Tuesday, 2021-09-14

yasufumhi, tacker team08:01
yasufum#startmeeting tacker08:02
opendevmeetMeeting started Tue Sep 14 08:02:14 2021 UTC and is due to finish in 60 minutes.  The chair is yasufum. Information about MeetBot at http://wiki.debian.org/MeetBot.08:02
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.08:02
opendevmeetThe meeting name has been set to 'tacker'08:02
yasufumThere are two topic on etherpad today.08:02
yasufum#link https://etherpad.opendev.org/p/tacker-meeting08:02
yasufumThe first one is my topic08:03
yasufumSo, let get start it from the first item.08:03
yasufumAs discussed in the previous vPTG, dropping lower-constraints was stopped because it’s still useful sometimes.08:05
yasufumI remember this discussion was only for tacker, right?08:06
yasufumdidn’t for tacker-horizon and python-tackerclinet.08:06
takahashi-tscI think so.08:07
uehaAccording to the xena ptg etherpad, it doesn't mention the sub project.08:08
ueha https://etherpad.opendev.org/p/tacker-xena-ptg#L19308:08
yasufumNow, ropping lower constraints test has proposed.08:08
yasufumSo, I would like to know your thinking of about this proposal.08:09
yasufumIn my opinion, we don’t need to do so strict tests tacker-horizon and python-tackerclient actually.08:09
yasufumsimple and light-weight implementation comparing with tacker.08:10
uehaIt is useful to leave the lower-constraints test of the tacker, but not the sub project. right?08:11
yasufumI’m not sure it’s surely useful actually, but may be yes :)08:13
uehaIf so, other projects have dropped lower-constraints, so I think we can drop it as well for tacker-horizon and python-tackerclient.08:13
yasufumAny other comment or objection?08:15
manpreetkJust a thought could we check other horizon plugin as well.08:15
manpreetkAs horizon still has lower-constraint file.08:16
yasufumSure, I’ve just found it.08:17
yasufumThere are many horizon plugin projects, such as xxx-horizon or xxx-dashboard, and some of them already might dropped lower constraints.08:22
yasufumI think we don’t need to consider the decision of horizon now because it’s not related with tests in tacker-horizon.08:25
yasufumBut might need to have a look about other horizon related projects as examples for making our decision.08:26
manpreetkTotally agree with you, considering above points we could drop l-c from tacker-horizon.08:26
manpreetkI could check/investigate other plugins and update with opinion. WDYT?08:29
yasufumOK, if you have any other comment, please tell me on the patch review.08:29
yasufummanpreetk: thanks08:29
yasufum#link https://review.opendev.org/c/openstack/tacker-horizon/+/78911408:30
manpreetkSure, thats from my side. Thanks08:31
yasufumFor python-tackerclient, no update for dropping has not suggested yet, but I will do when appropriate.08:31
yasufumOK, go to the next topic.08:32
yasufumh-asahina: could you share your topic?08:32
h_asahinathanks, sure.08:33
h_asahinaAs I mentioned on the etherpad, I'd like to confirm whether or not the patch https://review.opendev.org/c/openstack/tacker/+/799022 supports multi-tenant policy completely in LCM operations.08:34
h_asahinaIn other words, do we have to do any additional work to support multi-tenancy in LCM operation or we don't have to do anything?08:35
h_asahinaDoes anyone know the answer?08:36
takahashi-tscw-juso, do you make the plan as spec. So you can explain the answer base on the spec.08:39
yasufumthis one, right?08:39
yasufum#link https://review.opendev.org/c/openstack/tacker-specs/+/78713508:39
w-jusoI has implementing the 1 and 2 of proposed change int https://review.opendev.org/c/openstack/tacker-specs/+/787135.08:39
w-jusowhile, there are support still 3 of proposed change.08:41
w-jusoThese are the only three plans that I am planning.08:42
h_asahinaI suppose those plans include supporting multi-tenant in LCM operations, right?08:44
