Tuesday, 2015-12-15

openstackgerritMerged openstack/senlin: Add unit tests to engine.parser  https://review.openstack.org/25724000:28
*** Qiming has joined #senlin00:42
*** Qiming has quit IRC01:04
*** openstack has joined #senlin01:09
*** zhenguo has joined #senlin01:19
*** Yanyanhu has joined #senlin01:50
*** cschulz has joined #senlin01:53
*** elynn has joined #senlin01:57
*** Tiancheng has joined #senlin01:57
*** elynn has quit IRC02:02
*** elynn has joined #senlin02:03
*** Qiming has joined #senlin02:18
*** gongysh has joined #senlin02:43
*** Liuqing has joined #senlin02:53
elynnHi Qiming03:11
elynnhttps://review.openstack.org/#/c/247429/6/heat/engine/resources/openstack/senlin/cluster.py L18303:12
elynnDo I need to put self.CLUSTER_WARNING here? or just cluster_error is ok?03:12
Qimingfor deletion, the expectation is that the operation (DELETE) is successful03:13
openstackgerritxu-haiwei proposed openstack/senlin: Change String to Boolean type when listing cluster policies  https://review.openstack.org/25766403:14
elynnyes, I remember that you said cluster might not go into cluster_warnning03:14
Qimingcluster may enter a warning status03:14
xuhaiweihi, I got a very strange pep8 error: senlin/profiles/os/nova/server.py:276:1: C901 'ServerProfile.do_create' is too complex (20)     def do_create(self, obj):03:15
Qimingthis has to be treated case by case when operating it03:15
elynnSo I should mark as delete failed, right?03:15
Qimingyes, your DELETE operation has failed03:15
Qimingbut the cluster may be still operational03:15
elynnOk, I will keep my codes as it is, and reply to KM03:16
Qimingthis is a big difference between Heat and Senlin03:16
QimingWe return the cluster/node status, not the status of the last operation03:16
Qimingthere won't be a status called 'CREATE_COMPLETE', that is only about the CREATE operation, it is not about the cluster/node itself03:17
elynnQiming: yes, but we can add check_live_status later to monitor status when convergence is on.03:17
Qimingxuhaiwei, the complexity is a PEP8 measurement03:17
Qimingit measures how many if..else ... kind of control statements in a function03:18
Qimingif it is becoming too complex03:18
Qimingyou are expected to split it into several smaller functions03:18
Qimingit is a nice check by the way03:19
Qimingelynn, yes03:19
*** yuanying has quit IRC03:20
*** pratikmallya has quit IRC03:36
*** elynn has quit IRC04:00
*** yuanying has joined #senlin04:07
xuhaiweiQiming, I have never met it before, it seems Jenkins doesn't value this measurement04:07
*** openstackstatus has quit IRC04:24
*** openstack has joined #senlin04:28
*** pratikmallya has joined #senlin04:36
*** pratikmallya has quit IRC04:41
*** elynn has joined #senlin04:42
*** elynn has quit IRC04:47
*** elynn has joined #senlin04:47
Qimingxuhaiwei, it is part of pep8 test04:54
Qimingsee section [flake8] max-complexity=20 in tox.ini04:55
xuhaiweiwhy is doesnt fail in the gate?04:55
Qimingyou mean the same code failed locally, but passed the gate?04:56
Qimingthat is weird04:56
xuhaiweiit doesn't happen in your env?04:56
Qimingno, senlin code was not that complex so far04:57
xuhaiweiit happened only today04:57
xuhaiweinever met it before04:57
xuhaiweibut senlin/profiles/os/nova/server.py:276:1: C901 'ServerProfile.do_create'  is indeed a little long04:57
QimingI used to do simple 'flake8' test locally before commit04:58
Qimingmaybe 'tox -e pep8' will give different results04:58
Qimingdo you have code not yet committed?04:58
xuhaiweiI did tox -epep804:58
Qimingokay, I'm building pep8 tox env now04:58
xuhaiweiit seem this happened after "Add block_device_mapping_v2 support for 'os.nova.server' profile." is merged05:01
Qimingwell... my tox -epep8 passed05:03
Qimingxuhaiwei, it is just an alarm05:20
Qiming[tengqm@node1 senlin]$ flake8 --max-complexity=1905:20
Qiming./senlin/profiles/os/nova/server.py:276:1: C901 'ServerProfile.do_create' is too complex (20)05:20
