| *** yamamoto_ has joined #openstack-karbor | 00:18 | |
| *** yamamoto_ has quit IRC | 00:24 | |
| *** jiaopengju has joined #openstack-karbor | 00:30 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/karbor master: Updated from global requirements https://review.openstack.org/495182 | 00:37 |
|---|---|---|
| *** yamamoto_ has joined #openstack-karbor | 01:21 | |
| *** yamamoto_ has quit IRC | 01:26 | |
| *** zengyingzhe has quit IRC | 01:31 | |
| *** liujiong has joined #openstack-karbor | 02:08 | |
| *** yamamoto_ has joined #openstack-karbor | 02:22 | |
| *** yamamoto_ has quit IRC | 02:28 | |
| *** yamamoto has joined #openstack-karbor | 02:56 | |
| openstackgerrit | chenying proposed openstack/karbor master: Spec: Add checkpoint verify API for Karbor https://review.openstack.org/495855 | 02:58 |
| *** yamamoto has quit IRC | 03:02 | |
| *** afanti has joined #openstack-karbor | 03:23 | |
| *** yamamoto has joined #openstack-karbor | 04:03 | |
| *** yamamoto has quit IRC | 04:09 | |
| *** yamamoto has joined #openstack-karbor | 05:05 | |
| *** yamamoto has quit IRC | 05:10 | |
| *** yamamoto has joined #openstack-karbor | 06:06 | |
| *** yamamoto has quit IRC | 06:12 | |
| jiaopengju | hi chenying yuval | 06:15 |
| chenying | jiaopengju hi | 06:16 |
| jiaopengju | I’m testing karbor’s pike branch recently, found some issues. https://bugs.launchpad.net/karbor/+bug/1712059 | 06:18 |
| openstack | Launchpad bug 1712059 in Karbor "Image boot nova instance do not backup the user data" [Undecided,New] | 06:18 |
| jiaopengju | In some of our product environment, we use image boot to create an instance. IMO, we should also protect the users’ data of an image boot instance | 06:20 |
| *** dixiaoli has joined #openstack-karbor | 06:21 | |
| chenying | ping yuval | 06:29 |
| chenying | jiaopengju I see. You want to protect the bootable image of the server including user's data? | 06:31 |
| jiaopengju | chenying yes | 06:31 |
| jiaopengju | In some of our environment, we provide two types of vms. boot from volume and boot from image (for users, they can choose create the nova instance locally or using ebs) | 06:33 |
| chenying | jiaopengju we can discuss it with yuval. In the original implementation of server plugins. The snapshot of server's image will be created in plugins. When restoring a new server, the server will be creating with the sanpshot of the iamge, so the new server will incule user's data. | 06:36 |
| jiaopengju | chenying ok | 06:37 |
| yuval | hey chenying jiaopengju | 06:54 |
| jiaopengju | hi yuval | 06:55 |
| yuval | protect & restore a server booting from image is important and should be supported. AFAIK if you choose to protect servers, karbor dependecy graph protects the base image as well | 06:55 |
| jiaopengju | yuval yes, but karbor do not protect the users’ data | 06:56 |
| yuval | jiaopengju: what do you mean by 'user data'? where is it stored? | 06:56 |
| chenying | hi yuval But now in karbor, only the original image in glance has been protected. The sanpshot of the server's image has not been protected. So some use's data has been lost. | 06:57 |
| yuval | jiaopengju: oh, you want to take a whole snapshot of ephemeral data? | 06:57 |
| jiaopengju | yuval: First, I created a image boot instance, then I write some data or install some softwares in the created VM | 06:58 |
| jiaopengju | yuval: Yes | 06:58 |
| chenying | yuval a whole snapshot of ephemeral data Yes. From the usecase of public cloud, the user indeed has the request to protect the snapshot of the server's image with ephemeral data. | 06:59 |
| jiaopengju | yuval: some of our product environments are like this | 06:59 |
| yuval | jiaopengju: but data saved on top of an image is by definition ephermal, and can get lost if the machine is powered down, for example | 06:59 |
| chenying | yuval AFAIK, the definition ephermal data will not be lost when the machine is powered down. Is that right? jiaopengju | 07:00 |
| jiaopengju | yuval: the data will not be lost | 07:02 |
| jiaopengju | yuval the data will be written to disk | 07:03 |
| yuval | jiaopengju: and if a user protect 10 servers, each booted from an ubuntu image, then 10 new snapshot images will be created? | 07:11 |
| jiaopengju | yuval: maybe yes | 07:13 |
| jiaopengju | yuval chenying: In some of our environments, we use snapshot to protect the image boot instances | 07:14 |
