*** Yanyanhu has joined #senlin | 01:10 | |
*** mathspanda has joined #senlin | 01:50 | |
*** Tiancheng has joined #senlin | 01:52 | |
*** Qiming has joined #senlin | 02:03 | |
*** Zhenqi has joined #senlin | 02:07 | |
Qiming | morning, all | 02:10 |
---|---|---|
xuhaiwei | I will have lunch soon :) | 02:22 |
Yanyanhu | morning | 02:29 |
Yanyanhu | have a good breakfast ;) | 02:29 |
*** Tennyson has joined #senlin | 02:30 | |
xuhaiwei | Qiming, about 'cluster-resize' I found a problem | 03:09 |
*** Zhenqi has quit IRC | 03:09 | |
xuhaiwei | if you just a -m option without -c option, even max_size is smaller than capacity, there is no exception happens | 03:10 |
xuhaiwei | because 'capacity' only gets value from tha args not database | 03:11 |
Qiming | xuhaiwei, the expected behavior would be that the cluster will be shrinked to the new max-size | 03:12 |
Qiming | there won't be exception | 03:12 |
*** Zhenqi has joined #senlin | 03:13 | |
xuhaiwei | ok | 03:14 |
Qiming | it was designed to be a multi-purposed command | 03:15 |
Qiming | maybe it was a bad design | 03:15 |
Qiming | but splitting it into many sub-commands will make things even worse I think | 03:15 |
xuhaiwei | make -c -m -n all required options will solve this problem.. | 03:16 |
Qiming | yes, there are many corner cases that need to be covered | 03:17 |
Qiming | no, we cannot assume that | 03:17 |
Qiming | it is much easier for the code, but more restrictive to users | 03:17 |
xuhaiwei | yes, also think so | 03:18 |
Yanyanhu | maybe we can start work on functional test with this as a startpoint :) | 03:18 |
Qiming | yes, ... | 03:19 |
Qiming | but I'd like to vote some basic functional tests first | 03:19 |
Yanyanhu | will try to see something about this | 03:19 |
xuhaiwei | I will go for lunch now, see you later | 03:19 |
Yanyanhu | yep | 03:19 |
Yanyanhu | ttyl | 03:19 |
Qiming | bon appetie | 03:19 |
Yanyanhu | will try to see something about functional test after finishing the test case of lb policy | 03:20 |
Qiming | cool | 03:20 |
Yanyanhu | guess need to refer to some existing works in other services ;) | 03:20 |
Qiming | it won't be hard | 03:25 |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Refactor plugin name checking logic https://review.openstack.org/207281 | 03:25 |
Qiming | just do a RC sourcing and assume you are working in a functional devstack environment | 03:25 |
Qiming | whatever real life functions can be verified in that way | 03:25 |
Qiming | Yanyanhu, xuhaiwei, please help review 207281 | 03:26 |
Qiming | need it for moving on | 03:26 |
*** mathspanda has quit IRC | 03:27 | |
Yanyanhu | ok, no problem | 03:28 |
Yanyanhu | Qiming, I noticed that the gate become very slow recently | 03:52 |
Yanyanhu | is that our problem? | 03:52 |
Qiming | gate crashed again last night | 03:52 |
Qiming | it is recovering | 03:52 |
Yanyanhu | is it possible that some bad-behavior test cases cost too much time? | 03:52 |
Qiming | this time due to disk full | 03:52 |
Yanyanhu | oh | 03:52 |
Yanyanhu | sigh... | 03:52 |
Qiming | it now has a long backlog | 03:53 |
Yanyanhu | how you get this issue... | 03:53 |
Yanyanhu | this information | 03:53 |
Yanyanhu | I was afraid that some test cases that include 'sleep' didn't work as expected... | 03:53 |
*** mathspanda has joined #senlin | 04:19 | |
*** Tiancheng has quit IRC | 04:34 | |
*** Yanyanhu has quit IRC | 04:42 | |
*** Yanyanhu has joined #senlin | 04:43 | |
*** Tiancheng has joined #senlin | 05:22 | |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Refactor plugin name checking logic https://review.openstack.org/207281 | 05:49 |
Zhenqi | http://paste.openstack.org/show/406337/ | 06:05 |
