| *** krtaylor has joined #openstack-qa | 00:15 | |
| *** nosnos has joined #openstack-qa | 00:16 | |
| *** matsuhashi has joined #openstack-qa | 00:35 | |
| *** malini is now known as malini_afk | 00:43 | |
| *** marun has quit IRC | 00:47 | |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: driver.py should log errors when ssh operations fail. https://review.openstack.org/48168 | 00:51 |
|---|---|---|
| *** bdperkin has quit IRC | 01:00 | |
| *** bdperkin has joined #openstack-qa | 01:00 | |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack/tempest: Ensure server is in ACTIVE state https://review.openstack.org/47494 | 01:01 |
| *** sthaha has joined #openstack-qa | 01:05 | |
| *** sthaha has joined #openstack-qa | 01:05 | |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: driver.py should log errors when ssh operations fail. https://review.openstack.org/48168 | 01:11 |
| openstackgerrit | David Kranz proposed a change to openstack/tempest: Don't log 'Not Found' ERROR during image cleanup https://review.openstack.org/48143 | 01:12 |
| *** dkranz has joined #openstack-qa | 01:25 | |
| openstackgerrit | Steve Baker proposed a change to openstack/tempest: Explicity specify network for heat slow tests https://review.openstack.org/48170 | 01:35 |
| openstackgerrit | Jun Jie Nan proposed a change to openstack-dev/devstack: Pass no proxy to skip proxy using for wget and curl https://review.openstack.org/46447 | 01:46 |
| *** krtaylor has quit IRC | 01:56 | |
| *** marun has joined #openstack-qa | 02:03 | |
| *** SergeyLukjanov has joined #openstack-qa | 02:03 | |
| *** SergeyLukjanov has joined #openstack-qa | 02:04 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/tempest: Add positive tests for os-floating-ip-pools https://review.openstack.org/47794 | 02:22 |
| *** dims has quit IRC | 02:45 | |
| *** ken1ohmichi has joined #openstack-qa | 02:46 | |
| *** SergeyLukjanov has quit IRC | 03:05 | |
| *** nosnos has quit IRC | 03:40 | |
| *** nosnos has joined #openstack-qa | 03:41 | |
| *** nosnos has quit IRC | 03:42 | |
| *** jhenner has joined #openstack-qa | 03:44 | |
| *** nosnos has joined #openstack-qa | 03:44 | |
| *** dtroyer has quit IRC | 04:25 | |
| *** marun has quit IRC | 04:26 | |
| *** dtroyer has joined #openstack-qa | 04:31 | |
| *** zhikunliu has joined #openstack-qa | 04:38 | |
| openstackgerrit | minsel proposed a change to openstack/tempest: Handling network resources in tenant isolation https://review.openstack.org/45578 | 05:13 |
| *** prekarat has joined #openstack-qa | 05:15 | |
| *** SergeyLukjanov has joined #openstack-qa | 05:18 | |
| *** afazekas has joined #openstack-qa | 05:41 | |
| *** jhenner has quit IRC | 05:42 | |
| openstackgerrit | Mh Raies proposed a change to openstack/tempest: VPNaas vpnservice test cases https://review.openstack.org/47339 | 05:44 |
| openstackgerrit | Sirushti Murugesan proposed a change to openstack-dev/devstack: Nomalise RECLONE flag to True Or False. https://review.openstack.org/48184 | 06:06 |
| *** sthaha has quit IRC | 06:07 | |
| openstackgerrit | Zhi Kun Liu proposed a change to openstack/tempest: add volume list tests https://review.openstack.org/47508 | 06:44 |
| openstackgerrit | A change was merged to openstack/tempest: Add locking to test_list_hosts_with_zone https://review.openstack.org/47599 | 06:45 |
| *** bashok has joined #openstack-qa | 06:46 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-dev/devstack: Update user_attribute_ignore for LDAP Identity config https://review.openstack.org/48189 | 06:49 |
| *** tkammer has joined #openstack-qa | 07:02 | |
| *** jhenner has joined #openstack-qa | 07:12 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/tempest: Enhance Neutron ports update test https://review.openstack.org/46894 | 07:17 |
| *** thomasbiege has joined #openstack-qa | 07:25 | |
| *** thomasbiege has quit IRC | 07:37 | |
| *** yassine has joined #openstack-qa | 07:53 | |
| *** jlibosva has joined #openstack-qa | 07:57 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/tempest: Enhance Neutron ports update test https://review.openstack.org/46894 | 07:58 |
| openstackgerrit | Sirushti Murugesan proposed a change to openstack-dev/devstack: Normalise RECLONE flag to True Or False. https://review.openstack.org/48184 | 08:05 |
| openstackgerrit | Tal Kammer proposed a change to openstack/tempest: Initial basic setup of openstack and tempest config file https://review.openstack.org/42920 | 08:06 |
| *** fbo_away is now known as fbo | 08:07 | |
| *** madan has joined #openstack-qa | 08:11 | |
| tkammer | guys, is there something wrong with jenkins? I'm getting jenkins to FAIL my patch on modules that it doesn't even touch... anyone knows something about this? | 08:44 |
| *** SergeyLukjanov has joined #openstack-qa | 08:58 | |
| *** giulivo has joined #openstack-qa | 08:59 | |
| openstackgerrit | Mate Lakat proposed a change to openstack-dev/devstack: Fix typo in lib/heat https://review.openstack.org/48207 | 09:05 |
| giulivo | ping cyeoh | 09:11 |
| cyeoh | giulivo: hi | 09:11 |
| *** fbo has quit IRC | 09:18 | |
| *** fbo has joined #openstack-qa | 09:19 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/tempest: Enhance Neutron ports update test https://review.openstack.org/46894 | 09:22 |
