*** vishwanathj has quit IRC | 01:48 | |
*** apoorv has joined #openstack-heat-translator | 05:01 | |
*** apoorv has quit IRC | 05:03 | |
*** apoorv has joined #openstack-heat-translator | 05:10 | |
*** openstackgerrit has quit IRC | 06:17 | |
*** openstackgerrit has joined #openstack-heat-translator | 06:18 | |
*** apoorv has quit IRC | 07:51 | |
*** apoorv has joined #openstack-heat-translator | 08:46 | |
*** apoorv has quit IRC | 12:07 | |
*** bobh has joined #openstack-heat-translator | 12:12 | |
*** bobh has quit IRC | 12:25 | |
*** shangxdy has joined #openstack-heat-translator | 12:58 | |
*** bobh has joined #openstack-heat-translator | 13:26 | |
*** shangxdy has quit IRC | 14:46 | |
*** spzala has joined #openstack-heat-translator | 15:18 | |
*** shangxdy has joined #openstack-heat-translator | 16:22 | |
*** sridhar_ram has joined #openstack-heat-translator | 16:24 | |
spzala | bobh: sridhar_ram: tosca-parser 0.5.0 is just out. I did sent an email to you with some detail related to Tacker. | 16:30 |
---|---|---|
bobh | spzala: thanks! | 16:37 |
spzala | bobh: no problem | 16:37 |
sridhar_ram | spzala: thanks... | 16:38 |
sridhar_ram | spzala: bobh: so the conclusion is, going forward, we use == <exact version> in tacker requirements.txt for these two deps ? | 16:39 |
bobh | sridhar_ram: I think we need to put this into global-requirements - it appears that we can't do project specific requirements that are a subset of global requirements | 16:39 |
bobh | sridhar_ram: spzala so the question is how do we pin the global requirements to a specific version - there aren't too many entries that use '==' in global-requirements | 16:40 |
sridhar_ram | bobh: that is my understanding as well... so if this method (== exact version) goes into global requirement.. is that acceptable for other projects dependent on tosca-parser and heat-translator ? | 16:40 |
bobh | sridhar_ram: good question! | 16:41 |
spzala | I believe it's good idea to talk to someone in release team | 16:41 |
sridhar_ram | spzala: agree... | 16:41 |
spzala | sridhar_ram: we did think about that question yesterday for a bit, and I think we shouldn't force global req to have == | 16:41 |
sridhar_ram | spzala: the chellenge here is, as I understand, we have two deps that needs to go hand in hand.. and they are out of sync (at least by few days to couple of weeks).. how to do we handle ? (paraphrazing the question to release team) | 16:42 |
bobh | spzala: it looks like we can't have a different requirement in the project than the requirement specified in global-requirements | 16:42 |
spzala | let we have heat-translator 0.5.0 out by tomorrow or asap and then we try to change only tacker and see if that works going forward and we take advice from release team as well | 16:43 |
spzala | bobh: interesting, yes that's what it seems like based on testing you did but that's frustrating .. ideally a specific project should override global req | 16:44 |
spzala | sridhar_ram: yes, that's a good question | 16:44 |
bobh | spzala: I would think a local project could require a subset of global-requirements but it looks like that doesn't work | 16:45 |
spzala | bobh: true, so let's ask release team. I am taking a quick break for lunch but will be back soon .. if that's OK? | 16:46 |
bobh | spzala: ok with me - I'm going to be out for a bit too | 16:46 |
spzala | It would be good if we are all available when ping release team | 16:46 |
sridhar_ram | spzala: bobh: lets have this clarified in #openstack-release while we are here..? | 16:46 |
sridhar_ram | spzala: +1 | 16:46 |
spzala | sridhar_ram: ok, sure | 16:47 |
sridhar_ram | bobh: you've few mins ? | 16:47 |
sridhar_ram | we can reconvene later in the day.. | 16:47 |
sridhar_ram | .. if needed | 16:47 |
*** bobh has quit IRC | 16:48 | |
sridhar_ram | spzala: okay, bobh answered :) | 16:48 |
spzala | sridhar_ran: let's meet after an hour? seems like bobh is gone for lunch already | 16:48 |
spzala | sridhar_ram: hehe | 16:48 |
sridhar_ram | spzala: sure.. in a hr | 16:48 |
spzala | sridhar_ram: cool | 16:48 |
*** bobh has joined #openstack-heat-translator | 16:49 | |
*** tbh has joined #openstack-heat-translator | 16:57 | |
*** tbh has quit IRC | 17:22 | |
*** spzala has quit IRC | 17:30 | |
*** spzala has joined #openstack-heat-translator | 17:30 | |
*** spzala has quit IRC | 17:31 | |
*** spzala has joined #openstack-heat-translator | 17:32 | |
spzala | bobh: sridhar_ram: I am back, no rush ..just fyi | 17:38 |
sridhar_ram | spzala: bobh: I'm in standby | 17:41 |
bobh | spzala: sridhar_ram I'm back | 17:54 |
spzala | bobh: hi | 17:54 |
sridhar_ram | bobh: spzala: so, let's head over to #openstack-release | 17:55 |
spzala | sure | 17:55 |
spzala | I m there | 17:55 |
sridhar_ram | bobh: can you join #openstack-release ? | 18:15 |
bobh | sridhar_ram: on my way | 18:19 |
sridhar_ram | bobh: I think we mostly concluded.. the short answer is tosca-parser library release needs to be backward compatibile with older version of heat-translator. Meanwhile, we need to manage by making quick heat-translator release following a tosca-parser release. spzala - is this your understanding as well ? | 18:21 |
