| *** sridhar_ram has joined #senlin | 00:00 | |
| *** Qiming has quit IRC | 00:23 | |
| *** zzxwill has joined #senlin | 00:30 | |
| *** zzxwill has quit IRC | 00:30 | |
| *** zzxwill has joined #senlin | 00:31 | |
| *** zzxwill has quit IRC | 00:42 | |
| *** zzxwill has joined #senlin | 00:43 | |
| *** zzxwill has quit IRC | 00:52 | |
| *** zzxwill has joined #senlin | 00:56 | |
| *** zzxwill has quit IRC | 01:04 | |
| *** sridhar_ram has quit IRC | 01:10 | |
| *** sridhar_ram has joined #senlin | 01:10 | |
| *** zzxwill has joined #senlin | 01:11 | |
| *** sridhar_ram has quit IRC | 01:15 | |
| *** Qiming has joined #senlin | 01:19 | |
| *** zzxwill has quit IRC | 01:25 | |
| *** gongysh_ has joined #senlin | 01:46 | |
| *** gongysh_ has quit IRC | 01:50 | |
| *** elynn has joined #senlin | 01:52 | |
| *** elynn has quit IRC | 01:57 | |
| *** elynn has joined #senlin | 01:58 | |
| *** elynn_ has joined #senlin | 02:03 | |
| *** elynn has quit IRC | 02:03 | |
| *** gongysh_ has joined #senlin | 02:04 | |
| elynn_ | Morning | 02:05 |
|---|---|---|
| Qiming | morning | 02:06 |
| *** elynn__ has joined #senlin | 02:08 | |
| *** zzxwill has joined #senlin | 02:09 | |
| *** elynn_ has quit IRC | 02:11 | |
| *** zzxwill has quit IRC | 02:12 | |
| *** Yanyanhu has joined #senlin | 02:16 | |
| *** elynn_ has joined #senlin | 02:21 | |
| *** elynn__ has quit IRC | 02:23 | |
| *** zzxwill has joined #senlin | 02:35 | |
| *** shu-mutou-AFK is now known as shu-mutou | 02:40 | |
| *** sridhar_ram1 has joined #senlin | 02:41 | |
| *** sridhar_ram1 has quit IRC | 02:45 | |
| *** zzxwill has quit IRC | 02:53 | |
| *** elynn_ has quit IRC | 04:07 | |
| *** gmann has joined #senlin | 04:11 | |
| xuhaiwei | Hi Qiming | 04:12 |
| xuhaiwei | are you there? | 04:12 |
| *** elynn_ has joined #senlin | 04:41 | |
| *** elynn_ has quit IRC | 04:46 | |
| *** elynn_ has joined #senlin | 04:46 | |
| *** zzxwill has joined #senlin | 04:48 | |
| *** zzxwill has quit IRC | 04:52 | |
| *** elynn__ has joined #senlin | 05:00 | |
| *** zzxwill has joined #senlin | 05:01 | |
| *** elynn_ has quit IRC | 05:02 | |
| Qiming | xuhaiwei, yes, I'm back | 05:16 |
| xuhaiwei | welcome | 05:16 |
| xuhaiwei | about api, stress, sceranio test, I asked gmann who is a tempest core | 05:17 |
| xuhaiwei | he said we need to use tempest external plugin to do this | 05:17 |
| Qiming | and the code will live in senlin, right? | 05:17 |
| xuhaiwei | the details are like what we had in mid-cycle, Congress is an example | 05:17 |
| xuhaiwei | yes | 05:18 |
| Qiming | as part of functional tests | 05:18 |
| *** Yanyanhu has quit IRC | 05:18 | |
| *** Yanyanhu has joined #senlin | 05:18 | |
| xuhaiwei | yes, the source layer we can manage them by ourselves | 05:18 |
| *** zzxwill has quit IRC | 05:18 | |
| xuhaiwei | https://github.com/openstack/congress/tree/master/congress_tempest_tests this is what congress does | 05:19 |
| xuhaiwei | if we want to user to run these tests from tempest, we can add this support | 05:19 |
| xuhaiwei | but if just ran from senlin side, that is not necessary IMO | 05:20 |
| Qiming | partially understood | 05:21 |
| Qiming | how would tempest know we have tests for it to run? | 05:21 |
| xuhaiwei | gmann is in this channel, if you have questions | 05:21 |
| xuhaiwei | that is a good question, I think they don't know :) | 05:22 |
| gmann | Qiming: xuhaiwei hi | 05:23 |
| Qiming | hi, gmann | 05:23 |
| gmann | actually if you implement tempest tests in senlin using tempest external plugin | 05:23 |
