| *** mtanino has quit IRC | 00:03 | |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/python-magnumclient: Use os-client-config in shell https://review.openstack.org/338742 | 00:05 |
|---|---|---|
| *** vijendar1 has joined #openstack-containers | 00:07 | |
| *** vijendar has quit IRC | 00:07 | |
| *** Drago has quit IRC | 00:11 | |
| *** vijendar1 has quit IRC | 00:12 | |
| *** Drago has joined #openstack-containers | 00:14 | |
| *** vijendar has joined #openstack-containers | 00:20 | |
| openstackgerrit | Vijendar Komalla proposed openstack/magnum: Support for async bay operations https://review.openstack.org/330771 | 00:20 |
| *** banix has joined #openstack-containers | 00:36 | |
| *** jerrygb has joined #openstack-containers | 00:36 | |
| *** dimtruck is now known as zz_dimtruck | 00:37 | |
| *** mbound has joined #openstack-containers | 00:38 | |
| *** jerrygb has quit IRC | 00:41 | |
| *** vijendar has quit IRC | 00:41 | |
| *** mbound has quit IRC | 00:43 | |
| *** kevin_zh has joined #openstack-containers | 00:46 | |
| *** harshs has quit IRC | 00:47 | |
| *** shu-mutou-AFK is now known as shu-mutou | 00:52 | |
| openstackgerrit | Merged openstack/python-magnumclient: Updated from global requirements https://review.openstack.org/343382 | 00:53 |
| *** trananhkma has joined #openstack-containers | 01:06 | |
| *** hongbin has joined #openstack-containers | 01:10 | |
| *** banix has quit IRC | 01:11 | |
| hongbin | yuanying: ping | 01:13 |
| *** mtanino has joined #openstack-containers | 01:17 | |
| openstackgerrit | Merged openstack/python-magnumclient: Display baymodel info with bay-show command https://review.openstack.org/342025 | 01:17 |
| yuanying | hongbin: pong | 01:19 |
| hongbin | yuanying: Just comment on your patch: https://review.openstack.org/#/c/338742/6/magnumclient/v1/client.py | 01:19 |
| yuanying | hongbin: hmm, ok | 01:20 |
| yuanying | hongbin: but actually,, I don't like this style.. | 01:23 |
| yuanying | Anyway I'll fix | 01:24 |
| *** muralia has quit IRC | 01:24 | |
| *** muralia has joined #openstack-containers | 01:24 | |
| hongbin | yuanying: :) | 01:25 |
| eliqiao | I found a bug of gerrit, there is no patch set 7 on https://review.openstack.org/#/c/338742/8 | 01:26 |
| yuanying | eliqiao: About patch set 7, It seems my push is failed at client side. | 01:28 |
| eliqiao | yuanying: :) haha, good to know | 01:28 |
| yuanying | "error: failed to push some refs to 'ssh://yuanying@review.openstack.org:29418/openstack/python-magnumclient.git'" | 01:28 |
| *** hongbin has quit IRC | 01:33 | |
| *** harlowja has joined #openstack-containers | 01:35 | |
| *** harlowja has quit IRC | 01:35 | |
| *** harlowja has joined #openstack-containers | 01:36 | |
| openstackgerrit | PanFengyun proposed openstack/magnum: Fix get_removal_nodes() of scale_manager.py https://review.openstack.org/343311 | 01:36 |
| *** jerrygb has joined #openstack-containers | 01:37 | |
| *** daneyon has joined #openstack-containers | 01:39 | |
| *** jerrygb has quit IRC | 01:42 | |
| *** daneyon has quit IRC | 01:43 | |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/python-magnumclient: Use os-client-config in shell https://review.openstack.org/338742 | 01:49 |
| *** EricGonczer_ has quit IRC | 02:06 | |
| *** abelur_ has joined #openstack-containers | 02:08 | |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/python-magnumclient: Use os-client-config in shell https://review.openstack.org/338742 | 02:09 |
| *** praneshp has joined #openstack-containers | 02:10 | |
| openstackgerrit | Merged openstack/magnum: Add check on docker_volume_size https://review.openstack.org/344102 | 02:16 |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/python-magnumclient: Use os-client-config in shell https://review.openstack.org/338742 | 02:20 |
| *** jerrygb has joined #openstack-containers | 02:21 | |
| *** zz_dimtruck is now known as dimtruck | 02:21 | |
| *** praneshp has quit IRC | 02:21 | |
| *** jberkus has quit IRC | 02:34 | |
| *** Drago has quit IRC | 02:35 | |
| *** jberkus has joined #openstack-containers | 02:42 | |
| *** praneshp has joined #openstack-containers | 02:42 | |
| *** hongbin has joined #openstack-containers | 02:44 | |
| *** hongbin has quit IRC | 02:44 | |
| *** yamamoto_ has joined #openstack-containers | 02:46 | |
| *** harlowja has quit IRC | 02:46 | |
| *** yuanying has quit IRC | 02:46 | |
| *** anushkrishnamurt has joined #openstack-containers | 02:55 | |
| *** praneshp has quit IRC | 02:55 | |
| *** praneshp has joined #openstack-containers | 02:56 | |
| *** praneshp has quit IRC | 03:00 | |
| *** anushkrishnamurt has quit IRC | 03:19 | |
| *** kushal has joined #openstack-containers | 03:24 | |
| *** dave-mccowan has quit IRC | 03:27 | |
| *** harshs has joined #openstack-containers | 03:27 | |
| *** yamamoto_ has quit IRC | 03:31 | |
| *** julim has quit IRC | 03:31 | |
| *** vimal has joined #openstack-containers | 03:43 | |
| *** julim has joined #openstack-containers | 03:44 | |
| *** yuanying has joined #openstack-containers | 03:48 | |
| *** harshs has quit IRC | 03:57 | |
| *** dimtruck is now known as zz_dimtruck | 04:01 | |
| openstackgerrit | Shu Muto proposed openstack/magnum-ui: Use Horizon's Angular registry of Actions and Views https://review.openstack.org/326289 | 04:01 |
