| *** smoriya has joined #openstack-ironic | 00:09 | |
| *** sinval has joined #openstack-ironic | 00:23 | |
| *** jamielennox is now known as jamielennox|away | 00:26 | |
| *** jamielennox|away is now known as jamielennox | 00:30 | |
| *** Haomeng has joined #openstack-ironic | 00:44 | |
| *** Haomeng|2 has quit IRC | 00:47 | |
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Adds port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 00:52 |
|---|---|---|
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Add port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 00:59 |
| *** zhenguo has joined #openstack-ironic | 01:16 | |
| *** puranamr has quit IRC | 01:37 | |
| *** ThomasPB has joined #openstack-ironic | 01:38 | |
| *** puranamr has joined #openstack-ironic | 01:45 | |
| *** chenglch has joined #openstack-ironic | 01:48 | |
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Adds port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 01:54 |
| *** Haomeng has quit IRC | 01:54 | |
| *** Sukhdev has joined #openstack-ironic | 02:01 | |
| *** sinval has quit IRC | 02:08 | |
| *** Haomeng has joined #openstack-ironic | 02:39 | |
| *** ramineni has joined #openstack-ironic | 02:40 | |
| *** zz_natorious is now known as natorious | 02:52 | |
| *** achanda has joined #openstack-ironic | 03:11 | |
| *** achanda has quit IRC | 03:18 | |
| *** saripurigopi has joined #openstack-ironic | 03:34 | |
| saripurigopi | good morning Ironic... | 03:35 |
| *** saripurigopi has quit IRC | 03:43 | |
| *** Nisha has joined #openstack-ironic | 03:47 | |
| *** davideagnello has joined #openstack-ironic | 03:51 | |
| *** puranamr has quit IRC | 03:51 | |
| *** coolsvap|away is now known as coolsvap | 03:52 | |
| *** Nisha_away has joined #openstack-ironic | 03:54 | |
| *** Nisha has quit IRC | 03:54 | |
| *** sinval has joined #openstack-ironic | 03:54 | |
| *** davideagnello has quit IRC | 03:56 | |
| *** Sukhdev has quit IRC | 04:01 | |
| *** saripurigopi has joined #openstack-ironic | 04:05 | |
| *** Marga_ has joined #openstack-ironic | 04:06 | |
| *** Marga_ has quit IRC | 04:07 | |
| *** Marga_ has joined #openstack-ironic | 04:07 | |
| *** puranamr has joined #openstack-ironic | 04:11 | |
| *** Nisha_brb has joined #openstack-ironic | 04:13 | |
| *** Nisha_away has quit IRC | 04:13 | |
| *** puranamr has quit IRC | 04:16 | |
| *** saripurigopi has quit IRC | 04:18 | |
| *** Marga_ has quit IRC | 04:20 | |
| *** Marga_ has joined #openstack-ironic | 04:21 | |
| *** Marga_ has quit IRC | 04:25 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add ability to filter nodes by provision_state via API https://review.openstack.org/196529 | 04:36 |
| openstackgerrit | Anusha Ramineni proposed stackforge/proliantutils: Add RIS support for firmware update https://review.openstack.org/193952 | 04:50 |
| *** sinval has quit IRC | 04:55 | |
| *** rameshg87 has joined #openstack-ironic | 04:55 | |
| *** davideagnello has joined #openstack-ironic | 04:57 | |
| *** davideagnello has quit IRC | 05:03 | |
| *** davideagnello has joined #openstack-ironic | 05:03 | |
| *** davideagnello has quit IRC | 05:03 | |
| *** puranamr has joined #openstack-ironic | 05:08 | |
| *** saripurigopi has joined #openstack-ironic | 05:18 | |
| *** radek__ has joined #openstack-ironic | 05:20 | |
| *** yuanying_ has joined #openstack-ironic | 05:22 | |
| *** radek_ has quit IRC | 05:23 | |
| *** Nisha_brb has quit IRC | 05:23 | |
| *** yuanying has quit IRC | 05:25 | |
| *** puranamr has quit IRC | 05:26 | |
| Haomeng | saripurigopi: morning:) | 05:28 |
| saripurigopi | morning Haomeng o/ | 05:29 |
| Haomeng | saripurigopi: :) | 05:29 |
| *** stegranet has joined #openstack-ironic | 05:30 | |
| *** Marga_ has joined #openstack-ironic | 05:35 | |
| naohirot | Haomeng: saripurigopi: good morning :) | 05:38 |
| Haomeng | naohirot: morning:) | 05:38 |
| naohirot | rameshg87: good morning | 05:39 |
| naohirot | rameshg87: or Haomeng: | 05:39 |
| Haomeng | naohirot: :) | 05:39 |
| naohirot | rameshg87: or Haomeng: I have question about rest api | 05:39 |
| Haomeng | naohirot: sure | 05:39 |
| *** Marga_ has quit IRC | 05:40 | |
| naohirot | Haomeng: If rest api is async, that api returns 202 if request is accepted. | 05:40 |
| *** Marga_ has joined #openstack-ironic | 05:40 | |
| naohirot | Haomeng: how can user know if error happened after api returned 202? | 05:41 |
| naohirot | Haomeng: for instance power control. | 05:41 |
| Haomeng | naohirot: yes, 202 just means that api request is accepted already | 05:41 |
| Haomeng | naohirot: have to check status api if have I thinnk | 05:41 |
| naohirot | Haomeng: okay, I'm looking source code too, node.py in api. | 05:42 |
| Haomeng | naohirot: we can call node-show api to get power status | 05:43 |
| naohirot | Haomeng: Okay, I believe that is a solution, but please look at Jim's comment here | 05:44 |
| Haomeng | naohirot: GET http://9.119.58.234:6385/v1/nodes/<NODE_UUID> | 05:44 |
| Haomeng | naohirot: for example | 05:44 |
| Haomeng | naohirot: and Lucas has a patch to support get subset for node properties | 05:44 |
| naohirot | Haomeng: https://review.openstack.org/#/c/186700/4/specs/liberty/enhance-power-interface-for-soft-reboot-and-nmi.rst | 05:45 |
| Haomeng | naohirot: so you can try to get node's power_state only | 05:45 |
| naohirot | Haomeng: line 156 | 05:45 |
| Haomeng | naohirot: ok, let me check | 05:45 |
| *** puranamr has joined #openstack-ironic | 05:46 | |
| Haomeng | naohirot: yes, this is concern, for user, have to call another api to check the status, and if user run with cli, that is easy to call another api to get result to return user | 05:47 |
| naohirot | Haomeng: This is a way of async call, so do we have a solution without making another call? | 05:49 |
| Haomeng | naohirot: sorry, no more idea so far:) | 05:49 |
| *** yog__ has joined #openstack-ironic | 05:50 | |
| Haomeng | naohirot: but I think we can follow the ansync call mode in openstack | 05:50 |
| Haomeng | naohirot: do you want to make it a sync call? | 05:50 |
| naohirot | Haomeng: Okay, I should ask to Jim, but timezone is different. so what do you think what kind of answer he expected? | 05:50 |
| Haomeng | naohirot: I understand Jim just think from user view, how to tell user if soft-power failed | 05:51 |
| *** puranamr has quit IRC | 05:51 | |
| Haomeng | Haomeng: so we can have another api call to get status, I think it make sense:) | 05:51 |
| naohirot | Haomeng: Yes, we can have another api to get status, but not error code of the previous call. | 05:52 |
| Haomeng | naohirot: same behavior with current existing PUT v1/nodes/<NODE_UUID>/states/power api call | 05:53 |
| naohirot | Haomeng: It's very difficult to keep track of the previous call. | 05:53 |
| Haomeng | naohirot: yes | 05:53 |
| Haomeng | naohirot: let me post these into your spec as my comments, thanks for your spec:) | 05:54 |
| naohirot | Haomeng: I understand that making another call is a way of async call. thanks! | 05:54 |
| Haomeng | naohirot: yes | 05:54 |
| naohirot | Haomeng: Yes, please, that is absolutely helpful for me.:-) | 05:55 |
| Haomeng | naohirot: so dont worry, that should be common behavior that we run api in async mode, I think:) | 05:55 |
| naohirot | Haomeng: Yep, :) | 05:56 |
| Haomeng | naohirot: jim's concern is "For "soft reboot", "get power state" will *always* return POWER_ON, I think. This makes it impossible to tell if the call succeeded or not. The only way to check is to log on and look at uptime." | 05:57 |
| Haomeng | naohirot: so can we add more status to identify the soft reboot status | 05:57 |
| Haomeng | naohirot: such as soft-reboot-pending status, make sense? | 05:58 |
| *** gridinv_ has quit IRC | 05:58 | |
| Haomeng | naohirot: So can we add more pending status, such as 'SOFT_REBOOTING' status as indicatior for user? | 05:59 |