| *** bashok has quit IRC | 09:29 | |
| raies | anyone please tell me how to make a external public network using CLI ? | 09:32 |
| raies | ?? | 09:48 |
| raies | ^^ | 09:48 |
| openstackgerrit | Mh Raies proposed a change to openstack/tempest: VPNaas vpnservice test cases https://review.openstack.org/47339 | 09:49 |
| openstackgerrit | Mh Raies proposed a change to openstack/tempest: VPNaas IKE policies tests https://review.openstack.org/47348 | 09:51 |
| openstackgerrit | Mh Raies proposed a change to openstack/tempest: VPNaas IPSec policies tests https://review.openstack.org/47816 | 09:51 |
| *** ken1ohmichi has quit IRC | 09:59 | |
| *** afazekas_ has joined #openstack-qa | 10:01 | |
| *** afazekas has quit IRC | 10:02 | |
| *** afazekas_ has quit IRC | 10:17 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/tempest: Add positive tests for os-floating-ip-pools https://review.openstack.org/47794 | 10:26 |
| *** prekarat has quit IRC | 10:33 | |
| *** afazekas_ has joined #openstack-qa | 10:37 | |
| openstackgerrit | A change was merged to openstack/tempest: add negative volume tests https://review.openstack.org/44610 | 10:44 |
| openstackgerrit | GuoHui LIu proposed a change to openstack/tempest: Skip test_volume_boot_pattern test for known issue https://review.openstack.org/48217 | 10:50 |
| *** dims has joined #openstack-qa | 11:13 | |
| openstackgerrit | Jinhe Fang proposed a change to openstack/tempest: Add test case for booting server from snapshot https://review.openstack.org/48219 | 11:15 |
| raies | afazekas_: if you have time then please visit https://review.openstack.org/#/c/47339/ I have incorporated this with given comments of people. | 11:16 |
| afazekas_ | ok | 11:18 |
| raies | One more request is that at least once look at https://review.openstack.org/#/c/47348/ and https://review.openstack.org/#/c/47816/ | 11:19 |
| raies | if you have time then only | 11:20 |
| mkoderer | giulivo: ping | 11:21 |
| *** zhikunliu has quit IRC | 11:23 | |
| giulivo | pong mkoderer | 11:24 |
| mkoderer | giulivo: https://review.openstack.org/#/c/48121 | 11:24 |
| giulivo | ehi thanks for jumping on that | 11:24 |
| giulivo | I wanted to setup a proper enforcement for those skip lines | 11:25 |
| openstackgerrit | Mh Raies proposed a change to openstack/tempest: VPNaas IPSec Site Connection API tests https://review.openstack.org/48220 | 11:25 |
| giulivo | but I'm not sure about the message formatting, I don't have any particular preference but whatever we decide to use, will have to be updated in the code | 11:25 |
| mkoderer | giulivo: I would rather spend the effort tp change this testtools.skip decorator instead of proofing that all lines duplicated in the same way ;) | 11:26 |
| mkoderer | giulivo: but I think we already talked about it ;) | 11:26 |
| giulivo | you mean define a custom skip which accepts a bug as argument? | 11:26 |
| mkoderer | giulivo: yes | 11:27 |
| mkoderer | giulivo: I didn't had time to do it | 11:27 |
| *** madan has quit IRC | 11:28 | |
| raies | afazekas: from https://wiki.openstack.org/wiki/Neutron/VPNaaS/HowToInstall will you please let me know what would be the topology for single host | 11:28 |
| giulivo | that is fine to me, so please add that in the review and I can eventually ask during the next meeting | 11:28 |
| mkoderer | ok cool :) | 11:28 |
| giulivo | the meeting will be tomorrow so it looks like the easiest way to ask other people | 11:28 |
| giulivo | can I add it to the agenda and point out it was your suggestion? | 11:29 |
| raies | afazekas: I meant it as <net1>--<subnet1>--------------<router1>--<external-net>-<router1>---------<subnet2>--<net2> | 11:29 |
| giulivo | (which I agree with btw) | 11:29 |
| mkoderer | giulivo: sure do it... | 11:29 |
| mkoderer | giulivo: it was already on a agenda but we didn't had time to speak about it | 11:30 |
| raies | The objective for this is to "how to update vpnservice and ipsecconnection site" | 11:30 |
| giulivo | doh, when was that? :( | 11:30 |
| giulivo | oh 29th of august , I see | 11:30 |
| giulivo | thanks | 11:30 |
| raies | afazekas_: are you there ? | 11:31 |
| *** madan has joined #openstack-qa | 11:31 | |
| *** tkammer has quit IRC | 11:33 | |
| rahmu | mkoderer: I have a question about the review you just gave my patch here https://review.openstack.org/#/c/44287/ | 11:33 |
| mkoderer | rahmu: ok | 11:38 |
| mkoderer | rahmu: IMHO a client function should do a client call. So helper functions should be implemented somewhere else | 11:39 |
| *** krtaylor has joined #openstack-qa | 11:40 | |
| *** weshay has joined #openstack-qa | 11:40 | |
| rahmu | mkoderer: so you mean each function in this class (or a client class) should send a request? | 11:40 |
| mkoderer | rahmu: for me this is the purpose of a client class .. to make client actions | 11:41 |
| rahmu | fair enough | 11:42 |
| rahmu | so where do you think I should move this function to? The test itself? | 11:42 |
| mkoderer | So far I saw you only using this function in your test right? | 11:43 |
| rahmu | nvm, I just saw that you already suggested that. Okay, will do | 11:43 |
| rahmu | mkoderer: yes. | 11:43 |
| mkoderer | rahmu: ok :) | 11:43 |
| rahmu | thanks for the feedback | 11:43 |
