| *** guoshan has quit IRC | 00:01 | |
| *** zhurong has joined #senlin | 00:19 | |
| *** zhurong has quit IRC | 00:40 | |
| *** guoshan has joined #senlin | 00:51 | |
| *** Drago1 has joined #senlin | 00:51 | |
| *** root___11 has joined #senlin | 00:52 | |
| *** Drago has quit IRC | 00:52 | |
| *** root___11 has left #senlin | 00:54 | |
| *** guoshan has quit IRC | 00:55 | |
| *** Drago1 has quit IRC | 00:56 | |
| *** miaohb has quit IRC | 00:58 | |
| *** miaohb has joined #senlin | 01:02 | |
| *** ruijie has joined #senlin | 01:12 | |
| *** yanyanhu has joined #senlin | 01:17 | |
| *** yanyanhu has quit IRC | 01:20 | |
| *** yanyanhu has joined #senlin | 01:21 | |
| *** guoshan has joined #senlin | 01:34 | |
| *** elynn has joined #senlin | 01:44 | |
| *** zhurong has joined #senlin | 01:51 | |
| *** elynn_ has joined #senlin | 01:53 | |
| *** shu-mutou-AWAY is now known as shu-mutou | 01:53 | |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: api support policy-type-list2 https://review.openstack.org/397591 | 01:53 |
|---|---|---|
| *** elynn has quit IRC | 01:55 | |
| *** EldonZhao has joined #senlin | 01:55 | |
| openstackgerrit | Merged openstack/senlin: fields support policy-type-list2 https://review.openstack.org/397565 | 01:56 |
| openstackgerrit | Merged openstack/senlin: engine service support policy-type-list2 https://review.openstack.org/397569 | 01:56 |
| openstackgerrit | gengchc2 proposed openstack/senlin: Translate info-level log messages for LOG.error https://review.openstack.org/398695 | 02:09 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: remove dead code about policy-type-list https://review.openstack.org/397595 | 02:17 |
| openstackgerrit | Merged openstack/senlin: Fix environment unit test https://review.openstack.org/397555 | 02:26 |
| openstackgerrit | Merged openstack/senlin: Move approved spec into the 'approved dir https://review.openstack.org/397475 | 02:26 |
| *** lvdongbing has joined #senlin | 02:39 | |
| Qiming | reminder | 02:53 |
| Qiming | this week is ocata-1 release deadline | 02:54 |
| ruijie | hi, Qiming. | 03:01 |
| Qiming | yes | 03:01 |
| ruijie | For the improvement of batch policy | 03:02 |
| ruijie | Do I need to keep the BP in processing or mark it as completed | 03:02 |
| Qiming | as far as that bp goes, it can be marked as completed | 03:03 |
| Qiming | there could still be things for improvement, but those can be left for future work | 03:04 |
| Qiming | my original expectation is that the policy can just focus on cluster update | 03:04 |
| Qiming | but we have already added cluster-delete support, that is fine | 03:05 |
| ruijie | yes, Qiming. | 03:05 |
| EldonZhao | https://github.com/openstack/senlin/blob/master/senlin/profiles/os/nova/server.py#L93 | 03:05 |
| ruijie | And in fact, I thinking if we could let the user decide whether use the policy for resizing or scaling | 03:06 |
| ruijie | am | 03:06 |
| Qiming | it is up to you though, whether you want to revisit all cluster actions and see if there are some actions missing support for batch policy | 03:06 |
| EldonZhao | what does this id mean? all the servers in this cluster use the same volume? | 03:06 |
| Qiming | em, interesting question, EldonZhao | 03:07 |
| Qiming | my answer is that "I don't know" | 03:08 |
| Qiming | for a guest Linux system, volume can be mounted at different location | 03:08 |
| EldonZhao | In cmcc product, we create volume before creating every node. | 03:09 |
| Qiming | it could be a shared CDROM | 03:09 |
| Qiming | mounted by every VM for shared installation/update source | 03:09 |
| EldonZhao | Shared storage? | 03:09 |
| Qiming | it could be | 03:09 |
| Qiming | we let the users choose | 03:09 |
| EldonZhao | Maybe we should support creating one storage for every node. | 03:10 |
| Qiming | having many VMs boot from the same disk volume may not make senses, but I think users should know it | 03:10 |
