08:03:40 <dkushwaha> #startmeeting tacker
08:03:41 <openstack> Meeting started Tue Dec 17 08:03:40 2019 UTC and is due to finish in 60 minutes.  The chair is dkushwaha. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:03:42 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:03:44 <openstack> The meeting name has been set to 'tacker'
08:03:53 <dkushwaha> #topic Roll Call
08:04:08 <tpatil> Hi
08:04:12 <takahashi-tsc> Hi
08:04:25 <joxyuki> hi
08:04:42 <keiko-k> Hello
08:04:50 <nitinuikey> Hi
08:05:46 <dkushwaha> hello all
08:06:20 <dkushwaha> #chair joxyuki
08:06:21 <openstack> Current chairs: dkushwaha joxyuki
08:07:21 <dkushwaha> #topic BP
08:07:29 <dkushwaha> BP support-etsi-nfv-specs
08:07:42 <dkushwaha> tpatil, any update from your side?
08:08:19 <dkushwaha> i think keiko-k working on it
08:08:20 <tpatil> Working on updating the specs, we will push a new PS today e.o.d or tomorrow morning
08:09:24 <dkushwaha> tpatil, nice
08:12:08 <dkushwaha> moving next..
08:12:37 <dkushwaha> #topic attach-existing-volume
08:12:46 <dkushwaha> #link https://review.opendev.org/#/c/698477/
08:13:47 <dkushwaha> joxyuki, I am agree on your comment
08:14:01 <joxyuki> nice feature. please check my comment.
08:14:03 <dkushwaha> and hopefully, will push updated patch by today
08:14:17 <joxyuki> nice!
08:14:53 <dkushwaha> A dependent patch is pushed in tosca-parser
08:14:58 <dkushwaha> #link https://review.opendev.org/#/c/698439/
08:15:15 <dkushwaha> Please review
08:16:05 <dkushwaha> i am hoping to merge that dependent patch, but seems  no reviewer from tosca-parser
08:16:43 <dkushwaha> will try to ping them (if available)
08:17:37 <joxyuki> yes. anyway I will review the patch.
08:18:16 <dkushwaha> thanks
08:18:33 <dkushwaha> moving next..
08:19:10 <dkushwaha> joxyuki, do you have some update?
08:19:41 <joxyuki> https://review.opendev.org/#/c/697616/
08:20:08 <joxyuki> I would like you to review this patch.
08:20:48 <joxyuki> Tacker lower-constraitns job doesn't work properly.
08:22:11 <dkushwaha> joxyuki, yea, I looked an overview earlier. will review it
08:22:41 <dkushwaha> joxyuki, just a query, why to override install_command command?
08:23:15 <dkushwaha> I am not sure(haven't seen in other project) if it is ok to override it
08:25:21 <joxyuki> because in current tox.ini, install command in [testenv] includes -c upper-constraints.txt so this results in install command is issued with both upper-constraints and lower-constraints.
08:27:15 <joxyuki> I checked some logs of load patches and found that tacker lower-constraints job is actually tested against upper-constraints.
08:27:27 <joxyuki> sorry, load -> old
08:28:12 <dkushwaha> joxyuki, i see.
08:28:57 <joxyuki> I am not sure pip behavior when specifying multiple -c but the logs shows that.
08:29:01 <dkushwaha> Ok, i will review the patch(and hope to find out some way if we can avoid this overriding)
08:29:08 <keiko-k> dkushwaha Do you have any update regarding functional test?
08:29:45 <keiko-k> It seems that installation of devstack controller has failed in FT.
08:29:54 <keiko-k> https://zuul.opendev.org/t/openstack/build/eefa8600e75942289822ff9f2db3e4bd
08:32:30 <joxyuki> I think another fix is moving "-c upper constraints" to deps from install_command in [testenv]. My change follows nova tox.ini.
08:33:14 <dkushwaha> keiko-k, no update as of now, But I just trying to setup my env with devstack and  getting issues as: RTNETLINK answers: Permission denied
08:33:51 <dkushwaha> keiko-k, will look in to gate issue
08:34:23 <keiko-k> Okay.
