| *** rfolco has quit IRC | 00:23 | |
| *** adam_g has quit IRC | 00:24 | |
| *** adam_g has joined #openstack-qa | 00:24 | |
| *** adam_g has joined #openstack-qa | 00:24 | |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/61964 failed tempest with an unrecognized error | 00:32 |
|---|---|---|
| *** adam_g has quit IRC | 00:35 | |
| *** adam_g has joined #openstack-qa | 00:35 | |
| *** adam_g has joined #openstack-qa | 00:35 | |
| *** jhenner has joined #openstack-qa | 01:36 | |
| *** jhenner has quit IRC | 01:36 | |
| *** jhenner has joined #openstack-qa | 01:37 | |
| openstackgerrit | huangtianhua proposed a change to openstack/tempest: add tests for security_group updating https://review.openstack.org/53044 | 01:57 |
| *** jhenner has quit IRC | 01:58 | |
| openstackrecheck | openstack/glance change: https://review.openstack.org/64087 failed tempest with an unrecognized error | 02:08 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/59146 failed tempest because of: https://bugs.launchpad.net/bugs/1256106 | 02:08 |
| openstackrecheck | openstack/cinder change: https://review.openstack.org/64512 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 02:25 |
| *** zhikunliu has joined #openstack-qa | 02:34 | |
| openstackrecheck | openstack/swift change: https://review.openstack.org/64537 failed tempest with an unrecognized error | 02:40 |
| *** adam_g has quit IRC | 02:48 | |
| *** adam_g has joined #openstack-qa | 02:48 | |
| *** adam_g has joined #openstack-qa | 02:48 | |
| *** zhikunliu has quit IRC | 02:53 | |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip flavor_access_add/remove related tests https://review.openstack.org/64634 | 02:58 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/64590 failed tempest with an unrecognized error | 02:58 |
| openstackrecheck | openstack/horizon change: https://review.openstack.org/61391 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 02:58 |
| *** zhikunliu has joined #openstack-qa | 03:02 | |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip simple_tenant_usage related tests https://review.openstack.org/64637 | 03:11 |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip simple_tenant_usage related tests https://review.openstack.org/64637 | 03:13 |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip simple_tenant_usage related tests https://review.openstack.org/64637 | 03:16 |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip flavor_access_add/remove related tests https://review.openstack.org/64634 | 03:17 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/64605 failed tempest with an unrecognized error | 03:26 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/58450 failed tempest with an unrecognized error | 03:29 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Set default_network in tempest.conf https://review.openstack.org/63493 | 03:39 |
| openstackgerrit | A change was merged to openstack-dev/grenade: Fix spelling mistake occured=>occurred https://review.openstack.org/63861 | 03:39 |
| openstackgerrit | A change was merged to openstack-dev/grenade: Point readme to git.o.o instead of github https://review.openstack.org/63860 | 03:39 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/58444 failed tempest with an unrecognized error | 03:45 |
| openstackrecheck | openstack/cinder change: https://review.openstack.org/64635 failed tempest because of: https://bugs.launchpad.net/bugs/1257626 | 03:46 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/58443 failed tempest with an unrecognized error | 04:05 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/58218 failed tempest with an unrecognized error | 04:05 |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip flavor_access_add/remove related tests https://review.openstack.org/64634 | 05:14 |
| openstackgerrit | Alex Xu proposed a change to openstack/tempest: Skip simple_tenant_usage related tests https://review.openstack.org/64637 | 05:15 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/59931 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 05:16 |
| *** ivanzhu has joined #openstack-qa | 05:19 | |
| *** adam_g has quit IRC | 05:21 | |
| *** adam_g has joined #openstack-qa | 05:21 | |
| *** adam_g has joined #openstack-qa | 05:21 | |
| *** adam_g has quit IRC | 05:24 | |
| *** adam_g has joined #openstack-qa | 05:24 | |
| *** adam_g has joined #openstack-qa | 05:24 | |
| *** Madan has joined #openstack-qa | 05:43 | |
| *** yfried has quit IRC | 05:46 | |
| *** psharma has joined #openstack-qa | 05:46 | |
| *** prekarat has joined #openstack-qa | 06:55 | |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: enable volume list tests for cinder v2 - part1 https://review.openstack.org/64402 | 06:55 |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: enable volume list tests for cinder v2 - part2 https://review.openstack.org/59733 | 06:55 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63607 failed tempest with an unrecognized error | 06:55 |
| *** zigo has joined #openstack-qa | 06:56 | |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: enable volume list tests for cinder v2 - part2 https://review.openstack.org/59733 | 06:58 |
| *** prekarat has quit IRC | 06:59 | |
| *** yfried has joined #openstack-qa | 06:59 | |
| yfried | salv-orlando: here? | 07:04 |
| openstackrecheck | openstack/cinder change: https://review.openstack.org/64642 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 07:06 |
| *** prekarat has joined #openstack-qa | 07:14 | |
| *** yrabl has joined #openstack-qa | 07:30 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64402 failed tempest with an unrecognized error | 07:42 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/59733 failed tempest with an unrecognized error | 07:42 |
| *** nmagnezi has joined #openstack-qa | 07:54 | |
| *** adam_g has quit IRC | 08:07 | |
| *** adam_g has joined #openstack-qa | 08:08 | |
| *** adam_g has joined #openstack-qa | 08:08 | |
| *** fbo_away is now known as fbo | 08:12 | |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: enable volume list tests for cinder v2 - part2 https://review.openstack.org/59733 | 08:14 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/55146 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 08:14 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64622 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 08:16 |
| openstackgerrit | Ivan-Zhu proposed a change to openstack/tempest: port nova v2 images related tests into nova v3 part2 https://review.openstack.org/64361 | 08:17 |
| openstackgerrit | Ivan-Zhu proposed a change to openstack/tempest: port nova v2 images related tests into nova v3 part1 https://review.openstack.org/64360 | 08:17 |
| *** vponomaryov has joined #openstack-qa | 08:18 | |
| *** ivanzhu has quit IRC | 08:22 | |
| openstackrecheck | openstack/ceilometer change: https://review.openstack.org/64025 failed tempest with an unrecognized error | 08:23 |
| yfried | salv-orlando: ping | 08:23 |
| *** ivanzhu has joined #openstack-qa | 08:23 | |
| *** tkammer has joined #openstack-qa | 08:25 | |
| *** jlibosva has joined #openstack-qa | 08:28 | |
| *** vponomaryov has quit IRC | 08:30 | |
