| *** Qiming_ has quit IRC | 00:11 | |
| openstackgerrit | Merged stackforge/senlin: Revised design document for webhooks https://review.openstack.org/190071 | 00:13 |
|---|---|---|
| *** Yanyanhu has joined #senlin | 01:20 | |
| *** Yanyan has joined #senlin | 01:22 | |
| *** Qiming_ has joined #senlin | 01:22 | |
| *** Yanyanhu has quit IRC | 01:25 | |
| xuhaiwei | morning! | 01:35 |
| *** jdandrea has quit IRC | 01:42 | |
| *** Tennyson has joined #senlin | 01:42 | |
| Yanyan | xuhaiwei, morning :) | 01:45 |
| Yanyan | how is your vacation? | 01:45 |
| xuhaiwei | mainly stayed at home | 01:46 |
| xuhaiwei | not bad, anyway | 01:46 |
| Yanyan | nice, stay with family is always good ;) | 01:47 |
| xuhaiwei | :) | 01:47 |
| *** Tennyson has quit IRC | 02:03 | |
| xuhaiwei | hi, what is the exect version we should install, it seems 0.11 doesn't work | 02:43 |
| xuhaiwei | some package needs 'pbr<2.0,>=1.3' | 02:43 |
| xuhaiwei | some package still needs 'pbr!=0.7,<1.0,>=0.6' | 02:44 |
| Yanyan | xuhaiwei, you reinstall your environment? | 02:49 |
| Yanyan | pbr 0.11 works in my local env | 02:50 |
| Yanyan | hi, Qiming_ , how did you solve the pbr issue? | 02:50 |
| xuhaiwei | I reinstalled it before the vocation, but it works at that time | 02:50 |
| Yanyan | now you can't start engine/api? | 02:51 |
| xuhaiwei | I just installed some packages by pip install -r requirement.txt | 02:51 |
| xuhaiwei | yes, can't start engine and api | 02:51 |
| Yanyan | can you check your pbr version using pip list | grep -i pbr | 02:51 |
| Yanyan | I guess it was updated | 02:52 |
| xuhaiwei | yes, now 1.3 is installed by default | 02:52 |
| xuhaiwei | but some package still need 'pbr!=0.7,<1.0,>=0.6' | 02:52 |
| Yanyan | what I did is manually downgrade it by performing 'pip install pbr==0.11' | 02:53 |
| Yanyan | sudo pip install pbr==0.11 | 02:53 |
| xuhaiwei | if I reinstall pbr by pip install pbr==0.11, some package will complain 'pbr<2.0,>=1.3' | 02:53 |
| Yanyan | ah | 02:53 |
| xuhaiwei | there is no version to satisfy both of them | 02:53 |
| Yanyan | which packages | 02:53 |
| xuhaiwei | Requirement.parse('pbr!=0.7,<1.0,>=0.6'), set(['sqlalchemy-migrate'])) | 02:54 |
| Yanyan | maybe you can check which packages were installed when you executed 'pip install -r requirement.txt' | 02:55 |
| xuhaiwei | ok | 02:55 |
| Qiming_ | your sqlalchemy-migrate version? | 02:56 |
| xuhaiwei | sqlalchemy-migrate (0.9.6) | 02:56 |
| Qiming_ | sudo pip install sqlalchemy-migrate==0.9.7 | 02:56 |
| xuhaiwei | ok, works for me:) | 02:57 |
| Yanyan | maybe I should also upgrade my sqlalchemy-migrate, I found it is still 0.9.6 in my env | 02:58 |
| Yanyan | Qiming_, lunch time? | 03:47 |
| Qiming_ | y | 03:47 |
| openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for LoadBalancer driver https://review.openstack.org/203913 | 04:11 |
| openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for LoadBalancer driver(1) https://review.openstack.org/203913 | 04:11 |
| *** Tennyson has joined #senlin | 05:31 | |
| *** Zhenqi has joined #senlin | 05:54 | |
| openstackgerrit | xu-haiwei proposed stackforge/senlin: Add node module test case part1 and send some events for action https://review.openstack.org/199803 | 06:59 |
| openstackgerrit | xu-haiwei proposed stackforge/senlin: Add node module test case part1 and send some events for action https://review.openstack.org/199803 | 07:03 |
| openstackgerrit | xu-haiwei proposed stackforge/senlin: Add node module test case part1 and send some events for action https://review.openstack.org/199803 | 07:43 |
| openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for LoadBalancer driver(2) https://review.openstack.org/203956 | 07:49 |
| openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for policy base module and fix a bug https://review.openstack.org/197894 | 08:16 |
| openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for policy base module and fix a bug https://review.openstack.org/197894 | 08:17 |
| openstackgerrit | Yanyan Hu proposed stackforge/senlin: Add test case for policy base module and fix a bug https://review.openstack.org/197894 | 08:56 |
| Yanyan | hi, Qiming_, since the issue of plugin cannot be loaded correctly is still there, I tried to provide a temporary workaround in this patch https://review.openstack.org/#/c/197894/ | 08:56 |
| Yanyan | I just mocked the get_policy method of environment.Environment to avoid the policy type NotFound error which caused by plugin loading failure | 08:57 |
| Yanyan | I think xuhaiwei's patch of test case for node module also suffers this issue | 08:57 |
| Yanyan | https://review.openstack.org/#/c/199803/ | 08:58 |
| Yanyan | after plugin loading in gate can work correctly, we will remove this mock | 08:58 |
| xuhaiwei | yes, my patch faces the problem, I think, the test case ran correctly in my local env | 08:59 |
| Yanyan | em, this is caused by unstable plugin loading which sometimes could fail | 09:00 |
| Qiming_ | ok | 09:01 |
| Qiming_ | it could be something wrong with the environment initialization | 09:10 |
| Qiming_ | I'm working on a possible fix | 09:19 |
| Yanyan | Qiming_, yes, even when I manually invoke environment.initialze() before each test case, it could fail... | 09:19 |
| Yanyan | ok | 09:19 |
| Qiming_ | could be something related to test cases | 09:19 |
| Yanyan | you mean python's unittest module? | 09:20 |
| Yanyan | testtools | 09:21 |
| Qiming_ | possibly these lines: http://git.openstack.org/cgit/stackforge/senlin/tree/senlin/engine/environment.py#n188 | 09:21 |
| *** Tennyson has quit IRC | 09:21 | |
| Yanyan | mapping_func is always None in our use cases | 09:24 |
| Yanyan | you mean maybe we should provide it when doing test? | 09:24 |
| Qiming_ | ... | 09:24 |
| Qiming_ | no | 09:24 |
| Qiming_ | since it is related to a global env | 09:24 |
| Qiming_ | when some test cases set it to something and never set it back, other test cases may faile | 09:24 |
| Qiming_ | why would you provide a mapping func ... | 09:25 |
| Yanyan | just not sure why there is an input parameter defined for initialize, guess we can just provide a simple mapping_func for specific policy/profile? | 09:27 |
| Yanyan | so you mean the global _environment could be wrong? | 09:27 |
| *** Zhenqi has quit IRC | 09:28 | |
| Yanyan | since the judgement statement in line 183? | 09:32 |
| openstackgerrit | Qiming Teng proposed stackforge/senlin: Fix environment intialization logic https://review.openstack.org/203978 | 09:33 |
| Qiming_ | the "mapping_func" was a hack used to test inner functions | 09:33 |
| Qiming_ | the patch above is an attempt to fix that behavior, hopefully it will help | 09:34 |
| Yanyan | ok, let merge it first and haiwei can rebase his patch and submit it again | 09:35 |
| Yanyan | then we can check whether it works | 09:35 |
| Qiming_ | anyway, it does no harm I think | 09:36 |
| Yanyan | yes :) | 09:36 |
| Yanyan | hi, Qiming_, just noticed you remove this option when creating extention.ExtentionManager() 'verify_requirements=False' | 09:42 |
