| *** catintheroof has joined #senlin | 00:20 | |
| *** catintheroof has quit IRC | 00:26 | |
| *** elynn has joined #senlin | 01:18 | |
| *** elynn_ has joined #senlin | 01:23 | |
| *** elynn has quit IRC | 01:25 | |
| *** guoshan has joined #senlin | 02:12 | |
| *** XueFeng has joined #senlin | 03:10 | |
| yuanbin | elynn_, about senlin security_groups format in cirros_basic.yaml, I change how to write this file ? | 03:23 |
|---|---|---|
| yuanbin | elynn_, I use this file http://paste.openstack.org/show/603357/ | 03:24 |
| *** zhurong has joined #senlin | 03:25 | |
| yuanbin | elynn_, But the security_groups will be list, if security_groups:3acfdc18-22e0-4ff6-a153-4db846f5da18, the code will be change to ['3','a','c','f',...] format | 03:26 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin master: failed to dump 'WARNING' notification https://review.openstack.org/447353 | 03:37 |
| elynn_ | Hi yuanbin | 03:38 |
| elynn_ | security_groups is a list | 03:38 |
| elynn_ | Use security_groups: ["xxx"] | 03:38 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin master: failed to dump 'WARNING' notification https://review.openstack.org/447353 | 03:39 |
| elynn_ | For profile type os.nova.server , networks didn't support security_groups yet. | 03:41 |
| elynn_ | yuanbin, ^ | 03:41 |
| Qiming | elynn_, I think he was asking a different question | 03:55 |
| Qiming | his question is mainly about how to describe a list in YAML | 03:55 |
| elynn_ | Qiming, I see... I was thinking is it okay to introduce a strlist schema, which accept a string or a list, but return a list. | 03:57 |
| elynn_ | I was thinking to use it at 'fixed_ip', and let it accept a list or a string without breaking compatibility . | 03:59 |
| *** guoshan has quit IRC | 04:00 | |
| *** XueFeng has quit IRC | 04:07 | |
| openstackgerrit | Merged openstack/senlin master: Added CORS support to Senlin https://review.openstack.org/444697 | 04:08 |
| *** zhurong has quit IRC | 04:16 | |
| *** elynn_ has quit IRC | 04:17 | |
| openstackgerrit | liyi proposed openstack/senlin master: Add access log for devstack. https://review.openstack.org/447361 | 04:50 |
| *** zhurong has joined #senlin | 04:59 | |
| *** elynn has joined #senlin | 05:03 | |
| *** elynn_ has joined #senlin | 05:07 | |
| *** elynn has quit IRC | 05:08 | |
| *** zhurong has quit IRC | 05:14 | |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin master: failed to dump 'WARNING' notification https://review.openstack.org/447353 | 05:23 |
| *** ruijie__ has quit IRC | 05:27 | |
| *** XueFeng has joined #senlin | 05:38 | |
| *** ruijie has joined #senlin | 05:53 | |
| ruijie | hi, got a senlin client problem, not sure if this is a bug | 05:56 |
| ruijie | http://paste.openstack.org/show/603371/ | 05:56 |
| ruijie | http://paste.openstack.org/show/603372/ | 05:59 |
| ruijie | plz see the 2nd one... sorry :( | 05:59 |
| elynn_ | Create cluster failed? | 06:10 |
| ruijie | yes elynn_ | 06:11 |
| ruijie | looks like client failed to parse the err msg | 06:11 |
| elynn_ | FINAL-TEST is cluster name? | 06:12 |
| ruijie | its the profile name | 06:12 |
| elynn_ | Seems like sdk failed to parse the error. | 06:13 |
| *** ruijie_ has joined #senlin | 06:13 | |
| Qiming | looks like a SDK version | 06:13 |
| Qiming | problem | 06:13 |
| elynn_ | messages = [obj.get('message') for obj in resp.json().values()] | 06:14 |
| Qiming | http://git.openstack.org/cgit/openstack/python-openstacksdk/tree/openstack/exceptions.py | 06:14 |
| elynn_ | Obj here seems like a unicode | 06:14 |
| Qiming | check line 129 | 06:14 |
