14:31:00 #startmeeting release daily sync 14:31:00 Meeting started Tue Sep 23 14:31:00 2014 UTC. The chair is gzhao. Information about MeetBot at http://ci.openstack.org/meetbot.html. 14:31:00 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:31:00 The meeting name has been set to 'release_daily_sync' 14:31:13 #topic roll call 14:31:17 gzhao, dShall we do roll call then have Defense4All do their Release review? 14:31:20 Madhu: Also, has anyone tried the 1997 fix? 14:31:26 Madhu: That went in yesterday 14:31:58 phrobb1: ok 14:32:15 #info gzhao for release 14:32:27 phrobb1: Apologies... I may be confused (and behind)... I thought we were talking about Release Reviews starting on Wed (I ask not because I object, but to be sure I haven't misplaced in my mind when I need to do the controller release review ) 14:32:37 #info edwarnicke 14:32:51 #info colindixon (TSC member and for TTP) 14:32:58 The Defense4All team has a holiday starting tomorrow through the end of the week so they asked to go today 14:33:09 edwarnicke: def4a is doing the one that doesn’t fit that 14:33:11 #chair colindixon tbachman phrobb1 edwarnicke 14:33:11 Current chairs: colindixon edwarnicke gzhao phrobb1 tbachman 14:34:00 #info Phil Robb 14:34:01 #info abhijitkumbhare for OpenFlow plugin 14:34:04 so, we probably don’t need all the projects to info in today? 14:34:34 yes, D4A will take ten minutes, then it on to release topics 14:34:37 #info AsafOron & BennyR for Defense4All 14:35:27 so, are we doing IRC-only release reviews or are we also spinning up a WebEx (I have no preference, just curious) 14:35:28 #info tbachman for groupbasedpolicy 14:35:39 IRC only colindixon 14:35:42 #topic Defense4All release review 14:35:47 phrobb1: I am 100% in favor :) Just wanted to make sure I hadn't dropped the ball on controller ;) 14:36:09 ok 14:36:15 Asaf: Do you have a link for your Release Review wiki page? 14:36:17 do we have a wiki page to link? 14:36:28 https://wiki.opendaylight.org/view/Defense4All:Helium 14:36:37 ok, switched topic to defense4all release view 14:36:58 #link https://wiki.opendaylight.org/view/Defense4All:Helium#Other:_Release_Review_2014-09-23 the release review fro Defense4All 14:37:19 Asaf: https://wiki.opendaylight.org/view/Defense4All:Helium Appears to be your Release Plan, not your Release Review 14:37:20 section Other: Release Review 2014-09-23 at the bottom of the page 14:37:38 colindixon: Many thanks :) 14:37:39 Asaf: yup, got it in the #link above 14:37:40 * colindixon scans 14:37:43 Asaf: Apologies, I read to fast :) 14:38:10 colindixon: do you have link for requirement for release view? 14:38:39 #link https://wiki.opendaylight.org/view/Sample_Release_Review this is the sample template with a link to place in the bylaws that calls for it 14:38:45 #info thomas packetcable 14:38:58 Thanks colindixon for the sample release review 14:39:19 Asaf: quick question, do you have docs and where can we find them? 14:39:31 Asaf: same for any release notes 14:40:19 also, where are you tracking bugs? I don’t see any here: https://bugs.opendaylight.org/buglist.cgi?component=General&product=defense4all&resolution=--- 14:40:28 https://wiki.opendaylight.org/view/Defense4All:Main was updated for helium 14:41:04 #link https://wiki.opendaylight.org/view/Defense4All:User_Guide documentation is here including information about installation, APIs, and a tutorial 14:41:07 impressive 14:41:29 #info oflibMichal for openflowjava 14:41:53 # info Hi Asaf: This isn't explicit on the release review, but can you confirm for us D4A's ability to fully participate in the Stable branch maintenance and QA for Helium? 14:42:24 #info phrobb1 asks “This isn't explicit on the release review, but can you confirm for us D4A's ability to fully participate in the Stable branch maintenance and QA for Helium?” 