| gmann | and make entry of that plugin in setup.cfg | 05:24 |
| gmann | then tempest will detect that plugin and set up all tests env as instructed by plugin | 05:24 |
| gmann | it will run same way as devstack plugin works | 05:24 |
| Qiming | okay, this line: https://github.com/openstack/congress/blob/master/setup.cfg#L39 | 05:24 |
| gmann | and once tempest detect the plugin, you can run tests from tempst | 05:25 |
| gmann | Qiming: yes | 05:25 |
| Qiming | tempest searches for the 'tempest.test_plugins' name space | 05:25 |
| Qiming | one more question | 05:25 |
| gmann | all detected plugin ca be seen by tempest list-plugins command | 05:25 |
| gmann | Qiming: yea | 05:25 |
| Qiming | how do we trigger tempest? is it supposed to be set up as a gate? | 05:26 |
| *** zzxwill has joined #senlin | 05:26 | |
| gmann | Qiming: you can trigger tempest using tox env or directly with testr | 05:26 |
| Qiming | ah, I see | 05:26 |
| Qiming | thank you very much, got it | 05:26 |
| gmann | http://docs.openstack.org/developer/tempest/overview.html#legacy-run-method | 05:27 |
| gmann | there are multiple way to trigger | 05:27 |
| gmann | tox -eall-plugin senlin etc | 05:28 |
| gmann | Qiming: this is detailed doc for plugin - http://docs.openstack.org/developer/tempest/plugin.html | 05:28 |
| Qiming | that is helpful | 05:28 |
| gmann | and as example you can refer congress, Manila etc | 05:28 |
| Qiming | okay | 05:29 |
| Qiming | will spend time on them | 05:29 |
| Qiming | just curious about the role of tempest | 05:29 |
| gmann | Qiming: cool, let me or any QA core know if you face any issue | 05:29 |
| Qiming | I know defcore is using tempest for testing the api surface | 05:29 |
| gmann | Qiming: yea | 05:29 |
| Qiming | from the README.rst file, I saw it is also about scenario tests and stress tests | 05:30 |
| gmann | Qiming: Yea, along with API we can implement scenario tests etc also | 05:30 |
| Qiming | am I understanding this correctly? | 05:30 |
| gmann | Qiming: stress tests framework is deprecated as of now but we will make it available in separate repo | 05:31 |
| Qiming | for API tests, we will try find out if a specific API does its job correctly and fails gracefully as designed | 05:31 |
| gmann | Qiming: yea you are right. and any scenario tests you can implement | 05:31 |
| Qiming | for scenario tests, we will need to test much more complicated use cases | 05:31 |
| gmann | yea | 05:31 |
| gmann | tempest has lot of scenario tests with core projects - https://github.com/openstack/tempest/tree/master/tempest/scenario | 05:32 |
| gmann | Qiming: mainly tempest is more for those tests - integration tests | 05:32 |
| Qiming | so ... we will need to create tempest plugin, register them, add some/all api test cases, add scenario tests which can be based on api tests | 05:32 |
| gmann | API tests even can be just tested as functional tests in project repo itself | 05:33 |
| gmann | Qiming: exactly | 05:33 |
| Qiming | much clearer about this now | 05:33 |
| xuhaiwei | thanks, gmann | 05:34 |
| Qiming | any other things we should pay attention to before getting our hands dirty? | 05:34 |
| gmann | xuhaiwei: np! | 05:34 |
| Qiming | e.g. some common mistakes people make before? | 05:34 |
| gmann | Qiming: I think that would be enough, its easy to implement and try with few tests in starting | 05:34 |
| Qiming | good! | 05:34 |
| Qiming | many thanks, gmann | 05:35 |
| gmann | Qiming: np! | 05:35 |
| gmann | Qiming: xuhaiwei take care about deprecated configuration cred option | 05:36 |