| Qiming | it was added recently | 06:14 |
| Qiming | this logic was broken by a developer when adding some neutron specific error parsing into sdk | 06:14 |
| *** ruijie has quit IRC | 06:16 | |
| elynn_ | It's in 0.9.14, so I guess latest version won't break it. | 06:16 |
| elynn_ | BTW, I failed to enable senlin-dashboard following the guide of senlin-dashboard. https://bugs.launchpad.net/senlin/+bug/1674216 | 06:17 |
| openstack | Launchpad bug 1674216 in senlin "senlin dashboard can not enable" [Undecided,New] | 06:17 |
| ruijie_ | I am using 0.9.13, let me update it and check it again | 06:17 |
| XueFeng | dashboard has bug recently | 06:23 |
| ruijie_ | 0.9.14 is fine :) | 06:23 |
| elynn_ | XueFeng, is this bug that lead to the failure I met? | 06:24 |
| *** guoshan_ has joined #senlin | 06:25 | |
| *** guoshan has joined #senlin | 06:29 | |
| *** guoshan_ has quit IRC | 06:29 | |
| XueFeng | We discuss two problem. One is ruijie report, this is cause by sdk. | 06:37 |
| XueFeng | And it maybe only the problem when exception handling. | 06:38 |
| XueFeng | About dashboard, which I met is a install problem. | 06:40 |
| XueFeng | When I enable senlin-dashboard, I can't install devstak sucessfully | 06:40 |
| *** 7IZAATF07 has joined #senlin | 06:42 | |
| *** 7GHAALN3P has joined #senlin | 06:42 | |
| *** 7GHAALN3P has quit IRC | 06:42 | |
| *** guoshan has quit IRC | 06:45 | |
| *** guoshan has joined #senlin | 06:45 | |
| elynn_ | XueFeng, How do you enable senlin-dashboard? | 07:13 |
| elynn_ | enable_plugin senlin-dashboard https://xxx/senlin-dashboard ? | 07:13 |
| openstackgerrit | liyi proposed openstack/senlin master: Add access log for devstack. https://review.openstack.org/447361 | 07:26 |
| openstackgerrit | liyi proposed openstack/senlin master: Add access log for devstack. https://review.openstack.org/447361 | 07:28 |
| elynn_ | Hi Qiming , around? | 07:32 |
| *** Jeffrey4l__ has quit IRC | 07:33 | |
| *** Jeffrey4l__ has joined #senlin | 07:36 | |
| Qiming | yes | 07:44 |
| *** elynn__ has joined #senlin | 08:02 | |
| elynn__ | Qiming, I'm about to back port vdu , should I create a new server profile and set type as os.nova.server-2.0? | 08:03 |
| *** elynn_ has quit IRC | 08:04 | |
| elynn__ | Or just override the current server profile? | 08:04 |
| Qiming | em, I'd suggest we just add properties to the existing one | 08:29 |
| Qiming | but you will bump the profile version in place | 08:30 |
| elynn__ | Okay, but if user use os.nova.server-1.0, will we break it? | 08:30 |
| Qiming | you may want to check this: http://git.openstack.org/cgit/openstack/senlin/tree/senlin/common/schema.py#n63 | 08:31 |
| Qiming | for schemas, we have min_version and max_version support | 08:31 |
| Qiming | although not complicated as versioned objects, we do have versioning support | 08:32 |
| *** XueFeng has quit IRC | 08:32 | |
| Qiming | A spec has a version, you will compare the schema version against it to determine if a property is acceptable | 08:33 |
| Qiming | e.g. nova server 1.0 will not accept a new property if you have labeled property with min_version 1.1 | 08:33 |
| elynn__ | Okay, I see how it work now... | 08:35 |
| ruijie_ | Hi Qiming around? | 09:01 |
| *** openstackgerrit has quit IRC | 09:02 | |
| Qiming | yes | 09:03 |
| ruijie_ | https://github.com/rackerlabs/otter/blob/master/otter/convergence/planning.py#L432-L521 | 09:03 |
| ruijie_ | is this process too complex for us ? | 09:04 |
| ruijie_ | as far as I concerned, line 432-466 is for creation process | 09:05 |
| ruijie_ | and the rest is for scaling, resizing and recover | 09:05 |
