| *** anniec has quit IRC | 00:03 | |
| *** matsuhashi has joined #openstack-ironic | 00:18 | |
| *** nosnos has joined #openstack-ironic | 00:34 | |
| *** datajerk has quit IRC | 00:35 | |
| *** datajerk has joined #openstack-ironic | 00:38 | |
| *** anniec has joined #openstack-ironic | 00:41 | |
| *** lnxnut has joined #openstack-ironic | 01:20 | |
| *** anteaya has quit IRC | 01:43 | |
| *** anniec has quit IRC | 01:51 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/python-ironicclient: Support port-list by node_id https://review.openstack.org/54065 | 01:56 |
|---|---|---|
| *** rloo has quit IRC | 02:01 | |
| *** anniec has joined #openstack-ironic | 02:05 | |
| *** jianingy has joined #openstack-ironic | 02:12 | |
| *** anniec has quit IRC | 02:34 | |
| *** anniec has joined #openstack-ironic | 02:35 | |
| *** anniec has quit IRC | 02:48 | |
| *** anniec has joined #openstack-ironic | 03:11 | |
| *** matsuhashi has quit IRC | 03:20 | |
| *** matsuhashi has joined #openstack-ironic | 03:21 | |
| *** matsuhashi has quit IRC | 03:25 | |
| *** anniec has quit IRC | 03:30 | |
| *** lnxnut has quit IRC | 03:43 | |
| *** lnxnut has joined #openstack-ironic | 03:43 | |
| *** lnxnut has quit IRC | 03:48 | |
| *** dguitarbite has joined #openstack-ironic | 04:24 | |
| *** dguitarbite has quit IRC | 04:27 | |
| *** dguitarbite has joined #openstack-ironic | 04:28 | |
| *** openstackgerrit has quit IRC | 04:36 | |
| *** openstackgerrit has joined #openstack-ironic | 04:36 | |
| *** matsuhashi has joined #openstack-ironic | 04:42 | |
| *** matsuhashi has quit IRC | 05:02 | |
| *** matsuhashi has joined #openstack-ironic | 05:03 | |
| *** matsuhashi has quit IRC | 05:08 | |
| *** rameshg87 has quit IRC | 05:53 | |
| *** sjing has joined #openstack-ironic | 06:22 | |
| *** r-mibu has quit IRC | 06:28 | |
| *** GheRivero has quit IRC | 06:28 | |
| *** GheRivero has joined #openstack-ironic | 06:30 | |
| *** r-mibu has joined #openstack-ironic | 06:33 | |
| *** prekarat has joined #openstack-ironic | 06:38 | |
| *** prekarat has quit IRC | 06:42 | |
| *** prekarat has joined #openstack-ironic | 06:56 | |
| *** bauzas has joined #openstack-ironic | 07:39 | |
| *** martyntaylor has joined #openstack-ironic | 07:52 | |
| *** martyntaylor has left #openstack-ironic | 07:52 | |
| *** nick_long has joined #openstack-ironic | 08:04 | |
| *** nick_long has quit IRC | 08:04 | |
| *** ndipanov has joined #openstack-ironic | 08:04 | |
| *** jianingy_ has joined #openstack-ironic | 08:22 | |
| *** jianingy has quit IRC | 08:24 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/python-ironicclient: Enable created_at/updated_at for port-show/chassis-show https://review.openstack.org/53991 | 08:25 |
| *** romcheg has joined #openstack-ironic | 08:27 | |
| *** nosnos_ has joined #openstack-ironic | 08:29 | |
| *** nosnos has quit IRC | 08:30 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/python-ironicclient: Support port-list by node_id https://review.openstack.org/54065 | 08:51 |
| *** jistr has joined #openstack-ironic | 08:57 | |
| *** nosnos has joined #openstack-ironic | 09:02 | |
| *** nosnos_ has quit IRC | 09:03 | |
| *** prekarat has quit IRC | 09:04 | |
| *** derekh has joined #openstack-ironic | 09:09 | |
| *** prekarat has joined #openstack-ironic | 09:12 | |
| *** sjing has quit IRC | 09:18 | |
| *** lucasagomes has joined #openstack-ironic | 09:24 | |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Add wrapper for keystone service catalog https://review.openstack.org/52601 | 09:25 |
| openstackgerrit | A change was merged to openstack/ironic: Add mailmap file https://review.openstack.org/54090 | 09:33 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Add wrapper for keystone service catalog https://review.openstack.org/52601 | 09:49 |
| *** epim has joined #openstack-ironic | 09:52 | |
| *** epim has quit IRC | 09:56 | |
| *** prekarat has quit IRC | 09:59 | |
| *** jianingy has joined #openstack-ironic | 10:06 | |
| *** jianingy_ has quit IRC | 10:08 | |
| *** prekarat has joined #openstack-ironic | 10:12 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Register API options under the 'api' group https://review.openstack.org/54277 | 10:29 |
| *** nosnos has quit IRC | 10:31 | |
| *** nosnos has joined #openstack-ironic | 10:32 | |
| *** prekarat has quit IRC | 10:35 | |
| *** nosnos has quit IRC | 10:36 | |
| *** prekarat has joined #openstack-ironic | 10:38 | |
| *** prekarat has quit IRC | 11:11 | |
| *** anniec has joined #openstack-ironic | 11:32 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Expose /drivers on the API https://review.openstack.org/53157 | 11:35 |
| *** romcheg1 has joined #openstack-ironic | 11:36 | |
| *** romcheg has quit IRC | 11:41 | |
| *** anniec has quit IRC | 11:45 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/python-ironicclient: Enable created_at/updated_at for port-show/chassis-show https://review.openstack.org/53991 | 12:20 |
| *** jdob has joined #openstack-ironic | 12:23 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Required fields on nodes https://review.openstack.org/53664 | 12:23 |
| *** lucasagomes is now known as lucas-hungry | 12:27 | |
| *** romcheg has joined #openstack-ironic | 12:29 | |
| *** jbjohnso has quit IRC | 12:31 | |
| *** romcheg1 has quit IRC | 12:33 | |
| *** linggao has joined #openstack-ironic | 12:57 | |
| *** jdob has quit IRC | 12:57 | |
| *** jdob has joined #openstack-ironic | 12:57 | |
| *** romcheg has left #openstack-ironic | 12:59 | |
| *** anteaya has joined #openstack-ironic | 13:09 | |
| *** romcheg has joined #openstack-ironic | 13:10 | |
| *** jdob has quit IRC | 13:11 | |
| *** jdob has joined #openstack-ironic | 13:12 | |
| *** jbjohnso has joined #openstack-ironic | 13:28 | |
| *** rloo has joined #openstack-ironic | 13:34 | |
| *** dguitarbite has left #openstack-ironic | 13:52 | |
| *** jianingy_ has joined #openstack-ironic | 13:57 | |
| *** jianingy_ is now known as jianingy_afk | 13:58 | |
| *** lucas-hungry is now known as lucasagomes | 13:59 | |
| *** jianingy has quit IRC | 14:00 | |
| NobodyCam | good mornining iRonic | 14:28 |
| *** lnxnut has joined #openstack-ironic | 14:30 | |
| *** blamar has quit IRC | 14:30 | |
| *** blamar has joined #openstack-ironic | 14:34 | |
| lucasagomes | NobodyCam, morning | 14:35 |
| NobodyCam | morning lucasagomes :) how was your holiday | 14:35 |