14:42:39 phrobb1: Good question :) 14:42:43 phrobb1: I confirm 14:42:58 Asaf: Thank you 14:43:26 We plan to be an active part of the project from now on. no more lapses 14:43:35 #Info Asaf confirmst that they will be able to participate in the stable branches and stable releases of Helium 14:43:39 Asaf: :-) 14:43:58 This is looking pretty clean. Do any others have questions for the Defense4All team? 14:44:25 phrobb1: no, I’m good 14:44:29 colindixon: regarding bugs. We track internally. gzao helped me undertsand the odl bug procedure. hope to fit in in the near future 14:44:44 … and thank you colindixon for keeping a clean record of events 14:44:57 #info Asaf notes that they are tracking bugs internally, but hope to migrate to using the ODL bugzilla soon 14:45:03 Asaf: Many thanks on working with gzhao to get your bugs in bugzilla :) 14:45:29 I’m just going to note that in the future, we should figure out how we want to treat the security issues question 14:45:38 gzhao & phrobb1 , I am stepping out to drop the kid to school 14:45:46 I’m not faulting d4a at all, but “none” is probably not a valid answer :p 14:45:49 back shortly 14:45:56 abhijitkumbhare: sure 14:46:25 very impressive user guide, dev guide and tutorial for D4A 14:46:45 and again, not to beat on d4a, but schedule really didn’t have “no change from original”, I’m not sure if we want to document that 14:46:46 gzao: thank you :-) 14:46:48 gzhao: +10 to that 14:47:31 edwarnicke, phrobb1, gzhao: I’m mostly just asking to see if we want to try to establish the schdule provides an accurate description of what happened during the helium release 14:48:04 colindixon: can you elaborate on your schedule question for D4A? 14:49:18 phrobb1: maybe I’m mis-remembering, but didn’t code freeze happen later than 9/1 14:49:43 Asaf: do you plan to re-arrange the release plan using sample release review format? I know most contents are already there. https://wiki.opendaylight.org/view/Sample_Release_Review 14:49:48 actually no 14:49:54 nevermind 14:49:56 they’re right 14:50:20 I’m happy 14:50:29 gzhao: using the template collidixon sent ? yes. it looks nice 14:50:47 yes 14:50:48 gzhao, we shoudl also aggregate the release reviews on the Helium Release Plan page 14:51:05 colindixon: that's a very good idea 14:51:09 colindixon: ok 14:51:21 Any last questions for D4A… going once… 14:51:27 #action gzhao aggregate the release reviews on the Helium Release Plan page 14:51:32 * phrobb1 counts to 10 14:51:40 going twice…. 14:51:53 * phrobb1 counts to 10 again 14:52:20 Done… gzhao shall we switch topics back to the normal sync schedule? 14:52:30 thanks Asaf 14:52:47 #topic autorelease 14:52:56 thanks Asaf! 14:52:58 Thank you Asaf and Benny for your release review of D4A 14:53:30 Thank you all very much . Benny & I are very excited. its a big moment for us. looking forward to see you next week in Santa Clara 14:53:51 Same here Asaf. Safe travels! 14:53:59 Asaf: Congratulations :) 14:54:10 see you then 14:54:10 ed: Thanks !!! 14:55:44 ok, for autorelease RC2 was cut on time yesterday, and so far we have successfully streaks 14:56:02 #info ChristineH for SNMP4SDN. 14:56:05 if no questions we can switch to next topic 14:56:08 #link http://nexus.opendaylight.org/content/groups/staging/org/opendaylight/integration/distribution-karaf/0.2.0-Helium/ this is the RC2 artifacts, version “Helium” :p 14:56:27 #action colindixon to link to them from the how to test RCs section 14:56:35 #undo 14:56:35 Removing item from minutes: 14:56:49 #action colindixon to link to the RC2 release from the how to test RCs section 14:57:08 #info goldavberg for lispflowmapping 14:57:44 #info Vijay for SNBI 14:57:55 any questions regarding to autorelease? 14:58:09 #link https://jenkins.opendaylight.org/odlautorelease/job/autorelease-helium-worker/ autorelease is on a streak of 5 successful runs in a row, so hopefully now it’s about bugs in the release, not in the release process :-) 14:58:20 gzhao: I think we can move on to bugs and bug hunting 14:58:37 #topic blocking issues 14:58:52 #link https://bugs.opendaylight.org/show_bug.cgi?id=2051 <= new blocking bug for GBP 14:59:22 #info groupbasedpolicy has a new bug where they fail to create flows needed in the POC demo 14:59:30 This is all on us 14:59:32 :( 14:59:46 ChristineH: maybe you can give an update for 1986, 1987 as well 14:59:52 aaa https://bugs.opendaylight.org/show_bug.cgi?id=2011 and the follow on