| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: WIP: Fix test requirements for py27 https://review.opendev.org/720317 | 00:50 |
|---|---|---|
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: WIP: Fix test requirements for py27 https://review.opendev.org/720317 | 01:21 |
| kota_ | tkajinam: today, I bring up the py3 related conversation at openstack swift weekly meeting. | 02:01 |
| kota_ | brought | 02:01 |
| tkajinam | kota_, ack | 02:01 |
| * tkajinam is checking the meeting log | 02:02 | |
| kota_ | then, I confirmed swift still want to keep py2 compatible for a while, because known production clusters have not yet moved to py3 envrionment at this morment. | 02:02 |
| tkajinam | kota_, understood | 02:03 |
| kota_ | but timburke also said "Storlets can decide to drop py2 support according to the product decision" so we may handle the plan when we should complete the py3 movement. | 02:03 |
| kota_ | project decision so we have the ball for now. | 02:04 |
| tkajinam | kota_, I think it's still one of our available options to drop py2 | 02:04 |
| kota_ | yup | 02:04 |
| tkajinam | with saying that if you want to use py2 you shouldn't update storlets to ussuri | 02:04 |
| kota_ | hmm | 02:06 |
| tkajinam | kota_, I proposed patch to add py2 test back as gmann suggested, but am currently facing issue caused by required verion of hacking | 02:06 |
| tkajinam | https://review.opendev.org/#/c/720317/ | 02:06 |
| tkajinam | I'm not really sure why it still requires hacking >=3.0.0 | 02:06 |
| kota_ | lemme see | 02:06 |
| tkajinam | kota_, thx | 02:10 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: WIP: Fix test requirements for py27 https://review.opendev.org/720317 | 02:12 |
| tkajinam | ^^^ fixed keystoneauth1 version, which I've found so far | 02:12 |
| kota_ | oic | 02:19 |
| kota_ | almost of openstack submodules won't work with py2 | 02:19 |
| kota_ | currently i have a couple of questions | 02:20 |
| kota_ | 1. which module requires 3.0>=hacking, 2. hacking 3.0.0 is available at pypi (https://pypi.org/project/hacking/3.0.0/) but why it failed with NOT FOUND> | 02:21 |
| tkajinam | 1. I'm not really sure. AFAIK it shouldn't included in requirements.txt, but not in test-requierements.txt, so nothing should require it AFAIK | 02:30 |
| tkajinam | 2. It's available, but it's only for py3 | 02:30 |
| tkajinam | not compatible with py2 | 02:30 |
| tkajinam | ^^^ kota_ | 02:31 |
| kota_ | i see | 02:32 |
| tkajinam | hmm... the logs says that constraing is coming from our own test-requirement.txt in storlets | 02:32 |
| kota_ | e.g. https://github.com/openstack/os-testr/commit/d5c66507d35685a96131014c8294516d49e08625 os-testr bumps the requirement. I'm assuming some required module did the same way. | 02:33 |
| tkajinam | kota_, but it's test-requirements. I don't think it will add dependencies while installing its actual package | 02:33 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Cap hacking for py2 tests https://review.opendev.org/720327 | 02:35 |
| tkajinam | kota_, do you think it's worth to try updating hacking requirement first ? | 02:36 |
| kota_ | tkajinam: perhaps https://github.com/openstack/swift/commit/96b56519bf5756022d495aedc0033cd44e0e3267 may be a hint? | 02:36 |
| tkajinam | ahhh | 02:36 |
| tkajinam | hmmm... | 02:37 |
| kota_ | should we have lower-constraints.txt independently like swift? | 02:37 |
| tkajinam | kota_, I don't think it helps actually here. because it doesn't have any change for installation | 02:38 |
| tkajinam | I'm talking about the patch | 02:38 |
| kota_ | ok | 02:38 |
| tkajinam | kota_, the only thing in my mind now is to revert hacking version first, because it seems that it is still referring original test-requirements.txt | 02:45 |
| tkajinam | though I'm not really sure about its mechanism | 02:45 |
| tkajinam | oh, wait | 02:47 |
| tkajinam | I think I found the reason... | 02:48 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "Cleanup py27 support" https://review.opendev.org/720131 | 02:50 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "[ussuri][goal] Drop python 2.7 support and testing" https://review.opendev.org/720314 | 02:50 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: WIP: Fix test requirements for py27 https://review.opendev.org/720317 | 02:50 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "Update hacking for Python3" https://review.opendev.org/720329 | 02:50 |
| kota_ | what??? | 02:51 |
| tkajinam | it was not rebased on top of master | 02:51 |
| kota_ | the master has the requirement??? | 02:51 |
| kota_ | omg | 02:51 |
| kota_ | but got it. we are able to merge such a patch if we don't have py2 gate. | 02:52 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Fix cap for hacking in python 2.7 https://review.opendev.org/720329 | 02:55 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "[ussuri][goal] Drop python 2.7 support and testing" https://review.opendev.org/720314 | 02:55 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: WIP: Fix test requirements for py27 https://review.opendev.org/720317 | 02:55 |
| tkajinam | hope that it will go well now ... | 02:57 |
| tkajinam | sorry, I shouldn't miss that ... :-( | 03:00 |
| tkajinam | kota_, seems like updating required version for hacking fixed ut. though I still see func tests are failing | 03:12 |
| tkajinam | I'll check the logs once current test run completes... | 03:13 |
| kota_ | thx | 03:13 |
| tkajinam | kota_, seems like it failed to install keystone... | 03:51 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "[ussuri][goal] Drop python 2.7 support and testing" https://review.opendev.org/720314 | 03:59 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: WIP: Fix test requirements for py27 https://review.opendev.org/720317 | 03:59 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Use stable/train keystone to run Python 2.7 job https://review.opendev.org/720337 | 03:59 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Skip TestPartitionsIdentityStorlet https://review.opendev.org/720354 | 06:07 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "Use SBusClient for execute command" https://review.opendev.org/720364 | 06:45 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: Revert "Use Python 3 by default to run storlet applications" https://review.opendev.org/720380 | 07:47 |
| *** tkajinam has quit IRC | 08:50 | |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: DNM: get leaked fds https://review.opendev.org/720421 | 09:55 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: DNM: get leaked fds https://review.opendev.org/720421 | 20:24 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: DNM: get leaked fds https://review.opendev.org/720421 | 21:03 |
| openstackgerrit | Takashi Kajinami proposed openstack/storlets master: DNM: get leaked fds https://review.opendev.org/720421 | 21:36 |
| *** tkajinam has joined #openstack-storlets | 22:39 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!