| *** zhurong has quit IRC | 00:04 | |
| *** elynn has quit IRC | 00:35 | |
| *** zhurong has joined #senlin | 01:17 | |
| *** yanyanhu has joined #senlin | 01:27 | |
| *** elynn has joined #senlin | 01:52 | |
| *** elynn_ has joined #senlin | 01:57 | |
| *** elynn has quit IRC | 02:00 | |
| *** guoshan has joined #senlin | 02:11 | |
| *** guoshan has quit IRC | 02:26 | |
| openstackgerrit | Merged openstack/senlin: api-ref: OpenStack-Request-Id and location https://review.openstack.org/358986 | 02:33 |
|---|---|---|
| yanyanhu | hi, Qiming, around? | 02:38 |
| yanyanhu | about the subscriber url, I'm considering to add a new receiver API interface, e.g. /v1/receivers/{receiver_id}/notify | 02:39 |
| yanyanhu | and also a new engine service call receiver_notify | 02:39 |
| yanyanhu | also add a simple description about trust usage in subscription to this etherpad: https://etherpad.openstack.org/p/senlin-message-type-receiver | 02:41 |
| yanyanhu | just read the code of zaqar's trust notification driver, I think we can apply this way to support authentication for receiver notify API access | 02:42 |
| Qiming | so the receiver-notify will only server authenticated and authorized requests? | 02:44 |
| Qiming | s/server/serve | 02:45 |
| yanyanhu | yes | 02:45 |
| yanyanhu | and it will be invisible for enduer | 02:46 |
| yanyanhu | will make some tries on this design to see whether it can work as expected, especially the trust subscription part | 02:49 |
| openstackgerrit | Qiming Teng proposed openstack/senlin: Fix docs about autoscaling https://review.openstack.org/359048 | 02:56 |
| Qiming | okay | 02:56 |
| *** yanyanhu has quit IRC | 02:56 | |
| *** yanyanhu has joined #senlin | 02:58 | |
| openstackgerrit | Nam Nguyen Hoai proposed openstack/senlin: Clean imports in code https://review.openstack.org/360876 | 03:54 |
| openstackgerrit | Merged openstack/senlin: Improve server metadata format https://review.openstack.org/356302 | 03:56 |
| *** gongysh_ has joined #senlin | 03:59 | |
| openstackgerrit | Merged openstack/senlin: Fix docs about autoscaling https://review.openstack.org/359048 | 04:04 |
| openstackgerrit | Qiming Teng proposed openstack/senlin: Fix api version mismatch return code https://review.openstack.org/360879 | 04:08 |
| *** xuhaiwei has quit IRC | 04:11 | |
| *** elynn_ has quit IRC | 04:31 | |
| openstackgerrit | XueFeng Liu proposed openstack/senlin: Fix policy command example https://review.openstack.org/360884 | 04:36 |
| *** elynn_ has joined #senlin | 04:56 | |
| *** elynn_ has quit IRC | 05:00 | |
| *** elynn_ has joined #senlin | 05:01 | |
| *** elynn_ has quit IRC | 05:06 | |
| *** elynn_ has joined #senlin | 05:08 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/senlin: Updated from global requirements https://review.openstack.org/360898 | 05:14 |
| Qiming | $ senlin cluster-run -u cirros -n private -t fixed -s r1.sh c1 | 05:34 |
| Qiming | node: f9274ccb-99fa-409d-8678-37dfd2645b85 | 05:34 |
| Qiming | status: SUCCEEDED | 05:34 |
| Qiming | output: | 05:34 |
| Qiming | bin | 05:34 |
| Qiming | dev | 05:34 |
| Qiming | etc | 05:34 |
| Qiming | home | 05:34 |
| Qiming | init | 05:35 |
| Qiming | lib | 05:35 |
| Qiming | linuxrc | 05:35 |
| Qiming | lost+found | 05:35 |
| Qiming | media | 05:35 |
| Qiming | mnt | 05:35 |
| Qiming | old-root | 05:35 |
| elynn_ | Wow! | 05:35 |
| Qiming | opt | 05:35 |
| Qiming | proc | 05:35 |
| Qiming | root | 05:35 |
| Qiming | run | 05:35 |
| Qiming | sbin | 05:35 |
| Qiming | sys | 05:35 |
| Qiming | tmp | 05:35 |
| Qiming | usr | 05:35 |
| Qiming | var | 05:35 |
| Qiming | node: cdd5c318-2c85-4cf8-be2b-fd353a2de8be | 05:35 |