Zhenqi | look at the 'role' | 06:05 |
Zhenqi | updated successfully | 06:06 |
Zhenqi | but report a exception | 06:06 |
*** Tiancheng has quit IRC | 06:09 | |
xuhaiwei | Zhenqi, seems like a bug | 06:10 |
*** Tiancheng has joined #senlin | 06:10 | |
openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for LoadBalancingPolicy(2) https://review.openstack.org/206952 | 06:25 |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Refactor plugin name checking logic https://review.openstack.org/207281 | 06:40 |
Qiming | Zhenqi, have you pinpoined the problem? | 06:40 |
Zhenqi | still looking for it | 06:43 |
Qiming | Yanyanhu, nova v2 driver test case | 06:43 |
Qiming | are you working on it? | 06:43 |
Qiming | or there is patch on that? | 06:44 |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Fix some time related function calls https://review.openstack.org/206318 | 06:46 |
Yanyanhu | hi, Qiming, yes I'm trying work on it. Just wait for the exception handling refactor | 06:48 |
Yanyanhu | in driver layer | 06:48 |
Qiming | okay, mind if I take it over? | 06:48 |
Yanyanhu | no problem | 06:48 |
Qiming | exception handling cannot get done over night I'm afraid | 06:48 |
Qiming | we have to keep the wheels rolling | 06:49 |
Yanyanhu | yes, I think so... | 06:49 |
Yanyanhu | ok | 06:49 |
xuhaiwei | Qiming, I am looking up to Zhenqi's report, it seems the server response is not in json format | 06:49 |
xuhaiwei | really strange | 06:49 |
Qiming | at the same time, your review on https://review.openstack.org/207079 would be helpful | 06:49 |
Qiming | xuhaiwei, so it looks like a client side problem? | 06:50 |
Qiming | xuhaiwei, you mean you can reproduce it at your side? | 06:50 |
Yanyanhu | sure, will watch it | 06:50 |
xuhaiwei | senlin's client is not perfect , but it is not the main problem | 06:50 |
xuhaiwei | SDK reported this to us | 06:50 |
Yanyanhu | I'm still trying to recover my env... now working on setup senlin service | 06:50 |
xuhaiwei | yes, I reproduced it | 06:50 |
Qiming | there some exceptions escaped from SDK? | 06:51 |
xuhaiwei | the exception comes from sdk | 06:51 |
xuhaiwei | senlin server updated the node info and response to SDK, but SDK found it is not a Json format, so sdk raised an exception | 06:52 |
xuhaiwei | sdk called it InvalidResponse | 06:52 |
Qiming | not JSON format? | 06:52 |
Qiming | ... | 06:52 |
Qiming | how could that happen? | 06:52 |
xuhaiwei | ('exception: ', JSONDecodeError('Extra data: line 1 column 5 - line 5 column 4 (char 4 - 58)',)) | 06:53 |
xuhaiwei | seems so, can't believe that | 06:53 |
Qiming | okay, it does seem happen frequently | 06:54 |
Qiming | need to look into it | 06:55 |
xuhaiwei | the exception happened in this line https://github.com/stackforge/python-openstacksdk/blob/master/openstack/transport.py#L272 | 06:55 |
Qiming | it is mainly a exception parsing problem, like the one you fixed before | 06:55 |
Qiming | okay, check the resp content then | 06:56 |
Qiming | maybe line 259 in that file is doing a poor job | 06:56 |
Yanyanhu | hi, Qiming, just noticed that sdk's latest release has been 6.0 | 06:59 |
Qiming | okay | 07:04 |
*** Zhenqi has quit IRC | 07:23 | |
openstackgerrit | LinPeiyu proposed stackforge/senlin: Fix cluster-resize operation param checking in server side https://review.openstack.org/207325 | 07:35 |
Qiming | cool, mathspanda | 07:35 |
Qiming | this is your first contribution to OpenStack? | 07:35 |
mathspanda | yeah, i am excited | 07:36 |
Qiming | I'm excited too!!! | 07:36 |
Yanyanhu | con :) | 07:36 |
xuhaiwei | mathspanda, good start | 07:38 |
