| *** kopecmartin has joined #refstack | 07:38 | |
| *** tosky has joined #refstack | 07:39 | |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Document installation process https://review.openstack.org/580191 | 08:10 |
|---|---|---|
| *** rmart04 has joined #refstack | 08:32 | |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Document installation process https://review.openstack.org/580191 | 08:41 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Fix http_image uploading to glance https://review.openstack.org/581344 | 12:07 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Fix http_image uploading to glance https://review.openstack.org/581344 | 12:07 |
| openstackgerrit | Merged openstack/python-tempestconf master: Fix documentation structure https://review.openstack.org/580172 | 12:09 |
| *** markvoelker has quit IRC | 12:18 | |
| *** markvoelker has joined #refstack | 12:21 | |
| *** markvoelker has quit IRC | 12:31 | |
| *** markvoelker has joined #refstack | 12:35 | |
| chkumar|ruck | kopecmartin: is the http image issue only coming in RHOS-14 only? | 12:46 |
| tosky | ahem, Rocky | 12:52 |
| kopecmartin | chkumar|ruck, yes | 13:04 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Document CLI options https://review.openstack.org/580981 | 13:22 |
| kopecmartin | chkumar|ruck, well not .. the package for queens is without the refactoring changes, however the package for Rocky contains the changes, which means that we broke it by refactoring | 13:41 |
| kopecmartin | and by "we" i mean "me" because i didn't understand the image.http_image option correctly :) | 13:42 |
| tosky | "we" because we reviewed and approved it | 13:42 |
| tosky | and having just one relevant breakage so far is not bad; it could have been much more complicated :) | 13:43 |
| kopecmartin | tosky, regarding to the comment on line 79 https://review.openstack.org/#/c/581058 | 14:09 |
| kopecmartin | can you please write me the whole sentence? | 14:10 |
| tosky | good point, let me rephrase the sentence (or sentences?) properly | 14:10 |
| tosky | comment added | 14:16 |
| kopecmartin | tosky, thanks, what about the sentence on line 79? | 14:31 |
| kopecmartin | should i just append "using the Story and Task footer respectively." ? | 14:31 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Fix http_image uploading to glance https://review.openstack.org/581344 | 15:03 |
| tosky | kopecmartin: yes, sorry, that was the idea, but I'm not sure now if's really needed, or that's implicit if you know how contribution works, or we can just reference the storyboard documentation | 15:06 |
| kopecmartin | tosky, I'd say it's implicit | 15:08 |
| tosky | oki, so no need to add it | 15:09 |
| tosky | if we say someone referencing the story and the task in a freeform way, after the -1 we add it | 15:10 |
| openstackgerrit | Martin Kopec proposed openstack/python-tempestconf master: Add Contributor Guide https://review.openstack.org/581058 | 15:10 |
| tosky | kopecmartin: about the CLI options change: is that automatically generated with some command? | 15:33 |
| kopecmartin | tosky, no, i copy pasted it | 15:34 |
| tosky | kopecmartin: did you check the autoprogram-cliff directive for sphinx? | 15:35 |
| tosky | kopecmartin: https://docs.openstack.org/cliff/2.8.0/user/sphinxext.html | 15:36 |
| tosky | it should avoid the problem of keeping the documentation synchronized with the program | 15:36 |
| kopecmartin | i saw it, but it wasn't very self-explained :D | 15:37 |
| tosky | you can steal the example of other programs who migrated to it, like this change: https://github.com/openstack/python-neutronclient/commit/423890a4950a725d6b493304d7b933d5b7c090d7 | 15:37 |
| kopecmartin | I was looking for something more copy-pastable solution | 15:40 |
| kopecmartin | in any case, we would still need to write the description somewhere | 15:40 |
| tosky | yes, but no need to manually document the options | 15:41 |
| kopecmartin | i can try to check some blogs on how to implement it | 15:43 |