| *** yuanying_ has joined #openstack-containers | 04:03 | |
| *** diga has joined #openstack-containers | 04:04 | |
| *** diga has quit IRC | 04:05 | |
| *** julim has quit IRC | 04:05 | |
| *** diga has joined #openstack-containers | 04:06 | |
| *** yuanying has quit IRC | 04:06 | |
| openstackgerrit | huang.huayong proposed openstack/magnum: fix bug for configure-kubernetes-minion.sh https://review.openstack.org/345148 | 04:07 |
| *** jberkus has quit IRC | 04:15 | |
| *** vimal has quit IRC | 04:22 | |
| *** yamamoto_ has joined #openstack-containers | 04:29 | |
| *** diga has quit IRC | 04:29 | |
| *** kushal has quit IRC | 04:31 | |
| *** vimal has joined #openstack-containers | 04:41 | |
| *** mkrai has joined #openstack-containers | 04:41 | |
| *** kdas__ has joined #openstack-containers | 04:43 | |
| openstackgerrit | Merged openstack/python-magnumclient: Update the magnum client to send the latest version. https://review.openstack.org/345032 | 04:44 |
| openstackgerrit | Merged openstack/python-magnumclient: Add shell command bay-config https://review.openstack.org/342153 | 04:45 |
| *** kdas__ has quit IRC | 04:48 | |
| *** yuanying has joined #openstack-containers | 05:03 | |
| *** yuanying_ has quit IRC | 05:06 | |
| *** jerrygb has quit IRC | 05:09 | |
| *** abelur__ has joined #openstack-containers | 05:20 | |
| *** abelur_ has quit IRC | 05:20 | |
| *** harlowja has joined #openstack-containers | 05:39 | |
| *** abelur__ is now known as abelur_ | 05:57 | |
| *** harlowja has quit IRC | 06:04 | |
| openstackgerrit | Merged openstack/magnum: Fix the permission of these files -rwxr-xr-x https://review.openstack.org/344110 | 06:06 |
| *** daneyon has joined #openstack-containers | 06:09 | |
| *** daneyon has quit IRC | 06:14 | |
| openstackgerrit | Merged openstack/magnum: Set swarm api_address protocol to tcp on all cases. https://review.openstack.org/344835 | 06:23 |
| *** eric_lopez has joined #openstack-containers | 06:35 | |
| *** elo has quit IRC | 06:39 | |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/python-magnumclient: Use os-client-config in shell https://review.openstack.org/338742 | 06:45 |
| *** mfalatic has quit IRC | 06:51 | |
| *** mkrai has quit IRC | 06:57 | |
| *** pcaruana has joined #openstack-containers | 07:03 | |
| *** rcernin has joined #openstack-containers | 07:07 | |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/magnum: Add functional test for k8s ironic https://review.openstack.org/336433 | 07:08 |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/magnum: Fix ironic template https://review.openstack.org/320968 | 07:08 |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/magnum: Fix copying logs from nodes https://review.openstack.org/345197 | 07:08 |
| *** rcernin has quit IRC | 07:10 | |
| *** rcernin has joined #openstack-containers | 07:10 | |
| openstackgerrit | Motohiro/Yuanying Otsuka proposed openstack/magnum: Add functional test for k8s ironic https://review.openstack.org/336433 | 07:13 |
| *** Ich has joined #openstack-containers | 07:15 | |
| *** ccz has joined #openstack-containers | 07:15 | |
| *** zenoway has joined #openstack-containers | 07:17 | |
| *** ccz has quit IRC | 07:19 | |
| *** Ich has quit IRC | 07:19 | |
| *** kushal has joined #openstack-containers | 07:21 | |
| *** hvlad has joined #openstack-containers | 07:24 | |
| openstackgerrit | Vipul Nayyar proposed openstack/python-magnumclient: Increase unit test coverage for module baymodels https://review.openstack.org/344058 | 07:26 |
| *** hvlad has quit IRC | 07:28 | |
| *** hvlad has joined #openstack-containers | 07:28 | |
| strigazi | yuanying: around? | 07:31 |
| yuanying | hi strigazi | 07:32 |
| openstackgerrit | Shu Muto proposed openstack/magnum-ui: Use Horizon's Angular registry of Actions and Views https://review.openstack.org/326289 | 07:32 |
| strigazi | I have issues creating baremetal bays on my local devstack | 07:32 |
| yuanying | Oh, what is? | 07:32 |
| strigazi | What is your local.conf? | 07:32 |
| strigazi | mine is | 07:33 |
| yuanying | wait a minittue | 07:33 |
| yuanying | http://paste.openstack.org/show/538929/ | 07:33 |
| strigazi | mine is: http://paste.openstack.org/show/538930/ | 07:34 |
| strigazi | Apart from the change on the db of ironic have done anything else? | 07:35 |
| strigazi | Apart from the change on the db of ironic have you done anything else? | 07:35 |
| yuanying | I fixed the issue which is copying log from nodes doesn't work | 07:36 |
| yuanying | https://review.openstack.org/#/c/345197/ | 07:36 |
| yuanying | We want to log message from nodes why functional test doesn't work | 07:36 |
| strigazi | ok | 07:37 |
| yuanying | but now, it seems doen't work | 07:37 |
| strigazi | in my case I can't create a bay manually | 07:37 |
| strigazi | the nodes don't spawn completely | 07:37 |
| yuanying | did you check nova scheduler logs? | 07:38 |
| strigazi | at some point nova starts deleting them | 07:38 |
| strigazi | It says that there is no memory, but I do have memory. Here is what happens | 07:39 |
| strigazi | the ironic nodes are on and after deploying are on "wait call" state | 07:39 |
| strigazi | and nova instances are on spawning | 07:40 |
| yuanying | hmm | 07:40 |
| strigazi | after a while (not sure why) nova starts deleting them | 07:40 |
| strigazi | The RamFilter fails | 07:40 |