| lucasagomes | NobodyCam, ah was great! I mostly tried to sorted out the things for HK | 14:36 |
| NobodyCam | :) I have to do that | 14:36 |
| lucasagomes | like buying a bag cause my last trip to seattle they broke my wheels on the airport | 14:37 |
| lucasagomes | yea | 14:37 |
| lucasagomes | so was productive I'd say :) | 14:37 |
| lucasagomes | how was ur weekend? | 14:37 |
| max_lobur | morning Ironic! :) | 14:38 |
| NobodyCam | was good. | 14:38 |
| lucasagomes | morning max_lobur :) | 14:44 |
| lucasagomes | NobodyCam, nice! | 14:44 |
| NobodyCam | mornig max_lobur | 14:45 |
| NobodyCam | hp has a 1 hour call at 8 so I may be slow to respond | 14:46 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Register API options under the 'api' group https://review.openstack.org/54277 | 14:49 |
| devananda | morning, all | 14:54 |
| romcheg | Morning devananda | 14:55 |
| NobodyCam | morning devananda | 14:55 |
| NobodyCam | devananda: you doing the hp call? | 14:56 |
| devananda | ya | 14:56 |
| NobodyCam | morning romcheg :) | 14:56 |
| romcheg | Morning NobodyCam and everyone else | 14:56 |
| lucasagomes | morning devananda romcheg | 14:56 |
| NobodyCam | :) | 14:56 |
| *** openstackgerrit has quit IRC | 15:00 | |
| *** openstackgerrit has joined #openstack-ironic | 15:01 | |
| openstackgerrit | A change was merged to openstack/ironic: Implement /nodes/UUID/vendor_passthru in the API https://review.openstack.org/53761 | 15:03 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Expose /drivers on the API https://review.openstack.org/53157 | 15:28 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Required fields on nodes https://review.openstack.org/53664 | 15:32 |
| openstackgerrit | A change was merged to openstack/ironic: Add context to TaskManager https://review.openstack.org/53145 | 15:32 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Fix test node manager https://review.openstack.org/53877 | 15:33 |
| openstackgerrit | A change was merged to openstack/ironic: Update mailmap for Joe Gordon https://review.openstack.org/54091 | 15:34 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Fix test node manager https://review.openstack.org/53877 | 15:35 |
| *** yuriyz has joined #openstack-ironic | 15:39 | |
| openstackgerrit | A change was merged to openstack/ironic: Expose /drivers on the API https://review.openstack.org/53157 | 15:40 |
| NobodyCam | call over | 15:45 |
| openstackgerrit | Kevin McDonald proposed a change to openstack/python-ironicclient: Fixes Auth Token being sent as lambda function https://review.openstack.org/53963 | 15:47 |
| *** jianingy_afk is now known as jianingy_ | 16:26 | |
| lucasagomes | NobodyCam, devananda any thoughts on squashing things into one like: https://review.openstack.org/#/c/53683/1/ironicclient/tests/v1/test_driver.py | 16:27 |
| lucasagomes | the fixtures one I agree, the second one I could argue both sides... | 16:28 |
| lucasagomes | I personally like the current style, but I also think that less lines of code is important | 16:29 |
| *** jianingy_ is now known as jianingy_afk | 16:29 | |
| NobodyCam | I like that I can follow the fixture quite easly in its current form | 16:29 |
| NobodyCam | oh bbt and make mo coffee brb | 16:31 |
| *** romcheg has quit IRC | 16:32 | |
| *** rloo has quit IRC | 16:38 | |
| linggao | Hi everyone, one question, can we deploy both vm and baremetal nodes in the same cloud? I thought there was restriction before, not sure about havana. | 16:38 |
| NobodyCam | current compute nodes can not do both. | 16:40 |
| NobodyCam | but youcould have >1 compute | 16:40 |
| NobodyCam | node | 16:40 |
| openstackgerrit | Kevin McDonald proposed a change to openstack/python-ironicclient: Fixes Auth Token being sent as lambda function https://review.openstack.org/53963 | 16:40 |
| linggao | Hi NobodyCam, by compute node do you mean the controller? | 16:41 |
| *** yuriyz has quit IRC | 16:41 | |
| NobodyCam | nova compute node | 16:41 |
| openstackgerrit | Kevin McDonald proposed a change to openstack/python-ironicclient: Fixes Auth Token being sent as lambda function https://review.openstack.org/53963 | 16:42 |
| linggao | NobodyCam, like for using nova_baremetal, you have to specify compute_driver = nova.virt.baremetal.driver.BareMetalDriver in the nova.conf file | 16:43 |
| linggao | does that mean this compute node cannot deploy vm nodes any more? | 16:43 |
| NobodyCam | yes that is the compute node. so that node can only do baremetal | 16:44 |
| lucasagomes | NobodyCam, thanks, yea me too, the fixture one I think even packing it into one line seems looks alright. Now the second one hmmm I personally like the current form | 16:44 |
| NobodyCam | coluds are made up of many compute nodes | 16:44 |
| linggao | got it. thanks. | 16:44 |
| NobodyCam | :) | 16:44 |
| lucasagomes | linggao, NobodyCam I think there's also a problem with the scheduler when mixing both drivers in the same cloud | 16:51 |
| lucasagomes | cause we need to use a diff nova scheduler for nova bm in the moment, to pick 100% of the resources when allocating it | 16:52 |
| devananda | lucasagomes: i agree with max' comment -- it's readable as is, and lower delta if we add more tests | 16:53 |
| devananda | also, afaik, lucasagomes is correct ^^ there's still problems with >1 scheduler per AZ | 16:54 |
| devananda | but I believe it's possible to mix bm & vm hypervisors in the same cloud by using separate AZs (though I haevn't done it myself) | 16:55 |
| linggao | devananda, what is AZ ? | 16:56 |
| lucasagomes | devananda, great, yea I see the benefits on both sides, having it "expanded" or in one line. And personally I also like more the way the expanded version looks like | 16:56 |
| devananda | availability zone | 16:56 |
| linggao | ok, thanks guys. | 16:57 |
| devananda | anyone seen rloo lately, or know if she's working on the power state stuff? I haven't seen a reply on the bug in some days | 16:59 |
| lucasagomes | devananda, she commented in one of my patches today | 17:02 |
| lucasagomes | not sure about the power state implementation tho | 17:02 |
| devananda | k | 17:03 |
| NobodyCam | she was on a while ago | 17:03 |
| NobodyCam | 09:41 |-INFO > yuriyz [~yuriyz@194.213.110.71] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] | 17:03 |
| NobodyCam | 09:41 |-INFO > yuriyz [~yuriyz@194.213.110.71] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] | 17:03 |