| naohirot | Haomeng: Yes, that's make sense. | 05:59 |
| Haomeng | naohirot: ok, I will add comments, and wait more response for this issue, and we can get solutions I think, dont worry | 05:59 |
| naohirot | Haomeng: actually, current power status are | 06:00 |
| Haomeng | naohirot: ? | 06:00 |
| naohirot | Haomeng: just second | 06:01 |
| Haomeng | naohirot: yes | 06:01 |
| naohirot | Haomeng: POWER_ON, POWER_OFF,REBOOT, POWER_OFF_SOFT, REBOOT_SOFT or INJECT_NMI | 06:01 |
| Haomeng | naohirot: so for our status, we missed some 'PENDING' status | 06:01 |
| naohirot | Haomeng: I incorporated rameshg87's advice | 06:02 |
| naohirot | Haomeng: so power status will be | 06:02 |
| Haomeng | naohirot: can not find ramesh comments, which revision? | 06:03 |
| naohirot | Haomeng: POWER_ON, POWER_OFF,REBOOT, POWER_OFF_SOFT, REBOOT_SOFT, INJECT_NMI, POWER_OFF_SOFT_IN_PROGRESS, SOFT_REBOOTING | 06:03 |
| naohirot | Haomeng: something like that? | 06:03 |
| naohirot | Haomeng: there are a lot :) | 06:04 |
| Haomeng | naohirot: I think that is good, but not sure if other guys has more throughts:) | 06:04 |
| Haomeng | naohirot: and IMO, we need add more 'PENDING' status for existing power actions, such as power on/off | 06:04 |
| Haomeng | naohirot: we need have power-on-inprogress, and power-off-inprogress for user to get understand the pending status which is | 06:05 |
| naohirot | Haomeng: I see, PENDING is good, we can use for all of them by one word. | 06:05 |
| Haomeng | naohirot: yes, i think so:) | 06:06 |
| naohirot | Haomeng: May be Jim may have expected just POWER_ERROR and POWER_PENDING | 06:07 |
| Haomeng | naohirot: yes:) | 06:07 |
| Haomeng | naohirot: dont worry | 06:07 |
| naohirot | Haomeng: we had a good discussion :) | 06:08 |
| Haomeng | naohirot: :) | 06:08 |
| *** yog__ has quit IRC | 06:12 | |
| *** Marga_ has quit IRC | 06:17 | |
| saripurigopi | naohirot : hi | 06:17 |
| *** Marga_ has joined #openstack-ironic | 06:17 | |
| saripurigopi | rameshg87 : hii | 06:17 |
| naohirot | saripurigopi: good morning ;) | 06:17 |
| openstackgerrit | Merged openstack/ironic: Do not use "private" attribute in AuthTokenMiddleware https://review.openstack.org/195000 | 06:18 |
| openstackgerrit | Merged openstack/ironic: Use oslo_log https://review.openstack.org/195989 | 06:20 |
| *** dasm|afk is now known as dasm | 06:23 | |
| *** chenglch has quit IRC | 06:25 | |
| naohirot | Haomeng: I got a solution, GET /v1/nodes/(node_ident)/states | 06:33 |
| naohirot | Haomeng: this rest api returns Return type: | 06:33 |
| naohirot | NodeStates | 06:33 |
| naohirot | Haomeng: NodeStates has last_error field. http://docs.openstack.org/developer/ironic/webapi/v1.html#nodes | 06:34 |
| naohirot | Haomeng: so we can avoid explosion of power states. | 06:35 |
| Haomeng | naohirot: yes, we can call htis api to get status | 06:36 |
| Haomeng | naohirot: you mean to set last_error if we soft-power fail? | 06:37 |
| naohirot | Haomeng: If we introduced POWER_ERROR, which is very problematic. | 06:37 |
| Haomeng | naohirot: yes | 06:38 |
| naohirot | Haomeng: because if SOFT POWER OFF failed, still node is POWER_ON state, not POWER_ERROR state. | 06:38 |
| naohirot | Haomeng: yes, if SOFT POWER OFF failed, we put error message into last_error. | 06:39 |
| naohirot | Haomeng: then user issues GET /v1/nodes/(node_ident)/states to know what happened. | 06:40 |
| *** mitchjameson has joined #openstack-ironic | 06:40 | |
| Haomeng | naohirot: yes, but last_error is common field, it will be overwritten by other error | 06:41 |
| naohirot | Haomeng: Yes, it is likely to happen | 06:41 |
| naohirot | Haomeng: I believe user trys to power (hard) off if power soft off failed. | 06:43 |
| naohirot | Haomeng: so the state of node has to be remained in POWER ON. | 06:43 |
| rameshg87 | naohirot: saripurigopi: Haomeng: morning. sorry was away from my desk | 06:45 |
| saripurigopi | rameshg87: o/ | 06:45 |
| naohirot | rameshg87: good morning | 06:45 |
| Haomeng | rameshg87: good morning:) | 06:49 |
| *** rameshg87 is now known as rameshg87-lunch | 06:51 | |
| *** davideagnello has joined #openstack-ironic | 06:52 | |
| *** yuanying_ has quit IRC | 06:53 | |
| *** athomas has joined #openstack-ironic | 06:53 | |
| *** yuanying_ has joined #openstack-ironic | 06:54 | |
| *** davideagnello has quit IRC | 06:57 | |
| *** mitchjameson has quit IRC | 07:13 | |
| *** yog__ has joined #openstack-ironic | 07:17 | |
| *** amotoki has joined #openstack-ironic | 07:17 | |
| *** stegranet has quit IRC | 07:20 | |
| *** lsmola has joined #openstack-ironic | 07:26 | |
| *** natorious is now known as zz_natorious | 07:32 | |
| *** jistr has joined #openstack-ironic | 07:32 | |
| *** ifarkas has joined #openstack-ironic | 07:35 | |
| *** Marga_ has quit IRC | 07:37 | |
| *** kfox1111 has quit IRC | 07:40 | |
| *** Marga_ has joined #openstack-ironic | 07:46 | |
| *** stegranet has joined #openstack-ironic | 07:59 | |
| *** lucasagomes has joined #openstack-ironic | 08:01 | |
| *** mitchjameson has joined #openstack-ironic | 08:01 | |
| *** dtantsur|afk is now known as dtantsur | 08:03 | |
| openstackgerrit | Merged openstack/bifrost: Add additional note dependent upon os_ironic_facts https://review.openstack.org/195111 | 08:03 |
| dtantsur | Morning Ironic! | 08:03 |
| *** bethelwell has joined #openstack-ironic | 08:05 | |
| lucasagomes | dtantsur, morning | 08:09 |
| dtantsur | lucasagomes, o/ you're earlier than me today :) | 08:09 |
| lucasagomes | yeah I've slept a lot on the weekend because I was a under the weather | 08:10 |
| lucasagomes | heh | 08:10 |
| *** Marga_ has quit IRC | 08:10 | |
| *** yuanying_ has quit IRC | 08:12 | |
| dtantsur | lucasagomes, good for you, I wasn't sleeping much this weekend :( | 08:12 |
| lucasagomes | dtantsur, hah I would say good for you for it too | 08:12 |
| lucasagomes | I mean you've enjoyed it more I bet | 08:13 |
| dtantsur | heh | 08:13 |
| zhenguo | morning dtantsur, lucasagomes | 08:16 |
| dtantsur | o/ | 08:16 |
| lucasagomes | zhenguo, hi there, good morning | 08:17 |
| zhenguo | o/ | 08:17 |
| *** MattMan has joined #openstack-ironic | 08:18 | |
| *** Marga_ has joined #openstack-ironic | 08:22 | |
| *** puranamr has joined #openstack-ironic | 08:22 | |
| *** dguerri` is now known as dguerri | 08:23 | |
| *** puranamr has quit IRC | 08:25 | |
| *** mitchjameson has quit IRC | 08:26 | |
| *** Marga_ has quit IRC | 08:27 | |
| *** stegranet1 has joined #openstack-ironic | 08:30 | |
| *** stegranet has quit IRC | 08:31 | |
| *** stegranet1 is now known as stegranet | 08:31 | |
| *** derekh has joined #openstack-ironic | 08:33 | |
| *** Marga_ has joined #openstack-ironic | 08:35 | |
| *** rkhanbikov has joined #openstack-ironic | 08:41 | |
| *** davideagnello has joined #openstack-ironic | 08:41 | |
| openstackgerrit | Merged openstack/ironic: Log configuration options on ironic-conductor startup https://review.openstack.org/196256 | 08:42 |
| *** Marga_ has quit IRC | 08:45 | |
| *** rkhanbikov has quit IRC | 08:45 | |
| *** Marga_ has joined #openstack-ironic | 08:45 | |
| *** Marga_ has quit IRC | 08:45 | |
| *** stegranet has quit IRC | 08:45 | |
| *** puranamr has joined #openstack-ironic | 08:45 | |
| *** stegranet has joined #openstack-ironic | 08:45 | |
| *** Marga_ has joined #openstack-ironic | 08:46 | |
| *** davideagnello has quit IRC | 08:46 | |
| *** kbyrne has quit IRC | 08:47 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Switch to oslo.service https://review.openstack.org/195008 | 08:49 |
| *** puranamr has quit IRC | 08:50 | |
| *** kbyrne has joined #openstack-ironic | 08:53 | |
| *** ndipanov has joined #openstack-ironic | 08:54 | |
| *** bethelwe_ has joined #openstack-ironic | 09:00 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Update ironic.conf.sample https://review.openstack.org/196002 | 09:01 |