| Qiming | I know this project, but I have never read its code | 09:06 |
| Qiming | I'm not quite sure how otter manages that tasks it is building | 09:06 |
| Qiming | similar to the "action" concept in senlin? | 09:06 |
| ruijie_ | sorry ... I saw it here ..https://etherpad.openstack.org/p/newton-senlin-ha | 09:07 |
| ruijie_ | I am trying to let our resizing and recovering process support desired_capacity and health check | 09:07 |
| Qiming | em ... | 09:09 |
| Qiming | I'm a little bit conservative about convergence | 09:09 |
| Qiming | if I'm understanding it correctly (at least in Heat), convergence is a word spaning three different jobs | 09:10 |
| ruijie_ | em ... if we do these processes during scaling and recovering ... I am actually worrying about the performence.. | 09:10 |
| Qiming | 1. make things parallel, this may be irrelevant in our current dicussion context | 09:10 |
| *** openstackgerrit has joined #senlin | 09:10 | |
| *** ChanServ sets mode: +v openstackgerrit | 09:10 | |
| openstackgerrit | Merged openstack/senlin master: Add access log for devstack. https://review.openstack.org/447361 | 09:10 |
| Qiming | 2. make sure the engine is can be made aware of the actual status of physical resources WHEN REQUIRED | 09:11 |
| openstackgerrit | Merged openstack/senlin master: failed to dump 'WARNING' notification https://review.openstack.org/447353 | 09:11 |
| openstackgerrit | Merged openstack/senlin master: Add get_stack_template support for heat https://review.openstack.org/446866 | 09:11 |
| openstackgerrit | Merged openstack/senlin master: Add get_env and get_files function to heat driver https://review.openstack.org/447001 | 09:11 |
| Qiming | this should not be a problem for nodes and clusters today, although I'm not 100% percent confident about its stability | 09:11 |
| Qiming | 3. derive a plan to converge the current states of resources to the "desired" states ... this is my top concern | 09:12 |
| Qiming | actually, the health policy is trying to introduce such a loop in to Senlin engine | 09:12 |
| Qiming | detection -> (notification ->) recovery | 09:13 |
| ruijie_ | yes Qiming, they do status check first if needed, and then based on the status of nodes to get the nodes needed to be created, updated or deleted | 09:13 |
| Qiming | my concern is mainly related to the reliability of detection and recovery | 09:13 |
| Qiming | if an environment is not reliable, we will get false alarms, mistakenly treating some healthy resources as unhealthy | 09:14 |
| ruijie_ | em, can we just do size check in resizing process and let recover action do status check | 09:14 |
| Qiming | this can be caused by network delay, missing notification, busy service ... for example | 09:14 |
| Qiming | em, you can never get a perfect solution in a very dynamice environment | 09:15 |
| Qiming | say you have just completed a status check, just before you started to recover some bad nodes, another 3 nodes are going offline | 09:16 |
| Qiming | should we handle them at all? | 09:16 |
| ruijie_ | yes Qiming, so what we can do is finding relatively good ones | 09:16 |
| Qiming | I'd prefer making them configurable | 09:16 |
| Qiming | or ideally, orthogonal | 09:16 |
| ruijie_ | Qiming you mean for both resizing and recovery ? | 09:17 |
| Qiming | let health be health, let scale be scale | 09:17 |
| Qiming | if you don't want to apply health policy, and you still want to do a status check before doing scaling, fine | 09:17 |
| Qiming | I have no objections on that | 09:17 |
| Qiming | but I'd warn you that detection node failure is very tricky | 09:18 |
