| *** shu-mutou has joined #senlin | 00:08 | |
| *** Qiming has quit IRC | 00:22 | |
| openstackgerrit | Cindia-blue proposed openstack/senlin: Add timeout of waitting for loadbalancer creation https://review.openstack.org/282864 | 00:50 |
|---|---|---|
| *** Qiming has joined #senlin | 01:19 | |
| *** Yanyanhu has joined #senlin | 01:25 | |
| openstackgerrit | Qiming Teng proposed openstack/senlin-dashboard: Enable translation jobs https://review.openstack.org/282824 | 01:38 |
| *** elynn has joined #senlin | 02:02 | |
| *** elynn has joined #senlin | 02:04 | |
| *** Liuqing has joined #senlin | 02:06 | |
| elynn | morning | 02:06 |
| Qiming | morning | 02:07 |
| R_lixh | morning | 02:10 |
| R_lixh | Yanyanhu, there? | 02:10 |
| Yanyanhu | hi | 02:10 |
| R_lixh | Morning, Yanyanhu | 02:10 |
| Yanyanhu | morning :) | 02:10 |
| R_lixh | in the past weekend, I tried LB policy | 02:10 |
| R_lixh | and I am suffering from the problem with member_add | 02:11 |
| R_lixh | http://paste.openstack.org/show/487686/ | 02:11 |
| R_lixh | this is the error | 02:11 |
| R_lixh | wonder if you ever met this before | 02:11 |
| Yanyanhu | ok, let me check it | 02:11 |
| Yanyanhu | I doubted the part_arg issue is still there in sdk | 02:12 |
| Yanyanhu | let me check which branch of sdk I'm using | 02:12 |
| R_lixh | okay :) | 02:13 |
| Yanyanhu | yes, I'm using a branch patched by my self | 02:13 |
| R_lixh | a bug there? | 02:13 |
| Yanyanhu | yes, there is a bug there | 02:13 |
| R_lixh | Please point me to the link if have | 02:14 |
| Yanyanhu | but not sure whether they have fixed this bug in latest patch | 02:14 |
| Yanyanhu | about property rework | 02:14 |
| R_lixh | oh | 02:14 |
| openstackgerrit | Merged openstack/senlin-dashboard: Add receiver creation form https://review.openstack.org/282684 | 02:14 |
| Yanyanhu | ok, let me find my patch for you. I have proposed a fix for this problem, but it's not a very solution. So that patch is hold there. | 02:14 |
| R_lixh | Thanks | 02:15 |
| Yanyanhu | https://review.openstack.org/187934 | 02:15 |
| Yanyanhu | this one | 02:15 |
| Yanyanhu | I recalled terry proposed a more complete solution for this problem but he abandoned that patch a week ago | 02:16 |
| R_lixh | okay | 02:17 |
| R_lixh | let me use your patch firstly to try the end2end functions | 02:17 |
| Yanyanhu | https://review.openstack.org/219840 | 02:17 |
| Yanyanhu | this one from Terry | 02:17 |
| Yanyanhu | R_lixh, ok | 02:17 |
| R_lixh | Got them, Thanks :) | 02:18 |
| Yanyanhu | no problem :) | 02:19 |
| Yanyanhu | honestly, I think they should at least make it work even using an simplest and maybe ugly way. Then try to find a better and complete solution for it. | 02:20 |
| Yanyanhu | not just leave it there like this :) | 02:20 |
| openstackgerrit | Merged openstack/python-senlinclient: Ensure cluster is specified when creating receivers https://review.openstack.org/282814 | 02:22 |
| openstackgerrit | Liuqing Jing proposed openstack/senlin-dashboard: Add receiver deletion table/row action https://review.openstack.org/282685 | 02:22 |
| R_lixh | yes, I agree to you, Yanyanhu | 02:23 |
| Yanyanhu | R_lixh, hope this problem can be resolved soon :) | 02:24 |
| R_lixh | :) | 02:25 |
| openstackgerrit | Liuqing Jing proposed openstack/senlin: Add devstack support for senlin dashboard https://review.openstack.org/282138 | 02:26 |
| Qiming | Yanyanhu, maybe we should rebase the patch and argue with them again | 02:27 |
| Yanyanhu | Qiming, I think we can. But I'm worried that they won't let it in since it's not a complete way to address this issue. | 02:28 |
| Yanyanhu | just as they said before | 02:28 |
| Yanyanhu | let me check the latest change of sdk to see whether they have provide better solution for this | 02:29 |
| Qiming | even if this is not the perfect solution, we still remind them: 1) this is important, 2) it is broken; 3) please work on it, or adopt this temp patch | 02:29 |
