42 Anything else? 14:11:57 hi guys. i have to run shortly. 14:12:08 So you want the floor first? 14:12:15 abhijitkumbhare: yes. 14:12:19 Sure 14:12:22 if you guys are okay with it 14:12:27 Sure 14:12:56 Go ahead - what are the issues Madhu_ ? 14:13:20 with michal_rehak's help this morning, i was able to get an external extensible bundle plugged into the OF extensibility framework 14:13:29 Great! 14:13:49 Welcome back michal_rehak ! 14:13:57 abhijitkumbhare: but we need https://git.opendaylight.org/gerrit/9855 to be merged in ofplugin 14:14:13 OK - let's check that 14:14:26 edwarnicke: as we discussed yesterday, 14:14:40 we can have an external bundle to participate in the extensibility now 14:14:57 abhijitkumbhare: though we made it work, i find it extremely complicated :) 14:14:59 Just a pom change - so should be easy to merge that 14:15:08 OK 14:15:13 maybe its my lack of expertise in that. but if u look at the changes 14:15:30 it is quite involved and am not sure how much it will motivate folks to do the right thing 14:15:43 but it does work once all the bells and whistles are properly aligned 14:16:56 Guys... any objections to merging https://git.opendaylight.org/gerrit/9855 14:17:01 abhijitkumbhare: i haven't tested the stats or other stuffs yet. will do so in the next few days 14:17:06 OK - any thoughts michal_rehak ? May not have too much time this release Madhu_ & michal_rehak to make it a lot better than what it is - but ... 14:17:09 It looks good to me 14:17:25 No objections - edwarnicke 14:17:36 abhijitkumbhare: yes. i spoke to michal_rehak about that this morning. 14:17:45 OK 14:17:47 abhijitkumbhare: and yes, not for this release for sure :) but for next 14:17:53 OK 14:18:25 edwarnicke: no objection - just need to test it 14:18:27 Madhu_: It can be made dead simple with a fix to the Match API... but that will have to wait for Lithium 14:18:42 edwarnicke: it is not the match api fix. 14:18:48 it is the overall strucutre 14:18:54 i will write a wiki on my experience with this 14:18:58 Madhu_: The structure is driven by the Match API 14:18:59 Madhu_ - do you want to run thru your recommendations on an email? 14:19:02 and we can see how we can improve 14:19:10 wiki will be great as well 14:19:10 abhijitkumbhare: yes. 14:19:12 Madhu_: With a refactor of the Match API, the structure becomes dead simple 14:19:18 (in fact better) 14:19:56 anyways.. am unblocked once the plugin gerrit is merged 14:20:02 Good 14:20:09 and i will explore further with new extensions to see if I hit any others 14:20:12 fingers crossed 14:20:14 Sure 14:20:22 btw... RESTCONF isn't working 14:20:28 hmm 14:20:38 I demonstrated it to michal_rehak and he said he will look into it 14:20:47 RESTCONF for extensions i mean 14:20:47 For the extensions - right? 14:20:52 abhijitkumbhare: yes 14:20:54 OK 14:20:58 i will file a bug if that will help 14:21:01 Sure 14:21:12 Will help 14:21:36 thanks guys. appreciate your help. and i have to sign off now 14:21:42 OK - we can start the bugs list then 14:22:27 #info Bugs list link: 14:22:32 #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&deadlineto=2014-09-30&list_id=13400&product=openflowplugin&query_format=advanced&resolution=--- 14:23:24 I need to be go for a moment - will be back in a min 14:26:14 back 14:26:16 michal_rehak: Are you wanting to do more testing for https://git.opendaylight.org/gerrit/#/c/9855 ? It doesn't look like it could hurt anything, it just exposing more packages 14:26:44 oh 14:27:04 edwarnicke: I think we can merge - I just wonder why didn't we need that sooner 14:28:09 So the first bug (going to the unassigned first): 14:28:25 #topic Bug list 14:28:47 #link https://bugs.opendaylight.org/show_bug.cgi?id=1300 14:29:32 Any thoughts? 