| *** tkammer has quit IRC | 08:36 | |
| *** vponomaryov has joined #openstack-qa | 08:39 | |
| openstackrecheck | openstack/glance change: https://review.openstack.org/64079 failed tempest with an unrecognized error | 08:49 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/59860 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 08:50 |
| *** tkammer has joined #openstack-qa | 08:51 | |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/49110 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 08:52 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/55146 failed tempest with an unrecognized error | 08:59 |
| *** jpich has joined #openstack-qa | 09:00 | |
| *** adam_g has quit IRC | 09:02 | |
| *** adam_g has joined #openstack-qa | 09:03 | |
| *** adam_g has joined #openstack-qa | 09:03 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64622 failed tempest because of: https://bugs.launchpad.net/bugs/1254872 | 09:04 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/57614 failed tempest with an unrecognized error | 09:06 |
| enikanorov__ | sdague: Hi. Could you explain your comment on https://review.openstack.org/#/c/64377/ ? | 09:16 |
| *** tkammer has quit IRC | 09:25 | |
| *** giulivo has joined #openstack-qa | 09:25 | |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: enable volume list tests for cinder v2 - part2 https://review.openstack.org/59733 | 09:28 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64402 failed tempest with an unrecognized error | 09:38 |
| *** psharma has quit IRC | 09:53 | |
| openstackrecheck | openstack/heat change: https://review.openstack.org/63839 failed tempest with an unrecognized error | 09:55 |
| vponomaryov | Does anyone know about weekly qa meeting for today? On the https://wiki.openstack.org/wiki/Meetings/QATeamMeeting no info since 19 december... | 09:56 |
| openstackgerrit | Fabien Boucher proposed a change to openstack/tempest: Remove Swift container-sync test skipping https://review.openstack.org/64548 | 10:01 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/61462 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 10:12 |
| *** yrabl has quit IRC | 10:14 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63826 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 10:22 |
| *** psharma has joined #openstack-qa | 10:22 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63827 failed tempest with an unrecognized error | 10:23 |
| *** tkammer has joined #openstack-qa | 10:26 | |
| openstackgerrit | Yair Fried proposed a change to openstack/tempest: Increase exception log details https://review.openstack.org/64654 | 10:33 |
| openstackgerrit | Yair Fried proposed a change to openstack/tempest: Increase exception log details https://review.openstack.org/64654 | 10:35 |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack-dev/devstack: Handle more nicely when role root is already here https://review.openstack.org/64655 | 10:35 |
| *** ivanzhu has quit IRC | 10:36 | |
| yfried | dkranz: ping | 10:43 |
| salv-orlando | hi yfried | 10:43 |
| salv-orlando | sorry for being unreachable in the past few days, I was on vacation | 10:44 |
| yfried | salv-orlando: how are you? I have some question if you have the time | 10:44 |
| yfried | salv-orlando: np - hope you enjoyed it | 10:44 |
| yfried | salv-orlando: so - about your comment here - https://review.openstack.org/#/c/63886/ | 10:46 |
| yfried | salv-orlando: do you think those scenarios should retrieve the network resources created by the tenant instead of creating their own? | 10:47 |
| salv-orlando | yfried - we need to talk to mtreinish for these patches. The reason is that I believe he did the tests they way they are on purpose | 10:47 |
| salv-orlando | I mean always creating new stuff | 10:47 |
| salv-orlando | rather than re-using it. | 10:48 |
| salv-orlando | If he's happy with reusing rather than creating new items, then one could have a single tenant for all scenarios. period. | 10:48 |
| salv-orlando | But I think using a different tenant with different resources for each test was one of the goals of the introduction of "tenant isolation". | 10:49 |
| salv-orlando | But since I did not participate in this activity, we need to talk to the relevant people. I surely recall mtreinish, not sure if there are other; perhaps mlavalle as well. | 10:49 |
| yfried | salv-orlando: I think we are not talking about the same thing | 10:49 |
| yfried | salv-orlando: in "tenant isolation" each scenario has it's own tenant rather than the default "demo" tenant, right? | 10:50 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64106 failed tempest with an unrecognized error | 10:51 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/61964 failed tempest with an unrecognized error | 10:51 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64283 failed tempest with an unrecognized error | 10:52 |
| yfried | salv-orlando: and the "tenant isolation" module is creating network resources (net, sub, router) for each tenant | 10:52 |
| salv-orlando | yes correct. | 10:53 |
| yfried | salv-orlando: here comes my issue: Since in "regular" mode these network resources don't exist - each test create them anyway, regardless of regula/isolation mode it's in | 10:54 |
| yfried | salv-orlando: so I'm thinking: | 10:54 |
| yfried | salv-orlando: Either we get tests to use these resources when they exist, or we disable their creation | 10:55 |
| yfried | salv-orlando: what do you think? | 10:55 |
| salv-orlando | yfried: help me understand. So there is a fundamental problem that the scenarios create those resources but then don't use them? | 10:56 |
| yfried | salv-orlando: that's what I'm thinking. do you know differently? | 10:57 |
| salv-orlando | yfried: I assumed differently. But mine is an assumption based on how I think the code should work, not based on how the code actually works. | 10:57 |
| yfried | salv-orlando: what did you assume? | 10:58 |
| salv-orlando | You can ask for confirmation to tempest-core devs, but my expectation is that when tenant isolation is enabled a new tenant with a new set of resources is created and used for each test. When tenant isolation is disabled, pre-defined credentials and resources should always be used. | 10:58 |
| salv-orlando | if the code creates resource but then just don't use them, this is incorrect. | 10:59 |
| yfried | salv-orlando: I know that these resources are created | 10:59 |
| salv-orlando | yfried: I know that too for a fact! They've been troubling me a bit with parallel testing | 11:00 |
| *** mkollaro has joined #openstack-qa | 11:00 | |
| openstackgerrit | Rohan Kanade proposed a change to openstack/tempest: Rename instance_actions v3 plugin tests to server_actions https://review.openstack.org/59187 | 11:03 |
| yfried | salv-orlando: they are created here: https://github.com/openstack/tempest/blob/bfae59d7687bb2353b939f10be85f3c1651ccf17/tempest/common/isolated_creds.py#L197 | 11:03 |
| *** yrabl has joined #openstack-qa | 11:04 | |
| salv-orlando | yfried: I'm a bit knowledgeable with that bit. I also have a patch related on review. I'm more interesting in understanding why they're not used. | 11:04 |