| *** pelix has joined #openstack-ironic | 09:03 | |
| *** romcheg has joined #openstack-ironic | 09:03 | |
| *** bethelwell has quit IRC | 09:03 | |
| *** joe__ has joined #openstack-ironic | 09:06 | |
| *** joe__ has quit IRC | 09:10 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic-specs: Enhance Power Interface for Soft Reboot and NMI https://review.openstack.org/186700 | 09:11 |
| *** bethelwe_ has quit IRC | 09:14 | |
| *** bethelwell has joined #openstack-ironic | 09:17 | |
| *** bethelwell has quit IRC | 09:19 | |
| *** bethelwell has joined #openstack-ironic | 09:19 | |
| *** puranamr has joined #openstack-ironic | 09:26 | |
| sambetts | Morning all o/ | 09:31 |
| *** puranamr has quit IRC | 09:32 | |
| dtantsur | sambetts, morning | 09:33 |
| sambetts | dtantsur: Hows it going? | 09:33 |
| dtantsur | pretty fine, except for inspector gate | 09:34 |
| *** Marga_ has quit IRC | 09:34 | |
| *** bethelwe_ has joined #openstack-ironic | 09:44 | |
| *** athomas has quit IRC | 09:46 | |
| *** romcheg has quit IRC | 09:47 | |
| *** bethelwell has quit IRC | 09:48 | |
| *** naohirot has quit IRC | 09:50 | |
| *** athomas has joined #openstack-ironic | 09:52 | |
| *** openstackgerrit has quit IRC | 09:53 | |
| *** puranamr has joined #openstack-ironic | 09:53 | |
| *** openstackgerrit has joined #openstack-ironic | 09:54 | |
| lucasagomes | sambetts, morning | 09:54 |
| *** bethelwe_ has quit IRC | 09:54 | |
| sambetts | lucasagomes: Morning o/ | 09:55 |
| *** bethelwell has joined #openstack-ironic | 09:55 | |
| sambetts | dtantsur: Something broke? | 09:55 |
| dtantsur | sambetts, yep, https://bugs.launchpad.net/devstack/+bug/1469160 | 09:56 |
| openstack | Launchpad bug 1469160 in devstack "Cinder tries to start even if it wasn't requested - and fails" [Undecided,In progress] - Assigned to Dmitry Tantsur (divius) | 09:56 |
| *** Marga_ has joined #openstack-ironic | 09:57 | |
| *** puranamr has quit IRC | 09:57 | |
| sambetts | ah, I better not pull devstack on my testbed then :-P | 09:58 |
| *** uschreiber_ has joined #openstack-ironic | 10:03 | |
| *** uschreiber_ has quit IRC | 10:05 | |
| *** romcheg has joined #openstack-ironic | 10:11 | |
| openstackgerrit | Merged stackforge/proliantutils: Adding RIS support for virtual media interfaces https://review.openstack.org/194570 | 10:13 |
| *** romcheg has quit IRC | 10:13 | |
| *** romcheg has joined #openstack-ironic | 10:18 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Move Python ramdisk code out of tree https://review.openstack.org/196615 | 10:23 |
| *** davideagnello has joined #openstack-ironic | 10:30 | |
| *** davideagnello has quit IRC | 10:34 | |
| *** Marga_ has quit IRC | 10:40 | |
| dtantsur | sambetts, answered on https://bugs.launchpad.net/ironic-inspector/+bug/1441117 and https://bugs.launchpad.net/bugs/1391865 | 10:40 |
| openstack | Launchpad bug 1441117 in Ironic Inspector "Provide way to append/prepend plugins to processing_hooks w/o overriding the defaults" [Low,Confirmed] | 10:40 |
| openstack | Launchpad bug 1391865 in Ironic Inspector "Client should try to fetch base_url from Keystone " [Low,Confirmed] | 10:40 |
| *** subscope has joined #openstack-ironic | 10:42 | |
| sambetts | dtantsur: awesome :D | 10:42 |
| *** jistr_ has joined #openstack-ironic | 10:49 | |
| *** dtantsur_ has joined #openstack-ironic | 10:49 | |
| *** dtantsur has quit IRC | 10:50 | |
| *** jistr has quit IRC | 10:51 | |
| *** subscope has quit IRC | 10:52 | |
| *** jistr_ has quit IRC | 10:54 | |
| *** dtantsur_ has quit IRC | 10:54 | |
| *** dtantsur has joined #openstack-ironic | 10:59 | |
| *** ramineni has quit IRC | 11:01 | |
| *** jistr_ has joined #openstack-ironic | 11:06 | |
| TheJulia | good morning everyone | 11:07 |
| dtantsur | TheJulia, morning | 11:07 |
| * rameshg87-lunch goes home | 11:08 | |
| *** rameshg87-lunch has quit IRC | 11:08 | |
| *** coolsvap is now known as coolsvap|away | 11:10 | |
| *** romcheg has quit IRC | 11:21 | |
| *** dguerri is now known as dguerri` | 11:28 | |
| *** bethelwell has quit IRC | 11:29 | |
| *** bethelwell has joined #openstack-ironic | 11:29 | |
| *** keekz has joined #openstack-ironic | 11:31 | |
| lucasagomes | TheJulia, good morning | 11:32 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow vendor methods to serve static files https://review.openstack.org/189716 | 11:32 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: WIP: Split GIT downloads and OpenStack CI logic out https://review.openstack.org/196398 | 11:37 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow vendor methods to serve static files https://review.openstack.org/189716 | 11:42 |
| *** saripurigopi has quit IRC | 11:48 | |
| *** dguerri` is now known as dguerri | 11:48 | |
| *** lucasagomes is now known as lucas-hungry | 11:48 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Make functional test importable and stop depending on DIB code https://review.openstack.org/196632 | 11:50 |
| *** trown|outttypeww is now known as trown | 11:53 | |
| openstackgerrit | Merged openstack/ironic: Add IPMI 1.5 support for the ipmitool power driver https://review.openstack.org/195157 | 11:58 |
| *** stegranet1 has joined #openstack-ironic | 11:59 | |
| dtantsur | lucas-hungry, ^^ \o/ | 12:00 |
| *** stegranet has quit IRC | 12:01 | |
| *** stegranet1 has quit IRC | 12:04 | |
| *** dprince has joined #openstack-ironic | 12:06 | |
| *** davideagnello has joined #openstack-ironic | 12:18 | |
| *** romcheg has joined #openstack-ironic | 12:23 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: WIP: Split GIT downloads and OpenStack CI logic out https://review.openstack.org/196398 | 12:23 |
| *** davideagnello has quit IRC | 12:23 | |
| *** Marga_ has joined #openstack-ironic | 12:25 | |
| *** dguerri is now known as dguerri` | 12:28 | |
| dtantsur | lucas-hungry, mind reviewing https://review.openstack.org/#/c/191710/ ? we already have futurist public release | 12:28 |
| *** dguerri` is now known as dguerri | 12:32 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: WIP: Split GIT downloads and OpenStack CI logic out https://review.openstack.org/196398 | 12:48 |
| *** lucas-hungry is now known as lucasagomes | 12:51 | |
| lucasagomes | dtantsur, w00t... will do | 12:51 |
| *** ukalifon1 has joined #openstack-ironic | 13:02 | |
| *** ukalifon3 has joined #openstack-ironic | 13:03 | |
| *** krtaylor has quit IRC | 13:04 | |
| *** smoriya has quit IRC | 13:05 | |
| *** ukalifon1 has quit IRC | 13:07 | |
| *** jlvillal has quit IRC | 13:08 | |
| lucasagomes | dtantsur, done, pretty good. Only one thing inline | 13:11 |
| lucasagomes | dtantsur, I think that reusing workers_pool_size for setting the max number of parallel tasks running at the same time is wrong | 13:11 |
| *** puranamr has joined #openstack-ironic | 13:11 | |
| dtantsur | lucasagomes, I had some reason to do it, don't remember right now, but probably we should just change all of tasks/jobs/etc to use one thread pool | 13:12 |
| dtantsur | lucasagomes, otherwise it's just too much settings affecting concurrency | 13:12 |
| lucasagomes | dtantsur, yeah... we could try to simplify | 13:13 |
| lucasagomes | but like setting that to max is wrong because, because 1 peridioc task could spawn more workers from that same pool | 13:13 |
| dtantsur | lucasagomes, yeah, right, why do you think we need 2 thread pools for Ironic? | 13:13 |
| lucasagomes | which now is max 8 | 13:13 |
| *** zhenguo has quit IRC | 13:13 | |
| lucasagomes | dtantsur, I don't htink we need 2 | 13:13 |
| lucasagomes | I think we need to set some maximum number of parallel periodic tasks running at the same time | 13:14 |
| lucasagomes | max != pool size | 13:14 |
| dtantsur | lucasagomes, it defaults to 100 btw: https://github.com/openstack/ironic/blob/master/ironic/conductor/manager.py#L128 | 13:14 |
| dtantsur | lucasagomes, no sorry, max == pool size | 13:14 |