https://bugs.opendaylight.org/show_bug.cgi?id=1948 15:00:14 #info AAA still has two blocking bugs open 15:00:24 #link https://bugs.opendaylight.org/show_bug.cgi?id=2011 the first one 15:00:32 #link https://bugs.opendaylight.org/show_bug.cgi?id=1948 and the follow on 15:00:45 https://bugs.opendaylight.org/show_bug.cgi?id=2019 might be... investigating 15:01:07 Madhu: mentioned he identified several blocking issue as well 15:01:59 gzhao: we need bugs there 15:02:12 SNMP4SDN. Bug 1986: solved. Bug 1987: not yet solved, but it's okay to just leave the code in repo, won't cause harm. 15:02:23 edwarnicke: will ask Madhu when he is back 15:02:25 #link https://bugs.opendaylight.org/show_bug.cgi?id=2019 edwarnicke notes that this might be blocking as well, but is looking into it (GBP bug) 15:03:06 ChristineH: I’m sorry, do you mean you could ship the code that way for Helium and it woudl be OK? 15:03:10 ChristineH: has 1986 fix been merged? 15:03:47 gzhao: if this fixes it, it’s merged 15:03:48 https://git.opendaylight.org/gerrit/#/c/11487/ 15:03:50 gzhao: yes, 1986 fix is merged 15:03:56 colindixon: ChristineH that means 1987 is not a release blocker 15:04:16 back phrobb1 and gzhao 15:04:19 #link https://git.opendaylight.org/gerrit/#/c/11487/ the fix for bug 1986, which has been merged 15:04:50 abhijitkumbhare: great, we are doing blocking issues 15:04:56 ok 15:04:59 ChristineH: so, the question is, does SNMP4SDN consider bug 1987 to be a release-blocking bug? 15:05:49 colindixon: bug 1987 is not a release-blocking bug. 15:06:09 #info ChristineH says that bug 1987 is not a release-blocking bug 15:06:39 Ok, then all blocking issue for snmp are clear 15:06:45 ChristineH: so that means, assuming the next respin tests well, SNMP4SDN is good 15:06:46 cool 15:06:49 * edwarnicke cheers 15:07:09 yes :D 15:07:49 colindixon: you had reported this.. https://gist.github.com/anonymous/4e5627237b949599ae30 15:08:15 I just verified with RC2 and I dont seem to hit this.. 15:08:21 vjanandr_: ok 15:08:27 just wondering if the issue is still seen.. 15:08:38 #action colindixon to verify that this issue is no longer present https://gist.github.com/anonymous/4e5627237b949599ae30 15:08:51 it’s non-blocking, but I’ll check 15:08:51 Any other blocking issues 15:09:11 #action gzhao to hunt Madhu_offline to get blocking bugs so we can understand them 15:09:48 #action colindixon to work with gzhao, edwarnicke, and tykeal to figure out how to respin the staging artifacts 15:10:26 dave_tucker: do you know anything about these blocking issues? 15:10:30 the ones in OVSDB 15:10:33 and if there are bugs? 15:10:42 colindixon yes there are bugs 15:10:57 colindixon 1 sec while i pull up bz 15:10:59 dave_tucker: do you have the bug numbers? 15:11:01 thanks 15:12:36 gzhao: blocking issues... I think one: snmp4sdn needs web GUI to show topology for user, but currently has difficulty to enter web GUI. 15:12:53 #link https://docs.google.com/spreadsheets/d/1TRYposNDFPaKcySlvwkOXvfR6Anx2EFujlIjoTthhRY/edit#gid=670022163 we’re tracking blocking issues here 15:12:58 colindixon: 1885, 2013, 2052 and perhaps some others relating to lbaas but need to confer with Madhu 15:13:23 gzhao: I believe snmp4sdn has collected topology, but just can't see them on GUI 15:13:58 because 1987? 15:14:39 ChristineH: sorry, this is a new bug 15:16:54 ChristineH: yes, if you have any issues blocking release, they need to have a bug number so we can work on it 15:18:35 colindixon: thanks, I see...I'll create on. And excuse me in Helium we only have dlux as web GUI? the simple web GUI is not working now? 15:19:23 edwarnicke, gzhao: do you know on this point? 