h_asahinaIn that spec, the proposed changes is just "Add tenant_id to VnfLcmSubscriptions and VnfLcmOpOccs". So, I can't see if this patch enables multi-tenant operation in actual LCM operations or just adding `tenant_id` field to DB.08:47
w-jusoThere are three works related to adding this tenant ID.08:52
w-jusothe first is filtering of VnfLcmSubscriptions show/list.08:53
w-jusothe second is filtering of VnfLcmOpOccs show/list.08:53
w-jusothe third is filter of send notification.08:54
w-jusoDo you think there is anything else you need to do?08:56
h_asahinaHow about instantion or termination?08:58
w-jusoI think intantiation and termination are no problem for adding this tenant ID. but, there are exist issue of 3 of proposed change in instantiation.09:02
h_asahinaYou mean "3) Prohibiting VIM and VNF association created in different tenants"?09:04
h_asahinais this proposed change included in the patch?09:06
w-jusothis proposed change hasn't include in the https://review.opendev.org/c/openstack/tacker/+/799022.09:08
h_asahinathat means Tacker allows users to create VNF instance on VIM of a different tenant even if the above patch is merged. Am I correct?09:10
w-juso"3) Prohibiting VIM and VNF association created in different tenants" isn't include in the https://review.opendev.org/c/openstack/tacker/+/799022.09:10
h_asahinaI got it. so we can't say Tacker supports complete multi-tenant LCM operation yet.09:12
h_asahinaThank you, but one more question. What is the reason that you did't implement the above proposed change to your patch? is there any problem?09:15
w-jusoSorry, just it needed much time. The fix was done in our local but UT/FT is not finished yet.09:17
h_asahinais difficult to include that proposed change to your patch before the relase?09:21
w-jusoUT is possible, but FT is hard to implement.09:23
w-jusoby RC1-release09:24
h_asahinaok. anyway you've already enables the proposed change "3) Prohibiting VIM and VNF association created in different tenants" in your local environment. if so, we can wait, can't we?09:25
yasufumw-juso: Do you have any idea for release target? such as mid of yoga or so...09:26
w-jusoI will release FT at Yoga cycle.09:28
w-jusoWe can finish it by mid of Yoga09:29
w-jusoI'd like to finish this work first at next Yoga cycle.09:30
yasufumI’d appreciate if you share status of your work at the next vPTG so we can discuss more about.09:31
yasufumAnyway, we’d better to wrap up this meeting...09:32
yasufumDo you have any other topic?09:32
masaki-uenoHi, i have one09:32
masaki-uenoNot a discussion, but a review request.09:32
yasufumOK, please go ahead09:32
masaki-uenoRecently, I've uploaded a patch that fixes document of k8s VIM installation.09:33
masaki-ueno#link https://review.opendev.org/c/openstack/tacker/+/80798809:33
yasufumI’ll check it, thanks.09:34
masaki-uenoI think this needs your review, so I appreciate any comments not only my change, but also any part in this document.09:34
masaki-uenoThat's all, thanks :)09:34
yasufumWe’ve asked to fix RC1 by 16th, so please help for accelarating to merge propsed patches.09:37
yasufum#link https://review.opendev.org/c/openstack/releases/+/80873309:37
yasufumI’d like to close this meeting if no more comment.09:38
yasufumThanks for joining, bye!09:39
h_asahinathanks, bye.09:39
uehathanks, bye.09:40
manpreetkthanks bye09:40
w-jusothanks! bye09:40
opendevmeetMeeting ended Tue Sep 14 09:40:07 2021 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)09:40
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tacker/2021/tacker.2021-09-14-08.02.html09:40
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tacker/2021/tacker.2021-09-14-08.02.txt09:40
opendevmeetLog:            https://meetings.opendev.org/meetings/tacker/2021/tacker.2021-09-14-08.02.log.html09:40