| Yanyanhu | yes, agree | 02:29 |
| Yanyanhu | I noticed terry abandoned all related patches last week. Not sure why he did that. | 02:30 |
| Yanyanhu | those 5 patches about path_arg | 02:30 |
| Yanyanhu | oh, 4 patches | 02:30 |
| openstackgerrit | Merged openstack/senlin-dashboard: Add receiver deletion table/row action https://review.openstack.org/282685 | 02:31 |
| Qiming | it could be related to the latest changes to use *_id or *_ids | 02:31 |
| Yanyanhu | that's possible. Since those 4 patches also made really huge change | 02:31 |
| openstackgerrit | Shu Muto proposed openstack/senlin-dashboard: Update URLs to Django 1.8 style https://review.openstack.org/274049 | 02:32 |
| Yanyanhu | R_lixh, which version sdk are you using now? | 02:32 |
| R_lixh | 0.7.4 | 02:32 |
| Yanyanhu | so this is the latest version before that property naming rework is in? | 02:33 |
| R_lixh | yes, it is, Yanyanhu | 02:33 |
| Qiming | yes, 0.7.4 is the latest 'working' version | 02:34 |
| R_lixh | :) | 02:34 |
| Yanyanhu | ok, let me try to make a rebase first and repropose it | 02:34 |
| R_lixh | Great! | 02:35 |
| Yanyanhu | Qiming, actually I have a question about how to work on this kind of issue :) | 02:35 |
| Yanyanhu | e.g. I know there is a problem which breaks the system. And a complete solution need huge rework. | 02:36 |
| Yanyanhu | but maybe currently, I don't have enough bandwidth to work on this huge change and no other people as well. | 02:37 |
| Yanyanhu | so a simple workaround should be acceptable for this situation, right? We shouldn't let a known bug just stay there and break the system. | 02:38 |
| Qiming | yes | 02:38 |
| Yanyanhu | ok, understand. | 02:39 |
| Qiming | Liuqing, could you help verify if https://review.openstack.org/#/c/282824 works? | 02:39 |
| Qiming | I'm not sure if the change to _50_senlin.py.example would work | 02:40 |
| Liuqing | OK | 02:40 |
| Qiming | I'm not familiar with that | 02:40 |
| Qiming | The only thing I want to solve at the moment is to get translation set up | 02:40 |
| Qiming | for senlin, senlinclient and senlin-dashboard | 02:41 |
| Qiming | Liuqing, I'm not sure if we are missing something in the 'run_tests.sh' file related to translations | 02:44 |
| Qiming | As I was checking the 'run_tests.sh' file in horizon, there are commands like these: | 02:45 |
| Qiming | echo " --makemessages Create/Update English translation files." | 02:45 |
| Qiming | echo " --compilemessages Compile all translation files." | 02:45 |
| Qiming | echo " --check-only Do not update translation files (--makemessages only)." | 02:45 |
| Qiming | echo " --pseudo Pseudo translate a language." | 02:45 |
| Qiming | maybe that is the way for horizon (plugins) to do translations? they are not doing it through the setup.cfg file? | 02:46 |
| Liuqing | Hi Qiming, i found a similar patch in murano-dashboard for translations | 03:00 |
| Qiming | okay | 03:01 |
| openstackgerrit | Qiming Teng proposed openstack/senlin-dashboard: Enable translation jobs https://review.openstack.org/282824 | 03:02 |
| Qiming | Liuqing, we need to get this up quickly so that we can propose a patch to project-config where we claim we support translations | 03:03 |
| Qiming | Liuqing, feel free to grab this patch and fix it | 03:03 |
| Liuqing | OK will fix the bug now | 03:04 |
| Yanyanhu | hi, Qiming, R_lixh, a rebased patch has been proposed https://review.openstack.org/187934 | 03:04 |
| Yanyanhu | and I added a note at the end of commit msg | 03:05 |
| Qiming | okay | 03:07 |
| Qiming | just wondering why we have to parse the base_path to figure out path_args? | 03:07 |
| Qiming | couldn't we pass in the 'path_args' from the 'create_by_id' function? | 03:08 |
| Qiming | that parsing is very very ugly, :( | 03:08 |
| Yanyanhu | yes, we can, but that will change the existing interfaces | 03:08 |
| Yanyanhu | I hoped this workaround won't influence the definition and usage of existing methods, so choosed this way | 03:09 |