| strigazi | but I'm on a box | 07:41 |
| strigazi | with 16GB of | 07:41 |
| strigazi | with nothing else but ubuntu server in it | 07:41 |
| yuanying | I have never seen this kind of fail about ironic instance due to RamFilter | 07:41 |
| yuanying | Could you start ironic instance without magnum? | 07:42 |
| yuanying | It means, nova boot --image cirros --flavor baremetal | 07:43 |
| strigazi | just a sec | 07:44 |
| *** mikelk has joined #openstack-containers | 07:48 | |
| strigazi | spawning | 07:49 |
| strigazi | the ironic node is on wait call-back | 07:50 |
| strigazi | and the instance on spawning | 07:50 |
| yuanying | Same as bay case? | 07:52 |
| strigazi | yes | 07:52 |
| strigazi | but it hasn't failed yet | 07:53 |
| yuanying | booting ironic instance time will be long | 07:53 |
| yuanying | What about 'virsh list"? | 07:56 |
| strigazi | runnning | 07:57 |
| yuanying | quiet strange, if instance doesn't run | 07:58 |
| *** daneyon has joined #openstack-containers | 07:58 | |
| yuanying | Suceeded to boot in my local env | 07:58 |
| strigazi | you're devstack is on a psysical host? (mine it is) | 07:59 |
| yuanying | no, my devstack is on kvm vm | 07:59 |
| yuanying | so I'm using nested-kvm | 08:00 |
| yuanying | to build bay | 08:00 |
| strigazi | ram? | 08:00 |
| yuanying | host is 32GB, devstack node is 24GB | 08:01 |
| *** daneyon has quit IRC | 08:02 | |
| strigazi | It's stuck, I have created baremetal insatces in before in much less time | 08:02 |
| strigazi | I'll try to do the ironic exercises on a clean environment | 08:05 |
| strigazi | thanks | 08:05 |
| yuanying | ok | 08:05 |
| jgrassler | Good morning. | 08:07 |
| jgrassler | I made some progress on https://review.openstack.org/#/c/326428/ but now I could use some help: | 08:08 |
| jgrassler | I figured out what the problem is at least - magnum uses its service user credentials to get a token in service.py | 08:08 |
| jgrassler | The only question is why now... | 08:08 |
| jgrassler | Might I be missing something other than https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/service/periodic.py#L97 ? | 08:09 |
| jgrassler | I thought creating an OpenStack client with a context (the trustee user context in this case) is enough. | 08:09 |
| jgrassler | Just to make sure that didn't pose a problem I even did a set_ctx() as the set_context decorator does it: https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/service/periodic.py#L95 | 08:11 |
| jgrassler | But when I pull the token output by https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/service/periodic.py#L105 from Keystone it's a token for the magnum user... | 08:11 |
| yuanying | I guess this token is trusted by trustee to magnum user | 08:13 |
| yuanying | Sorry confuseed | 08:14 |
| jgrassler | That was my first thought, but it definitely is the special trustee user in the trustee user domain: https://github.com/jgrassler/magnum/commit/494ee3332b0af84afb8fc19e86749833ec169f7c#diff-f8634d2ba6ecb1e6f296c48551e65137R102 | 08:15 |
| jgrassler | (I added debug output to verify that as well) | 08:15 |
| yuanying | Who is magnum user? he is trustee_domain_admin user? | 08:16 |
| jgrassler | No, `magnum` is the service user. | 08:18 |
| jgrassler | trustee_domain_admin_name = trustee_domain_admin | 08:18 |
| jgrassler | There's a separate user for the trustee domain admin. | 08:18 |
| yuanying | ok, you create a request context for trustee user, but you got token for magnum user, right? | 08:19 |
| jgrassler | (it's a mostly standard devstack setup, by the way) | 08:19 |
| jgrassler | Exactly, yes | 08:19 |
| eliqiao | yuanying: that seems make sence. | 08:19 |
| jgrassler | The Heat API log shows this very nicely, too: ""GET /v1/cb6fb389ca0143b7929c6b0810a604af/stacks/3ee9789e-4ee9-4f39-95ca-4a89abb72156 HTTP/1.1" 404 839 0.116773" | 08:19 |
| jgrassler | (`cb6fb389ca0143b7929c6b0810a604af` is the `servic` project) | 08:20 |
| jgrassler | s/servic/service/ | 08:20 |
| jgrassler | The correct path would be /v1/6f61a0ae38c34e5c9f4460e4e14468de/stacks/3ee9789e-4ee9-4f39-95ca-4a89abb72156 because 6f61a0ae38c34e5c9f4460e4e14468de is the project of the bay's creating user. | 08:22 |
| eliqiao | jgrassler: so the bocs you constructed is wrong? | 08:24 |
| jgrassler | Also, the context output by https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/service/periodic.py#L102 looks good, i.e. it contains the trustee user's credentials. | 08:24 |
| jgrassler | eliqiao: It must be, yes. I just don't know why at this point :-/ | 08:24 |
| yuanying | jgrassler: may be this is the reason https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/common/keystone.py#L116 | 08:25 |
| jgrassler | eliqiao: It gets bay_context as input, but the client it returns still uses the service user's credentials. | 08:25 |
| yuanying | is self.context.trust_id is exist, keystone client will use admin user (who may be magnum user) | 08:26 |
| jgrassler | yuanying: No, I only added this as an afterthought (to clean up after looping over the bays). | 08:26 |
| eliqiao | you expect the context is an user not serviceļ¼ right? | 08:28 |
| jgrassler | Exactly, yes | 08:28 |
| jgrassler | Something like fb42d352-b109-49e8-91c2-ccb23d4acce3 in this case (that's the trustee user for the bay in question) | 08:28 |