| jiaopengju | sanpshot and create a new imgae based on the running instance | 07:15 |
| yuval | jiaopengju: well, you could write such plugin, or add a configuration option to the current plugin. | 07:19 |
| chenying | yuval adding a configuration option sound good. I'm going to say that. :D | 07:20 |
| yuval | jiaopengju: I would advise that this is a bad practice, but I can not argue with your service demands. If this is a different plugin or an option where the complexities are explained - that's ok, I guess | 07:21 |
| yuval | (option in the current plugin) | 07:21 |
| chenying | jiaopengju yuval option in the current plugin is a good idea. | 07:21 |
| chenying | jiaopengju What's your oppion about it? | 07:22 |
| jiaopengju | yuval chenying: agree with you | 07:23 |
| jiaopengju | yuval use image boot instance is due to some ‘can not describable’ reasons | 07:25 |
| yuval | jiaopengju: that's ok | 07:28 |
| yuval | jiaopengju: just pointing out that fact | 07:28 |
| jiaopengju | yuval :) | 07:29 |
| yuval | chenying: any patch we need to cherry pick over stable/pike from master? | 07:46 |
| yuval | chenying: I can't think of any, wanted to see if you have anything | 07:47 |
| chenying | yuval The patch about bug fix could be cherry pick over stable/pike from master. | 07:48 |
| yuval | chenying: which bug fix? | 07:49 |
| chenying | yuval It seems that karbor project don't have any bug fix after the stable/pike branch. | 07:53 |
| yuval | chenying: yes. Only karborclient has, but it's not critical enough for stable/pike + release | 07:54 |
| chenying | yuva Are there any bugfix in karbor-dashboard? | 07:57 |
| chenying | https://review.openstack.org/#/c/493353/ | 07:59 |
| chenying | https://review.openstack.org/#/c/492393/ | 07:59 |
| chenying | yuval Are these two patches included in stable/pike branch? | 08:00 |
| chenying | ping yuval | 08:02 |
| yuval | chenying: yes | 08:06 |
| chenying | yuval It seems that no patches need be cherry pick over stable/pike from master now. | 08:10 |
| yuval | chenying: great, we are ready for release then :) | 08:11 |
| yuval | chenying: maybe just need to look at the recent bugs reported here... | 08:11 |
| yuval | chenying: maybe this: https://bugs.launchpad.net/karbor/+bug/1711881 | 08:12 |
| openstack | Launchpad bug 1711881 in Karbor "Acquire lease failed many times" [Undecided,New] | 08:12 |
| *** liujiong_lj has joined #openstack-karbor | 08:13 | |
| *** liujiong has quit IRC | 08:14 | |
| chenying | ping jiaopengju | 08:16 |
| jiaopengju | hi chenying | 08:16 |
| chenying | jiaopengju https://bugs.launchpad.net/karbor/+bug/1711881 About this bug, do you find the reason? | 08:17 |
| openstack | Launchpad bug 1711881 in Karbor "Acquire lease failed many times" [Undecided,New] | 08:17 |
| *** Natanbro has joined #openstack-karbor | 08:18 | |
| jiaopengju | chenying: according to the log, it seems that the lease object was already expired or deleted, but karbor still acquires it | 08:19 |
| jiaopengju | chenying: I have not analyzed it deeply yet. | 08:19 |
| chenying | jiaopengju Do you think that it need be include in pike branch? If not, we could fix it in master branch. | 08:21 |
| jiaopengju | chenying: not need | 08:22 |
| jiaopengju | chenying: I think this should be https://bugs.launchpad.net/karbor-dashboard/+bug/1711879 but it maybe a little big change | 08:24 |
| openstack | Launchpad bug 1711879 in karbor-dashboard "Failed to restore image boot server" [Undecided,New] | 08:24 |
| jiaopengju | this issue is not only about image boot server, volume boot type instance also have this issue. https://bugs.launchpad.net/karbor-dashboard/+bug/1711879 | 08:25 |
| openstack | Launchpad bug 1711879 in karbor-dashboard "Failed to restore server" [Undecided,New] | 08:25 |
| *** liujiong has joined #openstack-karbor | 08:25 | |
| *** liujiong_lj has quit IRC | 08:26 | |
| chenying | jiaopengju As we discussed, only the network metadat about the server should be protected when backuping the server. So we may need fix it first. | 08:27 |
| *** liujiong_lj has joined #openstack-karbor | 08:28 | |
| chenying | jiaopengju: I have a idea about it. | 08:28 |
| jiaopengju | chenying: Yes. I will try to do this | 08:29 |
| *** liujiong has quit IRC | 08:30 | |
| jiaopengju | chenying: It seems we should tell the neutron protection plugin about the resources’ parent resource | 08:30 |