| Qiming | so you mean a new cinder volume created and then attached to a VM ? | 03:11 |
| EldonZhao | Yes, I this so | 03:11 |
| *** yanyanhu has quit IRC | 03:11 | |
| Qiming | the nova server spec is not designed to support creation of other resources ... | 03:11 |
| EldonZhao | and my product has made it sense. | 03:11 |
| EldonZhao | We may call cinder first | 03:12 |
| Qiming | so ... can you tell nova to do this if without senlin? | 03:12 |
| Qiming | my guess is no | 03:12 |
| EldonZhao | we create pool in advance in lb_policy | 03:12 |
| Qiming | okay, that is a different story | 03:13 |
| EldonZhao | I think this is same as pool | 03:13 |
| Qiming | you have volumes created beforehand | 03:13 |
| EldonZhao | storage can be aotumatically created | 03:13 |
| EldonZhao | volumeS, that is the point | 03:13 |
| Qiming | "storage can be aotumatically created" --- not yet senlin's scope, right? | 03:14 |
| openstackgerrit | XueFeng Liu proposed openstack/python-senlinclient: Revise the help message of policy binding list https://review.openstack.org/393981 | 03:14 |
| EldonZhao | how could people write the profile to allocate volumeS to every node? | 03:14 |
| Qiming | okay, if you have a 1:1 map between the VM and the volume to be used | 03:15 |
| Qiming | you will have to create a profile per VM | 03:15 |
| EldonZhao | Hmm | 03:16 |
| EldonZhao | But how can this vms be a cluster? | 03:16 |
| *** yanyanhu has joined #senlin | 03:17 | |
| Qiming | for vol_id in my_vol_pool { pid = create_a_profile_with_vol(vol_id); nid = create_a_node(pid); cluster_add_node(n1, my_cluster) } | 03:17 |
| openstackgerrit | Merged openstack/senlin: Revise profile-validate2 object fields https://review.openstack.org/398210 | 03:17 |
| Qiming | the fundamental thing here is that, in a strict sense, your VMs are not identical to each other, thus not the "homogeneous" objects senlin meant to manage | 03:18 |
| openstackgerrit | Merged openstack/python-senlinclient: Add sort key "oid" in event list https://review.openstack.org/398191 | 03:18 |
| Qiming | but senlin does allow you to build a pool of "similar-but-not-so-identical" things | 03:18 |
| EldonZhao | ok, but that doesn't support receiver, as receiver is used in cluster | 03:20 |
| Qiming | why it doesn't support receiver? | 03:20 |
| Qiming | you can create VMs in whatever ways you want and add them into a single cluster | 03:21 |
| Qiming | you can still define receivers on the cluster | 03:21 |
| EldonZhao | .... | 03:22 |
| EldonZhao | When we call receiver url, it scale_out node based on the cluster's profile | 03:23 |
| Qiming | yes | 03:23 |
| EldonZhao | that profile is the same as node has already scaled out | 03:24 |
| Qiming | then you want to pretend that you can scale-out the cluster and the scale-out is smart enough to use a volume from a pool? | 03:24 |
| EldonZhao | not a pool.. | 03:24 |
| Qiming | it that case, the cluster's profile has already becoming not that useful | 03:25 |
| EldonZhao | I guess we can create a volume for the new node | 03:25 |
| Qiming | if you want a new volume to be created along with a VM, you will use heat stack profile, not a nova-server | 03:25 |
| openstackgerrit | lvdongbing proposed openstack/senlin: Revise versioned object fields for profile-create2 https://review.openstack.org/398727 | 03:26 |
| Qiming | that brings us back to the question, can you ask nova to create a volume for you when booting a VM? the answer is no... | 03:26 |
| EldonZhao | That's great~ | 03:26 |
| Qiming | if you want to combine a lot of things to be created together, you will write your own heat stack template | 03:27 |
| Qiming | there are some discussions in the community whether this process can be simplified | 03:27 |
| EldonZhao | Maybe I should review the heat profile first~ | 03:28 |
