05:32:52 #startmeeting tacker 05:32:53 Meeting started Wed Mar 1 05:32:52 2017 UTC and is due to finish in 60 minutes. The chair is sridhar_ram. Information about MeetBot at http://wiki.debian.org/MeetBot. 05:32:54 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 05:32:56 The meeting name has been set to 'tacker' 05:33:08 #chair gongysh 05:33:09 Current chairs: gongysh sridhar_ram 05:33:29 #topic Agenda 05:33:38 https://wiki.openstack.org/wiki/Meetings/Tacker 05:33:45 gongysh: take over 05:33:45 thanks. sridhar_ram 05:33:58 gongysh: sure 05:34:21 who are here for tacker meeting? 05:34:28 hi all 05:34:39 Hi 05:34:41 tbh1, hi 05:34:42 o/ 05:34:45 hi 05:34:56 KanagarajM, how is the OpenO? 05:35:02 diga, YanXing_an , hi 05:35:13 gongysh: Hi 05:35:28 gongysh, :) its nice ! 05:36:09 gongysh, we have next release sooner and is one more happening place 05:36:15 It seems we are on meeting without many core members. 05:36:16 Hi 05:36:37 Jimmy_ye and YanXing_an are from china mobile. 05:36:41 welcome. 05:36:44 +1 05:36:46 yes 05:37:04 Jimmy_ye: YanXing_an: glad to have you from the operator side! 05:37:21 I want guys to register their absence on meeting wiki. 05:37:51 #info core members should register their absence on meeting wiki. 05:38:11 ok , lets move to next topic 05:38:34 #topic milestone 05:38:55 https://launchpad.net/tacker 05:39:21 we have set up the milestone according to openstack release plan 05:39:42 so P1 should be https://launchpad.net/tacker/+milestone/pike-1 05:39:55 2017-04-10 05:40:31 It seems more than one month away. but time flys. 05:40:42 we have got many stuff. 05:41:10 #topic review stats 05:41:51 just as we have said. we should put more efforts to improve our reviews. 05:42:02 current data is not good. 05:42:15 http://stackalytics.com/?module=tacker-group 05:42:32 we are deep behind. 05:43:21 we need put more effort to review. and get the commits merged. 05:43:27 #topic features 05:43:49 https://blueprints.launchpad.net/tacker 05:44:45 we have decomposed-monitor-framework, vnf-policy-monitoring-worker, service-assurance-engine, vnffg-ns, tacker-api-framework 05:45:22 gongysh: how abt decouple scaling policy? 05:45:22 these are 5 BPs that are related to our cycle's focus: stable, scalable. 05:46:03 tung_doan, we need 05:46:56 the owners of these BP can start to draft the specs as soon as possible. 05:47:03 service-assurance-engine is too broad .. we need to scope of something direct and impleentable 05:47:14 s/of/it to/ 05:47:39 sridhar_ram, +1. a big name. 05:47:48 sridhar_ram: could you pls help review https://review.openstack.org/#/c/434974/ 05:48:19 tung_doan: will review, glad to see a spec is out 05:48:23 sridhar_ram: in case we narrow down the scope 05:49:14 diga, are you there? 05:49:40 gongysh: hi 05:49:40 gongysh: we need to stabilize existing features as well.. particularly ffg and auto-scaling 05:50:37 for tacker-api-framework, you can start to code to show us how much efforts we will need for new API. 05:50:49 gongysh: ttx sent me mail on Pike WSGI thread & he told me to ping him on irc 05:50:53 sridhar_ram, yes, that is bugs and tests topic. 05:51:08 gongysh: ack 05:51:13 gongysh: yeah, I will start on it 05:51:39 #topic tests 05:52:12 first I want to talk is the unit tests dir. 05:52:29 on server side, the code is https://review.openstack.org/#/c/429198/ 05:53:08 this will make our unit tests to cover more than just our feature codes. it also covers our framework. 05:53:29 and also a step to support python3. 