| *** Qiming has quit IRC | 00:00 | |
| *** lawrancejing has joined #senlin | 00:28 | |
| *** lawrancejing has quit IRC | 00:36 | |
| *** lvdongbing has joined #senlin | 00:39 | |
| *** Qiming has joined #senlin | 00:54 | |
| openstackgerrit | Qiming Teng proposed stackforge/senlin: Remove local version of ssl middleware https://review.openstack.org/225486 | 01:15 |
|---|---|---|
| *** ldb_ has joined #senlin | 02:05 | |
| *** lvdongbing has quit IRC | 02:06 | |
| *** elynn has joined #senlin | 02:12 | |
| *** Shijia has joined #senlin | 02:15 | |
| *** lvdongbing has joined #senlin | 02:29 | |
| *** ldb_ has quit IRC | 02:29 | |
| *** Zhenqi has joined #senlin | 02:35 | |
| *** Shijia has quit IRC | 02:42 | |
| *** Shijia has joined #senlin | 03:23 | |
| *** Zhenqi has quit IRC | 03:33 | |
| *** Zhenqi has joined #senlin | 03:33 | |
| openstackgerrit | Ethan Lynn proposed stackforge/senlin: Get 'count' from action.inputs instead of action.data https://review.openstack.org/225545 | 04:02 |
| *** Zhenqi has quit IRC | 04:06 | |
| *** Zhenqi has joined #senlin | 04:36 | |
| *** Shijia_ has joined #senlin | 04:48 | |
| *** Shijia has quit IRC | 04:52 | |
| *** Shijia_ has quit IRC | 05:00 | |
| *** Shijia has joined #senlin | 05:09 | |
| *** Zhenqi has quit IRC | 05:13 | |
| *** Shijia_ has joined #senlin | 05:33 | |
| *** Shijia has quit IRC | 05:36 | |
| *** Shijia__ has joined #senlin | 05:43 | |
| *** Shijia_ has quit IRC | 05:46 | |
| *** Shijia has joined #senlin | 05:59 | |
| *** Shijia__ has quit IRC | 06:00 | |
| *** Shijia_ has joined #senlin | 06:01 | |
| *** elynn_ has joined #senlin | 06:03 | |
| *** Shijia has quit IRC | 06:04 | |
| *** elynn has quit IRC | 06:05 | |
| *** Shijia__ has joined #senlin | 07:02 | |
| *** Shijia_ has quit IRC | 07:03 | |
| *** Shijia_ has joined #senlin | 07:04 | |
| *** Shijia__ has quit IRC | 07:07 | |
| *** Shijia_ has quit IRC | 07:23 | |
| *** Shijia has joined #senlin | 07:26 | |
| *** Shijia_ has joined #senlin | 07:36 | |
| *** Shijia has quit IRC | 07:40 | |
| *** lawrancejing has joined #senlin | 07:42 | |
| *** elynn has joined #senlin | 07:47 | |
| *** elynn_ has quit IRC | 07:47 | |
| *** LiuWei has joined #senlin | 08:02 | |
| elynn | Hi | 08:21 |
| elynn | A quick question, if I restart senlin-engine when cluster is during an action, will the lock release after senlin-engine restarted? | 08:22 |
| Qiming | no, elynn | 08:22 |
| Qiming | it won't get released | 08:22 |
| elynn | I run into a problem that this lock can not release forever... | 08:22 |
| elynn | All other actions are blocked. | 08:23 |
| Qiming | yep | 08:23 |
| elynn | so the only way is to modify senlin db? | 08:23 |
| Qiming | we need to do something when the engine is started | 08:23 |
| Qiming | elynn, that is true as of today | 08:24 |
| elynn | ok, got it. | 08:24 |
| Qiming | so you have got a cluster locked by an action? | 08:24 |
| elynn | yes | 08:24 |
| elynn | I do a scale-in on cluster and then restart senlin-engine, then all other actions are failed, even cluster-delete... | 08:26 |
| elynn | will try to hack senlin db ;) | 08:26 |
| Qiming | we need to do something here | 08:26 |
| Qiming | consider this: | 08:26 |
| Qiming | we have two levels of locks | 08:26 |
| Qiming | a cluster or a node can be locked by an action | 08:27 |
| Qiming | the other level is that an action can be locked by an engine (we have multi-engine support) | 08:27 |
| Qiming | when an engine starts, how can we reset those locks in a safe way? | 08:28 |
| Qiming | how about this ... | 08:30 |
| elynn | In heat, an engine can steal a lock from a dead engine | 08:30 |
| Qiming | is that safe? | 08:30 |
| Qiming | how can you be sure the other engine is 'dead'? | 08:31 |