| Yanyanhu | it just worked at the last step where req body is constructed | 03:09 |
| Qiming | I mean we make the function something like 'def _remove_path_args_from_attrs(cls, path_args, attrs)' | 03:10 |
| Qiming | what's the interface that will be impacted? | 03:10 |
| Yanyanhu | oh, you mean this way | 03:10 |
| Qiming | [11:07] <Qiming> just wondering why we have to parse the base_path to figure out path_args? | 03:11 |
| Yanyanhu | this is better | 03:11 |
| Qiming | [11:08] <Qiming> couldn't we pass in the 'path_args' from the 'create_by_id' function? | 03:11 |
| Yanyanhu | I misunderstood your question :0 | 03:11 |
| Qiming | ok | 03:11 |
| *** yuanying has quit IRC | 03:20 | |
| Yanyanhu | hi, Qiming | 03:25 |
| Yanyanhu | I found that way won't work... | 03:25 |
| Yanyanhu | since the path_args passed to create_by_id is actually the resource itself | 03:26 |
| Yanyanhu | http://git.openstack.org/cgit/openstack/python-openstacksdk/tree/openstack/resource.py#n582 | 03:26 |
| openstackgerrit | Liuqing Jing proposed openstack/senlin-dashboard: Enable translation jobs https://review.openstack.org/282824 | 03:26 |
| Yanyanhu | so all keys will be included there and we can tell which ones of them are real 'path argument' | 03:26 |
| Yanyanhu | s/can/can't | 03:26 |
| Qiming | sigh | 03:27 |
| Yanyanhu | yes, another historical problem | 03:28 |
| Qiming | I really dislike the interface where sdk allows passing an object into an interface function | 03:29 |
| Yanyanhu | yes, although flexible, but really makes thing complicated | 03:30 |
| R_lixh | Yanyanhu, I saw current LB policy has implemented great automation work and wonder any plan to add healthmonitr and security group too? | 03:35 |
| Yanyanhu | R_lixh, you mean the healthmonitor of lb and neutron security group? | 03:36 |
| R_lixh | yes | 03:36 |
| R_lixh | without healthmonitor, tyhe memebr status will not change | 03:37 |
| Yanyanhu | R_lixh, yes, that's true | 03:37 |
| R_lixh | yes | 03:37 |
| Yanyanhu | actually I haven't spent time on healthmonitor before :) | 03:38 |
| R_lixh | And without security group, ssl to the member will be controlled | 03:38 |
| Yanyanhu | let me check the code, I think it won't very difficult to add it | 03:38 |
| R_lixh | Okay, Yanyanhu :) | 03:39 |
| Yanyanhu | security group should be part of nova profile I think | 03:39 |
| R_lixh | oh, also reasonable | 03:39 |
| Yanyanhu | since usually people will specify the security group if needed when creating a nova server | 03:40 |
| Yanyanhu | and the creation of security group itself is not in Senlin's scope :) | 03:40 |
| R_lixh | Maybe healthmonitor is more usefull for senlin lb+as scenario | 03:40 |
| Yanyanhu | R_lixh, yes, agree | 03:40 |
| R_lixh | Yes, I see. Yanyanhu | 03:40 |
| Yanyanhu | hi, R_lixh, healthmonitor related has been added to neutron driver. I think we just need to make a little extension of lb_policy to support healthmonitor. | 03:44 |
| openstackgerrit | Merged openstack/senlin: Update unit test of profile for filters https://review.openstack.org/282674 | 03:44 |
| openstackgerrit | Merged openstack/python-senlinclient: Add OpenstackClient plugin for cluster node show https://review.openstack.org/282599 | 03:44 |
| Yanyanhu | but I think we need to test it to ensure it works as expected | 03:44 |
| R_lixh | Okay, Yanyanhu | 03:45 |
| Yanyanhu | I don't have a complete environment for this kind of test in my local development env | 03:45 |
| Yanyanhu | so if you guys have this kind of environment, that will be great :) | 03:46 |
| R_lixh | Do not worry about testing part | 03:46 |
| R_lixh | Yanyanhu | 03:46 |
| R_lixh | We can do that part :) | 03:46 |
| Yanyanhu | just need a host that can create nova server that you can run some simple http workload in it | 03:46 |
| R_lixh | ok | 03:47 |
| R_lixh | We have tested through lbaas v1 functions but not v2 + senlin | 03:48 |