05:53:51 on client side: https://review.openstack.org/#/c/429813/ 05:55:04 a functional test bug is reported by me for vnffg: https://bugs.launchpad.net/tacker/+bug/1668838 05:55:04 Launchpad bug 1668838 in tacker "add vnffg functional tests" [High,New] 05:55:21 this is very important. 05:56:04 who wants to take this bug? 05:56:10 gongysh: yes 05:56:10 Jimmy_ye is adding ffg ut 05:56:18 ft 05:56:29 also, i had ft 05:56:34 gongysh: I got several questions abt ping driver: https://answers.launchpad.net/tacker/+question/497384 05:56:38 sorry add 05:56:52 gongysh: and related bug: https://bugs.launchpad.net/tacker/+bug/1668587 05:56:52 Launchpad bug 1668587 in tacker "Failing to start multiple monitor policies on one VNF" [Undecided,Confirmed] - Assigned to Tung Doan (tungdoan) 05:56:57 Jimmy_ye, do you want to write functional tests for vnfffg? 05:57:02 yes 05:57:06 gongysh: it should be fixed i this cycle 05:57:22 gongysh: i would also mark all scaling related bugs to higher priority .. https://bugs.launchpad.net/tacker/?field.searchtext=scaling 05:57:24 ok, Jimmy_ye you can assign it to yourself. thanks 05:57:31 ok 05:57:40 sridhar_ram, +1 05:57:59 * sridhar_ram is looking at tung_doan to help :) 05:58:15 sridhar_ram: sure :)) 05:58:43 tung_doan: thanks .. of course, KanagarajM could help here as well 05:58:59 tung_doan, thanks for you take the bug 1668587 05:58:59 bug 1668587 in tacker "Failing to start multiple monitor policies on one VNF" [Undecided,Confirmed] https://launchpad.net/bugs/1668587 - Assigned to Tung Doan (tungdoan) 06:00:16 sridhar_ram, gongysh for scaling , shall we involve HT team as well 06:00:52 this would help to make the changes on both tacker and ht at same time 06:00:57 KanagarajM: yes, we should .. particularly for custom VDU target 06:00:58 KanagarajM, yes 06:01:06 yes 06:01:19 https://bugs.launchpad.net/tacker/+bugs?field.tag=multi-services 06:01:36 I have two bugs there 06:01:40 tbh, could you help here ? 06:02:02 tbh, for scaling related changes in ht, 06:02:34 sridhar_ram: KanagarajM gongysh : abt monitoring node in HT & TP, need your review too https://review.openstack.org/#/c/429371/ https://review.openstack.org/#/c/431920/ 06:03:13 seems tung_doan has too much blocked. 06:03:15 KanagarajM, am particularly not looking in scaling part in HT these days, but AFAIK spzala had a patch on this 06:03:16 tung_doan, sure. i will look in to them 06:03:31 tbh1, ok 06:03:34 tung_doan: pls help with monitoring node in HT & TP.. thanks :) 06:03:49 tung_doan, you can put me into your patch's reviewers list. 06:04:03 gongysh, i will check with spzala 06:04:12 1667641 06:04:35 https://bugs.launchpad.net/tacker/+bug/1667641 06:04:35 Launchpad bug 1667641 in tacker "Use db unique index to control duplicate name for NFV resources" [High,New] - Assigned to Kanagaraj Manickam (kanagaraj-manickam) 06:04:52 https://bugs.launchpad.net/tacker/+bug/1667652 06:04:52 Launchpad bug 1667652 in tacker "Remove VIM credential storage problem on local file system" [Medium,New] 06:05:06 gongysh: sure. tbh: pls help review abt Ht & TP too, thanks :) 06:05:30 tung_doan, sure I will do that 06:05:31 these are two bugs to impact the multi tacker server instances running. 