| NobodyCam | 09:41 |-INFO > yuriyz [~yuriyz@194.213.110.71] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] | 17:03 |
| lucasagomes | ? | 17:03 |
| NobodyCam | gah | 17:03 |
| NobodyCam | key stuck | 17:04 |
| NobodyCam | 09:38 |-INFO > rloo [~rloo@nat/yahoo/x-isdfgtetbdntrsqg] has quit [Quit: Leaving...] | 17:04 |
| NobodyCam | thats what I wanted to past | 17:04 |
| NobodyCam | paste | 17:04 |
| devananda | ty | 17:04 |
| *** blamar has quit IRC | 17:19 | |
| NobodyCam | reboot brb | 17:19 |
| *** blamar has joined #openstack-ironic | 17:20 | |
| openstackgerrit | linggao proposed a change to openstack/ironic: Supports get node by instance uuid in API https://review.openstack.org/53262 | 17:31 |
| *** ndipanov has quit IRC | 17:32 | |
| *** ndipanov has joined #openstack-ironic | 17:35 | |
| *** rloo has joined #openstack-ironic | 17:36 | |
| NobodyCam | Hi rloo | 17:36 |
| rloo | hi NobodyCam | 17:36 |
| NobodyCam | rloo: devananda was looking for you awhile ago | 17:37 |
| rloo | ok, i'm around now. thx. | 17:37 |
| *** derekh has quit IRC | 17:38 | |
| *** rloo has quit IRC | 17:41 | |
| *** rloo has joined #openstack-ironic | 17:42 | |
| *** jistr has quit IRC | 17:46 | |
| openstackgerrit | A change was merged to openstack/python-ironicclient: Fixes Auth Token being sent as lambda function https://review.openstack.org/53963 | 17:47 |
| devananda | rloo: hi! was wondering if you've made any progress on the powerstate bug. i haven't seen a reply in a bit | 17:55 |
| linggao | Hi max_lobur | 17:57 |
| max_lobur | Hi linggao | 17:57 |
| rloo | hi devananda. Yeah, got bogged down with non-ironic stuff. and also, my tripleo/ironic env doesn't seem to be working, and wanted to test some stuff out. | 17:57 |
| linggao | I am glad that you reviewed my patch 53485. | 17:57 |
| rloo | devananda. do you have a few cycles? I am thinking of breaking it down into two issues. 1. handle power_state/errors; 2. add reset api | 17:58 |
| linggao | max_lobur, can we discuss the CLI here so that we can get input from folks here? | 17:58 |
| NobodyCam | rloo: I'm working on the official tripleo element right now | 17:58 |
| rloo | NobodyCam: great. let me know when it works! | 17:59 |
| max_lobur | linggao, sure! any input is welcome =) | 17:59 |
| max_lobur | https://review.openstack.org/#/c/53485/5/ironicclient/v1/node_shell.py | 18:00 |
| devananda | rloo: i've got to jump on a conf call, but maybe in an hour i can | 18:00 |
| max_lobur | here are the comments | 18:00 |
| linggao | max_lobur, let me get attendtion from devananda, NobodyCam. | 18:00 |
| rloo | ok devananda, that's fine. | 18:00 |
| linggao | max_lobur, devananda, NobodyCam, this is for showing associated nodes from CLI. | 18:01 |
| linggao | The current syntax that I have implemented is: | 18:01 |
| * NobodyCam looks | 18:02 | |
| linggao | ironic show <uuid> | 18:02 |
| linggao | or ironic show --instance-uuid <uuid> | 18:03 |
| linggao | max_lobur suggested that it is changed to: | 18:03 |
| linggao | ironic show --node-id <uuid> | 18:03 |
| linggao | ironic show --instance-id <uuid> | 18:04 |
| linggao | any preference from you guys? | 18:04 |
| linggao | I am ok either way | 18:04 |
| NobodyCam | linggao: the comand is ironic node-show uuid --instance-id | 18:04 |
| NobodyCam | ? | 18:04 |
| max_lobur | ironic node-show | 18:04 |
| max_lobur | yeah | 18:04 |
| NobodyCam | --node-id seems unclear to me | 18:05 |
| NobodyCam | --nova-instance-id? | 18:05 |
| NobodyCam | but for us it is just a instance | 18:06 |
| linggao | that's a lot of typing for user :) | 18:06 |
| NobodyCam | i like --instance-uuid | 18:07 |
| max_lobur | I suggested this because changing the behavior of ID arg by the separate flag seems tricky to me | 18:07 |
| max_lobur | by default it's considered as node id | 18:07 |
| max_lobur | and if flag is mentioned then it's instance id | 18:07 |
| max_lobur | thats it | 18:07 |
| NobodyCam | max_lobur: ironic has it's own node-id that is not associated with nova at all | 18:08 |
| NobodyCam | so our node-id is not the instance id | 18:09 |
| NobodyCam | they are different fields in the db | 18:09 |
| linggao | currently "ironic node-show <id>" takes node id (like 1) or node's uuid. | 18:09 |
| linggao | when --instance-uuid is specified, then the <id> is the instance uuid. this is the part that bothers max. | 18:10 |
| max_lobur | yea, I know, I'm talking just about comment. I have no questions to relation with nova | 18:10 |
| max_lobur | yes, thanks linggao | 18:10 |
| NobodyCam | I have been trying to use only uuid's | 18:11 |
| rloo | can you take an instance id, or only an instance uuid? | 18:12 |
| NobodyCam | lol no | 18:12 |
| NobodyCam | I just try and use only uuid's and not the "id | 18:13 |
| NobodyCam | " | 18:13 |
| rloo | it is aasymetric (sp) no matter how you look at it, unless you want to add a new ironic XX command. | 18:13 |
| max_lobur | ironic node-show <some_uuid> --instance-id -> showing node by an instance-id | 18:13 |
| rloo | and I'm assuming we don't want to modify the existing ironic node-show <id> | 18:13 |
| max_lobur | ironic node-show <some_uuid> -> showing node by the node-id | 18:14 |
| max_lobur | thats what I found tricky, and proposed to have two separate args for instance and node | 18:14 |
| NobodyCam | node-show is a detalied look at a single node | 18:14 |
| rloo | ironic node-show <id> --instance ? | 18:15 |
| NobodyCam | ironic node-list --instance-id=blah | 18:15 |
| rloo | presumably the help will indicate that --instanceXX means that the <id> is an instance uuid or whatever | 18:15 |
| NobodyCam | get back list | 18:15 |
| linggao | user would use it this way: ironic node-show --instance-uuid <id> | 18:16 |
| NobodyCam | node-show node id | 18:16 |
| max_lobur | ironic node-show --instance-uuid <id> -> In that way this clear to me | 18:16 |
| linggao | yes, it does. | 18:16 |
| linggao | usage: ironic node-show [--instance-uuid] <id> | 18:17 |
| linggao | Show a node. | 18:17 |
| linggao | Positional arguments: | 18:17 |
| linggao | <id> ID, UUID or instance UUID of node | 18:17 |
| linggao | Optional arguments: | 18:17 |
| linggao | --instance-uuid The id is an instance UUID | 18:17 |
| max_lobur | I agree that ironic node-show [--instance-uuid] <id> is OK | 18:17 |
| NobodyCam | now /me is lost | 18:18 |
| max_lobur | I just thought of ironic node-show <id> [--instance-uuid] when wrote those comment =) | 18:18 |