| dtantsur | IIRC at least | 13:14 |
| lucasagomes | dtantsur, if we set 100 periodic tasks | 13:14 |
| dtantsur | lucasagomes, "If the pool is currently at capacity, spawn will block until one of the running greenthreads completes its task and frees up a slot. | 13:15 |
| dtantsur | " | 13:15 |
| lucasagomes | and they all run in paralell at the same time we will starve workers from doing other things | 13:15 |
| dtantsur | lucasagomes, correct, that's how pool work | 13:15 |
| dtantsur | that's the whole idea of pools, now you try to overcome it by demanding one more pool for tasks :) | 13:15 |
| lucasagomes | I think ironic won't wait, it will raise NoFreeConductorWorker | 13:16 |
| *** krtaylor has joined #openstack-ironic | 13:17 | |
| dtantsur | lucasagomes, we also limit number of threads run by periodic tasks to 8 per each task (periodic_max_workers setting) | 13:17 |
| dtantsur | I guess you're confusing these 2 | 13:18 |
| lucasagomes | dtantsur, hmm /me thinking | 13:18 |
| lucasagomes | yeah so we have only 1 pool right? default 100 | 13:18 |
| dtantsur | oh, this NoFreeConductorWorker is a non-sense, how came we even have it? Oo | 13:18 |
| dtantsur | we should clean up our async code definitely.... | 13:18 |
| lucasagomes | right now it's used for: async tasks (e.g power state changes, deploying etc...) and also periodic tasks can use up to 8 of those workers to do something | 13:19 |
| lucasagomes | with futurist we will use workers from that same pool that will be the periodic tasks | 13:19 |
| *** jlvillal has joined #openstack-ironic | 13:19 | |
| dtantsur | oh my god, this NoFreeConductorWorker even issues 503.... | 13:19 |
| lucasagomes | dtantsur, yeah it's a bit messy that part | 13:20 |
| dtantsur | so we have 2 sources of random failures actually... | 13:20 |
| lucasagomes | so I think we may want to cap the number of parallel periodic tasks running at the same time so we don't starve workers from the conductor to do the rest | 13:21 |
| dtantsur | lucasagomes, why? parallel tasks are just as well important | 13:21 |
| lucasagomes | yes but if max == pool size the conductor will become unusable | 13:21 |
| lucasagomes | it will only be able to run periodc tasks | 13:21 |
| lucasagomes | would be good to have a cap so people can tune it | 13:22 |
| dtantsur | lucasagomes, it will finish running periodic tasks, and will process other requests | 13:22 |
| lucasagomes | dtantsur, right, but currently it will raise that exception and request will fail | 13:22 |
| *** mjturek1 has joined #openstack-ironic | 13:22 | |
| dtantsur | lucasagomes, right, it is broken. Already is, it's not broken by my future change | 13:23 |
| dtantsur | the same thing will happen if a user issues >100 requests | 13:23 |
| lucasagomes | yes | 13:23 |
| lucasagomes | well perhaps one could just increase that number... | 13:23 |
| * lucasagomes thinks | 13:23 | |
| dtantsur | exactly | 13:24 |
| lucasagomes | yeah perhaps that's the way if should go for now :-/ | 13:24 |
| dtantsur | lucasagomes, and now I realize that we need ironicclient retry for error 503 as well... | 13:26 |
| lucasagomes | 503!? | 13:26 |
| lucasagomes | ouch | 13:26 |
| dtantsur | lucasagomes, NoFreeConductorWorker is error 503 | 13:26 |
| dtantsur | more vague errors for gods of vague errors \o/ | 13:26 |
| lucasagomes | yeah | 13:26 |
| lucasagomes | not great | 13:28 |
| * dtantsur writes a patch | 13:31 | |
| *** puranamr has quit IRC | 13:31 | |
| *** rloo has joined #openstack-ironic | 13:31 | |
| *** zz_jgrimm is now known as jgrimm | 13:33 | |
| *** absubram has quit IRC | 13:36 | |
| *** absubram has joined #openstack-ironic | 13:36 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Also retry on HTTP 503 (service unavailable) https://review.openstack.org/196664 | 13:37 |
| dtantsur | lucasagomes, done ^^ | 13:37 |
| *** mtanino has joined #openstack-ironic | 13:43 | |
| *** Marga_ has quit IRC | 13:51 | |
| *** igordcard_ has quit IRC | 13:55 | |
| *** boris-42 has joined #openstack-ironic | 13:59 | |
| *** amotoki has quit IRC | 13:59 | |
| *** subscope has joined #openstack-ironic | 13:59 | |
| *** saripurigopi has joined #openstack-ironic | 14:00 | |
| *** subscope has quit IRC | 14:04 | |
| *** davideagnello has joined #openstack-ironic | 14:07 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Disambiguate the roles of ci_testing, ci_testing_zuul etc https://review.openstack.org/195759 | 14:10 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Split GIT downloads and OpenStack CI logic out https://review.openstack.org/196398 | 14:10 |
| *** absubram has quit IRC | 14:11 | |
| TheJulia | hmm... rebased again | 14:12 |
| *** davideagnello has quit IRC | 14:12 | |
| *** yuikotakada has joined #openstack-ironic | 14:12 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow vendor methods to serve static files https://review.openstack.org/189716 | 14:12 |
| *** absubram has joined #openstack-ironic | 14:15 | |
| *** absubram has quit IRC | 14:15 | |
| *** r-daneel has joined #openstack-ironic | 14:19 | |
| *** ukalifon3 has quit IRC | 14:23 | |
| *** subscope has joined #openstack-ironic | 14:24 | |
| *** ukalifon has joined #openstack-ironic | 14:29 | |
| NobodyCam | good morning Ironicers | 14:33 |
| dtantsur | NobodyCam, morning | 14:34 |
| NobodyCam | morning dtantsur | 14:34 |
| rloo | Happy Monday ironickers, NobodyCam, dtantsur, TheJulia, lucasagomes :) | 14:36 |
| lucasagomes | rloo, NobodyCam TheJulia good ugt morning! | 14:36 |
| dtantsur | rloo, o/ | 14:36 |
| TheJulia | good morning rloo | 14:37 |
| rloo | hey dtantsur, you ok if we approve this? https://review.openstack.org/#/c/173674/ ? | 14:38 |
| dtantsur | rloo, yep | 14:39 |
| NobodyCam | morning rloo lucasagomes | 14:39 |
| rloo | dtantsur: done! | 14:40 |
| jroll | goooood morning NobodyCam TheJulia dtantsur lucasagomes rloo and anyone else :) | 14:40 |
| dtantsur | jroll, morning! | 14:40 |
| rloo | morning jroll! | 14:40 |
| lucasagomes | jroll, yo! good morning | 14:41 |
| NobodyCam | mornign jroll :) | 14:41 |
| NobodyCam | :-p | 14:41 |
| *** yog__ has quit IRC | 14:42 | |
| NobodyCam | meeting aganda looks light today. call for agenda Items? | 14:54 |
| *** Marga_ has joined #openstack-ironic | 14:55 | |
| *** rameshg87 has joined #openstack-ironic | 14:56 | |
| *** mitchjameson has joined #openstack-ironic | 14:56 | |
| *** pradipta has joined #openstack-ironic | 14:58 | |
| rameshg87 | rloo: hello | 14:58 |
| rloo | hi rameshg87! | 14:58 |
| rameshg87 | rloo: regarding https://review.openstack.org/#/c/194786/1/ironic/drivers/modules/iscsi_deploy.py | 14:59 |
| rameshg87 | rloo: what do we do next ? looks good to me as it is | 14:59 |
| rameshg87 | rloo: but my question is whether we can procliam driver_internal_info['instance'] as a place to store driver internal data related to instance | 15:00 |
| rloo | rameshg87: oh. what do you think we should do? | 15:00 |
| rloo | rameshg87: personally, i'm afraid to use 'instance' right now cuz it seems too generic. | 15:00 |
| rameshg87 | rloo: lucasagomes had a comment we should change to driver_internal_info['disk_layout'] | 15:00 |
| rameshg87 | lucasagomes: ^^ | 15:00 |
| *** ijw has joined #openstack-ironic | 15:00 | |
| rloo | rameshg87: yeah, i prefer 'disk_layout' but the original code was approved with 'instance'. that's why i left it :) | 15:01 |
| lucasagomes | rameshg87, rloo hi yeah. I just think "instance" is too generic for that | 15:01 |
| *** cdearborn has joined #openstack-ironic | 15:01 | |
| lucasagomes | it was a suggestion tho, since instance was used before | 15:01 |
| rameshg87 | rloo: lucasagomes: I was coming to that | 15:01 |
| rameshg87 | couldn't we just provide driver_internal_info['instance'] to the drivers as such (just like we give instance_info) | 15:01 |