| yuanying | jgrassler: but keystoneclient seems that there is no way to create auth_token from context | 08:30 |
| jgrassler | But why is the admin user being used when there is a trust_id token? | 08:30 |
| jgrassler | yuanying: Ah...so keystoneclient basically ignores the RequestContext being passed? | 08:30 |
| yuanying | jgrassler: simply it uses context.auth_token if it exist | 08:30 |
| yuanying | I guess so | 08:31 |
| jgrassler | Alright, so I'd basically have to invoke KeystoneClientV3 in make_admin_context() to acquire a token? | 08:33 |
| *** kdas__ has joined #openstack-containers | 08:34 | |
| jgrassler | Or, in other words, as long as there's an auth_token in the request context the rest doesn't matter? | 08:34 |
| *** adrian_otto has joined #openstack-containers | 08:34 | |
| yuanying | I think we should add other pass to get auth_token form username and password in keystoneclient | 08:34 |
| *** kushal has quit IRC | 08:35 | |
| jgrassler | https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/common/keystone.py#L115 - here you mean? | 08:37 |
| yuanying | yes | 08:37 |
| *** kdas__ is now known as kushal | 08:37 | |
| *** kushal has joined #openstack-containers | 08:37 | |
| jgrassler | Ok, I can do that :-) | 08:38 |
| jgrassler | One last question on https://github.com/jgrassler/magnum/blob/debug-periodic/magnum/common/keystone.py#L116 | 08:38 |
| jgrassler | Why does this line check for trust_id as well? | 08:38 |
| jgrassler | (The only trust delegations I'm aware of in Magnum are to the bay trustee users) | 08:39 |
| yuanying | jgrassler: Actually I'm not sure, this code come from another project so some spam are remained | 08:39 |
| yuanying | (after our refactoring | 08:39 |
| jgrassler | Ah, ok :-) | 08:39 |
| jgrassler | another project == heat? | 08:40 |
| yuanying | haha, I remember the first copy is from solum | 08:40 |
| jgrassler | I think Heat delegates trusts to its service user, so there it would definitely make sense... | 08:41 |
| *** trananhkma has quit IRC | 08:41 | |
| yuanying | Actually I copied from solum, I remember! | 08:41 |
| yuanying | oh, from ironic... | 08:41 |
| yuanying | https://github.com/jgrassler/magnum/commits/debug-periodic/magnum/common/keystone.py | 08:41 |
| jgrassler | Ok, maybe they copied from Heat in turn :-) | 08:41 |
| jgrassler | Anyway, I know where the problem is and I'll see about fixing it now. Thank you :-) | 08:43 |
| yuanying | np | 08:43 |
| *** adrian_otto has quit IRC | 08:43 | |
| *** manikanta_tadi has joined #openstack-containers | 08:47 | |
| openstackgerrit | Merged openstack/magnum: fix bug for configure-kubernetes-minion.sh https://review.openstack.org/345148 | 08:52 |
| *** clsacramento has quit IRC | 09:05 | |
| openstackgerrit | Wenzhi Yu proposed openstack/magnum: Enable keystone authentication for k8s bay https://review.openstack.org/326924 | 09:13 |
| *** jerrygb has joined #openstack-containers | 09:13 | |
| *** jerrygb has quit IRC | 09:18 | |
| strigazi | yuanying: still around? | 09:20 |
| *** wznoinsk_ has joined #openstack-containers | 09:20 | |
| *** wznoinsk- has joined #openstack-containers | 09:22 | |
| *** wznoinsk has quit IRC | 09:22 | |
| *** wznoinsk has joined #openstack-containers | 09:22 | |
| *** wznoinsk| has joined #openstack-containers | 09:23 | |
| *** wznoinsk- has quit IRC | 09:24 | |
| *** wznoinsk has quit IRC | 09:25 | |
| *** wznoinsk_ has quit IRC | 09:25 | |
| *** wznoinsk has joined #openstack-containers | 09:30 | |
| *** hvlad has quit IRC | 09:31 | |
| *** hvlad has joined #openstack-containers | 09:31 | |
| *** wznoinsk| has quit IRC | 09:32 | |
| *** hvlad has quit IRC | 09:44 | |
| *** hvlad has joined #openstack-containers | 09:44 | |
| *** kushal has quit IRC | 09:46 | |
| *** kushal has joined #openstack-containers | 09:49 | |
| *** jerrygb has joined #openstack-containers | 10:14 | |
| *** wznoinsk_ has joined #openstack-containers | 10:15 | |
| *** wznoinsk has quit IRC | 10:17 | |
| *** jerrygb has quit IRC | 10:20 | |
| *** mbound has joined #openstack-containers | 10:26 | |
| *** vipulnayyar has joined #openstack-containers | 10:34 | |
| *** manikanta_tadi has quit IRC | 10:49 | |
| *** manikanta_tadi has joined #openstack-containers | 10:55 | |
| *** maestropandy has joined #openstack-containers | 10:59 | |
| *** shu-mutou is now known as shu-mutou-AFK | 11:04 | |
| *** jerrygb has joined #openstack-containers | 11:16 | |
| *** yamamoto_ has quit IRC | 11:17 | |
| *** jerrygb has quit IRC | 11:21 | |
| *** sergmelikyan has joined #openstack-containers | 11:23 | |
| *** dave-mccowan has joined #openstack-containers | 11:23 | |
| *** dfflanders has joined #openstack-containers | 11:53 | |
| *** yamamoto has joined #openstack-containers | 11:54 | |
| *** vijendar has joined #openstack-containers | 11:59 | |
| *** yamamoto has quit IRC | 12:03 | |
| *** yamamoto has joined #openstack-containers | 12:03 | |
| *** vijendar has quit IRC | 12:04 | |
| *** yamamoto has quit IRC | 12:08 | |
| *** yamamoto has joined #openstack-containers | 12:09 | |
| *** jerrygb has joined #openstack-containers | 12:17 | |
| *** jerrygb has quit IRC | 12:21 | |
| *** EricGonczer_ has joined #openstack-containers | 12:22 | |
| *** eric_lopez has quit IRC | 12:24 | |
| *** eric_lopez has joined #openstack-containers | 12:24 | |