| linggao | ok. Let's move to node-list | 18:18 |
| max_lobur | yes | 18:18 |
| max_lobur | so | 18:19 |
| linggao | usage: ironic node-list [--associated] [--unassociated] | 18:19 |
| linggao | List nodes. | 18:19 |
| linggao | Optional arguments: | 18:19 |
| linggao | --associated Show the nodes that have instance UUID associated | 18:19 |
| linggao | --unassociated Show the nodes that do not have instance UUID associated | 18:19 |
| max_lobur | what would you expect to see when both args are specified? =) | 18:20 |
| NobodyCam | error | 18:20 |
| NobodyCam | :-p | 18:20 |
| max_lobur | agree, as an option =) | 18:20 |
| max_lobur | another - to see all nodes. But this is too tricky =) | 18:20 |
| max_lobur | so I proposed to switch to a single argument | 18:21 |
| linggao | ok, I'll add that. I hope there is a way to specify that these flags are mutual exclusive. | 18:21 |
| linggao | does anyone know how? | 18:21 |
| rloo | so in nova, they tend to do something like --associated=1, so you could do --associated=0 to mean unassociated? | 18:21 |
| max_lobur | yep - the third option is to make those flags mutual exclusive | 18:21 |
| linggao | yes, that's also okay with me. | 18:21 |
| NobodyCam | +1 on exclusive | 18:22 |
| max_lobur | yeah, I saw that flags like --associated=0 or --associated=1 | 18:22 |
| max_lobur | but | 18:22 |
| NobodyCam | --associated set error you can not set --unass.. | 18:22 |
| max_lobur | I we decide to go that way | 18:22 |
| max_lobur | Why don't we have a 'null', 'notnull' instead of 0, 1 | 18:23 |
| rloo | max_lobur, not sure what you mean. --associated=null? | 18:23 |
| NobodyCam | rloo: was there not a issues with 0 as a value? | 18:23 |
| rloo | NobodyCam. I have no idea. I only jumped into this discussion now. Isn't associated just a flag thing? | 18:24 |
| rloo | 0/1, true/false. whatever. | 18:24 |
| max_lobur | currently is | 18:25 |
| max_lobur | and I proposed to take sting argument | 18:25 |
| rloo | why a string arg? | 18:25 |
| linggao | if we have 2 mutual exclusive flages, then we do not have to worry about values ;-) | 18:25 |
| rloo | btw, --associated w/o specifying a value, would default to 1/on/True. | 18:25 |
| max_lobur | linggao, agree =) | 18:26 |
| rloo | so you want to go back to --associated and --unassociated? | 18:26 |
| linggao | then, how about this, if both flags are there, I'll add an error messages. | 18:26 |
| max_lobur | If it's hard to explain then it will be more hard to understand this to the user =) | 18:27 |
| max_lobur | linggao, I like that too | 18:27 |
| rloo | the nice thing about an option/arg that takes values, is that it makes it easy to programmatically generate the command. | 18:27 |
| linggao | I just hope we have exclution built in the arg structure. | 18:27 |
| rloo | the other thing we should note is that we ought to be consistent. so if we go with two mutually exclusive flags here, we should do that in all other cases where there is some sort of binary choice for a command. | 18:28 |
| max_lobur | rloo, good point | 18:28 |
| rloo | i am not familiar enough with ironic/api to know which way to go. | 18:29 |
| NobodyCam | lucasagomes: you around still? | 18:31 |
| linggao | For API, I have added /nodes/?instance_uuid=null or notnull | 18:31 |
| linggao | that's why max_lobur suggest we use thoese values for the CLI. | 18:32 |
| max_lobur | yes, the API approach seems more obvious to me than two flags | 18:33 |
| NobodyCam | ya now I'm kinda leaning to --associated= | 18:33 |
| linggao | but on the other hand, --assoicated is like a switch. | 18:34 |
| max_lobur | the problem is that this is an optional switch | 18:35 |
| linggao | 0 or 1 makes more sense. unless we do thisL ironic nodelist instance_uuid=null | 18:35 |
| rloo | --instance_uuid=null? | 18:35 |
| max_lobur | if this switch wasn't specified it doesn't mean that it's False. It means that it should not be used at all | 18:35 |
| max_lobur | so if we go with a single arg | 18:36 |
| linggao | brb, you guys keep talking. | 18:36 |
| lucasagomes | NobodyCam, hey yea | 18:36 |
| max_lobur | we can have --associated=1 -> get all associated | 18:36 |
| max_lobur | -- associated=0 -> get all which are not associated | 18:37 |
| max_lobur | [-- associated not specified] - get all nodes | 18:37 |
| max_lobur | or, if we swich to a string arg | 18:37 |
| max_lobur | like in comment | 18:37 |
| max_lobur | --instance_association = 'notnull' -> get all associated | 18:38 |
| max_lobur | --instance_association = 'null' -> get all that not associated | 18:38 |
| max_lobur | ... | 18:38 |
| * linggao back | 18:39 | |
| lucasagomes | I tend to agree with max_lobur on having only one argument, having two might lead people to think that they can specify both at the same time | 18:39 |
| rloo | i hate coming up with apis. | 18:39 |
| * linggao too | 18:40 | |
| max_lobur | rloo, =) | 18:40 |
| lucasagomes | not 1/0, or null/notnull, I would vote to be as much as close to the API implmentation as possible, if when filtering the instances_uuid in the api we use the null/notnull keyword | 18:41 |
| rloo | i am wondering about who will use the cli, and whether null/notnull is 'intuitive' or should be 'intuitive' to that user. | 18:41 |
| lucasagomes | I would do the same on the CLI | 18:41 |
| rloo | ah yes, lucasgomes, it does make sense to have the cli and api similar. | 18:42 |
| linggao | keep in mind, user do not have to put = there. --associated=1 and --asociated 1 is the same on CLI . | 18:43 |
| max_lobur | linggao, thanks | 18:43 |
| linggao | so is --instance_uuid=null and --instance_uuid null | 18:43 |
| max_lobur | I think we need to name it like --instance_association, because I would expect and uuid entered for --instance_uuid and not a 'null'/'notnull' | 18:45 |
| lucasagomes | +1 | 18:45 |
| max_lobur | or something like this | 18:45 |
| devananda | rloo: back | 18:45 |
| rloo | lucasagomes: thx for your reply to https://review.openstack.org/53664. My head hurts just reading it though. I will have to find some time to grok it, but as long as that error string makes sense to you that is fine. | 18:45 |
| rloo | devananda, hi. | 18:45 |
| NobodyCam | wb devananda | 18:45 |
| rloo | devananda: so i was going to tackle the power_state/error thing first, then the new reset api second. does that make sense? i wasn't sure what was more critical. | 18:46 |