| openstackgerrit | Merged openstack/python-senlinclient: Add OpenstackClient plugin for cluster node create https://review.openstack.org/282608 | 03:49 |
| Yanyanhu | R_lixh, I also have tested lbaas v1 before | 03:50 |
| Yanyanhu | about a year ago | 03:50 |
| Yanyanhu | but for v2, I just made simplest test to ensure some basic functions work well | 03:50 |
| Yanyanhu | like whether loadbalancer/listener/pool can be created successfully | 03:51 |
| R_lixh | Thanks for your sharing by email last time on ceilometer part :) | 03:51 |
| R_lixh | I see. | 03:51 |
| Yanyanhu | whether the given nova server ip can be added to specific pool | 03:51 |
| Yanyanhu | R_lixh, you're welcome :) | 03:51 |
| openstackgerrit | Merged openstack/python-senlinclient: Add OpenstackClient plugin for cluster node update https://review.openstack.org/282641 | 03:51 |
| R_lixh | So far to the lisneter, I testes works well but | 03:51 |
| R_lixh | ceilometer v2 montiroing seems still a problem | 03:52 |
| R_lixh | some one submit a patch to support v2 meters | 03:52 |
| Yanyanhu | oh, I guess we may need to do extra extension on ceilometer to support it | 03:52 |
| Yanyanhu | I doubt there is no builtin support yet | 03:52 |
| R_lixh | I am planning to use taht part but still need senlin to support healthmonitor. | 03:52 |
| R_lixh | ... | 03:53 |
| Yanyanhu | R_lixh, I'm not sure about that, but I guess so :) | 03:53 |
| Yanyanhu | I will try to work on healthmonitor stuff in coming week if I have time and if you guys want to do some tests on it, please just test me | 03:53 |
| openstackgerrit | Merged openstack/python-senlinclient: Add OpenstackClient plugin for cluster node delete https://review.openstack.org/282644 | 03:53 |
| R_lixh | if no buildin support, there will be more works... | 03:53 |
| R_lixh | Okay, Yanyanhu | 03:54 |
| Yanyanhu | R_lixh, that's true | 03:54 |
| openstackgerrit | Merged openstack/senlin-dashboard: Enable translation jobs https://review.openstack.org/282824 | 03:55 |
| *** elynn has quit IRC | 03:59 | |
| *** Liuqing has quit IRC | 04:03 | |
| *** yuanying has joined #senlin | 04:10 | |
| *** elynn has joined #senlin | 04:25 | |
| *** elynn_ has joined #senlin | 04:29 | |
| *** elynn has quit IRC | 04:30 | |
| *** Liuqing has joined #senlin | 05:44 | |
| openstackgerrit | Liuqing Jing proposed openstack/senlin-dashboard: Add receiver detail page https://review.openstack.org/282954 | 05:51 |
| *** Liuqing has quit IRC | 05:59 | |
| openstackgerrit | Ayush Garg proposed openstack/senlin: Add check for bad parameter in profile-list https://review.openstack.org/282882 | 06:08 |
| Qiming | Yanyanhu, how many services/resources are suffering from the path_args problem? | 06:15 |
| Yanyanhu | as of now, only lb member create | 06:16 |
| Yanyanhu | lb pool_member resource | 06:16 |
| Qiming | okay, can we solve that issue this way ... | 06:16 |
| Yanyanhu | nova seems can tolerate this kind of unused args | 06:16 |
| Qiming | instead of changing all resources behavior | 06:16 |
| Qiming | we override the _get_create_body() function in pool_member | 06:17 |
| *** Ayush_ has joined #senlin | 06:17 | |
| Qiming | in the overriding function, we remove the path args from attrs before return the attrs | 06:17 |
| Qiming | sounds a better alternative? | 06:18 |
| Yanyanhu | yes, this can significant reduce the influence | 06:18 |
| Qiming | hopefully, we can somehow reduce the global impact and avoid the ugly string parsing logic | 06:19 |
| Yanyanhu | just a little worried that _get_creat_body() looks like a tool function of resource base class | 06:20 |
| Yanyanhu | so is it ok to override it? | 06:20 |
| Qiming | yes | 06:20 |
| Yanyanhu | ok, I see | 06:20 |
| Qiming | I inroduced that function when I was fixing nova server | 06:20 |
| Yanyanhu | will try to fix using this way | 06:20 |
| Qiming | so far only nova server resource is overriding in order to pass the scheduler hints | 06:21 |