*** pdeore_ is now known as pdeore13:37
*** abhishekk is now known as abhishekk|away14:31
*** abhishekk|away is now known as abhishekk15:51
julianpAhoy. Looks like I'm the first one here.21:02
julianpHi trandles.21:04
julianpHi b1airo.21:04
b1airohi julianp !21:04
trandlesI'm in the right place!21:05
trandleslucky guess21:05
trandlesHi julianp b1airo 21:05
b1airohowdy trandles 21:05
julianpAre we waiting for martial?21:08
b1airoPerhaps, I'm in another zoom in parallel :-)21:09
julianpHeh. Gotcha.21:09
julianpI can't stay long, so with your blessing b1airo, I'll drop some Exosphere updates and run.21:11
b1airosounds great julianp , have been keen for an update21:12
julianpOk! Thank you.21:12
b1airowe're building our new Cloud/HPC thing at the moment and I'm thinking exosphere could be our default self-service vdesktop option...21:12
julianpExcellent. That fits nicely with our roadmap.21:13
julianp1. User-friendly, DigitalOcean-esque OpenStack image selection landed recently: https://gitlab.com/exosphere/exosphere/-/merge_requests/497#screenshots21:13
julianpMappings to OpenStack images is configurable per OpenStack cloud.21:13
julianp2. Launch Binder-compatible repositories on OpenStack instances: https://gitlab.com/exosphere/exosphere/-/merge_requests/496#screenshots21:14
b1airooooh nice21:15
julianpGood for workshops I hope.21:15
b1airoyeah, that sounds very interesting. we do a lot of training via JupyterHub at the moment...21:15
julianpThat makes sense.21:16
julianpBoth of these features are available to on the Jetstream Exosphere for anybody with accounts. Let me know if you don't and I can get you testing credentials. https://exosphere.jetstream-cloud.org/21:17
julianpI'll be working on making the Binder functionality easier to understand for new users. I'd also like to make it easier for cloud operators to launch their own proxies - so they don't have to use/trust ours to test Exosphere out.21:20
b1airosounds useful actually21:21
julianpI'd like to know the biggest hurdles are which prevent folks from kicking the tires, and make it easier get going. Feel free to send me a DM for any feedback.21:22
julianpOy. And my typos. Apologies. I have to run (literally) to the next meeting. Have a great day! o/21:23
b1airothanks julian21:27
b1airohow are things your end trandles ?21:27
trandlesway way too busy21:31
trandlesI'm staring into the abyss and it really does stare back21:31
trandleswe've lost some key people and that's not making anything better21:31
martial(Sorry for being late, kids pickup)21:32
trandlesHi martial 21:33
b1airooh dear, sounds a bit dire21:36
b1airowanna move to NZ ?! :-)21:36
trandlesHrm, not sure moving to NZ is an option, especially during covid ;)21:36
b1airojust takes longer than usual21:37
martial(picking kids from camp, there must be some english-ism I was missing :) )21:38
martialwhat did I miss so far?21:39
trandlessome exosphere updates from julianp21:39
trandlesI think b1airo is multitasking on another meeting at the same time21:41
trandlesmartial: you still planning to attend SC?21:41
martialno, that got taken off the table a while back21:41
martialbasically after we did not get a panel together21:41
b1airotwo other discussions now :-)21:42
trandlesb1airo is now manytasking21:42
b1airoyeah I'm disappointed in SC organisers, what a cluster ...21:42
martialwell hopefully next year21:43
trandlesit's a cluster ... for folks already in the US, I can't imagine it would be doable (or worth it) for people coming from overseas21:43
b1airoyeah :-(21:44
trandlesgotta drop off and get ready for the next meeting21:54
trandlescheers guys21:54
martialwe are reaching the end of the hour anyhow21:56
martial#topic AOB21:56
martialwell if nothing else, thanks everyone for joining 21:57

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!