06:06:24 https://bugs.launchpad.net/tacker/+bug/1667652 is not urgent since we can use shared file system to do it. 06:06:24 Launchpad bug 1667652 in tacker "Remove VIM credential storage problem on local file system" [Medium,New] 06:07:02 that is the bugs I want to talk from my side. 06:07:21 do you guys have other bugs to talk? 06:07:53 I will try to research & fix the bug 1667652 06:07:53 bug 1667652 in tacker "Remove VIM credential storage problem on local file system" [Medium,New] https://launchpad.net/bugs/1667652 - Assigned to Yan Xing'an (yanxingan) 06:08:21 YanXing_an, nice. 06:08:38 gongysh: YanXing_an: little back story on fernet keys - it was picked after reseaching with keystone team 06:09:13 YanXing_an: DB was not considered a secure place to store VIM credentials 06:10:12 YanXing_an: anyways, do research on alternatives.. i'm super curious on the options 06:10:46 sridhar_ram, thanks for the background. 06:10:47 sridhar_ram: ok, I will discuss it with gongys after i have any ideas 06:11:03 YanXing_an: sure 06:12:07 #topic open-discussion 06:12:26 * gongysh is driving the meeting crazily. 06:13:03 * sridhar_ram thinks the meeting is progressing quite well .... so far ;-) 06:14:22 dharmendra is not here 06:14:38 https://etherpad.openstack.org/p/tacker-pike-grooming 06:14:46 gongysh: just one more question :) do we have consensus on Tacker decomposition.. That could help other Tacker contributor look into that. 06:14:52 he has promised 5 days a week. 06:15:51 sridhar_ram, I got the comments on zookeeper specs 06:16:22 currently as I know dragonflow is using it 06:16:39 gongysh: zk is a wonderful component .. it pulling JVM dep is the bad bad thing 06:16:58 * sridhar_ram needs to lookup dragonflow 06:17:07 tung_doan, yes. the decomposition is a very base. 06:17:40 sridhar_ram has talk about his concern on JVM dep. 06:17:59 sridhar_ram, but the ODL, Onus, OpenO are all using JVM. 06:18:51 gongysh: nothing personal against Java / JVM ... but that becoming a dependency for tacker whcih is primarily python is going to create headaches 06:18:51 sridhar_ram, I will also read some thing about Consul. 06:19:22 gongysh: yes, Consul is a new age gossip protocol based component... 06:20:01 another thing that comes to my mind is celert 06:20:05 .. celery 06:20:07 sridhar_ram, I will think about pluggable coordination middleware.. 06:20:26 http://www.celeryproject.org/ 06:22:05 keep the deployment complexity in mind as you are looking into this 06:22:16 a new one. how about making the coordination middleware pluggable and make work forward. 06:22:21 ? 06:22:48 hmm.. that would complicate the project even further.. 06:23:00 let's get opinionated for a change .. 06:23:42 overall, you approach of looking for a "task scheduler" is a wonderful idea... 06:23:53 let's look for something that is available in the python world. 06:24:10 even if that means slightly less featured 06:25:16 gongysh: python + consul --> https://github.com/billyshambrook/taskman 06:26:29 sridhar_ram, ok, good to know it. but It seems we have to rewrite the taskman, it is a one man project. 06:26:45 s/man/person 06:26:52 gongysh: i was quoting it as an approach .. 06:27:51 sridhar_ram, yes, I will think about consul way. 06:28:03 do other guys have things to talk about? 06:28:04 gongysh: sounds good.. 06:28:06 YanXing_an? 06:28:49 nothing else from me 06:28:52 YanXing_an, can you bring the team some use cases? 06:29:21 * gongysh is watching clock: 1 min to end 06:29:43 https://bugs.launchpad.net/tacker/+bug/1664805 06:29:43 Launchpad bug 1664805 in tacker "RFE: VIP support with allowed_address_pairs" [Wishlist,In progress] - Assigned to futangw (futangw) 06:29:56 We have a requirement that hit this bug 06:30:22 time up 06:30:31 sridhar_ram, how to end meeting? 06:30:45 using "#endmeeting" 06:30:51 #endmeeting