| Qiming | in the same way, you can manipulate the body | 06:21 |
| Yanyanhu | nice, if it is designed for this purpose, it's much better to use this way to address path_arg problem | 06:22 |
| Qiming | yes | 06:22 |
| openstackgerrit | Merged openstack/senlin: Fix an error in openstack_test driver https://review.openstack.org/282238 | 06:29 |
| openstackgerrit | Merged openstack/senlin: Add functional test for cluster check recover https://review.openstack.org/282268 | 06:29 |
| openstackgerrit | Merged openstack/senlin: Add check for bad parameter in profile-list https://review.openstack.org/282882 | 06:35 |
| *** Liuqing has joined #senlin | 06:38 | |
| *** Liuqing has quit IRC | 06:45 | |
| *** zzxwill has joined #senlin | 07:45 | |
| *** dixiaoli__ has quit IRC | 07:50 | |
| openstackgerrit | Qiming Teng proposed openstack/senlin-dashboard: Fix translation setup https://review.openstack.org/282994 | 08:01 |
| *** dixiaoli__ has joined #senlin | 08:06 | |
| *** elynn__ has joined #senlin | 08:50 | |
| *** elynn_ has quit IRC | 08:52 | |
| *** elynn__ has quit IRC | 08:55 | |
| *** elynn__ has joined #senlin | 08:55 | |
| *** zzxwill has quit IRC | 09:05 | |
| openstackgerrit | Ethan Lynn proposed openstack/senlin: Convert count to int in cluster scale_out/in action https://review.openstack.org/283016 | 09:20 |
| *** elynn__ has quit IRC | 09:24 | |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Allow user to define region_name when setting up senlin service https://review.openstack.org/283020 | 09:54 |
| *** Yanyanhu has quit IRC | 10:10 | |
| *** Qiming has quit IRC | 10:11 | |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: Add OpenstackClient plugin for cluster policy list https://review.openstack.org/282816 | 10:13 |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: Add OpenstackClient plugin for cluster policy show https://review.openstack.org/282817 | 10:15 |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: Add OpenstackClient plugin for cluster policy create https://review.openstack.org/282818 | 10:16 |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: Add OpenstackClient plugin for cluster policy update https://review.openstack.org/282819 | 10:17 |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: Add OpenstackClient plugin for cluster policy delete https://review.openstack.org/282820 | 10:18 |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: OSC plugin for openstack cluster policy type list https://review.openstack.org/282865 | 10:19 |
| openstackgerrit | Di XiaoLi proposed openstack/python-senlinclient: Add OSC plugin for cluster policy type show https://review.openstack.org/282866 | 10:19 |
| *** Qiming has joined #senlin | 11:26 | |
| *** zhenguo_ has quit IRC | 11:40 | |
| *** zigo has quit IRC | 11:40 | |
| *** zigo has joined #senlin | 11:41 | |
| *** zhenguo_ has joined #senlin | 11:43 | |
| openstackgerrit | Merged openstack/senlin-dashboard: Add receiver detail page https://review.openstack.org/282954 | 14:16 |
| *** openstackgerrit has quit IRC | 14:17 | |
| *** openstackgerrit has joined #senlin | 14:18 | |
| *** sridhar_ram has joined #senlin | 14:53 | |
| *** Qiming has quit IRC | 15:51 | |
| *** sridhar_ram has quit IRC | 16:05 | |
| *** sridhar_ram1 has joined #senlin | 16:57 | |
| openstackgerrit | Ayush Garg proposed openstack/python-senlinclient: Add filters option to policy-list command https://review.openstack.org/283196 | 18:09 |
| openstackgerrit | Ayush Garg proposed openstack/senlin: Add check for bad parameter in policy-list https://review.openstack.org/283204 | 18:26 |
| *** sridhar_ram1 has quit IRC | 18:33 | |
| *** Ayush_ has quit IRC | 18:45 | |
| *** sridhar_ram has joined #senlin | 19:22 | |
| *** openstackgerrit has quit IRC | 20:47 | |
| *** openstackgerrit has joined #senlin | 20:48 | |
| *** Qiming has joined #senlin | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!