| salv-orlando | basically, if they're not used anyway, all the hustle and bustle tempest does with parallel testing, which brings the cpu on devstack-gate machine at 100%, could just be avoided | 11:05 |
| * yfried researches network usage | 11:05 | |
| yfried | salv-orlando: alright | 11:08 |
| yfried | salv-orlando: this is the only place the network is used according to my ide https://github.com/openstack/tempest/blob/bfae59d7687bb2353b939f10be85f3c1651ccf17/tempest/common/isolated_creds.py#L295 | 11:08 |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: cinder v2 api tests - fix volume client https://review.openstack.org/64660 | 11:09 |
| yfried | salv-orlando: and according to the same ide - this method is never called in the tempest project | 11:09 |
| openstackgerrit | Christian Schwede proposed a change to openstack/tempest: List Nova and Cinder extensions in debug log https://review.openstack.org/64283 | 11:09 |
| salv-orlando | yfried: ok. Let's wait a bit for somebody from the tempest team to come online. If they confirm your finding then there is definitely a problem to address and we might just stop creating stuff we do not use. | 11:11 |
| openstackgerrit | Fabien Boucher proposed a change to openstack/tempest: Remove Swift container-sync test skipping https://review.openstack.org/64548 | 11:14 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/62043 failed tempest with an unrecognized error | 11:16 |
| openstackrecheck | openstack/glance change: https://review.openstack.org/63707 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 11:16 |
| yfried | salv-orlando: ok - it is created and saved but not used anywhere | 11:28 |
| yfried | salv-orlando: assuming this is the case - should we start using them or should we remove them from the code? | 11:29 |
| *** jroovers has joined #openstack-qa | 11:30 | |
| openstackrecheck | openstack/python-neutronclient change: https://review.openstack.org/60324 failed tempest because of: https://bugs.launchpad.net/bugs/1256106 | 11:32 |
| yfried | salv-orlando: different issue - did you have time to look at the blueprint I submitted about reports? | 11:33 |
| yfried | salv-orlando: sorry - not what I wanted to ask | 11:34 |
| yfried | salv-orlando: that's what I wanted you to look at - https://review.openstack.org/#/c/64622/ Could you please take a quick look and tell me what you think in general? | 11:35 |
| *** jorisroovers has joined #openstack-qa | 11:39 | |
| *** jroovers has quit IRC | 11:41 | |
| *** tkammer has quit IRC | 11:50 | |
| *** tkammer has joined #openstack-qa | 11:50 | |
| *** zhikunliu has quit IRC | 11:50 | |
| yfried | salv-orlando: still here? | 11:51 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64283 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 11:59 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/64654 failed tempest because of: https://bugs.launchpad.net/bugs/1257641 and https://bugs.launchpad.net/bugs/1254890 | 12:03 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/61118 failed tempest with an unrecognized error | 12:03 |
| salv-orlando | yfried - I'm back. Give me 10 minutes and I will look at your patch | 12:07 |
| *** Madan has quit IRC | 12:29 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63617 failed tempest with an unrecognized error | 12:43 |
| *** prekarat has left #openstack-qa | 12:49 | |
| *** jorisroovers has quit IRC | 12:49 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63615 failed tempest with an unrecognized error | 12:57 |
| *** weshay has joined #openstack-qa | 13:01 | |
| sdague | morning folks, happy new year | 13:01 |
| *** sthaha has joined #openstack-qa | 13:04 | |
| *** sthaha has joined #openstack-qa | 13:04 | |
| *** sthaha has quit IRC | 13:06 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63614 failed tempest with an unrecognized error | 13:10 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63616 failed tempest because of: https://bugs.launchpad.net/bugs/1257626 | 13:10 |
| *** rfolco has joined #openstack-qa | 13:13 | |
| enikanorov__ | sdague: good morning. | 13:20 |
| enikanorov__ | sdague: could you please explain your comment: https://review.openstack.org/#/c/64377/ | 13:21 |
| sdague | so the init scripts in a neutron environment now assume a long list of config files get sent to them? | 13:22 |
| sdague | which changes mid release? | 13:23 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/63618 failed tempest with an unrecognized error | 13:24 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/58270 failed tempest because of: https://bugs.launchpad.net/bugs/1210483 | 13:24 |
| yfried | sdague: is there an online qa meeting today? | 13:24 |
| enikanorov__ | sdague: right, another conf file is being added. what cons do you see with that? | 13:24 |
| sdague | yfried: no, today is skipped | 13:25 |
| enikanorov__ | sdague: services like x-aas are going to store their configurations (including vendor-specific) in this file. | 13:25 |
| sdague | yfried: we'll restart next week | 13:25 |
| yfried | sdague: tnx. | 13:25 |
| sdague | enikanorov__: what happens to existing functional havana installs? | 13:25 |
| openstackgerrit | Fabien Boucher proposed a change to openstack/tempest: Remove Swift container-sync test skipping https://review.openstack.org/64548 | 13:26 |
| enikanorov__ | sdague: it's not mandatory for existing insallations | 13:26 |
| enikanorov__ | i mean services.conf is not mandatory | 13:26 |
| enikanorov__ | but afaik there was a consensus to store vendor-specific info in a file separate from neutron.conf | 13:27 |
| *** yfried has quit IRC | 13:30 | |
| sdague | is there a mailing list thread on this I can go take a look at. It just seems really operator hostile to need to pass, manually 3 config files to a bunch of neutron services. If the answer is > 1 this should just be a directory which is handled automatically by neutron in startup. | 13:30 |
| enikanorov__ | the answer is already > 1 | 13:31 |
| enikanorov__ | lib/neutron passes neutron.conf and core plugin-specific conf-file | 13:31 |
| enikanorov__ | sdague: in some cases even more than that. ml2 plugin may require conf file for each of its configured mechanism drivers | 13:32 |
| sdague | so how is this expected to be managed via normal init scripts? | 13:33 |
| sdague | I'd like to back up this discussion, because this seems to have gone in a very different direction than the way nova handles this successfully, which is just having optional sections in the main config file for drivers. So being different requires strong justification. | 13:36 |
| sdague | especially when that puts load on init scripts to ensure they conditionally link in all the right config parts depending on what the user wants to enable | 13:37 |
| enikanorov__ | well, the nova's direction is 3.5K lines of config file | 13:38 |
| enikanorov__ | i guess that is what neutron tried to avoid | 13:39 |
| *** anand has joined #openstack-qa | 13:39 | |
| giulivo | happy new year there sdague :) | 13:39 |
| anand | any docs for rdo storage part? | 13:40 |
| giulivo | anand, I think you should try on #rdo rather than here | 13:41 |
| anand | okay. may I ask for what this channel will be useful? | 13:41 |