| *** hvlad has quit IRC | 12:25 | |
| *** hvlad has joined #openstack-containers | 12:25 | |
| *** banix has joined #openstack-containers | 12:28 | |
| *** banix has quit IRC | 12:33 | |
| *** banix has joined #openstack-containers | 12:34 | |
| *** maestropandy has left #openstack-containers | 12:37 | |
| *** banix has quit IRC | 12:37 | |
| *** eliqiao has quit IRC | 12:38 | |
| *** vimal has quit IRC | 12:39 | |
| *** vipulnayyar has quit IRC | 12:41 | |
| *** sergmelikyan has quit IRC | 12:47 | |
| *** sergmelikyan has joined #openstack-containers | 12:47 | |
| *** sergmelikyan has quit IRC | 12:52 | |
| *** nicolasbock has quit IRC | 12:52 | |
| *** hvlad has quit IRC | 12:52 | |
| *** hvlad has joined #openstack-containers | 12:52 | |
| *** nicolasbock has joined #openstack-containers | 12:52 | |
| *** julim has joined #openstack-containers | 13:01 | |
| *** manikanta_tadi has quit IRC | 13:02 | |
| openstackgerrit | Johannes Grassler proposed openstack/magnum: Fix global stack list in periodic task https://review.openstack.org/326428 | 13:11 |
| jgrassler | This should finally do it... | 13:12 |
| jgrassler | Modulo pep8 maybe :-) | 13:12 |
| *** rlrossit has joined #openstack-containers | 13:17 | |
| *** jerrygb has joined #openstack-containers | 13:18 | |
| *** jerrygb has quit IRC | 13:22 | |
| *** yamamoto has quit IRC | 13:25 | |
| *** EricGonczer_ has quit IRC | 13:28 | |
| *** jerrygb has joined #openstack-containers | 13:34 | |
| *** harshs has joined #openstack-containers | 13:37 | |
| *** mikelk has quit IRC | 13:38 | |
| *** mikelk has joined #openstack-containers | 13:38 | |
| *** yamamoto has joined #openstack-containers | 13:43 | |
| *** yamamoto has quit IRC | 13:43 | |
| *** yamamoto has joined #openstack-containers | 13:44 | |
| *** ametts has joined #openstack-containers | 13:47 | |
| *** krotscheck_dcm is now known as krotscheck | 13:50 | |
| *** mtanino has quit IRC | 13:57 | |
| *** anush_ has joined #openstack-containers | 14:02 | |
| *** zz_dimtruck is now known as dimtruck | 14:04 | |
| *** harshs has quit IRC | 14:08 | |
| *** vijendar has joined #openstack-containers | 14:12 | |
| *** vijendar has quit IRC | 14:13 | |
| *** harshs has joined #openstack-containers | 14:15 | |
| *** muralia has quit IRC | 14:28 | |
| *** muralia has joined #openstack-containers | 14:29 | |
| *** muralia has quit IRC | 14:33 | |
| *** dane_leblanc has joined #openstack-containers | 14:38 | |
| *** Drago has joined #openstack-containers | 14:44 | |
| *** Drago has quit IRC | 14:44 | |
| *** Drago has joined #openstack-containers | 14:44 | |
| *** vijendar has joined #openstack-containers | 14:50 | |
| *** mtanino has joined #openstack-containers | 14:54 | |
| *** swatson has joined #openstack-containers | 14:58 | |
| *** pcaruana has quit IRC | 15:02 | |
| *** muralia has joined #openstack-containers | 15:05 | |
| *** daneyon has joined #openstack-containers | 15:11 | |
| *** sdake has joined #openstack-containers | 15:13 | |
| *** hongbin has joined #openstack-containers | 15:14 | |
| *** harshs has quit IRC | 15:15 | |
| *** daneyon has quit IRC | 15:16 | |
| *** kevin_zh has quit IRC | 15:16 | |
| *** sdake_ has joined #openstack-containers | 15:17 | |
| *** dfflanders has quit IRC | 15:17 | |
| *** sdake has quit IRC | 15:19 | |
| *** Conficker has joined #openstack-containers | 15:20 | |
| *** zenoway has quit IRC | 15:37 | |
| *** mikelk has quit IRC | 15:41 | |
| *** zenoway has joined #openstack-containers | 15:41 | |
| *** tango__ has joined #openstack-containers | 15:43 | |
| *** dimtruck is now known as zz_dimtruck | 15:44 | |
| *** slagle has quit IRC | 15:45 | |
| *** zenoway has quit IRC | 15:46 | |
| *** slagle has joined #openstack-containers | 15:47 | |
| openstackgerrit | Johannes Grassler proposed openstack/magnum: Fix global stack list in periodic task https://review.openstack.org/326428 | 15:50 |
| *** jberkus has joined #openstack-containers | 15:53 | |
| *** hvlad1 has joined #openstack-containers | 15:53 | |
| *** praneshp has joined #openstack-containers | 15:55 | |
| *** hvlad has quit IRC | 15:55 | |
| *** yamamoto has quit IRC | 15:55 | |
| *** sergmelikyan has joined #openstack-containers | 16:09 | |
| *** rcernin has quit IRC | 16:19 | |
| *** vijendar has quit IRC | 16:22 | |
| *** julim has quit IRC | 16:28 | |
| *** derekjhyang has joined #openstack-containers | 16:29 | |
| openstackgerrit | Bharath kumar proposed openstack/magnum: The directory need not be /opt/stack/devstack always https://review.openstack.org/345538 | 16:29 |
| *** Conficker has quit IRC | 16:30 | |
| *** bks has joined #openstack-containers | 16:34 | |
| bks | hi | 16:34 |
| bks | https://review.openstack.org/#/c/345538/ | 16:35 |
| *** sdake_ has quit IRC | 16:46 | |
| *** sdake has joined #openstack-containers | 16:49 | |
| *** Drago has quit IRC | 16:51 | |
| *** yamamoto has joined #openstack-containers | 16:56 | |
| *** EricGonczer_ has joined #openstack-containers | 16:56 | |
| *** yamamoto has quit IRC | 17:01 | |
| *** rafael__ has joined #openstack-containers | 17:03 | |
| *** swatson has quit IRC | 17:03 | |
| *** bks has quit IRC | 17:07 | |
| *** hvlad1 has quit IRC | 17:09 | |
| *** eric_lopez has quit IRC | 17:10 | |
| *** sergmelikyan has quit IRC | 17:11 | |