| lucasagomes | rloo, np :) yea the patch() part is a a bit confusing | 18:46 |
| devananda | rloo: "new reset api" ? | 18:47 |
| rloo | a reset command to reset states in case they are wedged. | 18:47 |
| rloo | isn't that what you also wanted? or was that lifeless. | 18:47 |
| devananda | ah | 18:48 |
| devananda | there was another conversation about reset, unrelated | 18:48 |
| devananda | rloo: so yes, the power state thing first makes more sense | 18:48 |
| devananda | rloo: to me, because if you implement a power-state-reset now, you'll have to change that when you change the power-staet | 18:49 |
| rloo | devananda : yup, sorry that was lifeless, 'allow overrides by admins'. | 18:49 |
| devananda | i think the way i was proposing to change it wouldn't require an override | 18:49 |
| devananda | but IMBW | 18:49 |
| rloo | devananda. wrt last_error: is 'last_error' ok? and what value (I was thinking of exception string?) | 18:49 |
| lifeless | rloo: I didn't ask for a reset API; I asked for the ability to fix things; perhaps you mean that when you say reset api. | 18:50 |
| rloo | lifeless: yes, in order to fix things like db setting being 'wrong', need some sort of api. | 18:50 |
| rloo | or modify db directly. | 18:50 |
| devananda | rloo: last_error as a VARCHAR(255) field in the DB. exception message (but not a stack trace) seems fine | 18:50 |
| lifeless | rloo: so reset is an overloaded term. | 18:51 |
| lifeless | rloo: it might mean reset on a machine | 18:51 |
| lifeless | rloo: it might mean start-a-task-over | 18:51 |
| rloo | lifeless: true, I shouldn't have used reset. | 18:51 |
| lifeless | rloo: it might mean <.. several other things> | 18:51 |
| rloo | devananda: i was thinking TEXT, cuz I had no idea how long msgs might be, but VARCHAR(255) is fine. | 18:52 |
| devananda | I think my comment actually addressed the needs of "reset the power state" | 18:52 |
| devananda | power_state -- always represents current power state. Any power operation sets this back to "actual" when done (whether successful or not). Also, we should add a periodic_task to sync this. Set to ERROR only when unable to get the power state from a node. | 18:52 |
| devananda | if we do ^, then the power_state field won't need a "reset", because just issuing a driver.power.get_power_state() will update it | 18:53 |
| devananda | and we should allow changing target_power_state any time the node is not locked | 18:53 |
| devananda | by clearing the target_power_state when ever any requested power state change finishes (whether it completes successfully or not) we prevent needing to "reset" that field, as well | 18:54 |
| devananda | and then storing the error in a third field (last_error) allows the user to still see the error, without needing to clear an "error" state from one of the other two fields | 18:54 |
| devananda | <EOL> | 18:54 |
| rloo | devananda: thx. and wrt the last_error. Do we need to add a new API to expose it? Or wait, I'll ask that question again after I get the plumbing done. | 18:55 |
| linggao | max_lobur so it looks like everyone agrees on using 1 flag instead of 2. | 18:55 |
| devananda | asking now is good | 18:55 |
| rloo | devananda: well, I still have lots to learn, so I figure maybe I'll have an opinion by the time I've coded the rest ;) | 18:56 |
| linggao | max_lobur, have we decided on the value to use? | 18:56 |
| devananda | rloo: we dont need a "new API". we will need to add a new property to ironic/api/controlers/v1/node.Node | 18:56 |
| devananda | rloo: and to the client to match | 18:56 |
| linggao | maybe we wait on devananda, after he is done with rloo. | 18:57 |
| devananda | rloo: this bug will require adding a field to the DB, adding a property to the RPC and API object models, and implementing some changes in the conductorManager | 18:57 |
| rloo | devananda. Ah. Thx. (I still need to come up-to-speed on existing api/cli.) | 18:57 |
| devananda | there'll be a small CLI change too (but very small) | 18:57 |
| devananda | linggao, max_lobur - what's up? parallel is fine. I need to go get breakfast soon :) | 18:58 |
| max_lobur | linggao, yea, the discussion subsided | 18:58 |
| devananda | linggao: is this re: ironic node-show --instance-uuid <xxx> | 18:58 |
| devananda | ? | 18:58 |
| max_lobur | discussion topic is 2nd comment https://review.openstack.org/#/c/53485/5/ironicclient/v1/node_shell.py | 18:58 |
| linggao | devanda: usage: ironic node-list [--associated] [--unassociated] | 18:58 |
| linggao | List nodes. | 18:58 |
| linggao | Optional arguments: | 18:58 |
| linggao | --associated Show the nodes that have instance UUID associated | 18:58 |
| linggao | --unassociated Show the nodes that do not have instance UUID associated | 18:58 |
| rloo | devananda: yeah, i have the field added to db already, and I think the properties done too. just that i want to test it to see what I got wrong. btw, if you need it done asap and want to just code it (faster for you), I am fine with that. | 18:58 |
| linggao | davananda, folks, do not like 2 flags | 18:59 |
| linggao | so we decided to use 1 flag. | 18:59 |
| linggao | then the choice would be: | 18:59 |
| linggao | ironic node-list | 18:59 |
| linggao | ironic node-list --associated=0 or 1 | 18:59 |
| linggao | ironic node-list --instance-association=null or notnull | 19:00 |
| NobodyCam | api is null notnull? | 19:00 |
| linggao | which flag do you prefer. | 19:00 |
| linggao | ? | 19:00 |
| max_lobur | NobodyCam, yea api uses null notnull | 19:01 |
| devananda | rloo: if you'd like to put up what you have as a draft (git review -D) I'm happy to provide soem feedback early | 19:01 |
| NobodyCam | I say match api | 19:01 |
| max_lobur | like ?instance_uuid=notnull | 19:01 |
| max_lobur | or ?instance_uuid=null | 19:01 |
| devananda | rloo: and it's OK to fix it in parts. doesn't need to be one giant patch | 19:01 |
| lucasagomes | assuming the API uses the null/notnull keyworkds, I would use it on the CLI as well | 19:01 |
| linggao | yes, that's for API. | 19:01 |
| NobodyCam | associated=null / associated=notnull | 19:02 |
| rloo | devananda: Let me try to get ironic working outside of unit tests first, to see where it blows up first. hopefully by end of today. thx. | 19:02 |
| NobodyCam | rloo: I'm happy to help with any issuse you come across in setting up a dib env | 19:02 |
| devananda | linggao: i don't like a boolean option :) | 19:03 |