| Qiming | status: SUCCEEDED | 05:35 |
| Qiming | output: | 05:35 |
| Qiming | bin | 05:35 |
| Qiming | dev | 05:35 |
| Qiming | etc | 05:35 |
| Qiming | home | 05:35 |
| Qiming | init | 05:35 |
| Qiming | lib | 05:35 |
| Qiming | linuxrc | 05:35 |
| Qiming | lost+found | 05:35 |
| Qiming | media | 05:35 |
| Qiming | mnt | 05:36 |
| Qiming | old-root | 05:36 |
| Qiming | opt | 05:36 |
| Qiming | proc | 05:36 |
| Qiming | root | 05:36 |
| Qiming | run | 05:36 |
| Qiming | sbin | 05:36 |
| Qiming | sys | 05:36 |
| Qiming | tmp | 05:36 |
| Qiming | usr | 05:36 |
| Qiming | var | 05:36 |
| Qiming | $ cat r1.sh | 05:36 |
| Qiming | ls / | 05:36 |
| Qiming | $ senlin cluster-node-list c1 | 05:36 |
| Qiming | WARNING (shell) "senlin cluster-node-list" is deprecated, please use "openstack cluster node members list" instead. | 05:36 |
| Qiming | +----------+-------------------+-------+--------+-------------+----------------------+ | 05:36 |
| Qiming | | id | name | index | status | physical_id | created_at | | 05:36 |
| Qiming | +----------+-------------------+-------+--------+-------------+----------------------+ | 05:36 |
| Qiming | | cdd5c318 | node-db59d9de-004 | 4 | ACTIVE | 342b31c1 | 2016-08-26T05:21:16Z | | 05:36 |
| Qiming | | f9274ccb | node-db59d9de-003 | 3 | ACTIVE | 4fb20876 | 2016-08-26T05:21:18Z | | 05:36 |
| Qiming | +----------+-------------------+-------+--------+-------------+----------------------+ | 05:36 |
| openstackgerrit | Merged openstack/senlin: Close health manager loop https://review.openstack.org/356303 | 05:54 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/senlin: Updated from global requirements https://review.openstack.org/360898 | 05:57 |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Rework receiver delete https://review.openstack.org/360911 | 06:03 |
| yanyanhu | Qiming, cool :) | 06:05 |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Rework receiver delete https://review.openstack.org/360911 | 06:06 |
| lixinhui | Qiming, Looks good | 06:47 |
| lixinhui | have you read the code of Nova about host failure part | 06:47 |
| Qiming | it was a client-only patch | 06:47 |
| lixinhui | as far as I know | 06:47 |
| Qiming | haven't figured out how to transfer private key securely to senlin server and let the server do the ssh later | 06:48 |
| lixinhui | Nova only provide heartbeat nowdays | 06:48 |
| lixinhui | yes, Qiming, that is the critical part | 06:48 |
| lixinhui | what I am asking is about next step of host failure fencing | 06:48 |
| Qiming | okay, have to dig the openstack-dev mailinglist | 06:49 |
| lixinhui | okay | 06:49 |
| Qiming | april 07 | 06:51 |
| Qiming | http://lists.openstack.org/pipermail/openstack-dev/2016-April/091666.html | 06:53 |
| Qiming | this was the response: http://lists.openstack.org/pipermail/openstack-dev/2016-April/091735.html | 06:53 |
| lixinhui | I think I can understand this thread from my reading of nova's code about service status notifinication part | 06:57 |
| lixinhui | When admin trigger host into Maintenance, he/she is supposted to disable the nova services firstly | 06:58 |
| lixinhui | and accordingly, the event will be sent out | 06:58 |
| Qiming | accidental compute down is actually detected, but not reported | 06:59 |
| Qiming | that was my impression | 06:59 |
| lixinhui | that does not mean any event will happen when heartbeat failed, heartbeat still only change the DB | 06:59 |
| lixinhui | you can confirm | 07:00 |
| lixinhui | then discuss what we should do for host failure fencing | 07:01 |
| Qiming | ideally, nova knows the status of nova-compute | 07:09 |
| Qiming | when a nova-compute service is down, that physical machine is useless | 07:09 |