| Qiming | ICMP is not trustworthy, heartbeat can be lost, Nova API may not be so responsive, ... | 09:18 |
| Qiming | a more reliable way to detect node failure is to inquire the Load Balancer | 09:19 |
| Qiming | but, as you may know, Ocativia is constantly rejecting our patches ... | 09:20 |
| Qiming | even they have some good points, we failed to follow up ... | 09:20 |
| ruijie_ | sign .. | 09:20 |
| Qiming | so ... in this community | 09:20 |
| Qiming | just as someone mentioned on mailing list | 09:20 |
| Qiming | it is do-mocracy, rather than democracy | 09:21 |
| Qiming | need some features, some business requirements, very good usage scenarios? | 09:21 |
| Qiming | good, jump on to it and get it implemented | 09:21 |
| Qiming | no one will do that for you , :) | 09:22 |
| Qiming | we really hope ocavia can be stable and usable soon | 09:23 |
| ruijie_ | that is .. | 09:23 |
| Qiming | but ... there is no way to drive it, unless we jump onto it ourselves | 09:23 |
| ruijie_ | anyway, we can focus on our project, if possible, maybe we can implement it ourselves .. | 09:24 |
| Qiming | see this: https://review.openstack.org/#/c/402296/ | 09:25 |
| ruijie_ | that is also common even in one company :( | 09:25 |
| Qiming | no one answers your question, :) | 09:25 |
| ruijie_ | that is unreasonable ... :) | 09:28 |
| ruijie_ | And for this support of health check and desired capacity check, I want to file a spec for it | 09:29 |
| ruijie_ | So that we can also discussed it offline | 09:30 |
| *** guoshan has quit IRC | 10:27 | |
| *** 7IZAATF07 has quit IRC | 11:07 | |
| *** yanyanhu_ has joined #senlin | 11:09 | |
| *** yanyanhu_ has quit IRC | 11:09 | |
| *** zhurong has joined #senlin | 11:17 | |
| *** elynn__ has quit IRC | 11:24 | |
| *** elynn__ has joined #senlin | 11:28 | |
| *** elynn_ has joined #senlin | 11:32 | |
| *** elynn__ has quit IRC | 11:32 | |
| *** Jeffrey4l_ has joined #senlin | 11:34 | |
| *** Jeffrey4l__ has quit IRC | 11:36 | |
| *** zhurong has quit IRC | 11:37 | |
| *** elynn_ has quit IRC | 12:39 | |
| *** catintheroof has joined #senlin | 12:57 | |
| openstackgerrit | JeffreyGuan proposed openstack/senlin stable/mitaka: update the typo in setup.cfg from "senlin.common.wsgi = senlin.common.wsgi:list_opts" to "senlin.api.common.wsgi = senlin.api.common.wsgi:list_opts" https://review.openstack.org/447486 | 13:18 |
| *** guoshan has joined #senlin | 13:39 | |
| *** guoshan has quit IRC | 13:58 | |
| *** guoshan has joined #senlin | 14:06 | |
| openstackgerrit | JeffreyGuan proposed openstack/python-senlinclient stable/mitaka: Fix the type error for "result" in senlinclient. https://review.openstack.org/447517 | 14:21 |
| openstackgerrit | JeffreyGuan proposed openstack/senlin stable/mitaka: update the typo in setup.cfg from "senlin.common.wsgi = senlin.common.wsgi:list_opts" to "senlin.api.common.wsgi = senlin.api.common.wsgi:list_opts" https://review.openstack.org/447486 | 14:33 |
| *** guoshan has quit IRC | 15:24 | |
| *** catintheroof has quit IRC | 18:08 | |
| *** catintheroof has joined #senlin | 18:08 | |
| *** zhenguo_ has joined #senlin | 20:18 | |
| *** chohoor_ has joined #senlin | 20:24 | |
| *** chohoor has quit IRC | 20:29 | |
| *** zhenguo has quit IRC | 20:29 | |
| *** zhenguo_ is now known as zhenguo | 20:30 | |
| *** Jeffrey4l_ has quit IRC | 21:35 | |
| *** Jeffrey4l_ has joined #senlin | 21:38 | |
| *** chohoor_ has quit IRC | 21:39 | |
| *** chohoor has joined #senlin | 21:40 | |
| *** catintheroof has quit IRC | 23:13 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!