| Qiming | for the network case, it has taken a long time to land "get me a network" feature | 03:28 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: remove some useless rpc code https://review.openstack.org/397597 | 03:35 |
| *** yanyanhu_ has joined #senlin | 03:39 | |
| *** yanyanhu has quit IRC | 03:41 | |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: api support policy-type-list2 https://review.openstack.org/397591 | 03:52 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: remove dead code about policy-type-list https://review.openstack.org/397595 | 03:52 |
| *** elynn_ has quit IRC | 04:46 | |
| *** guoshan has quit IRC | 04:50 | |
| *** guoshan has joined #senlin | 04:51 | |
| *** guoshan has quit IRC | 05:01 | |
| *** yanyanhu_ has quit IRC | 05:02 | |
| *** guoshan has joined #senlin | 05:03 | |
| *** yanyanhu_ has joined #senlin | 05:05 | |
| *** elynn has joined #senlin | 05:06 | |
| openstackgerrit | Merged openstack/python-senlinclient: Add "cluster_id" column to `openstack cluster event list` https://review.openstack.org/397655 | 05:08 |
| openstackgerrit | Merged openstack/python-senlinclient: Revise the help message of policy binding list https://review.openstack.org/393981 | 05:08 |
| *** elynn has quit IRC | 05:10 | |
| openstackgerrit | miaohb proposed openstack/python-senlinclient: Add sortkey "priority" in policy binding list's help message https://review.openstack.org/393985 | 05:20 |
| openstackgerrit | Merged openstack/senlin: Engine support for profile-delete2 https://review.openstack.org/398137 | 05:33 |
| *** guoshan has quit IRC | 05:34 | |
| *** yanyanhu_ has quit IRC | 05:49 | |
| *** yanyanhu_ has joined #senlin | 05:54 | |
| Qiming | EldonZhao, still there? | 05:55 |
| openstackgerrit | Merged openstack/senlin: api support policy-type-list2 https://review.openstack.org/397591 | 06:00 |
| openstackgerrit | Merged openstack/senlin: remove dead code about policy-type-list https://review.openstack.org/397595 | 06:01 |
| *** guoshan has joined #senlin | 06:13 | |
| openstackgerrit | Merged openstack/senlin: Remove unsupported sort key "priority" https://review.openstack.org/398055 | 06:57 |
| openstackgerrit | Merged openstack/senlin: Remove unsupported sort key "user" https://review.openstack.org/398091 | 07:02 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: remove some useless rpc code https://review.openstack.org/397597 | 07:04 |
| openstackgerrit | miaohb proposed openstack/senlin: Support sort key "oid" in event list https://review.openstack.org/398116 | 07:15 |
| *** yanyanhu_ has quit IRC | 07:23 | |
| openstackgerrit | Qiming Teng proposed openstack/senlin: Support sort key "oid" in event list https://review.openstack.org/398116 | 07:26 |
| openstackgerrit | gengchc2 proposed openstack/senlin: Translate info-level log messages for LOG.error https://review.openstack.org/398695 | 07:27 |
| *** guoshan has quit IRC | 08:31 | |
| *** guoshan has joined #senlin | 08:32 | |
| openstackgerrit | Merged openstack/senlin: Revise ex_lbas.yaml cannot be found https://review.openstack.org/398027 | 08:33 |
| *** yanyanhu_ has joined #senlin | 08:57 | |
| *** yanyanhu_ has quit IRC | 08:57 | |
| openstackgerrit | lvdongbing proposed openstack/senlin: Engine support for profile-validate2 https://review.openstack.org/398863 | 09:07 |
| Qiming | EldonZhao, online? | 09:12 |
| *** zhurong has quit IRC | 09:20 | |
| *** zhurong has joined #senlin | 09:21 | |
| *** yanyanhu_ has joined #senlin | 09:22 | |
| lvdongbing | Qiming, could you please help review this patch: https://review.openstack.org/#/c/398863/1 | 09:23 |
| openstackgerrit | Qiming Teng proposed openstack/senlin: Add TODO item about referencing existing pool https://review.openstack.org/398872 | 09:23 |
| yanyanhu_ | hi, Qiming, around? | 09:24 |
| Qiming | yes | 09:24 |
| yanyanhu_ | so the Senlin release version of ocata-1 should be 3.0.0.0b1? | 09:24 |
| yanyanhu_ | and I guess we don't need to cut the version of senlin-dashboard and senlinclient? | 09:25 |