| openstackgerrit | Nate Potter proposed openstack/magnum: Fix typo in baymodel param https://review.openstack.org/345561 | 17:12 |
| *** chetna has joined #openstack-containers | 17:13 | |
| *** charliekang has joined #openstack-containers | 17:20 | |
| *** zul has quit IRC | 17:25 | |
| *** zul has joined #openstack-containers | 17:26 | |
| *** julim has joined #openstack-containers | 17:34 | |
| *** EricGonczer_ has quit IRC | 17:41 | |
| *** zz_dimtruck is now known as dimtruck | 17:43 | |
| *** harlowja has joined #openstack-containers | 17:44 | |
| *** vijendar has joined #openstack-containers | 17:45 | |
| *** vijendar has quit IRC | 17:45 | |
| *** sidx64 has joined #openstack-containers | 17:50 | |
| *** elo has joined #openstack-containers | 17:51 | |
| *** elo has quit IRC | 17:52 | |
| *** elo has joined #openstack-containers | 17:54 | |
| *** elo has quit IRC | 17:56 | |
| *** vijendar has joined #openstack-containers | 17:58 | |
| *** sergmelikyan has joined #openstack-containers | 17:59 | |
| *** julim_ has joined #openstack-containers | 18:01 | |
| openstackgerrit | rajiv proposed openstack/magnum: Change heat template version for consistency https://review.openstack.org/345587 | 18:01 |
| *** julim has quit IRC | 18:04 | |
| *** EricGonczer_ has joined #openstack-containers | 18:11 | |
| openstackgerrit | Vijendar Komalla proposed openstack/magnum: Support for async bay operations https://review.openstack.org/330771 | 18:15 |
| openstackgerrit | Vijendar Komalla proposed openstack/magnum: Add microversioning support for methods https://review.openstack.org/343060 | 18:16 |
| *** hvlad has joined #openstack-containers | 18:18 | |
| *** Drago has joined #openstack-containers | 18:19 | |
| *** harshs has joined #openstack-containers | 18:20 | |
| *** EricGonczer_ has quit IRC | 18:21 | |
| *** zul_ has joined #openstack-containers | 18:24 | |
| *** swatson has joined #openstack-containers | 18:27 | |
| *** zul has quit IRC | 18:29 | |
| openstackgerrit | Vijendar Komalla proposed openstack/magnum: Support for async bay operations https://review.openstack.org/330771 | 18:30 |
| *** harlowja has quit IRC | 18:35 | |
| *** harlowja has joined #openstack-containers | 18:38 | |
| *** EricGonczer_ has joined #openstack-containers | 18:41 | |
| *** daneyon has joined #openstack-containers | 18:47 | |
| *** derekjhyang has quit IRC | 18:52 | |
| *** daneyon has quit IRC | 18:53 | |
| *** sergmelikyan has quit IRC | 18:56 | |
| *** sergmelikyan has joined #openstack-containers | 18:56 | |
| *** harlowja has quit IRC | 18:59 | |
| *** EricGonczer_ has quit IRC | 19:00 | |
| *** sergmelikyan has quit IRC | 19:01 | |
| *** sergmelikyan has joined #openstack-containers | 19:02 | |
| *** harshs has quit IRC | 19:05 | |
| *** banix has joined #openstack-containers | 19:06 | |
| *** EricGonczer_ has joined #openstack-containers | 19:06 | |
| *** banix has quit IRC | 19:07 | |
| *** sidx64 has quit IRC | 19:07 | |
| *** EricGonczer_ has quit IRC | 19:07 | |
| *** kevin_zh has joined #openstack-containers | 19:07 | |
| *** dfflanders has joined #openstack-containers | 19:11 | |
| *** sergmelikyan has quit IRC | 19:12 | |
| *** harshs has joined #openstack-containers | 19:14 | |
| *** harshs has quit IRC | 19:15 | |
| *** abelur_ has quit IRC | 19:28 | |
| *** hvlad has quit IRC | 19:28 | |
| *** hvlad has joined #openstack-containers | 19:28 | |
| *** harshs has joined #openstack-containers | 19:29 | |
| *** swatson has quit IRC | 19:32 | |
| *** jerrygb has quit IRC | 19:32 | |
| *** ramishra has quit IRC | 19:36 | |
| *** ramishra has joined #openstack-containers | 19:37 | |
| *** ByPasS has joined #openstack-containers | 19:37 | |
| *** sdake_ has joined #openstack-containers | 19:39 | |
| *** sdake has quit IRC | 19:40 | |
| *** jerrygb has joined #openstack-containers | 19:40 | |
| *** swatson has joined #openstack-containers | 19:53 | |
| jvgrant_ | Drago: you around? | 20:00 |
| Drago | jvgrant_: Yes, in a meeting for 30 minutes though. Responses might be slow. What's up? | 20:01 |
| jvgrant_ | Drago: was looking into updating the images for software deployment but I noticed in the user guide doc that their inclusion is already required. Do we know if they are already included for all the images? | 20:03 |
| Drago | jvgrant_: They are included on the ubuntu mesos image, but they're not on the atomic fedora image. | 20:03 |
| *** vijendar has quit IRC | 20:09 | |
| *** dfflanders has quit IRC | 20:21 | |
| *** agireud has quit IRC | 20:30 | |
| *** agireud has joined #openstack-containers | 20:31 | |
| *** EricGonczer_ has joined #openstack-containers | 20:35 | |
| *** hvlad has quit IRC | 20:36 | |
| *** hvlad has joined #openstack-containers | 20:36 | |
| *** harlowja has joined #openstack-containers | 20:39 | |
| *** vijendar has joined #openstack-containers | 20:51 | |
| *** muralia has quit IRC | 20:54 | |
| *** ametts has quit IRC | 20:59 | |
| chetna | Hi Drago.. Finally, I was able to build Fedora23 atomic image on ubuntu 14.04 though it was stuck at the same point on fedora. Thanks much for your help! | 21:01 |
| *** swatson has quit IRC | 21:02 | |
| *** hvlad has quit IRC | 21:06 | |
| *** hvlad has joined #openstack-containers | 21:06 | |
| Drago | chetna: Excellent! You're welcome | 21:07 |
| *** mikal has left #openstack-containers | 21:08 | |