| rameshg87 | to store internal attributes related to instance | 15:02 |
| rameshg87 | we could clear that off at the end of deploy | 15:02 |
| rloo | rameshg87: no. or yes. i'm afraid to say explicitly that we do that, cuz then it is sort of an 'api' contract with drivers. | 15:02 |
| rloo | rameshg87: and then it means we should think about what really goes in 'instance', and is it really then a new instance_internal_info. | 15:02 |
| rloo | rameshg87: which might be a good thing. I don't really know. yet. | 15:03 |
| dtantsur | yuikotakada, g'evening/night :) could you remove -1 from https://review.openstack.org/#/c/196050/ if you're ok with it now? | 15:03 |
| lucasagomes | yeah sounds like an instance_internal_info indeed | 15:03 |
| * lucasagomes is in a call 1 sec | 15:03 | |
| rameshg87 | ah may be a new field then .. | 15:04 |
| yuikotakada | dtantsur, hi, g'evening :) OK, I've done | 15:04 |
| rameshg87 | rloo: this is not the first time we have come across such a need for drivers | 15:04 |
| *** absubram has joined #openstack-ironic | 15:04 | |
| rloo | rameshg87: well, that's why we added driver_internal_info :) | 15:04 |
| dtantsur | yuikotakada, thanks! will try to grab someone to review it today.. | 15:04 |
| rloo | rameshg87: so it seems like we have two 'sets' of information: 1. the users provide; 2. what we actually use | 15:04 |
| rameshg87 | but the problem it is related to node rather than an instance | 15:04 |
| rameshg87 | yeah #2 is something we never wants user to know even | 15:05 |
| rloo | rameshg87: sorry, what is 'related to node rather than an instance'? | 15:05 |
| rloo | rameshg87: I mean WHAT? | 15:05 |
| rameshg87 | I meant some information is related to instance which should be cleared at the end of the life of the instance | 15:06 |
| rameshg87 | so rather fitting it into driver_internal_info requires manual care | 15:06 |
| openstackgerrit | Yuiko Takada proposed openstack/ironic-inspector: Migrate to oslo_db https://review.openstack.org/181905 | 15:06 |
| rameshg87 | so if we could get an agreement on instance_internal_info or a driver_internal_info['instance'] or whatever, which is a contract | 15:07 |
| rloo | rameshg87: well, it is all 'manual'? i haven't analyzed what info we have in driver_internal_info. is there a pattern? | 15:07 |
| rameshg87 | well there isn't really :) | 15:08 |
| rloo | rameshg87: aren't you going to have to save some RAID info too? | 15:08 |
| rameshg87 | rloo: but that got moved to a new field | 15:08 |
| rameshg87 | node.raid_config and node.target_raid_config | 15:08 |
| yuikotakada | dtantsur, yeah, please. I hope that will be merged soon, too | 15:09 |
| rloo | rameshg87: ohhh, I hate this. You're making me think, and it is only Monday morning! | 15:10 |
| rameshg87 | rloo: :) I don't want to drag you again into that conversation | 15:10 |
| rameshg87 | rloo: so my proposal is just this | 15:10 |
| rameshg87 | rloo: we can get along with https://review.openstack.org/#/c/194786/1/ironic/drivers/modules/iscsi_deploy.py in it's current way (it's only addressing your comments on the review and not changing anything) | 15:11 |
| rameshg87 | rloo: and may be come up with a proposal for something like a contract (may be node.instance_internal_info) to drivers for storing this information | 15:11 |
| rloo | rameshg87: so you think we need to distinguish 'driver_internal_info' vs 'instance_internal_info'? | 15:12 |
| rloo | rameshg87: what if we had just called it 'internal_info'? | 15:12 |
| rameshg87 | rloo: but some *internal* information needs to be cleared at the end of life of instance | 15:13 |
| rameshg87 | like this one we just had | 15:13 |
| rameshg87 | it will be good if drivers don't need to explicitly take care in removing it and it is automatically done at the end of tear down | 15:14 |
| rloo | rameshg87: I almost think that if we are going to same "some" instance info, why don't we just save all the 'deployinfo' or whatever we call it | 15:14 |
| rameshg87 | rloo: where do mean to save it ? | 15:14 |
| rloo | rameshg87: my only concern with generalizing it (which makes sense) is that odd case where some driver wants to keep the internal instance info around beyond tear down. | 15:14 |
| *** mitchjameson has quit IRC | 15:15 | |
| rameshg87 | rloo: oh, then they should goto driver_internal_info | 15:15 |
| rameshg87 | rloo: we already have that in one of the proposed patches | 15:15 |
| rameshg87 | ipmi_persistent_boot_device | 15:15 |
| rameshg87 | it should persist beyond tear down | 15:15 |
| rloo | rameshg87: I mean, if we're going to save some of the instance info (like root_gb, swap_mb, ephemeral_gb), why not just save all the deploy (in this case, the variable is called i_info) instead of picking out bits to save | 15:16 |
| rameshg87 | rloo: but this is just one implementation of deploy interface | 15:17 |
| rameshg87 | rloo: there are deploy drivers proposed for diskless nodes - it might never have above fields | 15:17 |
| rameshg87 | connecting directly to a iscsi lun | 15:17 |
| rameshg87 | rloo: so I think it will be hard to generalize the fields even | 15:18 |
| rloo | rameshg87: just looking, we have 'is_whole_disk_image', 'instance', 'clean_steps', 'agent_last_heartbeat', 'agent_url', 'root_uuid_or_disk_id' | 15:18 |
| rameshg87 | yes | 15:18 |
| rameshg87 | except root_uuid_or_disk_id, everything above should persist beyond tear down | 15:19 |
| rloo | rameshg87: I don't understand your point about 'there are deploy drivers ... might never have above fields'. what does that have to do with it? they have instance-info that you're arguing they might want to save? | 15:19 |
| rameshg87 | instance as well :) (though name is misleading ) | 15:19 |
| rameshg87 | rloo: they might have. my point was information that different deploy drivers might want to save might be different. | 15:21 |
| rloo | rameshg87: I thought you were arguing for generalizing/using 'instance'. ?? You aren't? I'm confused. | 15:21 |
| rameshg87 | rloo: no. I am not actually :) | 15:21 |
| rloo | rameshg87: and if 'instance' name is misleading, why are we using it? | 15:21 |
| rameshg87 | rloo: I was just suggesting we need one common place for a driver to store all internal info related to an instance (that shouldn't persist beyond tear down) | 15:22 |
| rloo | rameshg87: and yeah, i agree, different drivers might want to save (or not save) different info | 15:22 |
| rloo | rameshg87: what does that have to do with (I think) your argument with having some 'instance' field that gets cleared after teardown? | 15:22 |
| rameshg87 | rloo: ah no. I am sorry if I couldn't convey it right | 15:23 |
| *** e0ne has joined #openstack-ironic | 15:23 | |
| openstackgerrit | Merged openstack/python-ironicclient: Cache negotiated api microversion for server https://review.openstack.org/173674 | 15:24 |
| rameshg87 | rloo: I just meant there are some information that needs to be cleared on tear down (which we saw things like root_uuid_or_disk_id, root_gb, swap_gb, ...) AND some information that needs to persist even after tear down (agent_last_heartbeat, agent_url, ipmi_persistent_boot_device (proposed), etc) | 15:24 |
| rloo | rameshg87: just to step back. wrt that patch, I submitted it cuz the code is wrong. I most likely would have objected to using 'instance' in the first place, but it got approved and I don't have strong-enough opinion to change that now. Unless others want to change it too. | 15:25 |
| *** e0ne is now known as e0ne_ | 15:25 | |
| rameshg87 | rloo: I agree. | 15:25 |
| rloo | rameshg87: yes, I agree. and that info, when it gets saved and when it gets removed, is up to the driver. | 15:25 |
| yuikotakada | good night, ironic :) | 15:25 |
| *** yuikotakada has quit IRC | 15:25 | |
| *** e0ne_ is now known as e0ne | 15:25 | |