| gmann | Qiming: xuhaiwei use dynamic cred option from tempest. | 05:36 |
| gmann | many project still use those deprecated cred and will be broken in future :) | 05:37 |
| Qiming | okay | 05:39 |
| *** zzxwill_ has joined #senlin | 05:42 | |
| *** zzxwill has quit IRC | 05:42 | |
| *** elynn_ has joined #senlin | 05:44 | |
| *** zzxwill_ has quit IRC | 05:44 | |
| *** zzxwill has joined #senlin | 05:46 | |
| *** elynn__ has quit IRC | 05:47 | |
| *** zzxwill has quit IRC | 05:53 | |
| *** zzxwill has joined #senlin | 06:00 | |
| *** zzxwill has quit IRC | 06:10 | |
| *** zzxwill has joined #senlin | 06:12 | |
| *** zzxwill has quit IRC | 06:23 | |
| *** zzxwill has joined #senlin | 06:31 | |
| openstackgerrit | xu-haiwei proposed openstack/senlin: A spec for container cluster service https://review.openstack.org/281102 | 06:31 |
| *** zzxwill has quit IRC | 06:54 | |
| *** zzxwill has joined #senlin | 06:57 | |
| *** Yanyan has joined #senlin | 07:02 | |
| *** Yanyanhu has quit IRC | 07:03 | |
| xuhaiwei | hi all, for design summit session, I added a new one S7, which is about inviting the non-senlin-created nodes to senlin, welcome to add comments about it | 07:08 |
| xuhaiwei | https://etherpad.openstack.org/p/newton-senlin-sessions | 07:08 |
| *** zzxwill has quit IRC | 07:13 | |
| *** zzxwill has joined #senlin | 07:14 | |
| *** zzxwill has quit IRC | 07:20 | |
| *** zzxwill has joined #senlin | 07:20 | |
| *** zzxwill has quit IRC | 07:25 | |
| *** zzxwill has joined #senlin | 07:34 | |
| *** zzxwill has quit IRC | 07:36 | |
| *** zzxwill has joined #senlin | 07:41 | |
| *** zzxwill has quit IRC | 08:05 | |
| *** zzxwill has joined #senlin | 08:06 | |
| *** zzxwill_ has joined #senlin | 08:09 | |
| *** zzxwill has quit IRC | 08:10 | |
| *** zzxwill_ has quit IRC | 08:20 | |
| *** zzxwill has joined #senlin | 08:26 | |
| *** zzxwill has quit IRC | 08:39 | |
| *** zzxwill has joined #senlin | 08:43 | |
| *** openstackgerrit has quit IRC | 08:47 | |
| *** zzxwill has quit IRC | 08:48 | |
| *** openstackgerrit has joined #senlin | 08:48 | |
| *** zzxwill has joined #senlin | 08:51 | |
| *** zzxwill has quit IRC | 09:00 | |
| lixinhui | Qiming, Yanyan | 09:02 |
| lixinhui | the problem of "pending t create", do you have similer things in your log: | 09:03 |
| lixinhui | ERROR oslo_messaging.rpc.dispatcher [-] Exception during message handling: Failed to build compute instance. | 09:03 |
| lixinhui | ? | 09:03 |
| Qiming | did you check nova log? | 09:09 |
| lixinhui | nothing there | 09:12 |
| *** zzxwill has joined #senlin | 09:13 | |
| Qiming | if a nova vm cannot be created, there must be something wrong | 09:13 |
| Qiming | check n-cpu | 09:13 |
| Qiming | n-sch | 09:14 |
| elynn_ | https://github.com/openstack/octavia/blob/master/octavia/common/exceptions.py#L154 | 09:14 |
| elynn_ | I think this error throw from neutron lbaasv2 | 09:15 |
| elynn_ | check octavia log? | 09:16 |
| Qiming | ... octavia is doing compute create | 09:18 |
| Qiming | an ComputeBuildException will be thrown if the create() call returns a failure, or the wait task gets a ERROR status from the VM | 09:18 |
| Qiming | the octavia exception is thrown here: https://github.com/openstack/octavia/blob/master/octavia/compute/drivers/nova_driver.py#L138 | 09:19 |
| elynn_ | Not so sure how it works, I'm still using haproxy backend... | 09:19 |
| Qiming | so I believe there must be a VM creation failure | 09:20 |
| Qiming | and that failure has to be checked from nova side | 09:20 |