| Yanyan | I remeber this option was added to avoid gate failure caused by package version conflict? | 09:44 |
| Qiming_ | just checked other projects | 09:46 |
| Qiming_ | they have all removed it | 09:47 |
| Yanyan | em, seems we don't need it now | 09:47 |
| Yanyan | hi, Qiming_ , seem the verification failed for package downloading timeout | 09:51 |
| Yanyan | recheck it | 09:51 |
| Qiming_ | ok | 09:51 |
| Yanyan | will leave, see u guys tonight | 10:01 |
| xuhaiwei | see you | 10:01 |
| xuhaiwei | jenkins takes quite a long time now, since the test cases becomes more and more | 10:02 |
| *** Yanyan has quit IRC | 10:05 | |
| openstackgerrit | Merged stackforge/senlin: Fix environment intialization logic https://review.openstack.org/203978 | 10:14 |
| openstackgerrit | xu-haiwei proposed stackforge/senlin: Add node module test case part1 and send some events for action https://review.openstack.org/199803 | 10:18 |
| xuhaiwei | Hope it will work this time | 10:18 |
| xuhaiwei | I will leave too, see you Qiming_ | 10:18 |
| Qiming_ | okay | 10:18 |
| openstackgerrit | Qiming Teng proposed stackforge/senlin: Add a utility function for random name generation https://review.openstack.org/204014 | 10:44 |
| *** Qiming_ has quit IRC | 10:49 | |
| *** lixinhui_ has joined #senlin | 11:22 | |
| *** Qiming_ has joined #senlin | 11:41 | |
| lixinhui_ | which one is the install dir of openstacksdk under dist-package? | 11:44 |
| lixinhui_ | openstack/openstackauth/openstackclient | 11:44 |
| Qiming_ | openstack | 11:45 |
| lixinhui_ | is it safe for me to remove whole dir of openstack? | 11:45 |
| lixinhui_ | and install from source code | 11:46 |
| lixinhui_ | is there any other module installed to the same dir | 11:46 |
| *** Qiming_ has quit IRC | 12:08 | |
| *** yanyanhu has joined #senlin | 12:23 | |
| *** lkarm has joined #senlin | 12:31 | |
| lixinhui_ | has anyone use the latest code of openstacksdk? | 12:36 |
| yanyanhu | yes, I think I am | 12:36 |
| yanyanhu | maybe not the latest, I pulled the code about a week ago | 12:37 |
| lixinhui_ | okay, yanyan, have you met this error? | 12:37 |
| lixinhui_ | http://paste.openstack.org/show/395541/ | 12:37 |
| lixinhui_ | I have adopted latest code of senlin | 12:38 |
| yanyanhu | ok, let me have a look | 12:38 |
| *** haiwei has joined #senlin | 12:42 | |
| *** Qiming has joined #senlin | 12:42 | |
| Qiming | lixinhui_, no other code lives in that dir | 12:43 |
| Qiming | since it is just a library code | 12:43 |
| lixinhui_ | okay, qiming | 12:43 |
| Qiming | you can clone the source anywhere and do a install | 12:43 |
| yanyanhu | https://review.openstack.org/#/c/202949/ | 12:45 |
| yanyanhu | maybe need to update the configure file? | 12:46 |
| yanyanhu | or this should be service_project_name? | 12:46 |
| yanyanhu | http://git.openstack.org/cgit/stackforge/senlin/tree/senlin/drivers/openstack/sdk.py#n231 | 12:48 |
| Qiming | it should be service_project_name I think | 12:50 |
| Qiming | checking | 12:50 |
| yanyanhu | em, seems an error here | 12:50 |
| Qiming | something is broken? | 12:51 |
| yanyanhu | I think lixinhui_ meet an error when doing profile-type-list | 12:52 |
| lixinhui_ | http://paste.openstack.org/show/395541 | 12:52 |
| yanyanhu | maybe caused by this erroneous option name | 12:52 |
| Qiming | ah.. that is an error | 12:52 |
| *** jruano has joined #senlin | 12:54 | |
| Qiming | proposing a fix | 12:55 |
| yanyanhu | ok | 12:56 |
| *** mathspanda has joined #senlin | 12:56 | |
| openstackgerrit | Qiming Teng proposed stackforge/senlin: Fix sdk service credentials https://review.openstack.org/204072 | 12:56 |