| chetna | Thanks tango for your help.. I could build Fedora23 atomic image on ubuntu 14.04. | 21:10 |
| chetna | I will be trying to build Fedora24 one now. | 21:10 |
| *** elo has joined #openstack-containers | 21:11 | |
| tango__ | chetna: Sounds good. If there is any new steps you found, please consider updating the docs. | 21:13 |
| *** julim_ has quit IRC | 21:14 | |
| chetna | sure tango.. | 21:14 |
| *** banix has joined #openstack-containers | 21:20 | |
| *** EricGonczer_ has quit IRC | 21:20 | |
| Drago | jvgrant_: Ping | 21:20 |
| *** vijendar has quit IRC | 21:21 | |
| jvgrant_ | Drago: hi | 21:21 |
| Drago | jvgrant_: Apparently there's an atomic-specific swconfig agent: https://github.com/openstack/heat-templates/blob/master/hot/software-config/heat-container-agent/scripts/hooks/atomic | 21:22 |
| Drago | I haven't looked into it, but I happened to see a conversation mentioning atomic and software config | 21:22 |
| jvgrant_ | ok, it also looks like the k8 fedora-ironic image builds with it as well | 21:23 |
| Drago | jvgrant_: It does?? | 21:24 |
| jvgrant_ | at least according to the readme in the driver | 21:24 |
| jvgrant_ | https://github.com/openstack/magnum/tree/master/magnum/drivers/k8s_fedora_atomic_v1/templates/elements | 21:24 |
| jvgrant_ | i'm building a new fedora atomic image right now for vm vs ironic that will include it as well | 21:26 |
| Drago | jvgrant_: What is "baremetal"? | 21:27 |
| jvgrant_ | running directly on the system instead of a VM | 21:28 |
| Drago | jvgrant_: Oh no I meant in "diskimage-builder/bin/disk-image-create baremetal \ | 21:30 |
| Drago | fedora selinux-permissive \ ..." | 21:30 |
| *** daneyon has joined #openstack-containers | 21:30 | |
| *** charliekang has quit IRC | 21:31 | |
| jvgrant_ | same, i'm assuming since in the other commands for disk-builder i see that param as vm | 21:32 |
| *** daneyon has quit IRC | 21:35 | |
| Drago | jvgrant_: I don't know what I was confused about. Nevermind | 21:35 |
| jvgrant_ | ok :) | 21:35 |
| *** dave-mccowan has quit IRC | 21:37 | |
| Drago | jvgrant_: Yeah, okay. I was using this https://github.com/openstack/magnum/blob/master/magnum/drivers/common/image/fedora-atomic/README.rst | 21:39 |
| Drago | And it wasn't working with the software config part | 21:40 |
| Drago | I am now unsure if that's how the atomic-fedora-latest image is created | 21:41 |
| *** harlowja has quit IRC | 21:41 | |
| *** swatson has joined #openstack-containers | 21:42 | |
| *** rafael__ has quit IRC | 21:43 | |
| *** swatson has quit IRC | 21:44 | |
| jvgrant_ | i'm using that as well as what i found from the k8 driver area. I also found another guide on how to use it. trying a little of each for now | 21:45 |
| jvgrant_ | were you not able to create the image or the software config part wasn't working? | 21:45 |
| Drago | jvgrant_: The image creation would fail | 21:45 |
| Drago | And I thought, cool, let's see how to debug this, and that's where I left off | 21:46 |
| jvgrant_ | i have an image built now, not sure if is going to work yet though | 21:46 |
| Drago | jvgrant_: A baremetal or vm image? | 21:46 |
| jvgrant_ | vm | 21:46 |
| Drago | cool | 21:47 |
| jvgrant_ | what is the best way to figure out if it is working? | 21:47 |
| Drago | If DIB ran correctly, that's a pretty good sign | 21:47 |
| Drago | SSH into the vm and run os-collect-config | 21:47 |
| Drago | jvgrant_: Also if you try the template I sent you and it completes, then it's working | 21:48 |
| *** harlowja has joined #openstack-containers | 21:48 | |
| jvgrant_ | ok i'll give that a try | 21:48 |
| Drago | jvgrant_: I think you will only have to update the ssh key parameter and the image parameter | 21:48 |
| *** harlowja has quit IRC | 21:48 | |
| *** muralia has joined #openstack-containers | 21:54 | |
| chetna | Hi tango.. I have a question.. The other day when we were discussing about building Fedora atomic images. You mentioned that DIB does not support Fedora24 yet and I might be able to tweak the patch that adds Fedora23 to DIB for it to work for fedora24. Could you please help me with the patch you were talking about so that I can try to tweak it. | 21:55 |
| *** banix has quit IRC | 21:57 | |
| daneyon_ | i am testing a magnum stable/mitaka deployment (not devstack). I am seeing different behavior when using the atomic-latest and atomic 6 image (https://fedorapeople.org/groups/magnum/). What is the recommended image to use with magnum stable/mitaka? | 22:00 |
| daneyon_ | ^ yuanying tango__ hongbin ^ | 22:03 |
| tango__ | daneyon_: I would use atomic_latest. | 22:05 |
| hongbin | daneyon_: tango__ I think it is atomic-5 https://github.com/openstack/magnum/blob/stable/mitaka/doc/source/dev/quickstart.rst#building-a-kubernetes-bay---based-on-fedora-atomic | 22:06 |
| hongbin | Not latest | 22:07 |
| tango__ | hmm, I thought we have switched to latest for mitaka. I guess I am mistaken. | 22:08 |
| daneyon_ | tango__ i have tried atomic-latest and atomic-6 but had problems, the kubelet is trying to register using $KUBE_API_VIP:6443 (api secure port) and it fails b/c the neutron lb is not config'd for 6443 | 22:08 |
| daneyon_ | Let me try the v5 image | 22:08 |
| daneyon_ | hongbin tango__ do you know off hand the diff between v5/v6 atomic images? | 22:10 |
| hongbin | I have no idea about 6 | 22:12 |