| giulivo | anand, I don't think there is a strict ruling but this channel is more about the QA processes and projects; I thought you had a specific question for rdo/storage so #rdo seemed more appropriate | 13:44 |
| *** mriedem has joined #openstack-qa | 13:44 | |
| sdague | enikanorov__: yeh, but a bunch of config files, which have to be manually passed to a server, is something *no* other server does. | 13:44 |
| sdague | I could see an apache style model with includes being managed in config | 13:45 |
| enikanorov__ | i'm not sure; why are you saying "manually"? | 13:45 |
| anand | giulivo, okay thank you so much. | 13:45 |
| sdague | enikanorov__: you are passing them to the binary | 13:45 |
| enikanorov__ | sdague: right. either by init script or devstack does it | 13:45 |
| enikanorov__ | what does mean "manually" in this case? | 13:46 |
| sdague | enikanorov__: but they are conditionally sent there | 13:46 |
| sdague | --config-file /$Q_PLUGIN_CONF_FILE | 13:46 |
| enikanorov__ | in fact I'd made it a mandatory file for the future release | 13:46 |
| sdague | this is backing up the whole conversation, because we should have caught it earlier | 13:47 |
| enikanorov__ | does the number of config files matter? neutron-server and agents care about config options | 13:47 |
| enikanorov__ | which could be gathered from 1 file or from 10 files | 13:47 |
| sdague | enikanorov__: if it doesn't matter, put them all into one config file | 13:48 |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/devstack-gate: Add oslo.rootwrap to the devstack gate https://review.openstack.org/64675 | 13:48 |
| enikanorov__ | sdague: well, folks found that putting them in a separate file is more convenient | 13:48 |
| enikanorov__ | and that was the whole point | 13:48 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/64613 failed tempest with an unrecognized error | 13:48 |
| sdague | I think it does matter, because show examples of any server software on your linux environment that requires passing multiple config files to the server process | 13:49 |
| sdague | enikanorov__: then make neutron do all that intrinsically, so it only starts with a neutron.conf (which is a default), and the rest isn't something that needs to be managed on the command line | 13:49 |
| enikanorov__ | ok, you're implying setting conf dir in main config file | 13:50 |
| sdague | sure | 13:50 |
| sdague | like an apache style | 13:50 |
| sdague | and that I'd be fine with, or some include logic | 13:50 |
| sdague | but if config-file is being passed more than once to a binary, I really think we're doing this wrong. | 13:51 |
| enikanorov__ | that may make sense, but what is the problem of passing additional conf file to neutron server in devstack? | 13:51 |
| enikanorov__ | while conf-directory not there | 13:51 |
| sdague | because this patch takes us in the wrong direction, so I want the underlying issue fixed first | 13:52 |
| sdague | which also provides some urgency to it, as this feature won't land until that gets sorted | 13:52 |
| enikanorov__ | ok. i'll bring this up on the next neutron meeting | 13:52 |
| *** timello_ is now known as timello | 13:58 | |
| sdague | sure, or kick of an ML thread on it to deal with it sooner. Because it seems like there also might need to be some oslo config interaction here as well. I don't know that they support an include syntax at this point. | 14:01 |
| openstackgerrit | Thierry Carrez proposed a change to openstack-dev/devstack: Add oslo.rootwrap to devstack gate https://review.openstack.org/64679 | 14:08 |
| *** anand has quit IRC | 14:10 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/57614 failed tempest with an unrecognized error | 14:10 |
| *** psharma has quit IRC | 14:10 | |
| *** jasondotstar has joined #openstack-qa | 14:27 | |
| *** malini_afk is now known as malini | 14:34 | |
| *** mkollaro has quit IRC | 14:40 | |
| *** ruhe has joined #openstack-qa | 14:41 | |
| *** Anju has quit IRC | 14:42 | |
| *** mkollaro has joined #openstack-qa | 14:44 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Remove test_auth_token.py https://review.openstack.org/61192 | 14:44 |
| *** markmcclain has quit IRC | 14:54 | |
| *** ruhe has quit IRC | 14:55 | |
| *** Anju has joined #openstack-qa | 15:03 | |
| *** yfried has joined #openstack-qa | 15:04 | |
| yfried | salv-orlando: did you manage to take a look at my patch? | 15:07 |
| salv-orlando | yfried: yes, commented it on it and will do a thorough review in the next few hours. I agree on the approach - makes sense to me. | 15:07 |
| yfried | salv-orlando: when you're done, please ping me, if I'm still awake I would like to discuss it. This is a stub in the dark for me. It seemed easy on paper but having some issues with the actual implementation | 15:09 |
| openstackrecheck | openstack/nova change: https://review.openstack.org/61977 failed tempest with an unrecognized error | 15:13 |
| *** croberts has joined #openstack-qa | 15:17 | |
| *** mkollaro has quit IRC | 15:19 | |
| yfried | dkranz: sdague: https://review.openstack.org/#/c/64622/ feels to me like a major change to scenario tests. I would appreciate your comments on the general idea (and later the implementation) when you can. It is rather time consuming and I would like some core members' point of view before committing more time and work in this direction | 15:24 |
| openstackrecheck | openstack/horizon change: https://review.openstack.org/63403 failed tempest with an unrecognized error | 15:26 |
| *** markmcclain has joined #openstack-qa | 15:26 | |
| openstackrecheck | openstack/python-neutronclient change: https://review.openstack.org/63003 failed tempest with an unrecognized error | 15:26 |
| *** nmagnezi has quit IRC | 15:29 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add a discoverable_apis option for swift https://review.openstack.org/62063 | 15:31 |
| openstackgerrit | Malini Kamalambal proposed a change to openstack/tempest: Adds basic Marconi test https://review.openstack.org/63449 | 15:31 |
| *** bdperkin has quit IRC | 15:34 | |
| *** bdperkin has joined #openstack-qa | 15:34 | |
| openstackrecheck | openstack/horizon change: https://review.openstack.org/64272 failed tempest because of: https://bugs.launchpad.net/bugs/1257626 | 15:35 |
| yfried | dkranz: are you back from vacation? | 15:37 |
| dkranz | yfried: Yes, such that it was | 15:38 |
| yfried | dkranz: great to have you back :) can you spare some time to look at the patch. It's the work we discussed - simulating fixtures | 15:39 |
| dkranz | yfried: Yes, but maybe not until tomorrow. I am trying to push a negative test framework WIP today. | 15:40 |
| yfried | dkranz: is that the automatically generated negative tests? | 15:40 |
| dkranz | yfried: Yes | 15:40 |
| yfried | dkranz: sounds intersting - can I take a look? | 15:41 |
| yfried | mtreinish: ping | 15:41 |
| dkranz | yfried: Yes, I need to actually run it before pushing it up. | 15:41 |
| yfried | dkranz: :( | 15:41 |
| sdague | yfried: so I'll put comments in the review as well | 15:41 |
| dkranz | yfried: I am writing code for about another hour I think | 15:42 |
| dkranz | yfried: But then need to debug | 15:42 |