| rloo | rameshg87: so if you want it to still be 'instance', you can voice your opinion/discuss with lucasagomes. I'll change it or not change it based on the reviewers' wishes :) | 15:26 |
| rloo | rameshg87: lucasagomes was making a suggestion about that | 15:27 |
| *** david-ly_ is now known as david-lyle | 15:27 | |
| lucasagomes | rloo, rameshg87 I haven't read the scrollback. It was just a suggestion because I think it was more accurate | 15:27 |
| * lucasagomes reads the scrollback | 15:27 | |
| rameshg87 | lucasagomes: yeah, I agree. but I was just talking to rloo that if we want we could generalize it to a field (something like node.instance_internal_info) which is cleared automatically by the conductor at the end of deploy. I am voting up for it. | 15:29 |
| rameshg87 | lucasagomes: but if we feel we still don't have a reason for doing it now, I am +1 for changing it to driver_internal_info['disk_layout'] | 15:30 |
| *** davideagnello has joined #openstack-ironic | 15:31 | |
| rameshg87 | lucasagomes: which you and rloo anyway agree upon | 15:31 |
| * rameshg87 feels I managed to confuse both of them | 15:31 | |
| lucasagomes | rameshg87, right, yeah I think it would be beneficial to have a field that gets cleared up at the end | 15:31 |
| lucasagomes | we already have plenty of keys that could be cleanered up | 15:31 |
| lucasagomes | root_uuid_or_disk_id, is_whole_disk_image, that "instance" now | 15:31 |
| rameshg87 | root_uuid_or_disk_id is one | 15:32 |
| rameshg87 | yeah | 15:32 |
| lucasagomes | s/now// it was already there | 15:32 |
| lucasagomes | rameshg87, but it seems that if we opt of having such field, it should be done outside the patch that rloo proposed | 15:33 |
| rameshg87 | lucasagomes: yes | 15:33 |
| lucasagomes | which is just fixing something else | 15:33 |
| rameshg87 | yeah | 15:33 |
| lucasagomes | ack :-) | 15:33 |
| lucasagomes | rloo, I would be +1 to change to disk_layout, but if you prefer to keep instance to minimize the changes I'm good too | 15:33 |
| rloo | lucasagomes: I actually prefer 'disk_layout' but others approved 'instance' so I was hesitant to change. | 15:34 |
| rloo | rameshg87: are you good with 'disk_layout'? | 15:34 |
| rameshg87 | rloo: yes I am if we decide we never gonna have something like node.instance_internal_info | 15:35 |
| rameshg87 | rloo: but until then I am of the opinion to keep it as it is. wdyt ? | 15:35 |
| rloo | rameshg87: we cannot decide that now. 'never' isn't guaranteed | 15:35 |
| rloo | lucasagomes: rameshg87 likes 'instance' so I'm going to leave it unless others disagree too. | 15:36 |
| lucasagomes | ++ | 15:36 |
| lucasagomes | ack | 15:36 |
| openstackgerrit | Merged openstack/ironic: Switch to oslo.service https://review.openstack.org/195008 | 15:36 |
| rameshg87 | rloo: yeah, so can we just keep driver_internal_info['instance'] for now (same as your patch), and get along and decide on the next this | 15:36 |
| rloo | but yeah, lucasagomes, we're right :) | 15:36 |
| rameshg87 | lucasagomes: rloo: +1, you were right | 15:36 |
| rloo | rameshg87: I am held hostage by you reviewers :) | 15:37 |
| * rameshg87 relaxes it's not only me ;-) | 15:37 | |
| lucasagomes | (-: | 15:37 |
| rloo | rameshg87: ha ha | 15:37 |
| *** achanda has joined #openstack-ironic | 15:37 | |
| * rameshg87 brb | 15:38 | |
| *** rameshg87 is now known as rameshg870brb | 15:38 | |
| *** rameshg870brb is now known as rameshg87 | 15:41 | |
| *** davideagnello has quit IRC | 15:42 | |
| *** mitchjameson has joined #openstack-ironic | 15:42 | |
| *** zz_natorious is now known as natorious | 15:44 | |
| *** Marga_ has quit IRC | 15:44 | |
| *** Sukhdev has joined #openstack-ironic | 15:46 | |
| *** e0ne has quit IRC | 15:46 | |
| *** e0ne has joined #openstack-ironic | 15:48 | |
| *** radek__ has quit IRC | 15:48 | |
| NobodyCam | just checking on sub-team status...any updates for the white board? | 15:48 |
| dtantsur | oh, bug stats | 15:52 |
| *** e0ne is now known as e0ne_ | 15:53 | |
| *** mitchjameson has quit IRC | 15:53 | |
| *** davideagnello has joined #openstack-ironic | 15:53 | |
| devananda | g'morning, all | 15:55 |
| jroll | thanks for reminder NobodyCam | 15:55 |
| *** e0ne_ is now known as e0ne | 15:55 | |
| jroll | ohai devananda :) | 15:55 |
| rloo | dtantsur: do you think it is worth discussing the is_hostname_safe check at the meeting? https://review.openstack.org/#/c/193587/ ? | 15:55 |
| dtantsur | devananda, morning | 15:55 |
| rloo | morning devananda | 15:56 |
| dtantsur | rloo, hmm maybe | 15:56 |
| *** trown is now known as trown|lunch | 15:56 | |
| rloo | dtantsur: I feel bad giving Chris the runaround, and it seems like we should just decide what to do | 15:56 |
| NobodyCam | mornign devananda | 15:56 |
| jroll | rloo: dtantsur +1 | 15:57 |
| rloo | jroll: ok, will add it | 15:57 |
| jroll | I thought we were trying to relax the names but am not sure | 15:57 |
| *** romcheg has quit IRC | 15:57 | |
| rloo | jroll: i thought so too but devananda doesn't seem to want it | 15:57 |
| *** derekh has quit IRC | 15:58 | |
| jroll | rloo: where is that discussion, here in irc or? | 15:58 |
| *** davideagnello has quit IRC | 15:58 | |
| jroll | disclaimer: I'm still catching up from vacation :P | 15:58 |
| rloo | jroll: yeah, i think it was in irc. devananda had some good reasons. | 15:58 |
| rloo | jroll: will see if i can dig it up and add to the meeting agenda | 15:59 |
| devananda | rloo: doesn't want what? | 15:59 |
| * jroll will look too | 15:59 | |
| jroll | devananda: relaxed node.name validation | 15:59 |
| rloo | devananda: freeform names for nodes. the unicode stuff | 15:59 |
| *** jistr_ has quit IRC | 15:59 | |
| jroll | oh yeah, unicode sounds like a horrible idea | 15:59 |
| rloo | devananda: cuz of requests | 15:59 |
| devananda | right | 16:00 |
| jroll | but other than that is there opposition to relaxing it? | 16:00 |
| devananda | the node.name field is usable as a resource identifier, eg. in the URL | 16:00 |
| jroll | (or any push to relax it at all) | 16:00 |
| devananda | which makes certain characters unacceptable | 16:00 |
| lucasagomes | devananda, morning | 16:01 |
| jroll | devananda: https://www.panic.com/blog/the-worlds-first-emoji-domain/ | 16:01 |
| jroll | :P | 16:01 |
| *** mitchjameson has joined #openstack-ironic | 16:02 | |
| *** puranamr has joined #openstack-ironic | 16:02 | |
| devananda | jroll: :) | 16:03 |
| lucasagomes | jroll, lol | 16:03 |
| rloo | jroll: you're a source of ... information :D | 16:04 |
| jlvillal | Good morning Ironic | 16:04 |
| lucasagomes | jlvillal, morning | 16:04 |
| jroll | rloo: hah | 16:04 |
| NobodyCam | morning jlvillal | 16:04 |
| jlvillal | :) | 16:04 |
| NobodyCam | jlvillal: how many days now? | 16:05 |
| devananda | jroll: I didn't say "makes unicode unacceptable". but this: https://tools.ietf.org/html/rfc3986#page-12 | 16:05 |
| jlvillal | NobodyCam, 2 more work days :) | 16:05 |
| jlvillal | Which I guess is just two more days. | 16:06 |
| NobodyCam | :) | 16:06 |
| jroll | devananda: yeah, I'm fine with not relaxing things, I just thought there was some push for that and this was reverse of that | 16:06 |
| *** romcheg has joined #openstack-ironic | 16:07 | |
| devananda | jroll: afaik, this is just a result of rally guys running into inconsistency across projects and wanting to clean things up | 16:07 |
| lucasagomes | devananda, do we have already some sort of agenda for the midcycle? | 16:07 |
| jroll | devananda: btw, ironic/neutron meeting in meeting-4 in case you forgot :) | 16:07 |
| devananda | lucasagomes: nope | 16:07 |
| *** Marga_ has joined #openstack-ironic | 16:07 | |
| lucasagomes | or it will be decided there? | 16:07 |
| devananda | jroll: blarg. thanks. mondays at 9am is terrible | 16:07 |
| jroll | devananda: yeah, I guess I just don't care much about this being 100% hostname safe or whatever... it's on the agenda, let's talk at 10 about it :) | 16:07 |