spzala | sridhar_ram: bobh: yep | 18:22 |
bobh | sridhar_ram: so in other words "don't do that" | 18:23 |
*** shangxdy has quit IRC | 18:23 | |
spzala | as Doug and I mentioned point releases typically are okay with backward-compatibility but that was until now because Tacker wasn't a consumer of these libraries, but going forward we will make sure about backward-comp | 18:24 |
spzala | bobh: didn't get your question? | 18:24 |
bobh | spzala: it was a lame attempt at humor | 18:24 |
spzala | bobh: LOL | 18:25 |
sridhar_ram | spzala: great! | 18:25 |
spzala | bobh; my bad not understanding it :) | 18:25 |
bobh | spzala: it was funnier in my head than on the keyboard | 18:25 |
* bobh has that problem a lot | 18:25 | |
* sridhar_ram wonders who takes the responsibility if someone doesn't get the joke / humor | 18:25 | |
*** tbh has joined #openstack-heat-translator | 18:26 | |
bobh | sridhar_ram: doesn't the buck stop at the PTL? :-D | 18:26 |
spzala | sridhar_ram: np at all.. I was telling bobh: that the problem seems only 'one-liner' change in tosca-parser, if I knew of this issue that we can't make req. workaround I would have it backward-comp from this release :) | 18:26 |
spzala | sridhar_ram: lol | 18:26 |
spzala | bobh: lol | 18:26 |
sridhar_ram | bobh: didn't see that in the job description ;-) | 18:27 |
spzala | lol | 18:27 |
bobh | "in addition to your regular duties"... | 18:27 |
sridhar_ram | spzala: I less worried about this time.. I know you and bobh are co-ordinating smoothly here .. I was mostly concerned when this comes up late in Newton / during release crunch time.. it will be disruptive | 18:28 |
spzala | sridhar_ram: yup, totally agree | 18:28 |
sridhar_ram | spzala: .. it will be good to iron out this working model when there is no time pressure | 18:28 |
bobh | sridhar_ram: my only concern was testing during the gap | 18:28 |
spzala | Good that we ran into early phase | 18:29 |
spzala | and agree with bobh: it's hard to stay on top of backward-comp with two diff. projects and diff. contributors, if we can't find it with some testing before release | 18:29 |
sridhar_ram | bobh: the worst case is our gate is broken when we absorb a tosca-parser and we can't merge anything until a newer heat-translator shows up | 18:29 |
spzala | in the worst case, in future, I guess may be we should able to remove the troubled pypi release | 18:30 |
spzala | (I haven't done it yet, but I guess that should be possible) | 18:30 |
sridhar_ram | spzala: co-gate is an option, as mentioned by dhellman .. | 18:30 |
spzala | sridhar_ram: I was wondering without giving release team hard time on req, just remove ourselves | 18:32 |
spzala | sridhar_ram: co-gate is to not to use troubled version I guess right? I was thinking to remove that version at all. | 18:32 |
sridhar_ram | spzala: co-gate will allow you to block a tosca-parser / heat-translator version that would break tacker tests .. basically run tacker gate job in tosca-parser and heat-translator | 18:33 |
sridhar_ram | s/version/patchset/ | 18:33 |
spzala | sridhar_ram: ok, gotcha, thanks! | 18:35 |
spzala | tbh: Hi | 18:50 |
*** vishwanathj has joined #openstack-heat-translator | 18:50 | |
tbh | spzala, Hi | 18:51 |
spzala | tbh: how are you? just wanted to say hi, it's been long due to summit :) | 18:51 |
tbh | spzala, How it went? | 18:51 |
spzala | tbh: summit went well :-) busy but very productive as usual | 18:51 |
tbh | spzala, yeah, this summit has some interesting sessions I think | 18:52 |
spzala | tbh: yup | 18:52 |
spzala | tbh: I need to catch with few watching videos :) so many sessions/meetings, hard to stay on top of everything | 18:53 |
tbh | spzala, I can see more sessions in K8s this time | 18:55 |
spzala | tbh: yup, that's big one | 18:57 |
spzala | tbh: seems like it's gonna be all around it | 18:57 |
tbh | spzala, yeah it seems like that | 19:02 |
spzala | :-) | 19:03 |
spzala | tbh: good night :-) | 19:11 |
tbh | spzala, good night | 19:12 |
*** tbh has quit IRC | 19:23 | |
*** openstackgerrit has quit IRC | 19:48 | |
*** openstackgerrit has joined #openstack-heat-translator | 19:48 | |
openstackgerrit | Sahdev Zala proposed openstack/heat-translator: Move translator to use new TOSCA parser release https://review.openstack.org/312727 | 20:32 |
openstackgerrit | Sahdev Zala proposed openstack/heat-translator: Move translator to use new TOSCA parser release https://review.openstack.org/312727 | 20:34 |
spzala | bobh: hi | 20:43 |
spzala | bobh: when you get chance, can you please review the patch above, looks good but would be nice to have your quick look | 20:43 |
spzala | bobh: i have added patch for global req changes, so once that merge I guess this above patch will pass jenkins | 20:52 |
bobh | spzala: will do | 20:54 |
spzala | bobh: thanks! | 20:54 |
*** bobh has quit IRC | 21:28 | |
*** vishwanathj has quit IRC | 22:10 | |
*** sridhar_ram has quit IRC | 23:47 | |
*** bobh has joined #openstack-heat-translator | 23:47 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!