| *** rlrossit has quit IRC | 22:12 | |
| hongbin | 5 is based on fedora21 if I remembered correctly | 22:12 |
| tango__ | I had a README file there to explain but somehow it got deleted. | 22:13 |
| tango__ | Let me check my note | 22:13 |
| daneyon_ | hongbin tango__ One other thing I noticed in my mitaka non-devstack deployment is related to neutron secgroups. When the neutron lb is created for kube-api/etcd, the neutron lb port is associated to the neutron default security group. This group buy default does not allow 2379/8080. Have you seen this? | 22:16 |
| *** sergmelikyan has joined #openstack-containers | 22:18 | |
| tango__ | daneyon_: atomic-6 is the same as atomic-5, except kube-apiserver is custom built to fix 3 bugs for the external load balancer feature | 22:18 |
| hongbin | but the template will create a dedicated secgroup for api&etcd | 22:18 |
| hongbin | The secgroup created from the teamplte should open the right set of ports | 22:18 |
| daneyon_ | hongbin, that is correct, but i see the neutron lb port is assocaited to the neutron default sec group instead of the kube master sec group. The default sec group does not allow 2370/8080, so this traffic gets blocked st the lb. let me compare to my devstack setup | 22:20 |
| hongbin | Not sure why lb asscociate with the default secgroup | 22:22 |
| hongbin | It is very strange | 22:22 |
| *** sdake_ has quit IRC | 22:24 | |
| Drago | jvgrant_: Please review: https://blueprints.launchpad.net/magnum/+spec/bay-rebuild-operation | 22:29 |
| jvgrant_ | drago: will do | 22:29 |
| *** dimtruck is now known as zz_dimtruck | 22:30 | |
| *** dane_leblanc has quit IRC | 22:35 | |
| *** shubhams has quit IRC | 22:43 | |
| *** sergmelikyan has quit IRC | 22:44 | |
| *** sdake has joined #openstack-containers | 22:49 | |
| *** banix has joined #openstack-containers | 22:52 | |
| *** jerrygb has quit IRC | 22:55 | |
| daneyon_ | hongbin tango__ fyi.. these are the 2 bugs I'm hitting so far in my magnum deploy: https://bugs.launchpad.net/magnum/+bug/1605416 https://bugs.launchpad.net/magnum/+bug/1605415 | 23:04 |
| openstack | Launchpad bug 1605416 in Magnum "neutron loadbalancer uses incorrect security group" [Undecided,New] | 23:04 |
| openstack | Launchpad bug 1605415 in Magnum "Kubelet failure using Port 6443" [Undecided,New] | 23:04 |
| tango__ | daneyon_: Are these againts Mitaka stable only, or current master also? | 23:06 |
| tango__ | I guess we can check | 23:06 |
| daneyon_ | tango__ I am using stable/mitaka | 23:06 |
| *** hvlad has quit IRC | 23:06 | |
| *** hvlad has joined #openstack-containers | 23:06 | |
| hongbin | daneyon_: The second bug is related to the first bug? | 23:10 |
| hongbin | daneyon_: or they are independent issues | 23:11 |
| *** achanda has joined #openstack-containers | 23:11 | |
| achanda | trying to see if there is a way to set MTU on the docker0 bridge using magnum | 23:12 |
| achanda | looks like FLANNEL_MTU will set it | 23:12 |
| daneyon_ | hongbin independent issues | 23:12 |
| hongbin | daneyon_: ack | 23:12 |
| achanda | but how do I set that var using the CLI? | 23:12 |
| tango__ | achanda: Why do you need to set the MTU? This can lead to very strange network behavior. | 23:14 |
| achanda | tango__ if path mtu discovery is broken for some reason, not setting this can lead to weirdness | 23:14 |
| achanda | we had this in one of or envs | 23:15 |
| *** hongbin has quit IRC | 23:16 | |
| achanda | in our case, a firewall was dropping silently icmp replies | 23:16 |
| achanda | it was easier to change mtu on docker0 to get it to work | 23:17 |
| tango__ | achanda: I guess you are using K8s bay. Currently MTU is not exposed as a parameter. It needs to be set correctly to account for the extra header in Flannel. | 23:18 |
| *** daneyon has joined #openstack-containers | 23:18 | |
| achanda | yes, k8s | 23:19 |
| daneyon_ | hongbin https://bugs.launchpad.net/magnum/+bug/1605415 may be invalid. AT first I thought the neutron lb was load balancing 8080 and not 6443, but that does not appear to be the case. | 23:19 |
| openstack | Launchpad bug 1605415 in Magnum "Kubelet failure using Port 6443" [Undecided,New] | 23:19 |
| achanda | tango__: how do I set it though? | 23:19 |
| tango__ | achanda: To set the MTU, you would need to log into each VM and run ifconfig command on docker0 | 23:19 |
| achanda | tango__ :( | 23:19 |
| achanda | there is no way to set that FLANNEL_MTU flag? | 23:20 |
| tango__ | achanda: I guess if you have a clear use case, you can open a blueprint with all the details and we can discuss | 23:21 |
| achanda | tango__: will do, thanks | 23:21 |
| tango__ | OK | 23:21 |
| *** daneyon has quit IRC | 23:23 | |
| *** harshs has quit IRC | 23:37 | |
| *** hvlad has quit IRC | 23:38 | |
| *** harshs has joined #openstack-containers | 23:41 | |
| *** sergmelikyan has joined #openstack-containers | 23:45 | |
| *** sergmelikyan has quit IRC | 23:49 | |
| *** sergmelikyan has joined #openstack-containers | 23:50 | |
| *** anush_ has quit IRC | 23:57 | |
| *** harshs has quit IRC | 23:57 | |
| *** harshs has joined #openstack-containers | 23:57 | |
| *** sergmelikyan has quit IRC | 23:58 | |
| *** harshs has quit IRC | 23:58 | |
| *** sergmelikyan has joined #openstack-containers | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!