| lucasagomes | jroll, ++ to talk about it at the meeting | 16:08 |
| devananda | ++ | 16:08 |
| *** rwsu has joined #openstack-ironic | 16:08 | |
| *** puranamr has quit IRC | 16:12 | |
| *** radek__ has joined #openstack-ironic | 16:12 | |
| *** meghal has joined #openstack-ironic | 16:12 | |
| *** puranamr has joined #openstack-ironic | 16:13 | |
| *** lazy_prince has joined #openstack-ironic | 16:13 | |
| rameshg87 | dtantsur: hi | 16:17 |
| dtantsur | rameshg87, o/ | 16:17 |
| rameshg87 | dtantsur: I didn't get where it should str(e) here in https://review.openstack.org/#/c/196003/1/ironic/common/raid.py | 16:17 |
| dtantsur | rameshg87, eek, sorry, I meant "instead of e.message" | 16:18 |
| *** meghal has left #openstack-ironic | 16:18 | |
| rameshg87 | e.message is already a readable string as per http://python-jsonschema.readthedocs.org/en/latest/errors/ | 16:18 |
| rameshg87 | dtantsur: http://python-jsonschema.readthedocs.org/en/latest/errors/#jsonschema.exceptions.ValidationError.message | 16:18 |
| dtantsur | rameshg87, well, for standard python exceptions e.message is deprecated. you can probably ignore it for 3rdparty libraries, but "str(exc)" is kind of a pattern already :) | 16:19 |
| rameshg87 | dtantsur: oh okay. but since it's documented explicitly about e.message, I think it should be okay to go ahead with that. wdyt ? | 16:20 |
| dtantsur | yeah, fine with me | 16:20 |
| rameshg87 | okay, thanks | 16:20 |
| *** puranamr_ has joined #openstack-ironic | 16:21 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add node fields for raid configuration https://review.openstack.org/155230 | 16:21 |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Refactor check_allow_management_verbs https://review.openstack.org/196251 | 16:21 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Change release model to independent releases https://review.openstack.org/185171 | 16:21 |
| *** beekneemech has joined #openstack-ironic | 16:21 | |
| jroll | ^^ can we land this yet? :) | 16:21 |
| rloo | jroll: I was tempted to add that to the meeting agenda, but I've already added two items :) | 16:22 |
| jroll | rloo: then I'm going to add it :) | 16:23 |
| rloo | jroll: as long as they are *after* mine! | 16:23 |
| *** puranamr has quit IRC | 16:23 | |
| jroll | :D | 16:23 |
| rloo | jroll: just kidding, I don't really care. | 16:23 |
| *** bnemec has quit IRC | 16:23 | |
| *** puranamr_ has quit IRC | 16:26 | |
| *** puranamr has joined #openstack-ironic | 16:26 | |
| *** puranamr_ has joined #openstack-ironic | 16:27 | |
| *** puranamr has quit IRC | 16:27 | |
| *** puranamr has joined #openstack-ironic | 16:28 | |
| *** puranamr_ has quit IRC | 16:32 | |
| *** radek_ has joined #openstack-ironic | 16:33 | |
| *** radek__ has quit IRC | 16:34 | |
| *** krtaylor has quit IRC | 16:35 | |
| *** ifarkas has quit IRC | 16:36 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add RAIDInterface for RAID configuration https://review.openstack.org/196003 | 16:40 |
| *** dguerri is now known as dguerri` | 16:40 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add RPCAPIs for RAID configuration https://review.openstack.org/196006 | 16:40 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add APIs for RAID configuration https://review.openstack.org/196007 | 16:40 |
| *** maurosr has quit IRC | 16:41 | |
| *** radek_ has quit IRC | 16:41 | |
| * lucasagomes brb will be back for the meeting | 16:43 | |
| *** jgrimm has quit IRC | 16:43 | |
| *** athomas has quit IRC | 16:46 | |
| *** radek_ has joined #openstack-ironic | 16:47 | |
| *** maurosr has joined #openstack-ironic | 16:49 | |
| *** zz_jgrimm has joined #openstack-ironic | 16:49 | |
| *** e0ne has quit IRC | 16:59 | |
| devananda | reminder: meeting starting in #openstack-meeting-3 | 17:01 |
| *** mitchjameson has quit IRC | 17:03 | |
| *** krtaylor has joined #openstack-ironic | 17:04 | |
| *** davideagnello has joined #openstack-ironic | 17:07 | |
| *** bethelwell has quit IRC | 17:07 | |
| *** trown|lunch is now known as trown | 17:07 | |
| *** lsmola has quit IRC | 17:09 | |
| openstackgerrit | Merged openstack/ironic: Save disk layout information when deploying https://review.openstack.org/194786 | 17:10 |
| *** saripurigopi has quit IRC | 17:11 | |
| *** achanda has quit IRC | 17:17 | |
| *** thiagop has joined #openstack-ironic | 17:20 | |
| *** afaranha has joined #openstack-ironic | 17:21 | |
| *** dontalton has joined #openstack-ironic | 17:32 | |
| *** Marga_ has quit IRC | 17:32 | |
| *** Marga_ has joined #openstack-ironic | 17:32 | |
| *** Sukhdev has quit IRC | 17:46 | |
| *** achanda has joined #openstack-ironic | 17:46 | |
| *** lazy_prince has quit IRC | 17:47 | |
| *** pelix has quit IRC | 17:51 | |
| *** abrito has joined #openstack-ironic | 17:54 | |
| jlvillal | krtaylor, Are you interested in working on functional testing? I'm delighted to get all the help I can! :) | 17:55 |
| * jlvillal thinks lucasagomes replied in wrong channel ;) | 17:56 | |
| krtaylor | jlvillal, I would be if I could stuff it in, my day job is keeping me crazy busy right now - but, my team does have test case writing experience | 17:57 |
| * jlvillal thinks jlvillal doesn't know what he is talking about | 17:57 | |
| jlvillal | krtaylor, Great! :) | 17:57 |
| krtaylor | jlvillal, I am hoping I can free up in a few weeks and we can start a CI discussion | 17:57 |
| krtaylor | maurosr, mjturek1 ^^^^ re: test cases | 17:58 |
| jlvillal | krtaylor, Okay. If you want to start on things before I get back, feel free. So far only openstack/python-ironicclient is setup for functional testing. | 17:58 |
| lucasagomes | me? | 17:58 |
| jlvillal | lucasagomes, I was confused! | 17:58 |
| lucasagomes | jlvillal, :-D ack no worries | 17:58 |
| jlvillal | JoshNang, I wouldn't mind your feedback on: https://review.openstack.org/196251 Sort of related to your 'fail' patch | 18:00 |
| JoshNang | jlvillal: ah yeah i was looking at that earlier, got distracted | 18:00 |
| * rameshg87 runs to bed to sleep | 18:00 | |
| rameshg87 | good night folks | 18:00 |
| NobodyCam | night rameshg87 | 18:00 |
| jlvillal | JoshNang, Thanks. | 18:00 |
| devananda | rameshg87: good night! | 18:00 |
| NobodyCam | brb | 18:01 |
| jlvillal | rameshg87, Ciao | 18:01 |
| *** rameshg87 has quit IRC | 18:01 | |
| * dtantsur is going too | 18:01 | |
| dtantsur | g'night! | 18:01 |
| lucasagomes | dtantsur, night | 18:01 |
| jlvillal | dtantsur, Paka | 18:01 |
| dtantsur | :D | 18:01 |
| * jlvillal is not sure he spelled 'paka' correctly | 18:01 | |
| lucasagomes | JoshNang, devananda right yeah... I'm ++ for abort too | 18:01 |
| lucasagomes | I think it describes it better | 18:01 |
| dtantsur | jlvillal, пока! | 18:02 |
| lucasagomes | JoshNang, devananda re breaking the lock. You think that we should forcely break it if users ask for "abort" ? | 18:02 |
| JoshNang | so does deployfail automatically go to deleting? or does nova move it that way? | 18:02 |
| *** dtantsur is now known as dtantsur|afk | 18:02 | |
| devananda | JoshNang: aiui, nova exposes the failed instance to the user. the user may choose to 'rebuild' or 'delete' at that point | 18:03 |
| lucasagomes | JoshNang, I will give it a go tomorrow to confirm | 18:03 |
| devananda | lucasagomes: i dont think this has to do with breaking the lock | 18:03 |
| devananda | lucasagomes: the state machine currently has some unrecoverable states that dont hold any locks | 18:03 |
| devananda | lucasagomes: oh, and that never time out either | 18:04 |
| lucasagomes | devananda, if it's DEPLOYING it has a lock, so we can't change the state to fail | 18:04 |
| lucasagomes | without breaking it | 18:04 |
| JoshNang | lucasagomes: sweet. fwiw we don't run this downstream yet, so i'm not sure how well it'll work yet. i wanna test in devstack, but it was a quick patch before i took off for the day fri | 18:04 |