| mtreinish | yfried: what's up? | 15:42 |
| sdague | however, instead of doing all the logic with dirty validation and class level variables | 15:42 |
| sdague | why not just use setUp | 15:42 |
| sdague | and only build the resources there | 15:42 |
| sdague | setUpClass is just an optimization, but in this case I don't think we need or want that optimization, especially when the logic to keep it gets so complicated | 15:42 |
| yfried | dkranz: too bad. I was hoping to have a look. Purely out of interest. I don't think I can contribute much to this. Please tag me on the patch when you do push it. | 15:44 |
| mtreinish | ays vehicle information screens (performance gauges, climate controls, battery voltage, check engine codes and more) with compatible Kenwood radios.* Basic radio replacement features also available with other radios makes. | 15:44 |
| dkranz | yfried: ok. What is too bad? | 15:44 |
| mtreinish | oops finger slipped sorry about that | 15:44 |
| mtreinish | don't know what that was in my middle mouse paste buffer though... | 15:44 |
| yfried | dkranz: too bad that I can't look at it today | 15:44 |
| yfried | sdague: do you mean that I shouldn't declare the dirty flag in setUpClass? how else can I share the state between test_case instances? | 15:46 |
| sdague | so you shouldn't be sharing state between test case instances | 15:48 |
| yfried | mtreinish: salv-orlando and I had a discussion earlier about network resources in isolated_creds, and we were hoping for your input. Did you happen to see it? Would you like me to sum it up here? | 15:48 |
| sdague | because their execution order is not guarunteed | 15:48 |
| mtreinish | yfried: let me read the backlog one sec | 15:48 |
| yfried | sdague: but I need the "expensive" resources to be shared unless test_case fails | 15:48 |
| sdague | how expensive are they really? | 15:49 |
| sdague | I would rather see the non optimized version, then decide if it makes the test too long | 15:50 |
| yfried | sdague: boot a VM for every testcase??? | 15:52 |
| sdague | that's only 7 seconds | 15:52 |
| salv-orlando | yfried: thanks. Let me scroll back on catch up on IRC | 15:52 |
| yfried | sdague: you must have a state of the art setup. i can never get a pingable VM in 7 sec | 15:53 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/62063 failed tempest with an unrecognized error | 15:54 |
| salv-orlando | yfried: sorry - I thought you had already a chat with mtreinish. I seem to no be able to read anymore after the break. | 15:54 |
| mtreinish | yfried: so if a resource is needed for a test case but right now we're only creating it with isolated_creds then the test should explicitly create the resource needed | 15:54 |
| sdague | anyway, we're actually trying to evolve away from setupClass, so can we see what this looks like without all the complicated class caching, and see what the run time in the gate is | 15:54 |
| mtreinish | tenant isolation is there to provide isolation between tests running in parallel so they don't conflict with each other | 15:54 |
| yfried | salv-orlando: I am trying to get mtreinish into our chat. you are already up to speed | 15:55 |
| mtreinish | but if you need a network for the test you should create it | 15:55 |
| sdague | otherwise, if these are coupled, they should remain coupled in a single test, and just check point sanity more often | 15:55 |
| mtreinish | yfried: especially for scenario tests, you want to be testing the whole scenario which includes resource creation. I think a better solution would be to disable the network isolation for scenarios if you're trying to avoid duplicated effort | 15:58 |
| mtreinish | I think salv-orlando (or someone else) had a patch doing that | 15:58 |
| yfried | sdague: I will push another version without shared resources. Would you like it in a new patch, or on the same patch? Also - if a test case is skipped - can we avoide setup/teardown? Currently it seems like skip happens after setup | 15:58 |
| salv-orlando | mtreinish: I just has a patch for not creating routers unless needed | 15:59 |
| salv-orlando | I think yfried instead has a patch for reusing, rather than creating network resources. | 15:59 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/62687 failed tempest with an unrecognized error | 16:00 |
| salv-orlando | I am not sure if this is inline with the aim of tenant isolation, but it seems yfried observed that the resources creates by scenario tests are not used anyway - the 'default' credentials and resources are always used | 16:00 |
| salv-orlando | yfried: please correct me if I wrote anything wrong. | 16:00 |
| sdague | yfried: so that's a good point on per test script. In that case moving setup into something which is called in the tests, vs using setup at the test level would actually be better | 16:00 |
| sdague | then you can also use addCleanup instead of teardown, which will always get called | 16:01 |
| mtreinish | salv-orlando: yeah I was looking at it, it uses the isolated resources if they're available. But, I'm saying the test should be explicit if it's testing the neutron resources. | 16:01 |
| mtreinish | salv-orlando: things created by tenant isolation really weren't meant to be used for testing directly. It's more a way to enabling tests. If a resource needs to be created for a test, the test should do that. | 16:02 |
| yfried | sdague: salv-orlando: mtreinish I'm having trouble taking part in 2 chats at once. my fault for starting them both. Can I please get one of you to hold this for a few minutes? | 16:03 |
| mtreinish | so that the api semantics can be verified explicitly | 16:03 |
| salv-orlando | yfried: sure I'll be around for a while. | 16:03 |
| sdague | yfried: sure, I can hold | 16:03 |
| sdague | actually, I just put my comments in the review, we can take it there whenever is convenient for you | 16:04 |
| yfried | sdague: then why don't you join the isolated_cred issue for now? | 16:04 |
| yfried | sdague: then I will read them and get back to you shortly. Thanks for understanding | 16:05 |
| yfried | salv-orlando: mtreinish: I will recaup: | 16:06 |
| *** mkollaro has joined #openstack-qa | 16:09 | |
| yfried | iso_creds creates (upon first credential request) net_resources (net, subnet, router) for each tenant. As far as I can see, these resources are NEVER used. salv-orlando also observed that they put stress on the gate performance. We can easily drop them and current testing won't be affected. Did I miss anything? | 16:09 |
| yfried | salv-orlando: mtreinish ^ | 16:11 |
| salv-orlando | one thing to mention is that the "stress" was probably intentional. | 16:11 |
| yfried | salv-orlando: why intentional? | 16:11 |
| mtreinish | yes the stress was somewhat intentional (after we found that it was stressing things) | 16:11 |
| *** jgrimm has joined #openstack-qa | 16:12 | |
| mtreinish | yfried: if the extra resource creation is interfering with a test I'd say instead of just using the resources created by the isolated creds code, add a flag to disable there creation | 16:12 |
| mtreinish | if it's not getting in the way I don't see any harm in leaving things as they are | 16:13 |
| mtreinish | just for the | 16:13 |