Qiming    def do_create(self, obj):05:20
Qiming[tengqm@node1 senlin]$ flake8 --max-complexity=2005:20
xuhaiweiyou got the alarm?05:21
Qimingsee my paste above05:21
Qimingif I'm setting max complexity to 19 manually, I'm getting an error05:21
xuhaiweiby --max-complexity=19 it happens?05:21
Qimingif I'm setting it to 20, the test passed05:22
Qimingin tox.ini, we have max_complexity=2005:22
xuhaiweiERROR: InvocationError:05:22
Qimingthat is why I'm not seeing it in my local setup05:22
xuhaiweiI checked tox.ini just now, max_complexity is 20 in my env05:22
xuhaiweiwhen I changed it to 22, it passed05:23
Qimingmaybe it is caused by tox or flake8 versions?05:24
QimingI have tox==2.1.1, flake8==2.4.105:24
xuhaiweiyes, maybe05:24
Qimingmaybe the gate was not complaining about this because it is running different versions05:24
xuhaiweimy flake8 is 2.2.405:24
xuhaiweilower than you05:25
xuhaiweiI updated flake8 to 2.5.1, but it still happens05:26
Qimingyour python version?05:27
*** Liuqing has quit IRC05:28
xuhaiweiit will pass only if max_complaxity>= 2105:28
Qimingone mitigation I can think of is for you to propose a refactor of the do_create() method05:29
Qimingmaybe just move some logics out into a private function05:29
Qimingthen the code passes pep8 for you and everyone else, hopefully05:29
Qimingyou can tag the commit as 'for haiwei', ;)05:30
Qiminggongysh, bro, are you there?05:32
xuhaiweiQiming, I deleted .tox/pep8 and ran tox -epep8 again, it passed this time05:33
Qimingokay, something is too old in your tox env05:33
xuhaiweiI will split that do_create() method when I am a little free05:34
Qiminglet's focus on things that are of higher priorities05:35
*** elynn has quit IRC05:52
*** elynn has joined #senlin05:56
*** elynn_ has joined #senlin06:01
*** elynn has quit IRC06:01
QimingYanyanhu, about bug #152581106:01
openstackbug 1525811 in senlin "Bug happened when attaching policy" [Undecided,New] https://launchpad.net/bugs/152581106:01
openstackgerritMerged openstack/senlin: Fix list data type implementation  https://review.openstack.org/25722306:01
Qimingplease help check if patch #257223 is fixing it06:01
gongyshQiming, hi06:10
Qiminggongysh, you guys are doing things related to auditing?06:11
gongyshQiming, I don't think so, but you can ask liuqing06:12
QimingI see06:12
Qimingjust saw a paper on auditing06:12
Qiminganyway, it is not a good paper06:12
gongysho, can u paste the URL?06:13
Qimingoh, shit06:13
Qimingthe title is 'Comments on a Public Auditing Mechanism for Shared Cloud Data Service'06:14
QimingOn IEEE Transactions on Services Computing06:14
QimingI just downloaded it, ...06:14
Qimingit is only two pages, full of equations06:15
Qimingforget it06:15
gongyshyes, definitely06:16
QimingI'm shift-deleting it!06:16
Yanyanhuhi, Qiming will check it. Thanks06:16
openstackgerritlvdongbing proposed openstack/senlin: Update node status when fail to create nova instance  https://review.openstack.org/23875306:23
Yanyanhuhi, Qiming, it works06:26
Qimingthanks, Yanyanhu06:26
Yanyanhuno problem06:26
Qimingxuhaiwei, I like the patch #23875306:26
xuhaiweiso let's merge it :)06:27
Qimingwait a minute06:27
Qimingthe test case is ....06:27
xuhaiweiyes, need to improve it06:29
Qimingsigh, it is removing a whole test case06:29
Qiming... it looks fine06:29
xuhaiweiwait_for_server is not tested06:29
Qimingactually, the previous test case is bad06:30
xuhaiweilvdongbing is not in irc?06:30
Qiminghe is not06:30
Qimingso let's get it in06:31
xuhaiweiyou can comment in the patch06:31
Qimingand file a bug about the test case?06:31
Qimingwe need the patch itself urgently06:31
Qimingit has been there for about 20 days now06:31
Qimingoh, no, 40 days06:32
xuhaiweiyou comment in the patch, he will update soon I think06:32
Qimingokay, will try this06:33
