| *** tosky has quit IRC | 00:15 | |
| *** rcernin_ has quit IRC | 03:14 | |
| *** rcernin_ has joined #openstack-rally | 04:25 | |
| *** rcernin_ has quit IRC | 04:25 | |
| *** rcernin has joined #openstack-rally | 04:25 | |
| *** rcernin has quit IRC | 06:26 | |
| *** rcernin has joined #openstack-rally | 06:26 | |
| *** rcernin has quit IRC | 07:07 | |
| *** nikparasyr has joined #openstack-rally | 07:27 | |
| *** dave-mccowan has quit IRC | 07:30 | |
| *** tosky has joined #openstack-rally | 08:22 | |
| *** marosg has joined #openstack-rally | 11:21 | |
| marosg | Hi all. What happened to this project? Last commit is from Nov 2nd and I don't see andrey here. Is this still alive? | 11:27 |
|---|---|---|
| *** andreykurilin has joined #openstack-rally | 13:35 | |
| marosg | hi andreykurilin ! I asked this two hours ago "Hi all. What happened to this project? Last commit is from Nov 2nd and I don't see andrey here. Is this still alive?" | 13:53 |
| andreykurilin | heh. I just saw your message at linkedin and start answering you there | 13:54 |
| marosg | :D | 13:54 |
| andreykurilin | so yes, project is alive | 13:54 |
| andreykurilin | I forgot to start IRC client at some point | 13:54 |
| andreykurilin | anyway, there was no much messages here :) | 13:54 |
| marosg | I wanted to ask - I see when my tempest fails some tests, rally still returns code 0, it is returning 1 only if it is not able to setup environment. Is that on purpose? | 13:56 |
| andreykurilin | oh...let me try to remember | 13:57 |
| andreykurilin | I do not have a better answer than "it is done in such way just by historical issues" and can not find a reasonable explanation why it was implemented in such way :) | 14:00 |
| marosg | well, that is one the most common reasons everywhere. Thanks for update. | 14:02 |
| andreykurilin | Maybe, I was thinking that if we succeeded to execute a tool (verifier), rally made its work and and we should not fail | 14:02 |
| andreykurilin | but now it sounds unreasonable and I'm ok about changing this behavior | 14:03 |
| marosg | as far as I understand the code, rally shoudl check return code from tempest and return it. I did not check if tempest is returning different code depending on results | 14:06 |
| andreykurilin | It is not how it works. | 14:10 |
| andreykurilin | Rally ignores the tempest CLI , we are consider tempest as like a subset of tests that should be executed via testr/stestr. Triggering testr/stestr with proper argument gives subunit output that rally parses in real-time | 14:12 |
| andreykurilin | this gives the final dict with total and detailed numbers | 14:12 |
| andreykurilin | this dict has all that is needed for returning the correct exit code | 14:13 |
| andreykurilin | the proper exit code should be selected right after https://github.com/openstack/rally/blob/master/rally/cli/commands/verify.py#L527 based on results["totals"] | 14:14 |
| marosg | yes, that is the code I was looking at, that is where it returns 1 when env cannot be set and nothing after the test. And there I had no idea how is tempest really invoked | 14:17 |
| andreykurilin | it is invoked via run method of https://github.com/openstack/rally/blob/master/rally/plugins/verification/testr.py#L90 | 14:19 |
| andreykurilin | the exact command is produced by https://github.com/openstack/rally/blob/master/rally/plugins/verification/testr.py#L43 | 14:20 |
| andreykurilin | ^ It is for the case if you want to know more details:) | 14:20 |
| andreykurilin | Actually, the whole logic is implemented in 2 repos (rally repo includes unified things that can be theoretically be re-used and rally-openstack includes tempest specific stuff) :) | 14:23 |
| *** marosg has quit IRC | 18:05 | |
| *** nikparasyr has left #openstack-rally | 18:13 | |
| *** rcernin has joined #openstack-rally | 23:02 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!