| yfried | mtreinish: salv-orlando pointed out that a standard cloud usage requires each tenant to take create its own net_resources, and thus we are simulating this | 16:13 |
| mtreinish | "stess effect" it's nice side effect to make sure things can handle a reasonable load | 16:14 |
| yfried | mtreinish: ok. so let me propose this: | 16:14 |
| salv-orlando | perhaps we can just make sure the scenario test uses the resource being created | 16:15 |
| salv-orlando | I honestly thought this was how it worked | 16:15 |
| mtreinish | salv-orlando: there are 2 issues with that. 1. the resources aren't always created. (A non default config flag must be set) 2. You aren't explicitly testing the create operations anymore which I thought was part of the test scenario | 16:17 |
| yfried | mtreinish: we allow each test the choice of using the resources created, or creating its own (in case net-creation is part of the test). However, if the test creates its own tenant network, the previous net_resources will be disabled at test-setup and not created at all | 16:17 |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/61037 failed tempest with an unrecognized error | 16:18 |
| mtreinish | yfried: yes that's what I'm saying we should do | 16:18 |
| yfried | mtreinish: salv-orlando: I would also like to point out that using existing net_resources causes substantial difference between "default" and "isolated_creds" mode, and this should be addressed by the tests. | 16:19 |
| mtreinish | yfried: that is 1 of the reasons the network scenario tests create resources explicitly | 16:20 |
| *** markmcclain has quit IRC | 16:21 | |
| salv-orlando | cool. So perhaps we can have a first step in which a test case can tell the credentials manager whether it should create or not network resources? | 16:22 |
| yfried | mtreinish: AFAIK the 1st scenario net test was created before iso_creds was introduced. The 2nd was written by me, and I didn't know this option existed. | 16:22 |
| yfried | salv-orlando: +1 | 16:22 |
| salv-orlando | and then how can we work out the reuse/vs create thing? I'm not sure I understand how can we reuse net resource when tenant isolation is enabled. Every test will run with a different tenant... | 16:23 |
| mtreinish | yfried: it was, but when I added isolated creds to the scenario tests I considered all of this. | 16:23 |
| mtreinish | salv-orlando: it's more if the test doesn't specifically need to test any of the net resources created by tenant isolation then just use those resources | 16:24 |
| openstackrecheck | openstack/horizon change: https://review.openstack.org/64272 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 16:24 |
| mtreinish | but if it needs to test something created by it, then the test has to create it | 16:25 |
| yfried | salv-orlando: I don't see your concern. The point is to use already created net_resources. Not to share them with other tests | 16:25 |
| *** jlibosva has quit IRC | 16:25 | |
| salv-orlando | mtreinish, fried: forgive my dumbness. I don't know where these network resources are created… | 16:26 |
| mtreinish | salv-orlando: in the scenario tests or the isolated creds code? | 16:27 |
| salv-orlando | mtreinish: meh just ignore me then. I was moaning because I thought you wanted to create them once and then reuse for several scenarios | 16:28 |
| *** andreaf has joined #openstack-qa | 16:28 | |
| salv-orlando | sorry for wasting your time | 16:28 |
| *** jlibosva has joined #openstack-qa | 16:28 | |
| *** yrabl has quit IRC | 16:29 | |
| *** mlavalle has joined #openstack-qa | 16:29 | |
| mtreinish | salv-orlando: oh, ok. yeah, that's not going to happen. | 16:29 |
| yfried | salv-orlando: they are created for each tenant. each isolated "domain" has admin, primary, and sometimes alt tenant | 16:29 |
| yfried | salv-orlando: AFAIK "domain" is a Class. | 16:30 |
| salv-orlando | ok so there is nothing really to discuss beyond what we previously agreed | 16:30 |
| *** yfried is now known as yfried_brb | 16:30 | |
| yfried_brb | salv-orlando: mtreinish need to step away for 2 min | 16:31 |
| *** tkammer has quit IRC | 16:37 | |
| *** yfried_brb is now known as yfried | 16:42 | |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/64589 failed tempest because of: https://bugs.launchpad.net/bugs/1257626 | 16:46 |
| yfried | salv-orlando: I'm back to wonder how we know if a test should reuse the net_resources or not. I know about my the ones I wrote. but what of the others? We need a list of all the tests that create network resources. What about API tests - they don't use network at all (except for network api that create their own resources), so the only serve to stress the gate | 16:46 |
| salv-orlando | yfried: That isn't easy to know; some API tests use network resource even if they're not network test - think about the tests that create an instance | 16:48 |
| salv-orlando | But you're right that some do not need network resources at all. | 16:48 |
| salv-orlando | Finding all of them in a single patch is probably too cumbersome. | 16:48 |
| salv-orlando | And on the other hand, let's say that they add some "healthy" stress on Neutron as mtreinish pointed out. | 16:49 |
| yfried | salv-orlando: those that create the instance should definitely reuse resources | 16:49 |
| salv-orlando | I've just a patch for not creating the routers - as actually in many cases you do not need them, and setting up a router on the l3 agent takes time for ip and ovs commands | 16:49 |
| yfried | salv-orlando: we should combine our patches | 16:50 |
| salv-orlando | I've seen yours and mine and they can stay separated. | 16:50 |
| salv-orlando | no conflict. Reviewers also tend to prefer smaller patches. | 16:50 |
| salv-orlando | but I don't have strong opinions (well, I never have :)) | 16:51 |
| yfried | salv-orlando: just because I didn't know about yours. But the should be applied to the same place in the code, I think. | 16:52 |
| yfried | salv-orlando: can you post your patch here? I haven't seen it | 16:52 |
| salv-orlando | https://review.openstack.org/#/c/62962/ | 16:53 |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/59526 failed tempest because of: https://bugs.launchpad.net/bugs/1249065 | 16:55 |
| *** vponomaryov has quit IRC | 17:00 | |
| *** jasondotstar has quit IRC | 17:05 | |
| sdague | salv-orlando: so I have another idea about an approach on that last patch, which would be a little more extensive, but I think might serve us better. What if we annotated the classes that have tests that need a router with some attribute, and have the isolated creds call look at that instead. | 17:07 |
| *** zz_ajo is now known as ajo | 17:07 | |
| sdague | having it come as a parameter seems kind of odd | 17:07 |
| *** jasondotstar has joined #openstack-qa | 17:12 | |
| *** ruhe has joined #openstack-qa | 17:13 | |
| openstackrecheck | openstack/python-novaclient change: https://review.openstack.org/59861 failed tempest because of: https://bugs.launchpad.net/bugs/1257626 | 17:16 |
| *** jaypipes has joined #openstack-qa | 17:18 | |
| *** ChanServ sets mode: +o jaypipes | 17:18 | |
| *** jaypipes has quit IRC | 17:19 | |
| salv-orlando | sdague: that is functionally equivalent to me | 17:20 |