*** mathspanda has quit IRC | 07:50 | |
*** mathspanda has joined #senlin | 07:53 | |
Yanyanhu | hi, Qiming, just looked through this patch https://review.openstack.org/#/c/207079/ | 08:00 |
Yanyanhu | it's looks good, just one question, do we need to test the exception handling logic that wrapped in decorator? | 08:00 |
Qiming | that belongs to the sdk test case right? | 08:01 |
Yanyanhu | BTW, it's very nice coding to wrap this exception parsing into this decorator :) | 08:01 |
Yanyanhu | hmm, makes sense | 08:02 |
Yanyanhu | so that part will come later | 08:03 |
Qiming | yes | 08:03 |
Yanyanhu | ok | 08:03 |
*** mathspanda has quit IRC | 08:13 | |
*** mathspanda has joined #senlin | 08:15 | |
*** LinPeiyu has joined #senlin | 08:45 | |
xuhaiwei | Qiming, when updating a node/profile/cluster, the server will return 202 first? | 08:45 |
*** LinPeiyu has left #senlin | 08:45 | |
xuhaiwei | updating profile returns 200 directly | 08:45 |
*** Chris-IBM has joined #senlin | 08:46 | |
Qiming | yes I think so | 08:46 |
Qiming | 202 means things are still ongoing | 08:46 |
Qiming | 200 means it is done | 08:46 |
xuhaiwei | updating node/cluster returns 202, which is not a JSON format, and sdk complains it | 08:46 |
Qiming | http://git.openstack.org/cgit/stackforge/senlin/tree/senlin/api/openstack/v1/clusters.py#n219 | 08:48 |
xuhaiwei | yes, this is what I want to ask | 08:49 |
Qiming | we are following the webob convention to raise a HTTPAccepted exceptoin | 08:49 |
xuhaiwei | this is exception is caught by whom | 08:49 |
Qiming | it is not an exception | 08:49 |
Qiming | it is just a way to send back 202 | 08:49 |
Qiming | http://git.openstack.org/cgit/openstack/heat/tree/heat/api/openstack/v1/stacks.py#n446 | 08:50 |
xuhaiwei | the content of the response is "202 Accepted\n\nThe request is accepted for processing.\n\n {u'cluster': {u'timeout': None, u'profile_id': None, u'name': u'cluster1', u'parent': None, u'metadata': {}}} ") | 08:50 |
xuhaiwei | can we modify it? | 08:50 |
Qiming | it should be handled by SDK I think | 08:51 |
Qiming | it is actually formed here: https://github.com/Pylons/webob/blob/master/webob/exc.py#L385 | 08:54 |
Yanyanhu | hi, Qiming, currently, integration test in openstack is usually done using tempest? | 09:16 |
Qiming | don't think so | 09:16 |
Qiming | most projects are moving functional tests back to their own repo | 09:17 |
Yanyanhu | yes, saw related dirs | 09:17 |
Qiming | the CI cannot afford the workload to run all functional tests from all projects | 09:17 |
Yanyanhu | ok | 09:17 |
Yanyanhu | talk with Bo, back later | 09:17 |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Unit test cases for nova driver https://review.openstack.org/207353 | 09:17 |
Qiming | 16:57 NOTICE: CI is back online but has a huge backlog. Please be patient and if possible delay approving changes until it has caught up. | 09:20 |
Qiming | 15:48 NOTICE: CI system is broken and very far behind. Please do not approve any changes for a while. | 09:20 |
Qiming | 15:38 NOTICE: Our CI system is broken again today, jobs are not getting processed at all. | 09:21 |
Qiming | mathspanda, there? | 09:22 |
mathspanda | yes | 09:24 |
Yanyanhu | Hi, Qiming, I'm back, so the functional test is completely supported with code inside each project? | 09:27 |
Yanyanhu | with only tox's help? | 09:27 |
Qiming | yes | 09:27 |
Qiming | mathspanda, about your patch | 09:27 |
Qiming | two problems | 09:27 |
Qiming | one is trivial | 09:27 |
Qiming | we usually limit length of each line in the commit message to be <80 chars | 09:28 |
Qiming | this will make it easy for all kinds of terminals to check your commit message | 09:28 |