| *** mathspanda has quit IRC | 12:56 | |
| yanyanhu | will need test case for this driver :) | 12:56 |
| Qiming | please add agenda items here: https://wiki.openstack.org/wiki/Meetings/SenlinAgenda | 12:56 |
| Qiming | yes | 12:57 |
| Qiming | or we will be changing code freely ... | 12:57 |
| yanyanhu | yes | 12:57 |
| *** jdandrea has joined #senlin | 13:11 | |
| openstackgerrit | Merged stackforge/senlin: Fix sdk service credentials https://review.openstack.org/204072 | 13:38 |
| Qiming | haiwei, there? | 14:00 |
| haiwei | yes | 14:00 |
| Qiming | about this one: https://bugs.launchpad.net/python-senlinclient/+bug/1472557 | 14:00 |
| openstack | Launchpad bug 1472557 in python-senlinclient "cluster, profile and node can't be updated" [Undecided,New] | 14:00 |
| yanyanhu | lixinhui_, since this patch has been merged https://review.openstack.org/204072, you can pull the lastest code and try it again | 14:00 |
| Qiming | maybe we need to improve the code here: /usr/local/lib/python2.7/dist-packages/senlinclient/common/exc.py", line 234, in parse_exception | 14:00 |
| Qiming | the exc.details sometimes is not a valid JSON | 14:01 |
| Qiming | sdk might have missed that part | 14:01 |
| lixinhui_ | yanyanhu, before the error, I have adopt the patch | 14:01 |
| haiwei | ok, I will check it out tomorrow Qiming | 14:01 |
| Qiming | haiwei, please help look into the exc.details and see what we can do there | 14:01 |
| Qiming | maybe it should be handled in SDK, maybe it's our job | 14:02 |
| Qiming | thanks | 14:02 |
| Qiming | lixinhui_, you need to update your senlin.conf file, by adding a section named 'authentication' | 14:02 |
| yanyanhu | lixinhui_, you mean the patch Qiming proposed about the option 'service_project_name' | 14:02 |
| haiwei | I had checked the source when I found this for the first time, but didn't resolve it | 14:02 |
| haiwei | will check it again | 14:02 |
| lixinhui_ | yes, yanyanhu | 14:02 |
| Qiming | thanks, haiwei | 14:03 |
| yanyanhu | ok, lixinhui_ maybe you need update configure file as Qiming suggested | 14:03 |
| lixinhui_ | Qiming, give me a example about uthentication | 14:03 |
| lixinhui_ | an example | 14:03 |
| lixinhui_ | of authetication | 14:03 |
| haiwei | I will quit guys, good night | 14:03 |
| Qiming | good night, haiwei | 14:04 |
| Qiming | too late for you | 14:04 |
| Qiming | lixinhui_, http://git.openstack.org/cgit/stackforge/senlin/tree/devstack/lib/senlin#n112 | 14:04 |
| Qiming | the patch was about stop stealing keystonemiddleware data in our code | 14:04 |
| Qiming | those options are supposed to be used by the auth_token middleware | 14:05 |
| *** yanyanhu has quit IRC | 14:05 | |
| Qiming | if we need something for even similar purposes, we should create and use our own options | 14:05 |
| lixinhui_ | okay | 14:07 |
| *** haiwei has quit IRC | 14:07 | |
| Qiming | jruano, https://coreos.com/docs/running-coreos/platforms/openstack/ | 14:08 |
| *** lixinhui_ has quit IRC | 14:19 | |
| *** lixinhui_ has joined #senlin | 14:20 | |
| jruano | thanks qiming. looks straightforward. i will investigate further | 14:34 |
| Qiming | ok | 14:35 |
| *** Qiming has quit IRC | 16:09 | |
| *** lixinhui_ has quit IRC | 16:11 | |
| *** Qiming has joined #senlin | 16:44 | |
| *** Qiming has quit IRC | 16:59 | |
| *** xuhaiwei has quit IRC | 17:18 | |
| *** lkarm has quit IRC | 21:19 | |
| *** jruano has quit IRC | 23:33 | |
| *** jruano has joined #senlin | 23:45 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!