| salv-orlando | I can do that | 17:20 |
| sdague | salv-orlando: cool. I'd also like the default to be not creating the resource. Assume we only need the attribute if we are creating it | 17:21 |
| sdague | because reverse logic is harder for people to keep straight in their head on reviewing | 17:21 |
| salv-orlando | sdague: I agree with you - I did my patch the other way around because I'm lazy and it led to a shorter diff | 17:21 |
| sdague | :) | 17:21 |
| sdague | that's fair | 17:21 |
| salv-orlando | but yes, I can do as you suggest | 17:22 |
| sdague | cool, thanks | 17:22 |
| sdague | there was another patch of yours that I had a question with a -1 on. https://review.openstack.org/#/c/64217/ | 17:22 |
| sdague | because I don't actually understand how we are hitting the issue that is trying to prevent. | 17:23 |
| *** krtaylor has quit IRC | 17:24 | |
| *** markmcclain has joined #openstack-qa | 17:24 | |
| *** rwsu has joined #openstack-qa | 17:27 | |
| *** ruhe has quit IRC | 17:28 | |
| yfried | salv-orlando: great. so I will hold my patch until you push yours and work according to it. | 17:31 |
| *** mlavalle has quit IRC | 17:32 | |
| yfried | salv-orlando: mtreinish while on the subject - can we take a look at the methods used to handle isolated_creds? we have a different method for every resource and tenant, so that's something like get_(router, creds, net, subnet,...)x(admin,primary,alt) | 17:32 |
| *** jpich has quit IRC | 17:33 | |
| dansmith | I'm sure I'm being stupid, but how do I get tempest to run with tox and only run one test? seems like specifying the whole class path or a regex (like nova) doesn't work | 17:34 |
| yfried | mtreinish: seems like a better way would be to define a tenant_object and have it handle its resources | 17:37 |
| yfried | sdague: what do you think ^ | 17:37 |
| russellb | dansmith: compare tox.ini ... should have something to pass through args you provide | 17:38 |
| sdague | dansmith: the class path should work | 17:38 |
| dansmith | russellb: yeah, seems like it's there, that's the thing | 17:39 |
| *** DennyZhang has joined #openstack-qa | 17:39 | |
| russellb | then i've got nothin | 17:39 |
| dansmith | sdague: nose-style with a colon or testr style ? | 17:39 |
| sdague | testr style | 17:39 |
| dansmith | sdague: because regardless, it just runs everything, instead of not matching anything | 17:39 |
| sdague | ok, let me go test | 17:39 |
| sdague | we did do some fancy loader logic recently, maybe we broke args somewhere | 17:40 |
| dansmith | sdague: I'm using smoke-serial by the way, but it looks like it's the same as full | 17:40 |
| *** jaypipes has joined #openstack-qa | 17:40 | |
| *** ChanServ sets mode: +o jaypipes | 17:40 | |
| *** jlibosva has quit IRC | 17:41 | |
| sdague | dansmith: ok, I'm spinning up a rax devstack node, will poke there | 17:41 |
| mtreinish | dansmith: the issue is the regex on the tox jobs | 17:41 |
| *** jlibosva has joined #openstack-qa | 17:41 | |
| dansmith | mtreinish: use more words | 17:41 |
| mtreinish | dansmith: testr uses or's between multiple regexes on the cli | 17:42 |
| mtreinish | so when you add a more selective regex to tox and there is already a regex specified | 17:42 |
| mtreinish | yours doesn't do anything | 17:42 |
| dansmith | ah | 17:43 |
| dansmith | is there a tox job I should use that doesn't specify a regex? | 17:43 |
| dansmith | full does | 17:43 |
| mtreinish | use all | 17:43 |
| * dansmith tries | 17:43 | |
| dansmith | mtreinish: nice, it works, thanks :) | 17:46 |
| sdague | mtreinish: given that we are passing that through the helper script all the time, we could always drop the default regex if arg #2 doesn't start with -- | 17:46 |
| mtreinish | sdague: probably, I just get a bit twitchy touching those helper scripts | 17:47 |
| mtreinish | for something that simple I sure have caused a lot of problems with them | 17:47 |
| sdague | heh | 17:48 |
| sdague | yeh, let me put it on my queue. | 17:48 |
| *** harlowja has joined #openstack-qa | 17:49 | |
| *** prekarat has joined #openstack-qa | 17:57 | |
| *** DennyZhang has quit IRC | 17:59 | |
| *** mlavalle has joined #openstack-qa | 17:59 | |
| sdague | yfried: sorry, back to your question, an tenant_object model would be a good move forward as well. Create an object responsible for create and cleanup. | 17:59 |
| yfried | sdague: what do you mean? | 18:03 |
| sdague | <yfried> mtreinish: seems like a better way would be to define a tenant_object and have it handle its resources | 18:04 |
| sdague | yfried: +1 | 18:04 |
| yfried | sdague: I tried to do this locally for my cross_tenant scenario. if this would be done globally, that would be great. we could have the object handle either the default tenant, or the isolated_creds on runtime | 18:06 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Migrating trove to entry points https://review.openstack.org/64457 | 18:07 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Change the libvirtd log level to DEBUG https://review.openstack.org/63992 | 18:08 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Correct glance db_sync command https://review.openstack.org/63605 | 18:08 |
| openstackgerrit | A change was merged to openstack-dev/devstack: XenAPI: Fix bug with Xen ext4-using guests https://review.openstack.org/63795 | 18:08 |
| sdague | yfried: agreed | 18:13 |
| yfried | sdague: since I'm new - how do we translate these discussions to patches? The tenant object seems beyond my current abilities, but I would like to this added, and not just in the channel log | 18:16 |
| sdague | yfried: so I think this should probably become a blueprint | 18:16 |
| sdague | that would be the right way to record it | 18:16 |
| openstackgerrit | Steven Hardy proposed a change to openstack/tempest: Add basic read-only tests for heat cli https://review.openstack.org/64707 | 18:17 |
| yfried | sdague: So I will create an initial one. | 18:17 |
| sdague | yfried: great, thank you | 18:17 |
| yfried | sdague: keep in mind that it will probably require massive edit from you :) | 18:18 |
| sdague | yfried: no problem, but it's a start :) | 18:18 |
| sdague | and also does the right thing of making sure we don't completely forget it | 18:19 |
| openstackgerrit | A change was merged to openstack/tempest: keystone OS-TRUST extension, test list operations https://review.openstack.org/61176 | 18:19 |
| *** markmcclain has quit IRC | 18:21 | |
| yfried | sdague: https://blueprints.launchpad.net/tempest/+spec/tenant-menager please review and add relevant people | 18:33 |
| *** mlavalle has quit IRC | 18:35 | |
| *** markmcclain has joined #openstack-qa | 18:37 | |
| *** jroovers has joined #openstack-qa | 18:41 | |
| *** jroovers has quit IRC | 18:48 | |
| *** sileht has quit IRC | 18:59 | |
| *** sc68cal has quit IRC | 18:59 | |
| *** markmcclain has quit IRC | 19:04 | |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/59526 failed tempest because of: https://bugs.launchpad.net/bugs/1254890 | 19:04 |
| *** markmcclain has joined #openstack-qa | 19:05 | |
| *** sileht has joined #openstack-qa | 19:05 | |