| Qiming | I think so | 09:25 |
| yanyanhu_ | I see | 09:25 |
| yanyanhu_ | will propose the patch soon | 09:25 |
| Qiming | check the release model for them | 09:25 |
| openstackgerrit | Merged openstack/senlin: remove some useless rpc code https://review.openstack.org/397597 | 09:25 |
| yanyanhu_ | ok | 09:25 |
| lvdongbing | Qiming, not sure about that we should pass ProfileValidateRequest or ProfileValidateRequestBody from API to engine when deal with profile validate | 09:26 |
| Qiming | oh, seems the release model has been removed from governance ... | 09:26 |
| yanyanhu_ | they are "cycle-with-intermediary"? | 09:27 |
| yanyanhu_ | based on the release history | 09:27 |
| Qiming | yes | 09:27 |
| yanyanhu_ | if we didn't change it | 09:27 |
| Qiming | yep | 09:27 |
| yanyanhu_ | so we don't need to follow the milestone for them ? | 09:27 |
| Qiming | check this: https://review.openstack.org/#/q/project:openstack/releases | 09:27 |
| Qiming | then you will get an idea how server and client are supposed to be released | 09:27 |
| yanyanhu_ | got it | 09:28 |
| yanyanhu_ | will refer to it | 09:28 |
| Qiming | pulling release repo now | 09:29 |
| Qiming | yes, service is 'cycle-with-milestones', client and dashboard are 'cycle-with-intermediaries' | 09:32 |
| Qiming | lvdongbing, it all depends on how would you model the request from API side | 09:34 |
| Qiming | the current implementation seems a little bit plumpy but it leaves room for future extension | 09:37 |
| Qiming | there could be extensions such as a "deep" validation or a "shallow" validation | 09:38 |
| Qiming | where a "deep" validation checks if the image specified does exist, the port specified does exist, etc | 09:38 |
| yanyanhu_ | hi, Qiming, after read this guide, https://releases.openstack.org/reference/release_models.html, I guess we don't have to cut client and dashboard release based on milestone unless we think it is necessary? | 09:39 |
| Qiming | a "shallow" one could just do validation at senlin engine level, i.e. check if there are data type mistakes in the spec etc. | 09:39 |
| Qiming | correct, yanyanhu_, we don't have to | 09:39 |
| Qiming | we have to cut a release by the end of a cycle though | 09:39 |
| yanyanhu_ | I see | 09:40 |
| Qiming | most client projects are using this model I think | 09:40 |
| yanyanhu_ | will propose patch for service now | 09:40 |
| Qiming | any specific patches we want to merge ? | 09:42 |
| lvdongbing | Qiming, thanks, got it | 09:43 |
| Qiming | we are building a pipeline for review now: https://review.openstack.org/#/q/project:openstack/senlin+status:open | 09:43 |
| Qiming | yanyanhu_, need to be careful and see if there a high-priority patch that needs to be go in | 09:44 |
| yanyanhu_ | Qiming, ok, will make a check now | 09:44 |
| Qiming | I'd suggest this: https://review.openstack.org/398116 | 09:44 |
| Qiming | and this: https://review.openstack.org/396522 | 09:44 |
| Qiming | and this: https://review.openstack.org/397920 | 09:45 |
| Qiming | and this: https://review.openstack.org/398049 | 09:45 |
| Qiming | especially the global requirements one | 09:45 |
| yanyanhu_ | right | 09:46 |
| yanyanhu_ | that one should be included | 09:46 |
| yanyanhu_ | Qiming, we are allowed to specify the hash of an pending patch, right? | 09:47 |
| yanyanhu_ | although that patch should be merged asap | 09:47 |
| *** openstackgerrit has quit IRC | 09:48 | |
| yanyanhu_ | will wait for those patches and update the cut | 09:49 |
| *** openstackgerrit has joined #senlin | 09:49 | |
| *** ChanServ sets mode: +v openstackgerrit | 09:49 | |
| *** lvdongbing has quit IRC | 09:57 | |
| *** zhurong has quit IRC | 10:02 | |
| Qiming | no ... | 10:02 |
| Qiming | the commit has to appear in git history | 10:02 |
| Qiming | you are not supposed to ask those cores from release team to verify that the commit id is being merged | 10:03 |