| Qiming | also this line is stupid: | 09:20 |
| Qiming | https://github.com/openstack/octavia/blob/master/octavia/compute/drivers/nova_driver.py#L136 | 09:20 |
| Qiming | they should log the exception content for debugging's purpose | 09:21 |
| elynn_ | yes... | 09:21 |
| Qiming | got to see the full log for diagnosis | 09:22 |
| lixinhui | easy to reproduce | 09:23 |
| lixinhui | just set up devtsack from the latest code | 09:23 |
| Qiming | I cannot install devstack | 09:23 |
| Qiming | devstack is broken on rhel7 | 09:23 |
| Qiming | it cost me 3 hours yesterday | 09:23 |
| Qiming | https://github.com/openstack/octavia/blob/master/octavia/controller/worker/tasks/compute_tasks.py#L176-L184 | 09:24 |
| Qiming | the above is the only other case where the exception is thrown, it also points to the VM creation error | 09:24 |
| Qiming | so ... | 09:24 |
| lixinhui | yes | 09:25 |
| lixinhui | I think so | 09:25 |
| lixinhui | just nova boot can work pretty well | 09:25 |
| Qiming | how did you do nova-boot? | 09:26 |
| Qiming | from command line? | 09:26 |
| lixinhui | senlin node-create | 09:26 |
| Qiming | no | 09:26 |
| lixinhui | nova boot | 09:26 |
| lixinhui | cli | 09:26 |
| Qiming | octavia is secretly creating a vm | 09:26 |
| lixinhui | sure | 09:27 |
| Qiming | and it maybe passing some weird parameters | 09:27 |
| Qiming | you got to check the nova-compute log | 09:27 |
| lixinhui | ai | 09:27 |
| lixinhui | last night | 09:27 |
| lixinhui | you told me you have met this beforew | 09:27 |
| lixinhui | any experience there? | 09:27 |
| Qiming | yanyan has met some similar problems | 09:28 |
| Qiming | can you please paste your nova log? | 09:28 |
| Yanyan | hi | 09:29 |
| Yanyan | nova server creation failed? | 09:29 |
| Qiming | from the exception thrown, that is the only reason | 09:30 |
| Yanyan | hi, lixinhui, you mean the lbaas creation was pending there? | 09:31 |
| lixinhui | yes | 09:32 |
| lixinhui | but no error in nova log | 09:32 |
| Qiming | which process ? | 09:32 |
| Yanyan | is there any error message in lbaas's log? | 09:33 |
| lixinhui | no | 09:33 |
| Yanyan | I could be caused by incorrect configuration of haproxy | 09:33 |
| Yanyan | I recalled there are some driver related configure in lbaas configure file | 09:34 |
| Qiming | ... | 09:34 |
| Qiming | sigh | 09:34 |
| Qiming | this is not yet about haproxy based on the previous log pasted | 09:34 |
| Yanyan | yes... I met that before in lbaasv1 | 09:34 |
| Yanyan | oh | 09:34 |
| Yanyan | I missed your discussion | 09:34 |
| Yanyan | this one? | 09:35 |
| Qiming | cannot find the paste either | 09:35 |
| Yanyan | ERROR oslo_messaging.rpc.dispatcher [-] Exception during message handling: Failed to build compute instance. | 09:35 |
| Qiming | yes | 09:35 |
| Qiming | there is a stack trace for that exception | 09:35 |
| Yanyan | looks like nova's problem | 09:35 |
| lixinhui | that is the only thing in o-cw | 09:36 |
| Qiming | based on previous stack trace, I'm 100% sure it is a compute error: https://github.com/openstack/octavia/blob/master/octavia/controller/worker/tasks/compute_tasks.py#L176-L184 | 09:36 |
| Qiming | what is o-cw? | 09:37 |
| lixinhui | http://paste.openstack.org/show/489793/ | 09:37 |
| lixinhui | I am using octavia provider service | 09:38 |
| Qiming | great, let's look at the log | 09:38 |
| Qiming | that is the same thing you pasted me before | 09:39 |
| Qiming | it is a nova error | 09:39 |
| lixinhui | yes | 09:40 |