| raies | mkoderer: hi you please assist me on creation of public network ? | 11:47 |
| raies | I want to do set router gateway | 11:47 |
| raies | so I need an external interface | 11:47 |
| raies | Please assist me on creation of external network | 11:48 |
| raies | mkoderer: are you there ? | 11:53 |
| *** raies has quit IRC | 12:00 | |
| openstackgerrit | Joe Hakim Rahme proposed a change to openstack/tempest: Adds more test to cover Swift tempURL middleware https://review.openstack.org/44287 | 12:09 |
| *** jhenner has quit IRC | 12:11 | |
| *** madan has quit IRC | 12:12 | |
| *** jhenner has joined #openstack-qa | 12:13 | |
| *** bashok has joined #openstack-qa | 12:14 | |
| *** bdperkin is now known as bdperkin_gone | 12:16 | |
| *** hootjr has joined #openstack-qa | 12:25 | |
| *** dims has quit IRC | 12:36 | |
| *** annegentle has quit IRC | 12:42 | |
| *** annegentle has joined #openstack-qa | 12:43 | |
| *** dims has joined #openstack-qa | 12:43 | |
| *** krtaylor has quit IRC | 12:49 | |
| *** adam_g has quit IRC | 12:49 | |
| *** adam_g has joined #openstack-qa | 12:50 | |
| *** adam_g has quit IRC | 12:50 | |
| *** adam_g has joined #openstack-qa | 12:50 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Do not merge test of neutron tenant isolation. https://review.openstack.org/46330 | 12:53 |
| openstackgerrit | A change was merged to openstack/tempest: makes passing the client optional to utilities in scenario/manager.py https://review.openstack.org/47182 | 12:55 |
| *** mriedem has joined #openstack-qa | 12:57 | |
| *** krtaylor has joined #openstack-qa | 13:01 | |
| *** jasondotstar has joined #openstack-qa | 13:02 | |
| *** jhenner has quit IRC | 13:06 | |
| *** anteaya has joined #openstack-qa | 13:09 | |
| *** weshay has quit IRC | 13:09 | |
| *** prekarat has joined #openstack-qa | 13:10 | |
| *** prekarat has quit IRC | 13:10 | |
| openstackgerrit | Zhu Zhu proposed a change to openstack/tempest: Test for the nova diagnostics API https://review.openstack.org/48249 | 13:11 |
| *** SergeyLukjanov has quit IRC | 13:15 | |
| *** rfolco has joined #openstack-qa | 13:20 | |
| *** jhenner has joined #openstack-qa | 13:21 | |
| openstackgerrit | A change was merged to openstack-dev/devstack: make git show not display a diff https://review.openstack.org/48011 | 13:22 |
| *** weshay has joined #openstack-qa | 13:22 | |
| *** yassine has quit IRC | 13:36 | |
| *** yassine has joined #openstack-qa | 13:37 | |
| openstackgerrit | A change was merged to openstack/tempest: Added test to check list/show extensions-neutron https://review.openstack.org/45030 | 13:50 |
| *** DennyZhang has joined #openstack-qa | 13:55 | |
| openstackgerrit | Mate Lakat proposed a change to openstack-dev/devstack: xenapi: enable user interaction with stack.sh https://review.openstack.org/48092 | 13:59 |
| *** bdperkin_gone is now known as bdperkin | 14:06 | |
| *** matsuhashi has quit IRC | 14:07 | |
| *** nosnos has quit IRC | 14:08 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Update docs config https://review.openstack.org/46778 | 14:09 |
| *** dkranz has quit IRC | 14:10 | |
| *** malini_afk is now known as malini | 14:13 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/elastic-recheck: Reorganize files into subdirectories https://review.openstack.org/48264 | 14:18 |
| mtreinish | sdague: there bug with your devstack | 14:23 |
| mtreinish | wow finger slipped :) | 14:24 |
| mtreinish | sdague: there is an issue with your devstack change for printing out HEAD for each tree | 14:24 |
| sdague | mtreinish: ok, what is it? | 14:25 |
| mtreinish | it uses the default pager on my system (which is vimpager) so it opens up the pager for each tree | 14:25 |
| mtreinish | after each checkout | 14:25 |
| sdague | ah, gotcha | 14:26 |
| sdague | maybe I should just change it to head -1 | 14:26 |
| giulivo | mtreinish, where is the recheckbug bot gone? | 14:27 |
| giulivo | I just noticed a guy abusing of the no bug thing and wanted to see how the bot was doing | 14:28 |
| mtreinish | giulivo: it's probably stuck in a wait loop on ES (it's happens at least once a day) | 14:28 |
| mtreinish | let me restart it | 14:28 |
| giulivo | doh | 14:28 |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/devstack: change git show to | head -1 https://review.openstack.org/48265 | 14:28 |
| giulivo | mtreinish, I think mkoderer is right we should fix this skip thing with a proper decorator, I'd volunteer some work for that | 14:29 |
| giulivo | enforcing the message is only a 'temporary' measure | 14:29 |
| giulivo | what do you think? | 14:29 |
| mtreinish | giulivo: yeah I'm fine with that it would probably be cleaner | 14:30 |
| mtreinish | what about conditional skips though? | 14:30 |
| mtreinish | are you going to handle that too? | 14:30 |
| giulivo | I don't know how to cope with it, but we have one around | 14:30 |
| giulivo | a decorator could easily fix the multibug thing, but I'm not sure about the conditionals, ideas? | 14:31 |
| mtreinish | sdague: yeah that change should fix it | 14:31 |
| *** DennyZhang has quit IRC | 14:31 | |
| mtreinish | giulivo: you could just have an optional conditional argument for the decorator and do it that way | 14:31 |
| giulivo | you mean, passing as argument which skip to use? | 14:32 |