| Qiming | it should emit a huge notification (or in separate steps) containing the host info, hypervisor info, and the info about the list of vm instances running on that machine | 07:10 |
| Qiming | yanyan, any idea why 360911 fails? | 07:11 |
| Qiming | I don't think nova has that capability today | 07:14 |
| yanyanhu | Qiming, oh, didn't notice it failed, let me check it | 07:18 |
| Qiming | you forgot to add test case to the change of the service module | 07:18 |
| yanyanhu | ah, yes | 07:19 |
| yanyanhu | will add it | 07:19 |
| yanyanhu | BTW, about get_endpoint support in sdk session http://git.openstack.org/cgit/openstack/python-openstacksdk/tree/openstack/session.py#n206 | 07:19 |
| yanyanhu | I made a test and found region filtering doesn't take effect | 07:19 |
| yanyanhu | while service_type and interface filtering works | 07:20 |
| openstackgerrit | Merged openstack/senlin: Fix api version mismatch return code https://review.openstack.org/360879 | 07:32 |
| Qiming | yup | 07:32 |
| yanyanhu | Qiming, so may need a fix at sdk side? | 07:33 |
| Qiming | line 225 is invoking keystoneauth1 | 07:33 |
| Qiming | you mean the filter is not working there? | 07:33 |
| yanyanhu | yes, seems so | 07:33 |
| yanyanhu | I passed region=$REGION_NAME as kwargs | 07:34 |
| Qiming | no | 07:34 |
| yanyanhu | but I found kwargs was not used in get_endpoint | 07:34 |
| Qiming | in sdk, you set region via profile, right? | 07:34 |
| Qiming | look at line 222 | 07:35 |
| yanyanhu | yes, but the region of service I want to get endpoint for could be different from the one defined in profile I think | 07:35 |
| Qiming | you are supposed to do profile.set_region(...) before that? | 07:35 |
| Qiming | it may and may not make sense always, right? | 07:36 |
| yanyanhu | Qiming, sometimes, maybe, e.g. services from different regions registered to one keystone | 07:36 |
| Qiming | how would you assume keystone at one region knows the endpoint for all other regions? | 07:36 |
| Qiming | yes, that is possible | 07:37 |
| yanyanhu | so in that case, I think this won't work? | 07:38 |
| Qiming | to support that case ... you may have to fix sdk | 07:38 |
| yanyanhu | umm, guess so | 07:38 |
| yanyanhu | not sure whether keystoneauth1 provides this support | 07:39 |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Rework receiver delete https://review.openstack.org/360911 | 07:40 |
| Qiming | about keystoneauth1 support, you may want to try a quick fix | 07:42 |
| Qiming | pass the kwargs directly to keystoneauth.session.get_endpoint() | 07:42 |
| yanyanhu | Qiming, yes | 07:42 |
| yanyanhu | let me have a try | 07:42 |
| Qiming | the base authentication supports this: | 07:43 |
| Qiming | 159 def get_endpoint(self, session, service_type=None, interface=None, | 07:43 |
| Qiming | 160 region_name=None, service_name=None, version=None, | 07:43 |
| Qiming | 161 allow={}, **kwargs): | 07:43 |
| yanyanhu | oh, I guess that means region_name based filtering is supported | 07:43 |
| yanyanhu | hi, Qiming, if I add 'region' key to filt, it works | 07:57 |
| yanyanhu | e.g. after line 224, add if region: filt.region = region | 07:57 |
| yanyanhu | http://git.openstack.org/cgit/openstack/python-openstacksdk/tree/openstack/session.py#n224 | 07:57 |
| yanyanhu | but just as you said, not sure this is real use case | 07:57 |
| yanyanhu | from sdk's perspective | 07:58 |
| Qiming | you may want to add a region_name=None to the parameter list | 08:00 |
| Qiming | then the key on line 218 becomes (service_type, region_name, interface) | 08:00 |