14:30:49 abhijitkumbhare: looks like md-sal model change + adapt corresponding convertors in ofplugin 14:31:22 Should that be fixed (now that it is after the API freeze)? 14:31:33 For Helium I mean 14:32:24 abhijitkumbhare: in my opinion this could grow into large change 14:32:29 If not I can put the deadline as past Oct 1 - to be a Lithium or a stable Helium 14:32:39 I thought so as well 14:33:14 So I will put the deadline as Oct 15 (to make it for post Helium) 14:33:31 edwarnicke: I am sure you have some comments on how huge this change might get 14:34:34 I have already put it as post Helium :-) 14:34:42 ok 14:34:52 #link https://bugs.opendaylight.org/show_bug.cgi?id=1395 14:35:22 #info performance degradation 24 hour test 14:36:08 This looks important 14:37:16 #info Steady drop from 15k flows per sec to 10.5k flows per sec 14:37:39 Who wants to tackle this? 14:38:17 BTW Prasanna is on a PTO this week and Anil Vishnoi is not available today 14:39:37 This bug & performance in general is likely going to need a lot of time on whoever is looking into it 14:40:30 I can ping Muthu if he is up to it - if no one else is wanting to take a crack at it (will #action) 14:40:49 do we know via profiling , which would be area of focus? 14:41:25 I don't recollect what was the last update on the profiling 14:41:44 I mean the last datapoint from the profilining 14:42:00 jmedved - do you have that info? 14:43:05 kamal_rameshan: Are you around? 14:43:19 yes 14:43:42 joined in late 14:46:29 edwarnicke: do we have enough +1s for https://git.opendaylight.org/gerrit/#/c/9802/ to be merged? 14:47:45 OK - if no one is taking on https://bugs.opendaylight.org/show_bug.cgi?id=1395 right now I will start chasing down Muthu (but he may be on a PTO) 14:48:46 abhijitkumbhare: i can take on 1395 , but only after i tackle 1491 (VTN bug)) 14:49:02 #action abhijitkumbhare to check with Muthu if he can take on https://bugs.opendaylight.org/show_bug.cgi?id=1395 & performance. In the meantime - it will be good to have an alternate contact for performance 14:49:25 Good kamal_rameshan - but let me 14:50:03 meant let's check with Muthu or someone so you can be free for the other bugs 14:50:30 but if there is no one - then we will of course need to turn to you 14:50:48 abhijitkumbhare : ok :) 14:51:16 Next: 14:51:22 #link https://bugs.opendaylight.org/show_bug.cgi?id=1403 14:51:46 #info MD-SAL topology does not show link information 14:52:31 edwarnicke - who is the expert on this area? 14:52:57 Have you checked with Luis to see is thats still an issue? 14:53:02 there was already a similar bug on compatibility : https://bugs.opendaylight.org/show_bug.cgi?id=1468, wondering both are the same.... 14:53:04 Because its not showing up integration testing 14:53:04 Will check 14:53:19 OK (yes - that's true) 14:53:53 #action abhijitkumbhare check with Luis if https://bugs.opendaylight.org/show_bug.cgi?id=1403 still happening 14:55:37 kamal_rameshan - last comment on the AD-SAL topology bugs seems to be that its not reproducible 14:56:10 abhijitkumbhare - luis mentioned that he would retest and let me know 14:56:16 OK 14:57:58 Going on severity/priority now (unassigned are over): 14:58:03 #link https://bugs.opendaylight.org/show_bug.cgi?id=1421 14:58:33 Is this the VTN issue? kamal_rameshan 14:59:04 abhijitkumbhare:yes this is the VTN issue 14:59:17 i am working on it 14:59:34 OK 15:00:24 Time is up - we need to be faster next time :-( 15:00:59 Let me try to go thru them offline before the next meeting - and ask questions beforehand 15:01:41 I mean go thru the bug list beforehand - and get to this meeting only the bugs that need attention 15:01:51 that need action I mean 15:02:10 Tomorrow is a webex meeting 15:02:27 abhijitkumbhare: makes sense, also better if its a webex meet. As response is immediate 15:02:34 OK 15:02:46 So I will end meeting now 15:02:51 #endmeeting