| mtreinish | I think we only have bugs numbers listed with testtools.skip() and testtools.skipIf() | 14:32 |
| mtreinish | giulivo: no I meant you have a decorator like tempest_skip('1234', skip_cond=bool_var) | 14:34 |
| mtreinish | if skip_cond is set then you can use skipIf with the condition specified | 14:34 |
| mtreinish | else you use a normal skip | 14:34 |
| *** RecheckWatchBot has quit IRC | 14:36 | |
| sdague | mtreinish: well a +1 would be appropriate then :) | 14:37 |
| giulivo | mtreinish, got it, thanks | 14:38 |
| mtreinish | sdague: I was waiting for jenkins, but sure why not | 14:38 |
| openstackgerrit | Zhu Zhu proposed a change to openstack/tempest: Test for the nova diagnostics API https://review.openstack.org/48249 | 14:40 |
| sdague | mkoderer: do you have time to check out why the stress test nightlies are failing | 14:46 |
| mkoderer | sdague: I know why.. | 14:46 |
| mkoderer | sdague: https://review.openstack.org/#/c/46087/ | 14:46 |
| mkoderer | but didn't had time to fix your findings | 14:46 |
| mkoderer | hopefully tomorrow | 14:47 |
| mkoderer | sdague: I mean with the change we will have at least the logs to see what's happening | 14:48 |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: driver.py should log errors when ssh operations fail https://review.openstack.org/48168 | 14:48 |
| sdague | mkoderer: ok, can you open a bug to track the wait issue? | 14:50 |
| sdague | then I'll +2 on that patch | 14:51 |
| mkoderer | sdague: sure I will do it | 14:51 |
| sdague | and we can pick up wait for a follow on | 14:51 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/devstack-gate: Add ability to do smart backwards compat testing https://review.openstack.org/41945 | 14:52 |
| openstackgerrit | afazekas proposed a change to openstack/tempest: The debug configuration group is not registered https://review.openstack.org/48271 | 14:54 |
| *** DennyZhang has joined #openstack-qa | 15:00 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/tempest: Add security groups updating test for ports https://review.openstack.org/46894 | 15:02 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/devstack-gate: Add ability to do smart backwards compat testing https://review.openstack.org/41945 | 15:11 |
| *** dkranz has joined #openstack-qa | 15:19 | |
| openstackgerrit | Mate Lakat proposed a change to openstack-dev/devstack: fix tee errors https://review.openstack.org/48279 | 15:19 |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: add __str__ function to RestClient class https://review.openstack.org/47350 | 15:32 |
| *** mlavalle has joined #openstack-qa | 15:32 | |
| openstackgerrit | afazekas proposed a change to openstack/tempest: Print the lekaed 'boto' tags on Failure https://review.openstack.org/48283 | 15:36 |
| mlavalle | mtreinish: ping | 15:36 |
| openstackgerrit | Marc Koderer proposed a change to openstack/tempest: Fix problem with never stopping stress tests https://review.openstack.org/46087 | 15:37 |
| mkoderer | sdague: ok I uploaded a new patchset and created bug https://bugs.launchpad.net/tempest/+bug/1230357 | 15:37 |
| mkoderer | sdague: let's wait until jenkins is done | 15:38 |
| *** jasondotstar has quit IRC | 15:52 | |
| openstackgerrit | afazekas proposed a change to openstack/tempest: Add xml support to the floating ip and router https://review.openstack.org/46050 | 15:59 |
| *** jlibosva has quit IRC | 16:03 | |
| *** SergeyLukjanov has joined #openstack-qa | 16:05 | |
| dkranz | afazekas_: Would you be able to look at https://bugs.launchpad.net/tempest/+bug/1229949 ? | 16:06 |
| dkranz | afazekas_: I could not figure out where the ERROR message is coming from. I submitted a patch for getting rid of the other ERROR on success in tempest logs. | 16:06 |
| dkranz | afazekas_: It is in boto somewhere | 16:07 |
| *** yassine has quit IRC | 16:07 | |
| afazekas_ | dkranz: yes | 16:08 |
| dkranz | afazekas_: Thanks! I am working on reporting bugs for the other services. | 16:09 |
| afazekas_ | dkranz: are you using tox ? what is your python version ? | 16:10 |
| dkranz | afazekas_: I'm not using tox at the moment. WHy? | 16:10 |
| dkranz | afazekas_: python 2.7 | 16:10 |
| afazekas_ | can you reproduce it with an individual test case run ? | 16:12 |
| afazekas_ | In the tempest.log, not on the console :) | 16:13 |
| afazekas_ | sorry | 16:13 |
| dkranz | afazekas_: I have not tried | 16:13 |
| dkranz | afazekas_: Should I? | 16:14 |
| *** bashok has quit IRC | 16:14 | |
| afazekas_ | do you see it in the console or in the log file ? | 16:15 |
| dkranz | afazekas_: log file | 16:15 |
| dkranz | afazekas_: You can see it in the tempest log file for any passing gate job | 16:16 |
| afazekas_ | So you would like to monkey patch the boto library in order to change the log level on negative tests ? | 16:18 |
| afazekas_ | if the status code is above 400 the boto library logs the response as an error | 16:19 |
| afazekas_ | the mentined exception is from a negative | 16:19 |
| afazekas_ | the mentioned error log is from a negative test case | 16:19 |
| afazekas_ | Or, Do you want to deny to the boto lib to log above the warning level by tricky log filter ? | 16:21 |
| openstackgerrit | Mate Lakat proposed a change to openstack-dev/devstack: xenapi: enable user to specify FLAT_NETWORK_BRIDGE https://review.openstack.org/48296 | 16:25 |
