*** tommylikehu has joined #openstack-karbor | 01:18 | |
*** chenying has quit IRC | 03:04 | |
*** chenying has joined #openstack-karbor | 03:22 | |
*** tommylikehu has quit IRC | 04:25 | |
*** tommylikehu has joined #openstack-karbor | 04:26 | |
openstackgerrit | chenying proposed openstack/karbor: Generic database Bank implementation https://review.openstack.org/365823 | 06:25 |
---|---|---|
*** yamamoto has joined #openstack-karbor | 07:00 | |
*** yamamoto has quit IRC | 07:02 | |
*** gouthamr has joined #openstack-karbor | 07:05 | |
*** yamamoto has joined #openstack-karbor | 07:33 | |
*** yuval has joined #openstack-karbor | 07:34 | |
yuval | ping zhonghua | 07:35 |
zhonghua | pong yuval | 07:36 |
zhonghua | hey | 07:36 |
*** zengyingzhe__ has joined #openstack-karbor | 07:40 | |
*** yamamoto has quit IRC | 07:42 | |
chenying | ping yuval | 07:43 |
yuval | hey chenying | 07:43 |
*** zengyingzhe_ has quit IRC | 07:44 | |
*** yamamoto has joined #openstack-karbor | 07:44 | |
chenying | I am doing the work about bugfix. Have you seen this bug? https://bugs.launchpad.net/karbor/+bug/1574980 need a mechanism to check resource status in plan | 07:46 |
openstack | Launchpad bug 1574980 in Karbor "need a mechanism to check resource status in plan" [Medium,Triaged] | 07:46 |
*** gouthamr has quit IRC | 07:48 | |
*** gouthamr has joined #openstack-karbor | 07:51 | |
*** gouthamr has quit IRC | 07:58 | |
yuval | chenying: maybe extend the protectable plugin interface? | 07:59 |
chenying | yuval: What I want to say, Do we need check the status of resources in the plan creation stage or in the protection execution stage? | 08:01 |
chenying | yuval: Checking the status of resources could be done in karbor protection service. I think vendors will check the status of resource before protecting this resource in their plug-in implementations. | 08:05 |
chenying | yuval: maybe extend the protectable plugin interface? --- maybe not, this work could be done in the protect/delete/restore interface of plugins. | 08:07 |
yuval | chenying: both in the creation stage and in the protection execution | 08:18 |
yuval | chenying: there is also a problem with showing a plan when a resource was deleted | 08:21 |
yuval | chenying: if a user creates a plan with resource A, then deletes it, trying to '.show()' the plan gives an error | 08:22 |
yuval | chenying: this is why I'm for adding a resource status to protectable interface | 08:22 |
openstackgerrit | Yuval Brik proposed openstack/karbor: Enable heat plugin in devstack sample and docs https://review.openstack.org/415638 | 08:27 |
chenying | yuval: this is why I'm for adding a resource status to protectable interface ---- you mean that adding a status filed for resources database table? | 08:29 |
*** tommylikehu_ has quit IRC | 08:30 | |
yuval | chenying: no | 08:31 |
yuval | chenying: I mean, right now if you try so do 'show' on a protectable instance | 08:31 |
yuval | chenying: and it is deleted, you get an error | 08:31 |
yuval | chenying: there should be a way for the protectable api to handle that case | 08:32 |
chenying | yuval: The protectable plugins will return error if to show a deleted resource. | 08:32 |
yuval | chenying: I know. This is way trying to view a plan with a deleted resource fails | 08:33 |
yuval | chenying: and that is not desired. we want the plan to show, and let us know that the resource is deleted | 08:33 |
chenying | yuval: maybe the the back-end of protectable plugins also don't know that resource is invalid or is having been deleted. | 08:35 |
yuval | chenying: can you provide an example for such case? | 08:35 |
*** tommylikehu_ has joined #openstack-karbor | 08:42 | |
chenying | yuval: volume protectable plugin call cinderclient to show a provided volume id. If it can not be got form cinder just return error. we don't know the volume is deleted or is a invalid id. | 08:42 |
yuval | chenying: it doesn't matter whether it is invalid or deleted. It can be displayed as 'Not Found'. The bottom line is, if the protectable plugin can't find the resource - Karbor can not operate on it | 08:44 |