| Qiming | just posted -1 to the release patch | 10:04 |
| *** elynn has joined #senlin | 10:08 | |
| yanyanhu_ | Qiming, yes, I see. Have w-1 on it | 10:13 |
| yanyanhu_ | will wait for those patch merged and update the cut | 10:13 |
| openstackgerrit | Merged openstack/senlin: add batch policy spec https://review.openstack.org/398049 | 10:17 |
| openstackgerrit | Merged openstack/senlin: Support sort key "oid" in event list https://review.openstack.org/398116 | 10:17 |
| openstackgerrit | Merged openstack/senlin: Add toggle in devstack plugin to run senlin-api under Apache2 https://review.openstack.org/396522 | 10:17 |
| openstackgerrit | Merged openstack/senlin: Updated from global requirements https://review.openstack.org/397920 | 10:29 |
| *** yanyanhu_ has quit IRC | 10:31 | |
| *** EldonZhao has quit IRC | 10:36 | |
| *** guoshan has quit IRC | 10:42 | |
| *** ruijie has quit IRC | 11:00 | |
| *** guoshan has joined #senlin | 11:13 | |
| openstackgerrit | XueFeng Liu proposed openstack/senlin: Api support for event_list2 https://review.openstack.org/398341 | 11:16 |
| *** guoshan has quit IRC | 11:18 | |
| openstackgerrit | qinchunhua proposed openstack/senlin: Delete python bytecode file https://review.openstack.org/386894 | 11:28 |
| openstackgerrit | qinchunhua proposed openstack/python-senlinclient: Delete python bytecode file https://review.openstack.org/386897 | 11:30 |
| *** elynn has quit IRC | 11:32 | |
| *** guoshan has joined #senlin | 12:07 | |
| *** guoshan has quit IRC | 12:12 | |
| *** zhurong has joined #senlin | 12:24 | |
| openstackgerrit | XueFeng Liu proposed openstack/senlin: Revise action's raise catch https://review.openstack.org/398988 | 12:38 |
| openstackgerrit | XueFeng Liu proposed openstack/senlin: Remove dead code about action list https://review.openstack.org/396914 | 12:42 |
| openstackgerrit | XueFeng Liu proposed openstack/senlin: Remove dead code about action get https://review.openstack.org/396917 | 12:44 |
| *** catintheroof has joined #senlin | 12:46 | |
| *** guoshan has joined #senlin | 13:01 | |
| *** guoshan has quit IRC | 13:06 | |
| *** zhurong has quit IRC | 13:28 | |
| *** zhurong has joined #senlin | 13:28 | |
| *** zhurong has quit IRC | 13:43 | |
| openstackgerrit | Merged openstack/senlin: Remove dead code about action list https://review.openstack.org/396914 | 13:44 |
| *** zhurong has joined #senlin | 13:44 | |
| openstackgerrit | Merged openstack/senlin: Remove dead code about action get https://review.openstack.org/396917 | 13:48 |
| *** guoshan has joined #senlin | 13:56 | |
| *** guoshan has quit IRC | 14:01 | |
| *** zhurong has quit IRC | 14:21 | |
| *** Drago1 has joined #senlin | 14:36 | |
| *** guoshan has joined #senlin | 14:50 | |
| *** guoshan has quit IRC | 14:55 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-senlinclient: Updated from global requirements https://review.openstack.org/395377 | 15:09 |
| *** guoshan has joined #senlin | 15:45 | |
| *** guoshan has quit IRC | 15:49 | |
| *** guoshan has joined #senlin | 16:45 | |
| *** guoshan has quit IRC | 16:50 | |
| *** guoshan has joined #senlin | 18:18 | |
| *** guoshan has quit IRC | 18:22 | |
| *** guoshan has joined #senlin | 19:19 | |
| *** guoshan has quit IRC | 19:23 | |
| *** openstackgerrit has quit IRC | 19:48 | |
| *** openstackgerrit has joined #senlin | 19:49 | |
| *** ChanServ sets mode: +v openstackgerrit | 19:49 | |
| *** guoshan has joined #senlin | 21:20 | |
| *** guoshan has quit IRC | 21:25 | |
| *** guoshan has joined #senlin | 22:21 | |
| *** guoshan has quit IRC | 22:25 | |
| *** Drago1 has quit IRC | 22:40 | |
| *** catintheroof has quit IRC | 23:29 | |
| *** catintheroof has joined #senlin | 23:30 | |
| *** catintheroof has quit IRC | 23:35 | |
| *** shu-mutou is now known as shu-mutou-AWAY | 23:42 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!