| openstackgerrit | Salvatore Orlando proposed a change to openstack-dev/devstack: Nicira plugin: do not die if NVP gateway IP is missing https://review.openstack.org/47414 | 16:26 |
| *** hemnafk is now known as hemna | 16:28 | |
| *** fbo is now known as fbo_away | 16:32 | |
| afazekas_ | dkranz: Do we need to distinguish the negative operation from the positive operations when logging ? | 16:32 |
| dkranz | afazekas_: Went to get lunch. Not sure what you mean. | 16:42 |
| dkranz | afazekas_: The idea was to fail builds if there is ERROR in the log | 16:42 |
| dkranz | afazekas_: So we should only log an ERROR if the build should fail as a result :) | 16:43 |
| afazekas_ | dkranz: so think thing is now we are logging ERRORs in the negative tests | 16:43 |
| dkranz | afazekas_: Just in that one case. | 16:44 |
| afazekas_ | not | 16:44 |
| dkranz | afazekas_: You mean outside of boto?> | 16:44 |
| dkranz | afazekas_: THere are lots of negative tests outside of boto that don't log ERROR | 16:44 |
| afazekas_ | I see | 16:45 |
| afazekas_ | I need to go | 16:45 |
| dkranz | ok | 16:46 |
| mkoderer | afazekas_: dkranz: could you have a look on https://review.openstack.org/#/c/46087/ | 16:46 |
| afazekas_ | https://review.openstack.org/#/c/46087/4/tempest/stress/stressaction.py | 16:48 |
| afazekas_ | L35 | 16:48 |
| afazekas_ | please log the exception | 16:48 |
| afazekas_ | log.exception(exc) | 16:48 |
| afazekas_ | mkoderer: ^^ | 16:49 |
| *** afazekas_ has quit IRC | 16:54 | |
| mkoderer | this is not needed | 16:54 |
| *** mlavalle has quit IRC | 16:54 | |
| *** marun has joined #openstack-qa | 16:55 | |
| *** salv-orlando has joined #openstack-qa | 17:00 | |
| salv-orlando | hello - is this the correct room to ask questions about tempest development? | 17:00 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Clean up automated changes to requirements https://review.openstack.org/45478 | 17:12 |
| *** afazekas has joined #openstack-qa | 17:16 | |
| *** jhenner has quit IRC | 17:16 | |
| *** mkollaro has quit IRC | 17:18 | |
| openstackgerrit | Dean Troyer proposed a change to openstack-dev/devstack: Add meta-config via local.conf https://review.openstack.org/46768 | 17:22 |
| sdague | mkoderer: cool, thanks | 17:22 |
| sdague | salv-orlando: yes it is | 17:23 |
| salv-orlando | oh great sdague! | 17:23 |
| salv-orlando | I guess you know I want to talk about advanced scenarios for neutron | 17:23 |
| salv-orlando | lack of them is possibly one reason for which several buggy patches slipped in during this release cycle | 17:23 |
| sdague | salv-orlando: sure, I'm just really concerned on the complexity of spinning up the second vm to test connectivity | 17:24 |
| sdague | because there are so many ways that could fail | 17:24 |
| salv-orlando | Yep. Too many moving parts. | 17:24 |
| salv-orlando | Especially because of the remote command execution. | 17:25 |
| sdague | I think that was the only recent -1 I threw on one of your patches, right? | 17:25 |
| salv-orlando | The alternative would be too instruct tempest to add an explicit route to the target network | 17:25 |
| salv-orlando | the private network (10.100.0.0/24 by default) | 17:25 |
| sdague | salv-orlando: I'd be ok with that | 17:25 |
| sdague | that seems more robust | 17:25 |
| salv-orlando | ok, I'll then add the needed infrastructure code to tempest (I think we don't have the bits to manipulate the routing table) | 17:26 |
| salv-orlando | then I will sort out how to test connectivity across a router without having to fire up two VMs. | 17:26 |
| sdague | salv-orlando: great | 17:26 |
| sdague | dkranz: mtreinish said that you were starting in on white listing? | 17:26 |
| salv-orlando | sdague: thanks | 17:27 |
| afazekas | more VM for one test case is not unique | 17:27 |
| sdague | afazekas: in this case the second VM was started solely to test ping connectivity | 17:28 |
| afazekas | What if we create an advanced neutron scenrio with multiple VMs for an extended testing ? | 17:28 |
| sdague | with the first VM | 17:28 |
| sdague | and that seemed very fragile to me | 17:28 |
| afazekas | 'fragile', but it is expected to work | 17:28 |
| afazekas | I do not want to skip the basic ops because the basic opes fails with 2 vm, so may be we should create an advanced ops in the future | 17:29 |
| afazekas | Which should contain more extensive security rule check as well | 17:29 |
| dosaboy | mkoderer: ping | 17:32 |
| *** SergeyLukjanov has quit IRC | 17:32 | |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: add __str__ function to RestClient class https://review.openstack.org/47350 | 17:33 |
| *** mlavalle has joined #openstack-qa | 17:38 | |
| *** RecheckWatchBot has joined #openstack-qa | 17:38 | |
| mtreinish | mlavalle: pong | 17:38 |
| mtreinish | sorry I was in meetings | 17:39 |
| mlavalle | mtreinish: np. just a quick question…. | 17:39 |
| mlavalle | mtreinish: in that test run, did you replace manager.py as well? | 17:39 |
| mtreinish | mlavalle: the only change I made for running the test was replace ever use of the tenant_isolation config variable with True | 17:40 |
| mtreinish | https://review.openstack.org/#/c/46330/ | 17:40 |