| yanyanhu | if so, for the case using cache, region_name will be used to filter endpoint | 08:03 |
| yanyanhu | hi, Qiming, do we need to propose this kind of fix to sdk? | 08:04 |
| Qiming | i think it makes sense | 08:06 |
| Qiming | let's see if brian agrees to it | 08:06 |
| yanyanhu | I see. | 08:06 |
| yanyanhu | will propose a patch for it | 08:06 |
| yanyanhu | anyway, it won't be included into newton release. But I think it's ok us | 08:07 |
| yanyanhu | ok for us | 08:07 |
| Qiming | well, that is a good point | 08:07 |
| Qiming | maybe we should prepare for both | 08:07 |
| Qiming | ideally, the logic should go to sdk | 08:08 |
| yanyanhu | I see | 08:08 |
| Qiming | but if we are trying to retrieve service endpoint, that version based endpoint select logic is the right thing to do | 08:08 |
| Qiming | if we want this feature to be included in senlin newton release, we can keep a local copy of the logic | 08:09 |
| Qiming | and then remove it when sdk has the fix | 08:09 |
| yanyanhu | you mean overriding the get_endpoint of sdk session? | 08:09 |
| yanyanhu | oh, we just implement the version match logic in the get_base_url method? | 08:10 |
| yanyanhu | like what session.get_endpoint does | 08:11 |
| yanyanhu | rather than using hard coded 'v1' there | 08:11 |
| *** openstackgerrit has quit IRC | 08:18 | |
| *** openstackgerrit has joined #senlin | 08:19 | |
| *** ChanServ sets mode: +v openstackgerrit | 08:19 | |
| Qiming | yep | 08:21 |
| Qiming | avoid hardcoding would be great | 08:21 |
| Qiming | although we don't know for sure when we will have to upgrade to v2 | 08:21 |
| openstackgerrit | Qiming Teng proposed openstack/python-senlinclient: Add 'cluster-run' command https://review.openstack.org/360996 | 08:59 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/senlin: Updated from global requirements https://review.openstack.org/360898 | 09:18 |
| *** elynn_ has quit IRC | 09:29 | |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Minor tweak webhook receiver https://review.openstack.org/359636 | 09:41 |
| openstackgerrit | Merged openstack/senlin: Enable health policy to handle CLUSTER_SCALE_IN https://review.openstack.org/358454 | 09:43 |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Add version support to schema and spec https://review.openstack.org/348709 | 09:50 |
| openstackgerrit | lvdongbing proposed openstack/senlin: Fix reraise issue https://review.openstack.org/361048 | 09:55 |
| *** zhurong has quit IRC | 10:01 | |
| *** yanyanhu has quit IRC | 10:02 | |
| *** zzxwill has joined #senlin | 10:03 | |
| *** yanyanhu has joined #senlin | 10:03 | |
| *** yanyanhu has quit IRC | 10:03 | |
| openstackgerrit | Merged openstack/senlin: Updated from global requirements https://review.openstack.org/360898 | 10:27 |
| openstackgerrit | xu-haiwei proposed openstack/senlin: Fix some bugs for container creation https://review.openstack.org/349906 | 11:12 |
| *** elynn_ has joined #senlin | 11:45 | |
| *** elynn_ has quit IRC | 11:48 | |
| *** zzxwill has quit IRC | 11:49 | |
| *** elynn_ has joined #senlin | 11:51 | |
| openstackgerrit | Qiming Teng proposed openstack/senlin: Update api-ref configuration https://review.openstack.org/361160 | 13:03 |
| *** zhurong has joined #senlin | 13:04 | |
| *** zhurong has quit IRC | 15:03 | |
| *** Drago has joined #senlin | 15:47 | |
| *** Drago has quit IRC | 15:47 | |
| *** Drago has joined #senlin | 15:47 | |
| *** elynn_ has quit IRC | 16:15 | |
| openstackgerrit | xu-haiwei proposed openstack/senlin: Fix some bugs for container creation https://review.openstack.org/349906 | 16:26 |
| *** Drago has quit IRC | 17:26 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!