| elynn | Seems heat-engine will send out a signal to see if this engine response. | 08:32 |
| Qiming | that's something we need to do | 08:34 |
| Qiming | each engine, when starts up, need to call a cleanup function to release locks | 08:35 |
| Qiming | then that engine will register it self to 'engines | 08:35 |
| Qiming | 'engines' table | 08:36 |
| *** branw has joined #senlin | 08:36 | |
| Qiming | the cleanup function looks up the engines table to check if every engine response to a 'is_alive' query | 08:36 |
| Qiming | or else, it deletes all locks that 'dead' engine holds on actions, and all locks on cluster/node by those actions | 08:37 |
| Qiming | make sense? | 08:38 |
| elynn | yes | 08:39 |
| elynn | we need to do that. | 08:39 |
| elynn | I can work on it. | 08:39 |
| Qiming | thanks, wonderful! | 08:39 |
| Qiming | I'm not doing any useful work recently, reinstalling devstack all over again | 08:40 |
| Qiming | they are taking away the servers I was using ... :( | 08:41 |
| elynn | oops.... | 08:45 |
| elynn | Are they replace your thinkpad with macbook? | 08:45 |
| elynn | reinstall devstack is painful... Especially in a poor network connection under GFW. .. | 08:47 |
| *** Shijia has joined #senlin | 08:47 | |
| *** Shijia_ has quit IRC | 08:49 | |
| *** Shijia_ has joined #senlin | 09:08 | |
| *** Shijia has quit IRC | 09:11 | |
| *** lawrancejing has quit IRC | 09:12 | |
| *** lawrancejing has joined #senlin | 09:20 | |
| *** Shijia__ has joined #senlin | 09:35 | |
| *** Shijia_ has quit IRC | 09:39 | |
| *** Shijia__ has quit IRC | 09:40 | |
| *** lawrancejing has quit IRC | 09:43 | |
| *** lvdongbing has quit IRC | 09:48 | |
| *** lawrancejing has joined #senlin | 09:51 | |
| *** lixinhui has quit IRC | 09:52 | |
| Qiming | oh, not laptop | 09:59 |
| Qiming | I was using a pureflex node, now I'm requested to migrate to a dataplex box | 10:00 |
| *** elynn has quit IRC | 10:20 | |
| *** lawrancejing has quit IRC | 10:33 | |
| *** gongysh_ has quit IRC | 10:43 | |
| *** LiuWei has quit IRC | 10:46 | |
| *** Qiming has quit IRC | 10:48 | |
| *** Qiming has joined #senlin | 11:37 | |
| openstackgerrit | Qiming Teng proposed stackforge/senlin: Remove a TODO item in cluster detach policy https://review.openstack.org/225639 | 12:05 |
| *** lawrancejing has joined #senlin | 12:12 | |
| *** jdandrea has joined #senlin | 12:50 | |
| *** lawrancejing has quit IRC | 12:56 | |
| *** pratikmallya has joined #senlin | 13:06 | |
| *** pratikma_ has joined #senlin | 13:08 | |
| *** pratikmallya has quit IRC | 13:12 | |
| openstackgerrit | Merged stackforge/senlin: Get 'count' from action.inputs instead of action.data https://review.openstack.org/225545 | 13:33 |
| openstackgerrit | Merged stackforge/senlin: Remove a TODO item in cluster detach policy https://review.openstack.org/225639 | 13:39 |
| *** Qiming has quit IRC | 13:55 | |
| *** Qiming has joined #senlin | 13:57 | |
| *** lawrancejing has joined #senlin | 14:17 | |
| *** Qiming has quit IRC | 14:34 | |
| *** Qiming has joined #senlin | 14:35 | |
| *** lawrancejing has quit IRC | 14:38 | |
| *** pratikma_ has quit IRC | 14:46 | |
| *** pratikmallya has joined #senlin | 15:17 | |
| openstackgerrit | Monty Taylor proposed stackforge/senlin: Change ignore-errors to ignore_errors https://review.openstack.org/225901 | 15:29 |
| *** pratikma_ has joined #senlin | 15:30 | |
| *** pratikmallya has quit IRC | 15:33 | |
| *** Qiming has quit IRC | 15:46 | |
| openstackgerrit | Merged stackforge/senlin: Change ignore-errors to ignore_errors https://review.openstack.org/225901 | 16:49 |
| *** pratikma_ has quit IRC | 22:24 | |
| *** pratikmallya has joined #senlin | 22:57 | |
| *** Qiming has joined #senlin | 23:36 | |
| *** pratikmallya has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!