| salv-orlando | afazekas: let's keep basic ops with 1 vm | 17:41 |
| salv-orlando | all we need is to inject a static route in tempest | 17:41 |
| salv-orlando | than we can do more advanced use cases in another test scenario. | 17:41 |
| *** giulivo has quit IRC | 17:42 | |
| salv-orlando | multiple VMs with remote command execution probably should be avoided if we can do an equivalent test without | 17:42 |
| salv-orlando | but otherwise it's fine if the test is architected in a way to avoid sporadic failures due to timeouts/races | 17:42 |
| afazekas | salv-orlando: so you would like to have tempest to issue a command as root for adding route entry ? | 17:43 |
| salv-orlando | sounds a bit icky too - I have to say sdague was ok with that :) | 17:44 |
| mlavalle | mtreinish: I can see that…. but maybe you put your tempest/scenarin/manager.py on top of mine….. | 17:45 |
| afazekas | Isn't the neutron-debug probes could be usable more controlled way ? | 17:45 |
| mlavalle | mtreinish: the problem we've had since my previous patch is that test_network_basic_ops is creating the network with a non isolated tenant and the server with isolated credentials… so it obviously doesn't find the network and we get the 404 | 17:47 |
| mtreinish | mlavalle: oh I see now you added another use of the config variable in your latest patch | 17:47 |
| mtreinish | let me update the test patch | 17:48 |
| mtreinish | I just push the rebase button when you make a change | 17:48 |
| mlavalle | mtreinish: yeah, that's what I thought….;-) keep me posted in Gerrit…. thanks | 17:48 |
| salv-orlando | It might be possible to use them; I think devstack-gate enables the debug command | 17:49 |
| salv-orlando | so it should be possible to use them. | 17:49 |
| afazekas | sdague: I am unsure we want to manipulate the tested system as root for a test case, especially if it expects it is just one node system. However, I am fully ok to do root operation after a failure, in order to get more information about the situation. | 17:49 |
| salv-orlando | afazekas: however I was under the impression that test_network_basic_ops should work with and without neutron | 17:49 |
| salv-orlando | ie: also with nova-network | 17:49 |
| afazekas | it is neutron only | 17:50 |
| salv-orlando | ah ok. | 17:50 |
| salv-orlando | so the probes should be definitely more lightweight. | 17:50 |
| salv-orlando | that's what the devstack exercises use after all | 17:51 |
| openstackgerrit | Matthew Treinish proposed a change to openstack/tempest: Do not merge test of neutron tenant isolation. https://review.openstack.org/46330 | 17:51 |
| afazekas | nova network afaik does not know too much about routers :) | 17:51 |
| mtreinish | mlavalle: ok pushed. You know you can just make these changes yourself too? you just checkout 46330 with 'git review -d' and make any updates you need | 17:51 |
| mlavalle | mtreinish: cool, thanks…. | 17:52 |
| salv-orlando | afazekas: indeed routers test where, in my plans, for a neutron only scenario | 17:55 |
| salv-orlando | let me do some work around probes | 17:55 |
| salv-orlando | I might come back with an updated patch soon | 17:55 |
| afazekas | salv-orlando: are you sure you need new entry at tempest run time ? | 17:56 |
| salv-orlando | afazekas: a new entry for what? sorry I do not understand | 17:56 |
| afazekas | salv-orlando: route | 17:56 |
| salv-orlando | not yet. The probes will probably work without the route even with namespaces. I think it's ok to assume the neutron agent is running on the host where tempest runs, because I think that will be required for successful probe creation | 17:57 |
| afazekas | If you want to ping out from the vm , you can find ping-able address on the host | 17:59 |
| afazekas | I am unsure you really need to change the route table at the test case time, even if it is needed may be it would be better to do it at devstack time | 18:02 |
| * afazekas afazekas_away | 18:02 | |
| salv-orlando | ok see you later | 18:02 |
| *** afazekas is now known as afazekas_away | 18:02 | |
| sdague | well, lets see how bad the new thing would be, and we can sort it out later | 18:08 |
| salv-orlando | I don't mind doing and undoinf | 18:12 |
| dkranz | sdague: Sorry, was at meeting. | 18:14 |
| dkranz | sdague: I am working on generating bugs for the logs that have bogus ERRORs | 18:15 |
| dkranz | sdague: Then we can see if there is interest in fixing them. | 18:15 |
| sdague | ok, but not tooling around it? | 18:15 |
| dkranz | sdague: The infra group said they would put the code in devstack gate to check against a whitelist | 18:16 |
| dkranz | sdague: That still leaves creation of the whitelist and figuring out where it goes | 18:16 |
| dkranz | sdague: You had some thoughts about that, right? | 18:16 |
| sdague | yep | 18:16 |
| dkranz | sdague: So I was starting with the first necessary step | 18:16 |
| *** jaypipes has quit IRC | 18:17 | |
| dkranz | sdague: I've grepped the logs and am ready to start filing bugs | 18:17 |
| *** SergeyLukjanov has joined #openstack-qa | 18:17 | |
| sdague | dkranz: so I think we're on the hook for tooling, unless you got a specific volunteer out of infra on it | 18:23 |
| *** jasondotstar has joined #openstack-qa | 18:39 | |