| Qiming | need your nova-compute log during 16:50 - 16:54 | 09:40 |
| lixinhui | just complete log of the error | 09:40 |
| lixinhui | I have to leave soon for an urgent VIO thing | 09:41 |
| lixinhui | thanks for your help here | 09:41 |
| lixinhui | I will get you tomorrow again | 09:41 |
| Yanyan | will keep looking at it | 09:41 |
| Yanyan | ttyl | 09:41 |
| lixinhui | and get call logs need to continue | 09:41 |
| lixinhui | thanks and sorry again. | 09:41 |
| *** zzxwill has quit IRC | 09:43 | |
| *** zzxwill has joined #senlin | 09:55 | |
| openstackgerrit | zzxwill proposed openstack/senlin: Refine Senlin manual install doc https://review.openstack.org/290367 | 09:55 |
| zzxwill | Besides, I wonder whether we need to add some verification step to check whether we successfully install Senlin server. | 09:57 |
| zzxwill | I checked all "senlin *-list", I found "senlin profile-type-list" can do the verification. | 09:58 |
| Qiming | okay, that is a good idea | 09:59 |
| Qiming | senlin build-info can do the job | 10:00 |
| *** zzxwill has quit IRC | 10:03 | |
| *** zzxwill has joined #senlin | 10:08 | |
| *** elynn_ has quit IRC | 10:10 | |
| *** zzxwill has quit IRC | 10:10 | |
| *** zzxwill has joined #senlin | 10:11 | |
| zzxwill | Okay, i will add it to the doc. Thanks. | 10:17 |
| *** Qiming has quit IRC | 10:32 | |
| *** gongysh_ has quit IRC | 10:34 | |
| *** Yanyan has quit IRC | 10:45 | |
| openstackgerrit | zzxwill proposed openstack/senlin: Refine Senlin manual install doc https://review.openstack.org/290396 | 11:10 |
| *** Qiming has joined #senlin | 11:22 | |
| *** zzxwill has quit IRC | 12:30 | |
| *** zzxwill has joined #senlin | 12:56 | |
| *** cschulz has joined #senlin | 12:58 | |
| *** zzxwill has quit IRC | 14:20 | |
| *** zzxwill has joined #senlin | 14:24 | |
| *** zzxwill has quit IRC | 14:40 | |
| *** zzxwill has joined #senlin | 14:42 | |
| *** zzxwill has quit IRC | 15:21 | |
| *** zzxwill has joined #senlin | 15:23 | |
| *** zzxwill has quit IRC | 15:29 | |
| *** zzxwill has joined #senlin | 15:30 | |
| *** zzxwill has quit IRC | 15:36 | |
| *** zzxwill_ has joined #senlin | 15:36 | |
| *** zzxwill_ has quit IRC | 15:49 | |
| *** zzxwill has joined #senlin | 15:51 | |
| *** Qiming has quit IRC | 15:53 | |
| *** zzxwill has quit IRC | 16:13 | |
| *** zzxwill has joined #senlin | 16:14 | |
| *** sridhar_ram has joined #senlin | 16:34 | |
| *** zzxwill has quit IRC | 16:34 | |
| *** zzxwill has joined #senlin | 16:37 | |
| *** sridhar_ram has quit IRC | 17:25 | |
| *** zzxwill has quit IRC | 17:34 | |
| *** zzxwill has joined #senlin | 18:06 | |
| *** sridhar_ram has joined #senlin | 18:27 | |
| *** zzxwill has quit IRC | 18:37 | |
| *** zzxwill has joined #senlin | 19:12 | |
| *** zzxwill_ has joined #senlin | 19:14 | |
| *** zzxwill has quit IRC | 19:17 | |
| *** zzxwill_ has quit IRC | 19:40 | |
| *** sridhar_ram1 has joined #senlin | 19:58 | |
| *** sridhar_ram has quit IRC | 19:59 | |
| *** zzxwill has joined #senlin | 20:22 | |
| *** zzxwill has quit IRC | 20:42 | |
| *** zzxwill has joined #senlin | 20:42 | |
| *** sridhar_ram1 has quit IRC | 21:13 | |
| *** sridhar_ram has joined #senlin | 21:14 | |
| *** zzxwill has quit IRC | 21:42 | |
| *** sridhar_ram1 has joined #senlin | 22:02 | |
| *** cschulz has quit IRC | 22:02 | |
| *** zzxwill has joined #senlin | 22:03 | |
| *** sridhar_ram has quit IRC | 22:04 | |
| *** zzxwill has quit IRC | 22:40 | |
| *** sridhar_ram1 has quit IRC | 22:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!