| rloo | NobodyCam. ok thx. I'm in the middle of setting it up (seed running, going to do undercloud w ironic) so cross my fingers... | 19:03 |
| *** anteaya has quit IRC | 19:03 | |
| devananda | linggao: what about ironic node-list | node-list-associated | node-list-unassociated | 19:04 |
| devananda | linggao: three separate commands | 19:04 |
| NobodyCam | rloo: you using the current walk thru? | 19:04 |
| max_lobur | devananda, yea, thought about this too | 19:04 |
| max_lobur | looks cool to me | 19:04 |
| max_lobur | but searched through nova | 19:04 |
| lucasagomes | rloo, (I don't know if it helps) but the way ironic will manage the power state of the nodes is similar to the way the ibm pureapplication does it (having 3 fields for the power state, but in our case we have a dictionary instead of 3 attributes at the root of the resource): http://pic.dhe.ibm.com/infocenter/psappsys/v1r0m0/index.jsp?topic=%2Fcom.ibm.puresystems.appsys.1700.doc%2Fpsapsys_restapi%2Fmanagement_nodes.html | 19:04 |
| devananda | linggao: i don't want users of the CLI to have to know about our particular keywords "null" and "notnull", or use 0/1 true/false etc | 19:04 |
| rloo | NobodyCam. Yeah, but with an older version of tripleo that was working. I tried with the latest yest and things didn't work. | 19:04 |
| max_lobur | and there are no similar things | 19:05 |
| max_lobur | also imagine if we have something more =) | 19:05 |
| devananda | max_lobur: this is coming from nova's driver API | 19:05 |
| devananda | so you can find it there | 19:05 |
| devananda | not in the CLI | 19:05 |
| linggao | devananda, then how about the origina node-list [--associated|--unassociated] ? | 19:05 |
| devananda | linggao: I'm fine with that | 19:05 |
| devananda | node-list => all | 19:06 |
| linggao | yes | 19:06 |
| NobodyCam | lucasagomes: should we land 52902? | 19:06 |
| linggao | max_lobur ? | 19:06 |
| NobodyCam | or are you planning another version? | 19:06 |
| devananda | node-list [--option] => filtered list | 19:06 |
| lucasagomes | NobodyCam, oh no, I will add more things to it | 19:06 |
| rloo | lucasagomes: thx for the ptr to ibm purapplication. | 19:06 |
| devananda | taht seems totally reasonable to me | 19:06 |
| lucasagomes | lemme put as a WIP | 19:06 |
| NobodyCam | :) | 19:06 |
| max_lobur | devananda, linggao agree with original too | 19:07 |
| linggao | ok, I'll make sure the 2 flags are mutual exclusive. | 19:07 |
| max_lobur | yea, I gree with that | 19:07 |
| devananda | ++ | 19:07 |
| *** anteaya has joined #openstack-ironic | 19:08 | |
| linggao | thanks max_lobur, devananda, NobodyCam, rloo, lucasagomes for working on the CLI together. | 19:08 |
| max_lobur | Yes, thanks Everyone for the discussion | 19:09 |
| max_lobur | sorry if it was a little messy =) | 19:09 |
| lucasagomes | thanks u guys :) | 19:09 |
| NobodyCam | :) | 19:09 |
| lucasagomes | I'm done for today (starving now) | 19:09 |
| lucasagomes | have a good night everyone! | 19:09 |
| max_lobur | g'night lucasagomes | 19:09 |
| NobodyCam | lol Good nught lucasagomes | 19:10 |
| devananda | I think it's great ya'll had a good discussion while I was afk :) | 19:10 |
| max_lobur | I'm leaving too | 19:10 |
| devananda | cheers! | 19:10 |
| linggao | :) | 19:10 |
| devananda | i'm going to run out for food, too. bbiah | 19:10 |
| NobodyCam | :) | 19:10 |
| *** lucasagomes has quit IRC | 19:10 | |
| max_lobur | g'night All | 19:10 |
| NobodyCam | good night max_lobur | 19:10 |
| *** anniec has joined #openstack-ironic | 19:24 | |
| *** anniec has left #openstack-ironic | 19:24 | |
| rloo | hi NobodyCam. | 19:26 |
| NobodyCam | hi rloo | 19:26 |
| rloo | have you recently used tripleo+ironic? | 19:26 |
| NobodyCam | yes | 19:26 |
| rloo | i am not sure what I did, but things aren't working. I'm going to start from scratch. | 19:27 |
| NobodyCam | befor you do that what is happening? | 19:27 |
| rloo | so I should be able to use the latest and greatest tripleo + ironic. | 19:27 |
| NobodyCam | what is not working? | 19:27 |
| rloo | the baremetal vm is trying to 'spawn' itself. but it doesn't seem to be able to find the image to boot from. | 19:28 |
| NobodyCam | ok so seed vm is working | 19:28 |
| rloo | yes, this is using an older tripleo. i tried a new one yesterday and i don't remember now, what didn't work, so i thought i could go back. but you can never go back ;) | 19:29 |
| rloo | i'll just start again with new, and try to figure that out. as far as you know, the walkthrough instructions are still the same? | 19:30 |
| NobodyCam | ok so with a working seed vm build a undercloud vm with the ironic elelmet from the walkthru | 19:30 |
| NobodyCam | load that undercloud + ironic image and deploy it as the under cloud | 19:31 |
| rloo | yeah, using heat. and that is what is not working. | 19:32 |
| NobodyCam | oko what is the error? check seedvm:/var/log/upstart/heat-engine.log | 19:32 |
| rloo | Failed to read /etc/heat/environment.d | 19:34 |
| NobodyCam | humm | 19:35 |
| NobodyCam | ok I'm going to resetup my env right now. | 19:36 |
| rloo | I don't know if that is a red herring or not though, since I don't know anything about heat or what might be in that file. it does some other stuff. I may be out of sync. so I'm going to restart. if I have to debug, might as well use the latest and greatest. | 19:37 |
| rloo | thx! | 19:37 |
| NobodyCam | ok booting seed vm now | 19:39 |
| NobodyCam | gah downloading new cloudimage :-p | 19:41 |
| rloo | oh oh. Hopefully it is a more improved version :-) | 19:42 |
| NobodyCam | lol | 19:42 |
| NobodyCam | brb | 19:53 |
| *** harlowja has quit IRC | 20:04 | |
| *** harlowja has joined #openstack-ironic | 20:05 | |
| *** jmk__ has joined #openstack-ironic | 20:09 | |
| *** anniec has joined #openstack-ironic | 20:13 | |
| *** jmk__ has quit IRC | 20:16 | |
| NobodyCam | ok seed vm booted configed. ironic undercloud built | 20:30 |
| NobodyCam | heat stack creating now | 20:31 |
| *** jdob has quit IRC | 20:31 | |
| openstackgerrit | linggao proposed a change to openstack/python-ironicclient: Modifies CLI to show nodes by instance uuid https://review.openstack.org/53485 | 20:33 |
| NobodyCam | quick walkies brb | 20:36 |
| rloo | NobodyCam, before I forget. I noticed that tripleo/tripleo-heat-templates has an 'ironic-source.yaml' file. (whereas we're using ironic.yaml from you.) | 20:39 |
| *** anniec_ has joined #openstack-ironic | 20:42 | |