15:19:24 gzhao - about OpenFlow plugin bugs - the following are not blocking the release: https://bugs.opendaylight.org/show_bug.cgi?id=1491 (Hideyuki had indicated earlier this was not blocking; however patch is available https://git.opendaylight.org/gerrit/#/c/11371/ which needs to be tested by hideyuki); https://bugs.opendaylight.org/show_bug.cgi?id=2028 (everyone including the tester agrees its not blocking); https://bugs.opendaylight.o 15:19:24 rg/show_bug.cgi?id=1772 (SFC-OF-L2 is able to use a workaround & OF-plugin developers think this is a case of a wrong input to the API); 15:20:29 #info abhijitkumbhare says that there are 3 bugs in openflowplugin that are either non-blocking or awaiting review 15:20:43 colindixon: well it seems only 8181 works for me, maybe edwarnicke has more insight 15:21:25 #info bug 1491 is non-blocking but is fixed by gerrit 11371, bug 2028 is non blocking, bug 1772 has a work-around and may be just misuse of the API 15:21:48 ChristineH: I *think* that’s correct, that we’re not shipping the old WebUI, but I could be wrong 15:21:53 So the only blocking bug from OpenFlow plugin is https://bugs.opendaylight.org/show_bug.cgi?id=1997 (which was fixed yesterday but needs confirmation) 15:22:31 ChristineH: did you test to see if topology shows in dlux? 15:23:06 #link https://git.opendaylight.org/gerrit/#/c/11457/ fixes bug 1997 in openflowplugin, but needs to be verified 15:23:06 ChristineH: if you can send me the steps for starting snmp I can test it 15:23:40 I missed reporting on https://bugs.opendaylight.org/show_bug.cgi?id=1526 (which is also not blocking - Madhu has not been able to retest for month + - so assume that this is not blocking him) 15:24:06 #action hideyuki verify bug fix https://bugs.opendaylight.org/show_bug.cgi?id=1491 15:24:21 #link https://bugs.opendaylight.org/show_bug.cgi?id=2002 has a work around 15:24:29 gzhao: real subnet of Ethernet switches are necessary to test... 15:25:19 ChristineH: Sorry for interrupting during your update (and providing mine :) ) 15:25:58 colindixon, gzhao: I just create bug 2054 for being unable to login dlux WebUI 15:26:19 ChristineH: have you tried admin/admin 15:26:40 abhijitkumbhare: no problem :) 15:27:01 abhijitkumbhare: do you know the login credentials for dlux 15:27:02 ? 15:27:22 no - I thought were admin / admin 15:27:29 ChristineH: I am able to login using admin/admin 15:28:59 ChristineH: which version you are using? RC2? 15:30:09 ok, any other blocking issues? 15:30:18 gzhao: did you get all the ones mentioned here? 15:30:33 ChristineH: not sure if this is related to 2011 15:30:48 colindixon: I believe so, I will hold on for 2054 for a while. 15:30:58 gzhao ; phrobb1 - can you please reschedule OpenFlow plugin release review to be either between 9 and 9:30 am tomorrow or on Thursday (other than 7:45-8:05 am) 15:31:10 colindixon: will check the minutes and verify after our meeting 15:32:13 gzhao: yes, I use RC2. I sent this issue (can't login dlux) to controller-dev (and seems still under discussion hhttps://lists.opendaylight.org/pipermail/controller-dev/2014-September/006634.html) 15:32:16 abhijitkumbhare : what is your time now 15:32:24 abhijitkumbhare: I'll work on that with you via email 15:32:46 Currently 8:10 am Pacific Wednesday 15:33:01 gzhao: I just scanned and I think we recorded all the blocking bugs 15:33:02 Thanks phrobb1 & gzhao 15:33:18 abhijitkumbhare: ok 15:33:23 colindixon: thanks 15:33:46 I have this list of blocking bugs w/o merged fixes: 1948, 2019, 2051, 2052, 2013, 1885 15:33:54 and possibly 2052 15:34:13 ChristineH: have you tried logging in with admin/admin 15:34:39 #info colindixon list of blocking bugs w/o merged fixes: 1948, 2019, 2051, 2052, 2013, 1885 15:35:51 ChristineH: your email link hhttps://lists.opendaylight.org/pipermail/controller-dev/2014-September/006634.html is blank 15:36:16 gzhao: did you already take an action to work with ChristineH or should I? 15:36:41 gzhao: https://lists.opendaylight.org/pipermail/controller-dev/2014-September/006632.html 15:37:17 colindixon: I think her issue is the same as 2011 15:37:34 colindixon: I can follow up dlux login issue 15:39:03 #info gzhao believes that ChristineH’s issue with logging into dlux is the same as 2011 and will follow up with her about this 15:39:12 ChristineH: does this describe what you’re seeing https://bugs.opendaylight.org/show_bug.cgi?id=2011 15:40:34 do we have time to discuss possible respin and how do we do it? 15:42:56 I know colindixon has an action on working with help desk 15:43:23 gzhao: I think that it’s easy, but I’ll reach out 15:43:33 gzhao: seems bug 2054 is not the same with bug 2011. Bug 2054 is that I can't log in webGUI, bug 2011 is after user login the webGUI 15:43:59 ChristineH: have you tried admin/admin? 