Qiming | the other one is not significant either | 09:28 |
Qiming | in your code there is a trailing space | 09:29 |
Qiming | this is not acceptable by the PEP coding style, thus need to be removed | 09:29 |
Qiming | other than these two nits, the patch looks good | 09:29 |
mathspanda | oh, i know. i will fix them soon. | 09:30 |
Qiming | you can revise it and do a | 09:30 |
Qiming | git commit --amend | 09:30 |
Qiming | followed by a | 09:30 |
Qiming | git review | 09:30 |
mathspanda | ok. thanks | 09:30 |
Qiming | that will make sure the "Change-Id" remains the same, and gerrit will be happy about it | 09:30 |
mathspanda | OK | 09:31 |
Qiming | Chris is another name for you, mathspanda? | 09:31 |
Qiming | You may want to avoid using many different names | 09:32 |
Qiming | it would be difficult for people to recognize you, so you will have problems creating your identity | 09:33 |
mathspanda | no, he just use my virtualbox image | 09:33 |
Qiming | it is perfectly okay if you have a IRC name here, but when you are submitting patches in a different name | 09:33 |
*** Tennyson has quit IRC | 09:33 | |
mathspanda | and login xchart without modify username | 09:33 |
Qiming | if you have some problems logging in, just follow the conventions here | 09:34 |
Qiming | use mathspanda_, mathspanda__ etc | 09:34 |
Qiming | so people know what happened | 09:34 |
xuhaiwei | mathspanda is a good name i think | 09:34 |
Qiming | yes, it is a cool name | 09:34 |
mathspanda | yes, i use mathspanda everywhere | 09:35 |
Qiming | then please abandon 'Chris-IBM', :) | 09:35 |
mathspanda | Chris-IBM is another person | 09:38 |
Qiming | ... it shows to me the "real name" is LiuPeiyu as well, :) | 09:38 |
mathspanda | he just use my ubuntu image, and login irc without modifying any my personal data | 09:39 |
*** mathspanda has quit IRC | 09:43 | |
xuhaiwei | Qiming, it's not good to modify SDK's design I think, because they desire the update response having a body https://github.com/stackforge/python-openstacksdk/blob/master/openstack/resource.py#L745 | 09:46 |
*** Chris-IBM has quit IRC | 09:51 | |
Qiming | xuhaiwei, I don't agree | 10:01 |
xuhaiwei | you think the response is ok without 'body'? | 10:02 |
openstackgerrit | Merged stackforge/senlin: Fix some time related function calls https://review.openstack.org/206318 | 10:06 |
Qiming | something is wrong in sdk | 10:07 |
Qiming | they need to handle the 201 case correctly | 10:07 |
xuhaiwei | 202? | 10:07 |
Qiming | now it seems they are only relying on the requests package, and only handle 300, 301 ... redirect requests | 10:08 |
*** Yanyanhu has quit IRC | 10:09 | |
*** Tiancheng has quit IRC | 10:10 | |
Qiming | 202's response body is not a json | 10:16 |
xuhaiwei | yes, it's just a object, has no body | 10:18 |
Qiming | there is body | 10:19 |
Qiming | a response always has a body | 10:20 |
Qiming | the problem is that the body is not JSON | 10:20 |
xuhaiwei | I am afraid not | 10:21 |
xuhaiwei | in 202's case, it is called text | 10:22 |
Qiming | text is not a body? | 10:23 |
Qiming | in a http request/response | 10:23 |
Qiming | you have only header and body, right? | 10:23 |
xuhaiwei | header and content | 10:23 |
xuhaiwei | header is {'date': 'Thu, 30 Jul 2015 10:17:56 GMT', 'connection': 'keep-alive', 'content-type': 'text/plain; charset=UTF-8', 'content-length': '58', 'x-openstack-request-id': 'req-cd280953-bfeb-4d31-8102-fff5e2522ca9'}) | 10:24 |
xuhaiwei | content is '202 Accepted\n\nThe request is accepted for processing.\n\n ' | 10:24 |
Qiming | ... | 10:25 |
Qiming | so only a json has a body? | 10:26 |