| NobodyCam | yes that landed | 20:43 |
| *** anniec has quit IRC | 20:43 | |
| *** anniec_ is now known as anniec | 20:43 | |
| NobodyCam | I'm going drop off for about a 1/2 an hour ... need to move | 20:43 |
| rloo | NobodyCam: does it matter, which one to use? | 20:43 |
| rloo | NobodyCam: that's a fast move :-) | 20:44 |
| NobodyCam | I use the one in with the element jsut because It's been tested | 20:44 |
| NobodyCam | lol Life of a homeless person | 20:44 |
| NobodyCam | well not really homeless... | 20:45 |
| NobodyCam | bbaib | 20:45 |
| *** linggao has quit IRC | 20:54 | |
| devananda | back | 20:57 |
| *** rloo_ has joined #openstack-ironic | 20:59 | |
| *** anniec has left #openstack-ironic | 21:03 | |
| *** rloo has quit IRC | 21:07 | |
| *** creiht has quit IRC | 21:07 | |
| *** creiht has joined #openstack-ironic | 21:14 | |
| NobodyCam | house movied | 21:17 |
| NobodyCam | moved even | 21:17 |
| NobodyCam | wb devananda | 21:17 |
| *** lnxnut has quit IRC | 21:19 | |
| *** romcheg has joined #openstack-ironic | 21:27 | |
| openstackgerrit | Jenkins proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/54424 | 21:30 |
| openstackgerrit | Jenkins proposed a change to openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/54430 | 21:32 |
| *** harlowja has quit IRC | 21:36 | |
| NobodyCam | rloo_: are you getting a error sshing to the undercloud? | 21:37 |
| rloo_ | NobodyCam. Yes, how did you guess? | 21:37 |
| *** harlowja has joined #openstack-ironic | 21:37 | |
| NobodyCam | i am getting that too | 21:37 |
| rloo_ | i can ping, and it shows the vm as 'active'. | 21:38 |
| rloo_ | in a meeting now so haven't been able to look into it. | 21:38 |
| NobodyCam | I have console access | 21:38 |
| NobodyCam | I'm looking in to it now | 21:38 |
| *** jbjohnso has quit IRC | 21:49 | |
| anteaya | is there anyone in here that is in China right now? | 22:07 |
| *** anniec has joined #openstack-ironic | 22:07 | |
| anteaya | just trying to see if I will be able to access my gmail account from my tablet in hk | 22:07 |
| *** harlowja has quit IRC | 22:08 | |
| anteaya | if anyone is they are probably asleep | 22:08 |
| NobodyCam | :-p | 22:09 |
| NobodyCam | you "SHOULD" be able to | 22:10 |
| Haomeng|2 | anteaya:ping | 22:10 |
| Haomeng|2 | Iam fromChina:) | 22:10 |
| NobodyCam | there's always starbucks | 22:10 |
| NobodyCam | hey Haomeng|2 :) | 22:10 |
| Haomeng|2 | :) | 22:11 |
| * NobodyCam hopes his Galaxy s3 will just work | 22:11 | |
| *** anniec has quit IRC | 22:11 | |
| Haomeng|2 | anteaya:yes, wecan access gmail in China | 22:13 |
| anteaya | Haomeng|2: okay great thank you | 22:15 |
| anteaya | a friend suggested I use this sight: http://www.greatfirewallofchina.org | 22:15 |
| Haomeng|2 | :) | 22:15 |
| anteaya | to evaluate what will work in hk | 22:15 |
| anteaya | and my gmail won't | 22:15 |
| Haomeng|2 | but i cannot access http://www.greatfirewallofchina.org/ now | 22:16 |
| anteaya | Haomeng|2: can you ping this server? 15.185.154.235 | 22:16 |
| anteaya | ha ha ha | 22:16 |
| Haomeng|2 | if you worry about the china firewall | 22:16 |
| Haomeng|2 | you can install goagent | 22:16 |
| Haomeng|2 | In China, govement will block some website by official:) | 22:17 |
| Haomeng|2 | Oops! Google Chrome could not connect to 15.185.154.235 | 22:17 |
| Haomeng|2 | welcome | 22:17 |
| NobodyCam | lol | 22:18 |
| Haomeng|2 | anteaya: do you have google application engine account? | 22:18 |
| Haomeng|2 | I suggest you install goagent to be backup solutionfor internete accessing | 22:19 |
| Haomeng|2 | :) | 22:19 |
| devananda | random nifty stats for you guys ahead of my talk next week | 22:20 |
| devananda | in the six months since ironic was git filterbranch'd from nova | 22:21 |
| devananda | we have ~300 commits from 30 developers in 13 different companies | 22:21 |
| anteaya | Haomeng|2: I will look at goagent | 22:21 |
| anteaya | I will be on an android tablet | 22:21 |
| Haomeng|2 | devanada:great work | 22:21 |
| anteaya | so hopefully goagent will work on android | 22:22 |
| NobodyCam | woo | 22:22 |
| Haomeng|2 | goagent support android | 22:22 |
| anteaya | devananda: w00t | 22:22 |
| Haomeng|2 | devananda:do you have time,want to discuss with you about https://bugs.launchpad.net/ironic/+bug/1244348 solution | 22:23 |
| devananda | Haomeng|2: sure | 22:23 |
| Haomeng|2 | I want to add code to catch UniqueConstraint validation exception to do EAFP | 22:23 |
| devananda | i see your comments on bug -- i'll respond | 22:24 |
| *** romcheg1 has joined #openstack-ironic | 22:24 | |
| Haomeng|2 | and I think we can keep the current logic- checking existing one before portgreat, right? | 22:24 |
| Haomeng|2 | ok, devananda, no rush | 22:24 |
| devananda | basically yes :) | 22:24 |
| Haomeng|2 | I will commit the fix today | 22:24 |
| Haomeng|2 | and help me review | 22:24 |
| Haomeng|2 | :) | 22:24 |
| Haomeng|2 | ifyou have time' | 22:24 |
| Haomeng|2 | I understand you are busy:) | 22:24 |
| *** romcheg has quit IRC | 22:25 | |
| devananda | try: insert; except DBDupKey: raise PrettyError; | 22:25 |
| devananda | basically what you said is correct | 22:25 |
| Haomeng|2 | and for another one - https://bugs.launchpad.net/ironic/+bug/1244541, looks like there isno db level UniqueConstraint validation exception , so itis diffcult to do EAFP | 22:26 |
| Haomeng|2 | yes,I will create the UniqueConstraint validation exception case to have test to see the exception type andinstance | 22:26 |
| devananda | Haomeng|2: I'm nt sure there should be a UX on instance_uuid. but even if there were, taht wouldn't solve that bug | 22:27 |
| devananda | Haomeng|2: the problem with 1244541 is that the API will simply overwrite an existing instance_uuid value | 22:28 |
| devananda | rather than fail if the row is non-null | 22:28 |
| devananda | take a look at .... | 22:28 |
| Haomeng|2 | yes | 22:28 |
| anteaya | Haomeng|2: can you see this: https://etherpad.openstack.org/p/icehouse-summit-more-salt-in-infra | 22:28 |
| devananda | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/baremetal/driver.py#n226 | 22:29 |
| Haomeng|2 | I understand 1244541 case, just want to discuss with you thesolution | 22:29 |
| devananda | and | 22:29 |
| devananda | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/baremetal/db/sqlalchemy/api.py#n190 | 22:29 |
| devananda | that ^ is the solution | 22:29 |