15:44:21 colindixon: yes 15:44:22 ok 15:44:57 do we have a date (e.g Wed) for the latest date to respin to make 9/29 public release 15:45:34 colindixon: there seems a solution to log in webGUI successfully, from a response of the mailing list, but I don't understand it yet. 15:45:53 I think at this point, I think we respin when blocking bugs are fixed 15:46:42 colindixon: ok 15:46:50 ChristineH: for now I have it listed as a blocking bug for you, but I’ll hope to hear from you that things are good 15:47:53 colindixon: just try to figure out the latest time that can still make 9/29 morning GA 15:48:00 yeah 15:48:17 Wed night is the latest to get TSC blessing at the Thursday meeting 15:48:21 I’ll work with edwarnicke 15:48:25 colindixon: sure, thank you :) 15:49:16 colindixon: edwarnicke seems we need to do nightly for people to verify blocking issue fixes. 15:49:28 yeah 15:49:33 On respins, shouldn't we continue daily respins so that the portion of bugs fixed in a given day can be tested/verified as fixed in the new build?… as opposed to waiting until all bugs are fixed to respin? 15:49:40 gzhao: yep 15:49:50 phrobb1: I think more than daily 15:49:54 we should be doing the every 4 hours 15:49:59 yes, will every 4 hours 15:50:01 yep 15:50:02 gzhao: do you know how to safely start the 4 hour respin? 15:50:34 colindixon: I can do it not on staging 15:50:40 just like nightly build 15:50:44 all: no MD-SAL interest today, I’m guessing? 15:50:57 tbachman: I was voting no 15:51:06 * tbachman agrees with colindixon 15:51:24 gzhao: ok, if you run it just like the normal nightlies that would be perfect 15:51:33 we should probably check with edwarnicke just to make sure 15:51:48 * colindixon gets ready to work with tykeal and others on pushing go for release a second time 15:52:02 tbachman: I assume people are looking into the GBP bugs? 15:52:14 colindixon: ack :) 15:52:18 edwarnicke: your input on 4 hours nightly build? 15:52:53 colindixon: phrobb1 actually it is 288 minutes apart instead of 4 hours, since sometimes build took longer than 4 hours to finish 15:53:14 gzhao: noted 15:53:29 it seems like edwarnicke is otherwise occupied now 15:53:51 #action gzhao and colindixon to work with ChristineH to establish if 2054 is actually blocking 15:53:56 I think we’re done here 15:54:04 anyone have anything else they need to cover? 15:54:54 I will kick off 4 hours nightly build unless edwarnicke says differently 15:55:34 oh shall I wait till edwarnicke confirms 15:55:54 #action gzhao will continoue the “nightly” builds about every 4 hours, e.g., RC2-1, RC2-2, … wating on confirmation that it won’t break anything from ed 15:56:56 that is all I have 15:57:25 phrobb1, tbachman, hideyuki, ChristineH, oflibMichal, rovarga, edwarnicke, abhijitkumbhare, xsited 15:57:27 anything else 15:57:35 phrobb1: are you going to send out the schedule for project to verify their slots? or just 15:57:39 * colindixon tries to call out everyone who #infoed in and is still here 15:57:40 * tbachman is heads down in bug-fxing 15:57:46 move to next and make up later 15:57:47 i have nothing 15:57:55 (s/fixing/finding) 15:58:47 #link https://lists.opendaylight.org/pipermail/tsc/2014-September/001729.html bad formatting aside, that mail has the schedule 15:59:04 tbachman: should do fixing not finding, otherwise, can never release. -:) 15:59:09 lol 15:59:14 true dat ;) 15:59:21 #undo 15:59:21 Removing item from minutes: 15:59:28 #link https://lists.opendaylight.org/pipermail/tsc/2014-September/001729.html bad formatting aside, that mail has the release review schedule 15:59:30 ok 15:59:30 gzhao: the best way to ensure contact is to send an calendar invitation to each primary contact and cc the project's dev list… Not sure I have time to get that done today though 16:00:16 ok, I have to run for a bit, but I think we could safely #endmeeting now 16:01:32 well people here, does review time work for them? 16:01:46 ok, 16:01:55 #endmeeting