| dkranz | sdague: Could be. I'll check in the logs :) | 18:42 |
| dkranz | sdague: IRC logs, that is. | 18:42 |
| dkranz | sdague: Should not be to hard in any event if we know exactly where the check should go. | 18:42 |
| *** giulivo has joined #openstack-qa | 18:44 | |
| sdague | yep, I just want to make sure it wasn't falling through the cracks | 18:47 |
| *** atiwari has joined #openstack-qa | 18:50 | |
| mtreinish | mlavalle: It failed the neutron job again, but I'm not sure it's related. Logs are here: http://logs.openstack.org/30/46330/7/check/gate-tempest-devstack-vm-neutron/919561d/ | 18:50 |
| mtreinish | afazekas_away: ^^^ it was a boto failure so maybe it's a known issue | 18:50 |
| dkranz | sdague: There are still a bunch of older jobs in tempest that need to be kicked since the failure yesterday. | 18:58 |
| dkranz | sdague: Can I do recheck no bug or is something else supposed to be done? | 18:58 |
| mtreinish | dkranz: the pyparsing failure thing? or is it another failure? | 18:58 |
| mtreinish | dkranz: because there is https://bugs.launchpad.net/openstack-ci/+bug/1229797 which was the issue yesterday that stopped the gate | 18:59 |
| dkranz | mtreinish: That's what I thought it was but I just looked and it is more varied. | 19:00 |
| dkranz | mtreinish: I am going to send something to the list and have the authors recheck or do whatever is required | 19:00 |
| *** hootjr has quit IRC | 19:11 | |
| *** RecheckWatchBot has quit IRC | 19:13 | |
| *** RecheckWatchBot has joined #openstack-qa | 19:14 | |
| dkranz | sdague: BTW, I'm tagging the bugs as "error-in-log" | 19:17 |
| sdague | cool | 19:17 |
| *** samalba has quit IRC | 19:32 | |
| *** samalba has joined #openstack-qa | 19:35 | |
| *** SergeyLukjanov has quit IRC | 19:47 | |
| openstackgerrit | A change was merged to openstack/tempest: Fix json version of bootable check https://review.openstack.org/47935 | 19:52 |
| EmilienM | hi, if anyone could review "add neutron metering agent support" > https://review.openstack.org/#/c/48042/ - thank you :) | 19:52 |
| *** adalbas has quit IRC | 19:53 | |
| *** malini is now known as malini_afk | 20:00 | |
| *** adalbas has joined #openstack-qa | 20:05 | |
| mlavalle | mtreinish: great, we got test_network_basic_ops working with isolated networks | 20:12 |
| mlavalle | mtreinish: if nothing else, we are cleaning up a lot tests. test_network_basic_ops was really messed up in the way it handled credentials | 20:13 |
| *** rockygrober has joined #openstack-qa | 20:14 | |
| *** rockyg has quit IRC | 20:16 | |
| openstackgerrit | A change was merged to openstack/tempest: Don't log 'Not Found' ERROR during image cleanup https://review.openstack.org/48143 | 20:23 |
| mtreinish | mlavalle: ok it looks good now it was just a flaky fail | 20:23 |
| mtreinish | let me do a quick rereview of the code | 20:23 |
| mlavalle | mtreinish: thanks | 20:25 |
| mtreinish | sdague, dkranz: can you take a look at https://review.openstack.org/#/c/45578/ it adds tenant isolation for neutron | 20:33 |
| dkranz | mtreinish: Looking... | 20:34 |
| mtreinish | dkranz: thanks | 20:34 |
| *** fbo_away is now known as fbo | 20:37 | |
| *** jhenner has joined #openstack-qa | 20:37 | |
| *** afazekas_away is now known as afazekas | 20:46 | |
| afazekas | mtreinish: 2013-09-25 18:24:25,603 State transition "pending" ==> "error" 13 second | 20:48 |
| afazekas | http://logs.openstack.org/30/46330/7/check/gate-tempest-devstack-vm-neutron/919561d/logs/screen-n-cpu.txt.gz#_2013-09-25_18_24_23_668 | 20:50 |
| afazekas | https://bugs.launchpad.net/tempest/+bug/1223161 | 20:54 |
| *** jasondotstar has quit IRC | 20:56 | |
| openstackgerrit | A change was merged to openstack-dev/devstack: Replace OpenStack LLC with OpenStack Foundation https://review.openstack.org/47438 | 20:58 |
| openstackgerrit | A change was merged to openstack-dev/devstack: fake hypervisor plugin https://review.openstack.org/47908 | 21:00 |
| *** DennyZhang has quit IRC | 21:03 | |
| afazekas | http://logs.openstack.org/30/46330/7/check/gate-tempest-devstack-vm-neutron/919561d/logs/screen-q-svc.txt.gz it has some issue as well, but not at the same time, I am unsure about is the neutron using the sub-transactions properly | 21:07 |
| *** bdperkin is now known as bdperkin_gone | 21:20 | |
| *** malini_afk is now known as malini | 21:26 | |
| *** rfolco has quit IRC | 21:32 | |
| *** malini is now known as malini_afk | 21:39 | |
| openstackgerrit | Thomas Maddox proposed a change to openstack-dev/devstack: Add back rpc_notifier when ceilometer is enabled https://review.openstack.org/48351 | 21:39 |
| *** malini_afk is now known as malini | 21:47 | |
| *** fbo is now known as fbo_away | 21:50 | |
| afazekas | salv-orlando: from the guest VM, per default you can reach the host system's eth0 ip without any additional trick | 21:53 |
| afazekas | with additional trick (masquerade) you can reach any external address including the 'well know' 8.8.8.8 | 21:53 |
| afazekas | are those good target for single vm connectivity test ? | 21:54 |
| afazekas | instead of hard coding the address of the eth0 we can can configure the address of the interface which is used for accessing the default gw | 21:55 |
| *** dkranz has quit IRC | 21:58 | |