| devananda | atomic "set instance_uuid on one and only one row, or raise exception" | 22:30 |
| Haomeng|2 | ok, got it | 22:30 |
| Haomeng|2 | will try to fix,thankyou devananda | 22:30 |
| devananda | basically must be implemented at DB layer: UPDATE .. SET col=value WHERE node.id=xxx AND node.instance_uuid=NULL; if rows_affected != 1: raise | 22:30 |
| devananda | Haomeng|2: thank you! | 22:30 |
| Haomeng|2 | yes, should be working on db level | 22:31 |
| Haomeng|2 | anteaya, I can acces https://etherpad.openstack.org/p/icehouse-summit-more-salt-in-infra | 22:31 |
| anteaya | Haomeng|2: well that is good news | 22:31 |
| anteaya | hmmm, how to ssh into my hpcloud servers | 22:32 |
| *** romcheg1 has quit IRC | 22:32 | |
| anteaya | I don't have root access on my android tablet | 22:32 |
| NobodyCam | why do you need root? | 22:32 |
| Haomeng|2 | this is guideline for goangent on android https://play.google.com/store/apps/details?id=org.gaeproxy&hl=en | 22:33 |
| *** romcheg has joined #openstack-ironic | 22:33 | |
| NobodyCam | anteaya: https://play.google.com/store/apps/details?id=sk.vx.connectbot&hl=en | 22:33 |
| anteaya | yeah I have connectbot | 22:33 |
| anteaya | goangent on android needs to be root, says the instructions | 22:34 |
| NobodyCam | so what do you need root for? | 22:34 |
| NobodyCam | ahh | 22:34 |
| NobodyCam | :-p | 22:34 |
| Haomeng|2 | yes | 22:34 |
| anteaya | so I have sshed into my hpcloud servers all day on my tablet | 22:35 |
| anteaya | that is what I have been doing today | 22:35 |
| Haomeng|2 | cool | 22:35 |
| anteaya | but Haomeng|2 can't ping them | 22:35 |
| anteaya | :( | 22:35 |
| NobodyCam | ping I can see | 22:35 |
| anteaya | so my gmail backed email, good, etherpad, good, not my hpcloud servers | 22:35 |
| NobodyCam | Haomeng|2: can you try ssh to anteaya's addy | 22:36 |
| Haomeng|2 | ok | 22:36 |
| Haomeng|2 | I tested | 22:36 |
| NobodyCam | connect? | 22:37 |
| Haomeng|2 | ssh working for accessing 15.185.154.235 | 22:37 |
| anteaya | what was the output? | 22:37 |
| Haomeng|2 | yes | 22:37 |
| *** jbjohnso has joined #openstack-ironic | 22:37 | |
| anteaya | great | 22:37 |
| NobodyCam | awesome | 22:37 |
| anteaya | so I'm good then? | 22:37 |
| NobodyCam | icomp blocked | 22:37 |
| NobodyCam | no biggie | 22:37 |
| anteaya | what is icomp | 22:37 |
| NobodyCam | icmp | 22:37 |
| anteaya | ping uses it? | 22:37 |
| Haomeng|2 | output is "login as: " | 22:37 |
| anteaya | great | 22:37 |
| anteaya | thanks Haomeng|2 | 22:37 |
| NobodyCam | ya thats port 53 | 22:37 |
| NobodyCam | ssh is port 22 | 22:37 |
| Haomeng|2 | the previous try is http iused | 22:37 |
| NobodyCam | thats port 80 | 22:37 |
| anteaya | I am relieved | 22:37 |
| Haomeng|2 | I understand thatisweb first | 22:37 |
| Haomeng|2 | :) | 22:37 |
| anteaya | ah | 22:38 |
| Haomeng|2 | so canj not access | 22:38 |
| Haomeng|2 | :) | 22:38 |
| Haomeng|2 | sorry for confusing | 22:38 |
| *** anniec has joined #openstack-ironic | 22:38 | |
| anteaya | no worries | 22:38 |
| anteaya | thanks for being in china | 22:38 |
| anteaya | :D | 22:38 |
| Haomeng|2 | welcome all of you come toChina HK | 22:38 |
| anteaya | yes | 22:38 |
| NobodyCam | :) look forward to meeting | 22:38 |
| anteaya | getting ready | 22:38 |
| Haomeng|2 | but I will not attend HK meeting:) | 22:38 |
| anteaya | no | 22:38 |
| NobodyCam | ahh | 22:38 |
| Haomeng|2 | :) | 22:38 |
| anteaya | why not | 22:38 |
| NobodyCam | it close to you? | 22:38 |
| Haomeng|2 | not sure | 22:38 |
| NobodyCam | :( | 22:39 |
| Haomeng|2 | depeds on my boss:) | 22:39 |
| Haomeng|2 | maybe because I amnew guy :) | 22:39 |
| NobodyCam | you need to at least come thrusday | 22:39 |
| NobodyCam | thats the ironc track | 22:39 |
| Haomeng|2 | yes, expect, but cannot:) | 22:40 |
| NobodyCam | :-p | 22:40 |
| anteaya | I hope you get there for one day | 22:40 |
| anteaya | would be nice to meet you | 22:40 |
| anteaya | thanks for your help | 22:40 |
| Haomeng|2 | I haveanother team member SunJing, she is girl will attend Ironic BP discusstion about her firmware updating | 22:40 |
| Haomeng|2 | maybe you can see her | 22:40 |
| NobodyCam | :) yes we will :) | 22:41 |
| Haomeng|2 | yes, feel badly for | 22:41 |
| Haomeng|2 | and I will get feedback from her | 22:41 |
| Haomeng|2 | and some other guys | 22:41 |
| Haomeng|2 | maybe just one persion is her to focus onIronic | 22:42 |
| Haomeng|2 | project | 22:42 |
| NobodyCam | :) | 22:42 |
| NobodyCam | your doing a great job | 22:42 |
| Haomeng|2 | dont think, just fix some minor defects:) | 22:43 |
| *** anniec has quit IRC | 22:43 | |
| Haomeng|2 | need more time to understand indeep | 22:43 |
| NobodyCam | gota start somewhere :) | 22:43 |
| Haomeng|2 | :) | 22:43 |
| Haomeng|2 | thank you | 22:43 |
| *** anniec has joined #openstack-ironic | 22:47 | |
| *** anniec has quit IRC | 23:00 | |
| * NobodyCam looks for some food | 23:03 | |
| *** romcheg has left #openstack-ironic | 23:03 | |
| *** harlowja has joined #openstack-ironic | 23:08 | |
| *** anniec has joined #openstack-ironic | 23:33 | |
| NobodyCam | rloo_: check the line you booted seed vm with. | 23:43 |
| NobodyCam | for me the fix was : | 23:43 |
| NobodyCam | #boot-seed-vm -a $NODE_ARCH $NODE_DIST bm-dnsmasq | 23:43 |
| NobodyCam | boot-seed-vm -a $NODE_ARCH $NODE_DIST neutron-dhcp-agent | 23:43 |
| rloo_ | hmm. I used neutron-dhcp-agent. Let me check. | 23:44 |
| NobodyCam | I just updated all my local repos and made the above change | 23:44 |
| NobodyCam | just got to: Service baremetal created with password b70d1b18540869d48c844b9d5c482d36614e887c | 23:45 |
| NobodyCam | :-p | 23:45 |
| rloo_ | NobodyCam. When you booted the undercloud image. Did you use the walkthru instructions, or tripleo instructions? I used tripleo. | 23:45 |
| rloo_ | Yay NobodyCam! There is hope yet :-) | 23:46 |
| rloo_ | ie, I didn't do the if [ $DHCP_DRIVER" = "bm_dnsmasq" ] ; ... part. Hmm, you probably didn't either. | 23:46 |
| NobodyCam | rloo_: I guess the walk thru. its devtest with a slight diff to add ironic | 23:47 |
| NobodyCam | nope | 23:47 |
| *** rloo has joined #openstack-ironic | 23:49 | |
| *** devanand1 has joined #openstack-ironic | 23:53 | |
| *** rloo_ has quit IRC | 23:58 | |
| *** devananda has quit IRC | 23:58 | |
| *** lynxman has quit IRC | 23:58 | |
| *** lynxman has joined #openstack-ironic | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!