*** elynn_ has quit IRC07:12
*** elynn_ has joined #senlin07:23
*** pratikmallya has joined #senlin07:27
openstackgerritQiming Teng proposed openstack/senlin: Fix server get_details when server is not found  https://review.openstack.org/25771107:44
*** yuanying has quit IRC08:05
openstackgerritxu-haiwei proposed openstack/python-senlinclient: Don't show show_deleted column in node-list when not needed  https://review.openstack.org/25774208:52
Qimingxuhaiwei, about making _short_cluster_id a non-inner function09:03
QimingI'm wondering if we can make it a lamda function?09:03
xuhaiweii will think about it09:07
Qiminge.g. line 105509:07
Qiming'cluster_id': lambda x: x.cluster_id[:8] if x.cluster_id else ''09:08
Qimingnever tried it, but it may help remove the need for a function which is not used anywhere else09:08
xuhaiweiseems fine09:08
openstackgerritlvdongbing proposed openstack/senlin: Update node status when fail to create nova instance  https://review.openstack.org/23875309:12
Qimingem ... https://review.openstack.org/#/c/255983/09:19
xuhaiweiadded test not you wanted?09:21
Qimingwhich one?09:21
xuhaiweiI am wrong09:22
xuhaiweiwhat do you mean by 255983?09:22
Qimingcheck it09:22
*** gongysh has quit IRC09:23
*** xuhaiwei has quit IRC09:33
*** openstackgerrit has quit IRC09:47
*** openstackgerrit has joined #senlin09:48
openstackgerritQiming Teng proposed openstack/senlin: Implement purge_deleted command  https://review.openstack.org/25777609:49
Qiminghello, openstackgerrit, :)09:49
openstackgerritQiming Teng proposed openstack/senlin: Implement purge_deleted command  https://review.openstack.org/25777609:52
*** Tiancheng has quit IRC09:57
*** zhenguo has quit IRC10:08
*** Qiming has quit IRC10:10
*** elynn_ has quit IRC10:14
openstackgerritMerged openstack/senlin: Fix server get_details when server is not found  https://review.openstack.org/25771110:30
*** openstackgerrit has quit IRC10:32
*** openstackgerrit has joined #senlin10:33
*** Yanyanhu has quit IRC10:43
*** pratikmallya has quit IRC10:48
openstackgerritMerged openstack/senlin: Update node status when fail to create nova instance  https://review.openstack.org/23875310:52
*** Qiming has joined #senlin11:05
*** zhenguo has joined #senlin11:17
*** cschulz has quit IRC11:18
*** branw has quit IRC11:19
*** gongysh has joined #senlin11:53
*** gongysh has quit IRC11:54
*** Liuqing has joined #senlin12:09
*** lixinhui has joined #senlin12:16
*** Tiancheng has joined #senlin12:43
*** yanyanhu has joined #senlin12:48
*** elynn_ has joined #senlin12:58
*** elynn has joined #senlin12:58
*** Tiancheng has quit IRC13:19
*** Tiancheng has joined #senlin13:57
*** elynn has quit IRC14:03
*** yanyanhu has quit IRC14:03
*** lixinhui has quit IRC14:10
*** Qiming has quit IRC14:40
*** zhenguo has quit IRC14:41
*** Liuqing has quit IRC14:52
*** Liuqing has joined #senlin14:55
openstackgerritMerged openstack/senlin: delete __context__ that unused  https://review.openstack.org/25728315:06
openstackgerritMerged openstack/senlin: Change String to Boolean type when listing cluster policies  https://review.openstack.org/25766415:18
*** pratikmallya has joined #senlin15:30
*** pratikma_ has joined #senlin15:34
*** pratikmallya has quit IRC15:36
*** Liuqing has quit IRC15:37
*** Liuqing has joined #senlin15:39
*** Tiancheng has quit IRC16:10
*** Liuqing has quit IRC16:49
openstackgerritOpenStack Proposal Bot proposed openstack/senlin: Updated from global requirements  https://review.openstack.org/25807419:00
*** pratikma_ has quit IRC19:18
*** pratikmallya has joined #senlin19:26
*** openstackgerrit has quit IRC19:32
*** openstackgerrit has joined #senlin19:33
*** pratikma_ has joined #senlin20:07
*** pratikmallya has quit IRC20:09
*** yuanying has joined #senlin23:02
*** yuanying has quit IRC23:03
*** yuanying has joined #senlin23:04
*** xuhaiwei_ has joined #senlin23:35
*** Qiming has joined #senlin23:39

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!