chenying | yuval: It can be displayed as 'Not Found'. You mean just show the 'Not Found' flag in karbor-dashboard? don't need save this 'Not Found' flag to the resources database table? | 08:46 |
yuval | chenying: yes | 08:48 |
yuval | chenying: btw, not just the dashboard, the whole rest api of protectables | 08:49 |
chenying | yuval: The response of plan API, also show the resources list, may need display the resources as 'Not Found' when it can not been got from protectables API. | 08:54 |
yuval | chenying: I don't understand | 08:55 |
*** yamamoto has quit IRC | 09:13 | |
*** tommylikehu_ has quit IRC | 09:14 | |
*** yamamoto has joined #openstack-karbor | 09:15 | |
yuval | chenying: ? | 09:23 |
chenying | yuval: I mean that when we call plan get restapi to show the detail of a plan, do we need check the status of resources in this plan. | 09:26 |
yuval | chenying: yes | 09:27 |
chenying | yuval: I note that most of work about the volume plugin refactoring patch have been finished. | 09:28 |
chenying | Could we start the refactoring work about nova/glance plugins base on new interfaces? | 09:28 |
yuval | chenying: yes :) | 09:28 |
chenying | yuval:If you don't mind, I could start the patch about refactoring glance plugin first. | 09:28 |
yuval | chenying: the current nova protection plugin needs to be fixed, not just refactored | 09:28 |
*** tommylikehu_ has joined #openstack-karbor | 09:29 | |
chenying | yuval: what is the problom in the nova protection plugin? | 09:31 |
yuval | chenying: current nova plugin creates a snapshot of the image | 09:32 |
yuval | chenying: this is not desireable. If the server boots from a volume - use the already restored volume. If the server boots from an image - use the already restored image. | 09:33 |
chenying | yuval: You mean that we need consider this situation, the server boots from a volume. not from a image? | 09:37 |
chenying | yuval:? | 09:47 |
yuval | chenying: yes | 09:47 |
yuval | chenying: but also, currently nova protection plugin takes a snapshot of the image | 09:47 |
yuval | chenying: that's not what we want | 09:47 |
*** tonyb has quit IRC | 10:30 | |
*** lihi has quit IRC | 10:30 | |
*** tonyb has joined #openstack-karbor | 10:36 | |
*** lihi has joined #openstack-karbor | 10:36 | |
*** yamamoto has quit IRC | 10:40 | |
*** tommylikehu_ has quit IRC | 10:53 | |
*** tommylikehu_ has joined #openstack-karbor | 11:04 | |
*** tommylikehu_ has quit IRC | 11:31 | |
*** yamamoto has joined #openstack-karbor | 11:40 | |
*** tommylikehu_ has joined #openstack-karbor | 11:48 | |
*** tommylikehu_ has quit IRC | 12:02 | |
*** tommylikehu_ has joined #openstack-karbor | 12:12 | |
*** tommylikehu_ has quit IRC | 13:20 | |
*** tommylikehu_ has joined #openstack-karbor | 13:23 | |
*** tommylik_ has joined #openstack-karbor | 13:41 | |
*** tommylikehu_ has quit IRC | 13:43 | |
*** yamamoto has quit IRC | 14:55 | |
*** yamamoto has joined #openstack-karbor | 15:03 | |
*** yuval has quit IRC | 15:09 | |
*** tommylik_ has quit IRC | 15:27 | |
*** yamamoto has quit IRC | 15:42 | |
*** yamamoto has joined #openstack-karbor | 16:43 | |
*** yamamoto has quit IRC | 16:52 | |
*** gouthamr has joined #openstack-karbor | 18:31 | |
*** Jeffrey4l has quit IRC | 21:34 | |
*** tommylikehu_ has joined #openstack-karbor | 22:26 | |
*** tommylikehu_ has quit IRC | 22:31 | |
*** thx has joined #openstack-karbor | 22:38 | |
*** thx has left #openstack-karbor | 22:40 | |
*** tommylikehu_ has joined #openstack-karbor | 22:43 | |
*** gouthamr has quit IRC | 22:47 | |
*** tommylikehu_ has quit IRC | 22:47 | |
*** tommylikehu_ has joined #openstack-karbor | 22:48 | |
*** tommylikehu_ has quit IRC | 22:48 | |
*** tommylikehu_ has joined #openstack-karbor | 22:49 | |
*** tommylikehu_ has quit IRC | 22:55 | |
*** zhangshuai has quit IRC | 23:08 | |
*** zhangshuai has joined #openstack-karbor | 23:08 | |
*** Jeffrey4l has joined #openstack-karbor | 23:31 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!