| *** sc68cal has joined #openstack-qa | 19:05 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/61175 failed tempest with an unrecognized error | 19:13 |
| *** markmcclain has quit IRC | 19:15 | |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/63918 failed tempest with an unrecognized error | 19:17 |
| *** markmcclain has joined #openstack-qa | 19:17 | |
| *** mlavalle has joined #openstack-qa | 19:18 | |
| *** krtaylor has joined #openstack-qa | 19:18 | |
| mtreinish | sdague: how should we handle this? https://review.openstack.org/#/c/63062/5 | 19:27 |
| mtreinish | I'm inclined to say it should just be the newer versions | 19:27 |
| sdague | yeh, I agree | 19:36 |
| sdague | also, the previous behavior didn't support < 1.9 anyway | 19:37 |
| mtreinish | well there is that too :) | 19:41 |
| *** jlibosva has quit IRC | 19:45 | |
| *** markmcclain has quit IRC | 19:49 | |
| *** markmcclain has joined #openstack-qa | 19:53 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add a discoverable_apis option for swift https://review.openstack.org/62063 | 19:53 |
| *** jgrimm has quit IRC | 20:00 | |
| openstackrecheck | openstack/keystone change: https://review.openstack.org/59526 failed tempest with an unrecognized error | 20:02 |
| openstackgerrit | andrea-frittoli proposed a change to openstack/tempest: Input scenario capability for tempest https://review.openstack.org/60434 | 20:08 |
| *** markmcclain has quit IRC | 20:09 | |
| *** markmcclain has joined #openstack-qa | 20:11 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/60434 failed tempest with an unrecognized error | 20:27 |
| openstackrecheck | openstack-dev/devstack change: https://review.openstack.org/61854 failed tempest with an unrecognized error | 20:29 |
| *** ruhe has joined #openstack-qa | 20:38 | |
| *** fbo is now known as fbo_away | 20:47 | |
| *** malini is now known as malini_afk | 20:50 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/62063 failed tempest with an unrecognized error | 20:54 |
| *** croberts has quit IRC | 20:59 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Add a discoverable_apis option for swift https://review.openstack.org/62063 | 21:01 |
| *** croberts has joined #openstack-qa | 21:11 | |
| *** yfried has quit IRC | 21:13 | |
| *** syerrapragada1 has quit IRC | 21:19 | |
| openstackrecheck | openstack/heat change: https://review.openstack.org/64720 failed tempest with an unrecognized error | 21:25 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/59787 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 and https://bugs.launchpad.net/bugs/1244255 | 21:26 |
| *** croberts has quit IRC | 21:27 | |
| openstackgerrit | A change was merged to openstack/tempest: Add test for HEAD queries on Swift tempurl MW https://review.openstack.org/64467 | 21:39 |
| *** ruhe has quit IRC | 21:40 | |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/64205 failed tempest because of: https://bugs.launchpad.net/bugs/1210483 | 21:42 |
| *** croberts has joined #openstack-qa | 21:45 | |
| openstackgerrit | David Kranz proposed a change to openstack/tempest: First draft of negative test generation https://review.openstack.org/64733 | 21:51 |
| openstackrecheck | openstack/python-heatclient change: https://review.openstack.org/62925 failed tempest with an unrecognized error | 22:01 |
| *** jasondotstar has quit IRC | 22:02 | |
| *** seiflotfy_ has quit IRC | 22:19 | |
| openstackgerrit | andrea-frittoli proposed a change to openstack/tempest: Input scenario capability for tempest https://review.openstack.org/60434 | 22:21 |
| *** seiflotfy_ has joined #openstack-qa | 22:22 | |
| openstackgerrit | andrea-frittoli proposed a change to openstack/tempest: Input scenario capability for tempest https://review.openstack.org/60434 | 22:26 |
| *** croberts has quit IRC | 22:27 | |
| sdague | ianw, dkranz: when you guys get some time - http://lists.openstack.org/pipermail/openstack-dev/2014-January/023321.html | 22:27 |
| dkranz | sdague: I think ianw will look at this. | 22:28 |
| sdague | sure, I just don't know how irc active he is, so I figured I'd poke you to poke him if he didn't see it | 22:29 |
| *** markmcclain has quit IRC | 22:32 | |
| *** rwsu has quit IRC | 22:33 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/39998 failed tempest because of: https://bugs.launchpad.net/bugs/1254890 | 22:33 |
| dkranz | sdague, mtreinish , cyeoh : It would be nice to finally get this in (nova v3 item) https://review.openstack.org/#/c/60776/ | 22:33 |
| dkranz | sdague: Will do | 22:33 |
| *** mkoderer_ has quit IRC | 22:33 | |
| sdague | dkranz: +A | 22:34 |
| dkranz | sdague: Thanks | 22:34 |
| *** mkoderer_ has joined #openstack-qa | 22:36 | |
| *** rwsu has joined #openstack-qa | 22:40 | |
| *** seiflotfy_ has quit IRC | 22:41 | |
| *** mkoderer_ has quit IRC | 22:48 | |
| *** weshay has quit IRC | 22:49 | |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/64724 failed tempest with an unrecognized error | 22:51 |
| *** mkoderer_ has joined #openstack-qa | 22:52 | |
| *** rwsu has quit IRC | 22:59 | |
| *** mkoderer_ has quit IRC | 23:00 | |
| anteaya | normally I would ping jog0 about this but I don't see him around | 23:02 |
| *** mkoderer_ has joined #openstack-qa | 23:03 | |
| anteaya | I am aware that https://bugs.launchpad.net/tempest/+bug/1253896 is a top bug in elastci-recheck, salv-orlando has asked if the elastic-recheck stats for this are accurate in this latest comment | 23:03 |
| anteaya | is there anyone who is able to offer an opinion, and perhaps also update the bug report? | 23:03 |
| openstackgerrit | A change was merged to openstack/tempest: Kill finally, use addCleanUp https://review.openstack.org/62659 | 23:03 |
| openstackrecheck | openstack/neutron change: https://review.openstack.org/58017 failed tempest because of: https://bugs.launchpad.net/bugs/1253896 | 23:04 |
| clarkb | anteaya: it is easy to do a check by hand, run the elasticsearch query for the bug but add 'AND NOT build_queue:"experimental" | 23:05 |
| anteaya | I'll try that, thanks clarkb | 23:05 |
| clarkb | anteaya: 83 hits the last 7 days | 23:05 |
| anteaya | thanks | 23:05 |
| anteaya | airport wifi might not have given me that | 23:06 |
| anteaya | still waiting | 23:08 |
| openstackgerrit | A change was merged to openstack/tempest: keystone OS-TRUST extension, test with expiry https://review.openstack.org/61175 | 23:08 |
| *** nsaje has quit IRC | 23:09 | |
| openstackrecheck | openstack/tempest change: https://review.openstack.org/60434 failed tempest with an unrecognized error | 23:09 |
| *** nsaje has joined #openstack-qa | 23:12 | |
| *** rwsu has joined #openstack-qa | 23:12 | |
| *** mkollaro has quit IRC | 23:20 | |
| *** mlavalle has quit IRC | 23:33 | |
| openstackrecheck | openstack/python-neutronclient change: https://review.openstack.org/62130 failed tempest with an unrecognized error | 23:33 |
| *** sthaha has joined #openstack-qa | 23:35 | |
| *** sthaha has joined #openstack-qa | 23:35 | |
| openstackrecheck | openstack/nova change: https://review.openstack.org/60719 failed tempest because of: https://bugs.launchpad.net/bugs/1250168 and https://bugs.launchpad.net/bugs/1254890 | 23:36 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!