08:35:10 <tpatil> #link : https://review.opendev.org/#/c/618086
08:35:17 <tpatil> Delete VNF should fail with 409 error
08:35:47 <tpatil> I was finding what's the best place to add the vnf status check so that it can return 409 error
08:36:00 <tpatil> #link: https://github.com/openstack/tacker/blob/master/tacker/db/vnfm/vnfm_db.py#L581
08:36:16 <dkushwaha> joxyuki, sounds good
08:37:18 <tpatil> But here, it's checking if the vnf_id exists in the Ns and then it checks if NS status is not in [constants.PENDING_DELETE, constants.ERROR]
08:38:42 <tpatil> So this code is written from NS point of view. If we check if NS is not present for a given vnf_id, then it should look for vnf. If it exists, then it's status shouldn't be in PENDING_CREATE
08:39:42 <dkushwaha> tpatil, joxyuki my comment on https://review.opendev.org/#/c/618086 was regarding mainly to raise "VNFDeleteFailed" error as check_vnf_status_legality can also be called from scale or some other places
08:41:44 <joxyuki> make sense
08:41:56 <tpatil> dkushwaha: For that comment only, I was checking what's the next best place to add this check and I think it's _delete_vnf_pre method. What do you think?
08:42:23 <dkushwaha> link please
08:42:37 <tpatil> #link: https://github.com/openstack/tacker/blob/master/tacker/db/vnfm/vnfm_db.py#L581
08:43:49 <tpatil> as mentioned above, _delete_vnf_pre is written from deleting vnf from Ns, but VNF can exists as standalone as well.
08:44:04 <tpatil> s/from/for
08:44:56 <tpatil> I'm talking mainly about lines 588-590
08:46:45 <dkushwaha> tpatil, hmm, probably  yes,  _delete_vnf_pre method should be good place
08:47:25 <dkushwaha> request you to please re-verify from your end too
08:47:56 <tpatil> We will make the necessary changes and upload a new PS for review
08:48:20 <dkushwaha> tpatil, thanks
08:48:52 <dkushwaha> moving next..
08:50:05 <dkushwaha> takahashi-tsc, do you something to update ?
08:51:15 <takahashi-tsc> Sorry, now no update... But we discuss and make spec "Support event trigger alarm and vdu_autoheal by policy type alarm", I will push BP and spec this week.
08:51:57 <dkushwaha> takahashi-tsc, np.
08:52:15 <dkushwaha> #topic OpenDiscussion
08:52:29 <hyunsikyang> Hi I am late due to my meeting at my office. Now we are trying to make a test code for two features.
08:52:43 <hyunsikyang> I have a two question..
08:52:59 <joxyuki> hi hyunsickyang
08:53:02 <hyunsikyang> First, https://review.opendev.org/#/c/681157/
08:53:04 <hyunsikyang> Hi
08:53:06 <hyunsikyang> Jo
08:53:27 <hyunsikyang> About that one,. can we make a new review after abandon?
08:53:53 <hyunsikyang> We will update it soon with test code.
08:54:25 <hyunsikyang> or can I just update code to there?
08:55:07 <dkushwaha> hyunsikyang, yes, you can. suggested way is to update the same patch with updated code
08:55:28 <hyunsikyang> Ah ok:)
08:55:42 <dkushwaha> no needs to abandon the patch until it is completely out of scope
08:55:44 <hyunsikyang> we will update it to same patch
08:55:54 <hyunsikyang> OKie.
08:55:55 <hyunsikyang> Thanks
08:56:07 <hyunsikyang> another question is ...
08:56:41 <hyunsikyang> Actually, About C-VNF, there is no testcode..
08:57:02 <hyunsikyang> To make a test code for my patch..https://review.opendev.org/#/c/674761/
08:57:52 <hyunsikyang> Can I just make a test code for function?
08:59:24 <dkushwaha> hyunsikyang, we should add unit test at least
09:00:09 <hyunsikyang> OK. I just check it for that . And If I have a more time, I will think about C-VNF Test code.
09:00:16 <dkushwaha> functional test is also required but I hope it can be done in some other patch
09:01:26 <dkushwaha> time up..
09:01:38 <dkushwaha> Closing this meeting. Will be available on tacker channel
09:01:52 <dkushwaha> Thanks Folks for joining.
09:01:59 <dkushwaha> #endmeeting