| salv-orlando | that is true for the IP of any physical interface - of the host where the l3 agent is running | 22:06 |
| salv-orlando | and it doesn't work if the vm is not attached to a router | 22:10 |
| salv-orlando | afazekas ^^ | 22:10 |
| salv-orlando | I meant, of course, it does not work if the network is not attached to a router | 22:11 |
| afazekas | but the vm is attached in thet scenario | 22:11 |
| afazekas | that | 22:11 |
| salv-orlando | yup | 22:11 |
| salv-orlando | I guess is a valid connectivity test for all the plugins I know of | 22:11 |
| salv-orlando | not just for the gate - so it should be generally acceptable | 22:12 |
| afazekas | yes | 22:12 |
| salv-orlando | as long as the idea of running the command from the vm is acceptable | 22:12 |
| salv-orlando | because I can't ping the other way round without adding an explicit route | 22:12 |
| salv-orlando | seems weird, but that's because of namespaces | 22:12 |
| salv-orlando | If I want to ping from the tempest machine | 22:13 |
| salv-orlando | I need to do the ping from within the namespace | 22:13 |
| salv-orlando | eg: ip nents exec qr-<xxx> ping <vm internal ip> | 22:13 |
| salv-orlando | and that adds another assumptions that the l3 agent is colocated with the host running tempest | 22:13 |
| afazekas | I guess if the floating ip working on tcp connect scan the private ip is working as well | 22:14 |
| salv-orlando | what do you mean "scan the private ip"? | 22:14 |
| afazekas | Your above command is pinging in router namespace | 22:15 |
| afazekas | it is 'private' ip | 22:15 |
| salv-orlando | yes | 22:15 |
| salv-orlando | this kind of test actually would probably not add anything to public connectivity test we already have, as that segment (router - private interface) is already covered | 22:17 |
| salv-orlando | the public connectivity tests adds the floating ip - internal router interface segment | 22:17 |
| *** mriedem has quit IRC | 22:22 | |
| afazekas | https://review.openstack.org/#/c/22415/ this would add addition confidentially we ssh -right vm with the right meta-data | 22:22 |
| salv-orlando | yes but the bugs which in the commit message is mentioned as a blocker is still open | 22:25 |
| salv-orlando | afazekas: however, testing that you can reach your gateway, is probably not adding coverage. Testing that you have connectivity between two interfaces, be them vm or probes is what we would like to verify. Nevertheless, since we are running the tests with a wide open security group that does not make a lot of difference either. | 22:27 |
| salv-orlando | The original goal of the test, I think was to check that you had connectivity even if a router was not involved at all? | 22:28 |
| afazekas | sooner or later we will need two vm for connectivity test anyway | 22:29 |
| afazekas | the problem with extending the basic ops to be two vm test case, if it start collecting random failures, we might skip the whole test case | 22:31 |
| afazekas | so it would be an all or nothing test case, that's why I said to creating an advanced ops can be a better option | 22:32 |
| * afazekas afazekas_zz | 22:33 | |
| *** afazekas is now known as afazekas_zz | 22:33 | |
| *** rfolco has joined #openstack-qa | 22:36 | |
| salv-orlando | afazekas: I got your poiny | 22:37 |
| salv-orlando | * point | 22:37 |
| salv-orlando | then I am now thinking we might extract that test from basic ops | 22:38 |
| salv-orlando | and start moving it into advanced ops | 22:38 |
| salv-orlando | I did not do that before because I stupidly thought that suite was meant to be executed also for nova network | 22:38 |
| salv-orlando | because an internal connectivity check where you ping your gateway is not really adding a lot | 22:39 |
| salv-orlando | afazekas_zz ^^ | 22:39 |
| salv-orlando | one downside is that people running tempest on deployments without namespaces will not find anymore the vm connectivity test in basic ops - and therefore the solution might be to leave things as they are in basic_ops and do all the new stuff in advanced_ops | 22:41 |
| salv-orlando | and the new test_network_advanced_ops test might be added to the gate-tempest-devstack-vm-neutron only when we're comfortable the tests are stable enough | 22:46 |
| *** mlavalle has quit IRC | 22:48 | |
| *** atiwari has quit IRC | 22:52 | |
| *** weshay has quit IRC | 22:56 | |
| jog0 | mtreinish: elastic-recheck is now on openstack-infra | 23:03 |
| jog0 | and I added tempest-core and infra core as reviewers | 23:04 |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: fix typo in config.py https://review.openstack.org/48363 | 23:08 |
| *** dims has quit IRC | 23:14 | |
| *** salv-orlando has quit IRC | 23:28 | |
| *** dims has joined #openstack-qa | 23:31 | |
| openstackgerrit | DennyZhang proposed a change to openstack/tempest: fix typo in config.py https://review.openstack.org/48363 | 23:35 |
| *** bdperkin_gone is now known as bdperkin | 23:35 | |
| *** ken1ohmichi has joined #openstack-qa | 23:36 | |
| *** malini is now known as malini_afk | 23:37 | |
| *** rfolco has quit IRC | 23:40 | |
| *** salv-orlando has joined #openstack-qa | 23:48 | |
| *** hemna is now known as hemnafk | 23:52 | |
| *** malini_afk is now known as malini | 23:56 | |
| *** rockygrober has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!