| kopecmartin | because i don't like the documentation much | 15:44 |
| tosky | uhm, then maybe it's better to rebase "Add Contributor Guide", which is more ready, directly on top of "Document installation process", which is more ready | 15:51 |
| *** rmart04 has quit IRC | 15:52 | |
| tosky | oh, about the question on the integration with refstack and what to do with the flavor: finding a flavor is fine by me | 15:54 |
| tosky | if no flavors exist, and the user can't create them, well, it means that it's not possible to run (various) tests anyway, not our fault | 15:54 |
| kopecmartin | tosky, do we want to create a new argument for that? | 16:01 |
| kopecmartin | or maybe add the implementation of the smallest available flavor here https://github.com/openstack/python-tempestconf/blob/master/config_tempest/flavors.py#L77 | 16:03 |
| kopecmartin | it would be done automatically | 16:03 |
| tosky | kopecmartin: yeah, we already have the argument | 16:04 |
| tosky | so, if refstack does not want/can't do the autodetection, we can try it, and if we fail, well, no much to do | 16:04 |
| tosky | that said, it may be tricky to find a flavor small enough that can be used properly, but again, if we explain this properly, it should work | 16:05 |
| tosky | with a big warning (at least info log entry) when the flavor is autodetected | 16:05 |
| chkumar|ruck | tosky: kopecmartin we need to find two smallest flavor if not found flag the msg of using overrides? | 16:06 |
| tosky | chkumar|ruck: ehm, can you please rephrase the final part of the sentence? | 16:07 |
| chkumar|ruck | tosky: flag the msg as pass arguments compute.flavor and compute.flavor_alt with the cli | 16:08 |
| chkumar|ruck | tosky: I am introducing --overrides in refstack-client to pass all extras params which needs to be overridden | 16:08 |
| chkumar|ruck | in tempest.conf | 16:08 |
| kopecmartin | so when create is not allowed and micro and nano are not found, the tool will try to find the smallest flavor available (with a info msg that it's doing so), and then if not found, big message for a user, that some flavors need to be created and for example specified via overrides (so that the tool doesn't have to autodetect them) | 16:09 |
| tosky | chkumar|ruck: if I get it correctly, no: the other way; if you pass a flavor use it; autodetect as last resource | 16:09 |
| chkumar|ruck | tosky: pass a flavor means? passing as a extra overrides? | 16:11 |
| tosky | chkumar|ruck: yes, specified as user input | 16:11 |
| tosky | if you specify the flavors with overrides, that's what the user wants | 16:12 |
| chkumar|ruck | kopecmartin: tosky so the implementation would be if extra overrides is passed use it, if not then look for nano and macro, if not found, detect the lowest possible flavor if all the above is false, flash the big mesg? | 16:13 |
| kopecmartin | yes | 16:15 |
| tosky | almost | 16:15 |
| chkumar|ruck | let's move it to storyboard | 16:15 |
| tosky | after nano and macro, if not found detect the lower possible flavor AND send a warning, but continue; if detection fails, exit | 16:16 |
| kopecmartin | yes, agree | 16:17 |
| kopecmartin | great, I'll have a look later, now I'm going for a run, see ya | 16:17 |
| tosky | o/ | 16:17 |
| chkumar|ruck | tosky: kopecmartin https://storyboard.openstack.org/#!/story/2002932 | 16:21 |
| tosky | chkumar|ruck: I would replace, in the first point, "it will look for it" with "it will use it" | 16:22 |
| chkumar|ruck | tosky: done ,t hanks :-) | 16:23 |
| * chkumar|ruck headed out | 16:24 | |
| openstackgerrit | wes hayutin proposed openstack/python-tempestconf master: add jobs for python-tempestconf only https://review.openstack.org/581084 | 16:55 |
| *** kopecmartin has quit IRC | 18:14 | |
| *** rmart04 has joined #refstack | 19:25 | |
| *** rmart04 has quit IRC | 19:38 | |
| *** rmart04 has joined #refstack | 20:02 | |
| *** rmart04 has quit IRC | 20:18 | |
| *** tosky has quit IRC | 23:08 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!