xuhaiwei | I mean we can't read the response's body by resp.body | 10:28 |
Qiming | I am not talking about the code | 10:28 |
Qiming | the code can be changed any way | 10:28 |
Qiming | I'm focusing on the problem | 10:28 |
Qiming | the problem is that the ...."content" (okay for you?) is not a json, right? | 10:29 |
xuhaiwei | I think this is the first problem | 10:29 |
xuhaiwei | yes, this is the problem we meet right now | 10:30 |
xuhaiwei | the content is 'text/plain' format | 10:30 |
Qiming | ... | 10:32 |
Qiming | we can control that, right? | 10:32 |
xuhaiwei | yes | 10:32 |
Qiming | somewhere around here: http://git.openstack.org/cgit/stackforge/senlin/tree/senlin/common/wsgi.py#n567 | 10:32 |
Qiming | actually, maybe here: http://git.openstack.org/cgit/stackforge/senlin/tree/senlin/common/wsgi.py#n595 | 10:34 |
xuhaiwei | you mean the place where the exception happens? | 10:35 |
xuhaiwei | where to handle it? | 10:35 |
Qiming | need to do some serialization so that the response will form a JSON | 10:36 |
Qiming | "where to handle it?" ? what do you mean? | 10:36 |
xuhaiwei | I got you | 10:36 |
xuhaiwei | you want to make it a JSON before response to sdk? | 10:37 |
Qiming | yes | 10:37 |
Qiming | any other way out? | 10:37 |
xuhaiwei | I was try my best to fix sdk side... | 10:37 |
Qiming | that was my initial thought as well | 10:38 |
Qiming | but ... SDK's Transport class is simply a sub-class of the requests package | 10:38 |
Qiming | it would be difficult to do that, if possible | 10:38 |
Qiming | got to run | 10:39 |
xuhaiwei | ok | 10:39 |
Qiming | very hungry now | 10:39 |
xuhaiwei | me too | 10:39 |
xuhaiwei | I will think about it tomorrow | 10:39 |
Qiming | ok | 10:39 |
Qiming | maybe we need to resync the wsgi module from Heat | 10:39 |
xuhaiwei | I will check heat side also | 10:40 |
Qiming | the original code was from Heat, and maybe eventually from Glance | 10:40 |
Qiming | leaving | 10:40 |
Qiming | bye | 10:40 |
xuhaiwei | bye | 10:40 |
*** Qiming has quit IRC | 10:44 | |
*** Qiming has joined #senlin | 11:35 | |
*** mathspanda has joined #senlin | 11:50 | |
openstackgerrit | LinPeiyu proposed stackforge/senlin: Fix cluster-resize operation param checking in server side https://review.openstack.org/207325 | 11:58 |
*** Tiancheng has joined #senlin | 12:05 | |
*** mathspan_ has joined #senlin | 12:12 | |
*** mathspan_ is now known as mathspanda_ | 12:13 | |
*** mathspanda has quit IRC | 13:16 | |
*** Tiancheng has quit IRC | 14:08 | |
*** Qiming has quit IRC | 14:22 | |
*** mathspanda_ has quit IRC | 14:45 | |
*** jruano has joined #senlin | 15:40 | |
openstackgerrit | Julio Ruano proposed stackforge/senlin: Add test case for Nova Server Profile https://review.openstack.org/205845 | 19:07 |
openstackgerrit | Julio Ruano proposed stackforge/senlin: Add test case for Nova Server Profile https://review.openstack.org/205845 | 19:36 |
*** openstackgerrit has quit IRC | 19:46 | |
*** openstackgerrit has joined #senlin | 19:47 | |
*** openstackgerrit has quit IRC | 20:16 | |
*** openstackgerrit has joined #senlin | 20:17 | |
*** openstackgerrit has quit IRC | 20:46 | |
*** openstackgerrit has joined #senlin | 20:47 | |
*** jruano has quit IRC | 20:49 | |
*** openstack has joined #senlin | 21:11 | |
*** Qiming has joined #senlin | 22:55 | |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Unit test cases for nova driver https://review.openstack.org/207353 | 23:10 |
openstackgerrit | Qiming Teng proposed stackforge/senlin: Unit test case for ceilometer v2 driver https://review.openstack.org/207079 | 23:10 |
*** Qiming has quit IRC | 23:32 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!