| chenying | jiaopengju Now the network protectable plugin only return the protect_id as the network resource id. So when the network resource is the subresource of a server, it could return the server id as the network resource id in the network protectable pluin. | 08:30 |
| chenying | jiaopengju So that in the network protection plugin, we can ony protect the network metadata filtered by server id. When restoring a network resource, we shoud check it, don't neet create the resource repeatedly. | 08:33 |
| jiaopengju | chenying: good idea! | 08:34 |
| *** yamamoto has joined #openstack-karbor | 08:35 | |
| chenying | yuval: What's your oppion about it? | 08:35 |
| chenying | yuval Now the network plugin protect all the network metadata, even when backuping one server, all the network metadatas are protected. In this usecase, we only need backup the network metadata about the server. | 08:37 |
| yuval | chenying: what if you are restoring to a different site? | 08:44 |
| chenying | yuval In this usecase, if the network resource is independent resource. All the network metadata will be backuped and restored in network plugin. Only if the resource is the subresource of the server, the network metadata will be filtered by server id. | 08:49 |
| *** yamamoto has quit IRC | 08:59 | |
| *** yamamoto has joined #openstack-karbor | 08:59 | |
| openstackgerrit | chenying proposed openstack/karbor master: Spec: Add checkpoint verification API for Karbor https://review.openstack.org/495855 | 09:17 |
| *** liujiong_lj has quit IRC | 09:27 | |
| *** liujiong has joined #openstack-karbor | 09:31 | |
| *** yamamoto has joined #openstack-karbor | 10:00 | |
| *** dixiaoli has quit IRC | 10:04 | |
| *** yamamoto has quit IRC | 10:05 | |
| *** liujiong has quit IRC | 10:13 | |
| *** jiaopengju has quit IRC | 10:14 | |
| *** yamamoto has joined #openstack-karbor | 10:30 | |
| *** lihi has quit IRC | 11:01 | |
| *** lihi has joined #openstack-karbor | 11:05 | |
| *** yamamoto has quit IRC | 11:18 | |
| *** yamamoto has joined #openstack-karbor | 11:30 | |
| *** yamamoto has quit IRC | 11:46 | |
| *** yamamoto has joined #openstack-karbor | 11:48 | |
| openstackgerrit | chenying proposed openstack/karbor master: Spec: Add checkpoint verification API for Karbor https://review.openstack.org/495855 | 12:02 |
| *** yamamoto has quit IRC | 12:03 | |
| *** yamamoto has joined #openstack-karbor | 12:04 | |
| *** yamamoto has quit IRC | 12:10 | |
| *** yamamoto has joined #openstack-karbor | 12:12 | |
| *** jiaopengju has joined #openstack-karbor | 12:26 | |
| *** zhurong has joined #openstack-karbor | 12:56 | |
| *** yamamoto has quit IRC | 13:12 | |
| *** yamamoto has joined #openstack-karbor | 13:14 | |
| *** yamamoto has quit IRC | 13:22 | |
| *** jiaopengju has quit IRC | 13:57 | |
| *** zhurong has quit IRC | 14:19 | |
| *** yamamoto has joined #openstack-karbor | 14:23 | |
| *** yamamoto has quit IRC | 14:28 | |
| *** gouthamr has joined #openstack-karbor | 14:59 | |
| *** yamamoto has joined #openstack-karbor | 15:25 | |
| *** yamamoto has quit IRC | 15:30 | |
| *** dixiaoli has joined #openstack-karbor | 15:50 | |
| *** dixiaoli has quit IRC | 15:54 | |
| *** afanti has quit IRC | 15:55 | |
| *** Natanbro has quit IRC | 16:03 | |
| *** yamamoto has joined #openstack-karbor | 16:26 | |
| *** yamamoto has quit IRC | 16:32 | |
| *** yamamoto has joined #openstack-karbor | 17:28 | |
| *** yamamoto has quit IRC | 17:35 | |
| *** dixiaoli has joined #openstack-karbor | 18:21 | |
| *** dixiaoli has quit IRC | 18:25 | |
| *** yamamoto has joined #openstack-karbor | 18:31 | |
| *** yamamoto has quit IRC | 18:37 | |
| *** yamamoto has joined #openstack-karbor | 19:33 | |
| *** yamamoto has quit IRC | 19:38 | |
| *** gouthamr has quit IRC | 20:29 | |
| *** yamamoto has joined #openstack-karbor | 20:35 | |
| *** yamamoto has quit IRC | 20:38 | |
| *** yamamoto has joined #openstack-karbor | 20:38 | |
| *** yamamoto has quit IRC | 20:39 | |
| *** gouthamr has joined #openstack-karbor | 20:53 | |
| *** yamamoto has joined #openstack-karbor | 21:10 | |
| *** yamamoto has quit IRC | 21:15 | |
| *** yamamoto_ has joined #openstack-karbor | 22:11 | |
| *** yamamoto_ has quit IRC | 22:17 | |
| *** yamamoto_ has joined #openstack-karbor | 23:13 | |
| *** gouthamr has quit IRC | 23:16 | |
| *** yamamoto_ has quit IRC | 23:18 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!