| lucasagomes | JoshNang, yup, yeah I got that impression | 18:04 |
| devananda | lucasagomes, JoshNang: I think the issue is cleaning doesn't have cleanwait, tbh | 18:04 |
| lucasagomes | but it's great | 18:04 |
| lucasagomes | the goal is good | 18:04 |
| devananda | *the issue in my case. there could certainly be other issues | 18:05 |
| jroll | JoshNang: devananda: when nova sees a build failure (e.g. DEPLOYFAIL) it calls destroy() to clean it up or something like that | 18:05 |
| JoshNang | jroll: k that's what i thought. counts a reschedule and tries another node, right? | 18:05 |
| jroll | or when it reschedules a failure, I should say | 18:05 |
| jroll | yep | 18:05 |
| jroll | and last one gets "no valid host" | 18:05 |
| lucasagomes | devananda, yeah, that's because cleaning is only done by the agent right? | 18:05 |
| lucasagomes | which coordinates things via heartbeat | 18:06 |
| lucasagomes | the DEPLOYWAIT fits more into the DIB case than agent to be honest | 18:06 |
| devananda | lucasagomes: deploywait is needed by the agent just as much | 18:06 |
| JoshNang | ++ on cleanwait | 18:07 |
| devananda | lucasagomes: for the period of time before the agent boots up and "phones hime" | 18:07 |
| lucasagomes | right | 18:07 |
| lucasagomes | which is basically the heartbeat | 18:07 |
| devananda | things are breaking for my test env because the agent isn't calling back to ironic, either for deploy or clean | 18:07 |
| lucasagomes | but it's not like an state visible for the user, I'm ok having DEPLOYWAIT CLEANWAIT | 18:07 |
| devananda | node gets into deploywait, but I can recover from this easily with 'delete' | 18:07 |
| lucasagomes | tho it's nothing more than a signal (which heartbeat does) | 18:08 |
| devananda | there's no way out of "cleaning but the agent hasn't called home yet" | 18:08 |
| *** pradipta has quit IRC | 18:08 | |
| devananda | JoshNang: rather than implementing "fail" or "abort" verb -- what if you just implemented CLEANWAIT state? | 18:09 |
| devananda | would that also resolve your issue(s)? | 18:09 |
| JoshNang | i don't think so | 18:10 |
| lucasagomes | it doesn't solve all problems of having a node stuck in *ING state | 18:10 |
| lucasagomes | it solves some problems | 18:10 |
| JoshNang | the reason i did more than just cleaning was yeah, that ^ | 18:10 |
| lucasagomes | if something happens with the conductor while node is *ING | 18:10 |
| lucasagomes | it is stuck | 18:10 |
| JoshNang | like, i've had to go into the db to fix things more often than i should, and i'm hoping to avoid it & script it | 18:11 |
| devananda | ah, right | 18:12 |
| JoshNang | but i think adding cleanwait is the right thing anyway. it's way better than than the hack that is in there now | 18:12 |
| devananda | so an API that exposes the list of active conductor hosts, and then an API to 'abort' transitions on nodes that are locked by non-active conductors | 18:13 |
| lucasagomes | devananda, JoshNang you can see any major problem in allowing calling Node.release() and moving it to "fail" if user explicitly ask for "abort" ? | 18:13 |
| devananda | so an operator can get a view of things that went sideways after, say, a conductor host had crashed and they started a new conductor with a different host name | 18:14 |
| devananda | lucasagomes: any time there is a long-running operation (whether in or out of band) that does destructive things, like firmware updates, yes | 18:14 |
| lucasagomes | devananda, right, like for zapping it wouldn't be allowed | 18:14 |
| devananda | since cleaning and zapping are intended to do just that | 18:14 |
| devananda | right | 18:15 |
| devananda | well, same for cleaning | 18:15 |
| lucasagomes | yeah and cleaning | 18:15 |
| jroll | devananda: cleaning getting stuck can also happen without a conductor dying | 18:15 |
| jroll | just for the record | 18:15 |
| lucasagomes | I wonder if the zapping/cleaning tasks could somehow indicate "this task is not abortable" | 18:15 |
| lucasagomes | or something like that for certain tasks | 18:15 |
| devananda | jroll: I'm not surprised -- but could you describe how? | 18:15 |
| jroll | devananda: tftp fails | 18:16 |
| devananda | jroll: tftp fails to boot the agent? | 18:16 |
| jroll | yes | 18:16 |
| JoshNang | also, if a node suddenly died, it would get stuck, since it would never heartbeat again | 18:16 |
| devananda | jroll: so that's the case I already described, and would be solved if that was represented as CLEANWAIT | 18:16 |
| JoshNang | devananda: if you're rebooting during cleaning, you'd be going cleaning->cleanwait->cleaning over and over. not sure if that's a problem or not | 18:17 |
| devananda | JoshNang: and releasing the taskmanager lock between those | 18:17 |
| lucasagomes | JoshNang, I think that's fine/good | 18:17 |
| devananda | JoshNang: that would resolve my present issue | 18:17 |
| jroll | devananda: ok, fair. some firmware update occassionally causes networking to go away until a reboot happens :) | 18:17 |
| *** Marga_ has quit IRC | 18:17 | |
| lucasagomes | with locall boot we go deploying-deploywait-deploying-deploywait-deploying... | 18:17 |
| JoshNang | ah | 18:17 |
| JoshNang | alright, wfm then | 18:17 |
| lucasagomes | the combination of those 2 *WAIT and aborting certian *ING states, I think would cover it all | 18:18 |
| lucasagomes | certain* | 18:18 |
| devananda | I'm actually not seeing the *ING states needing "abort" at all right now | 18:19 |
| devananda | *ING should indicate 'there is an active conductor doing something' | 18:19 |
| lucasagomes | devananda, conductor dies? | 18:19 |
| devananda | - waiting for node to reboot / heartbeat? put it in *WAIT | 18:19 |
| jroll | devananda: you don't think disk erase, firmware updates, etc, can get stuck? | 18:19 |
| devananda | - conductor dies and restarts? it clears its own locks during __init__ | 18:19 |
| lucasagomes | if conductor has the same hostname | 18:19 |
| devananda | - conductor dies and does not restart? those locks should hit a timeout and/or be able to be removed manually (via the API) | 18:20 |
| lucasagomes | we can add a timeout too, but that involves breaking the lock | 18:20 |
| devananda | ok, so that last case is the only case where "abort" is needed | 18:20 |
| devananda | and it is a manual process taking the place of a smart[-enough] timeout system | 18:20 |
| *** gridinv_ has joined #openstack-ironic | 18:21 | |
| devananda | JoshNang: clearly this is going to require a spec, heh | 18:21 |
| devananda | JoshNang: also, have you seen my reworking of the state machine spec? | 18:21 |
| JoshNang | devananda: yeah, i figured that was going to happen :P | 18:21 |
| JoshNang | i saw a patch, haven't looked at it yet though | 18:22 |
| lucasagomes | JoshNang, I'm having a couple of problems with locks in the internal setup | 18:22 |
| lucasagomes | JoshNang, I'm more than glad to contribute | 18:22 |
| lucasagomes | with code and spec too | 18:22 |
| JoshNang | lucasagomes: sweet | 18:22 |
| lucasagomes | s/locks/nodes stuck | 18:23 |
| *** Marga_ has joined #openstack-ironic | 18:23 | |
| lucasagomes | devananda, JoshNang another thing re abort | 18:24 |
| *** Sukhdev has joined #openstack-ironic | 18:24 | |
| lucasagomes | is that nova allows you to do a delete on a "spawning" state instance | 18:24 |
| lucasagomes | which is similar to our abort here | 18:24 |
| *** ukalifon has quit IRC | 18:24 | |
| lucasagomes | so this is aligned with the interface we have in nova I believe | 18:24 |
| JoshNang | that's true | 18:25 |
| lucasagomes | https://review.openstack.org/#/c/182992/ | 18:26 |
| devananda | functionally speaking, interrupting a deploy should be non-destructive. practically speaking, our driver API doesn't support it | 18:26 |
| lucasagomes | this allows it for DEPLOYWAIT | 18:26 |
| devananda | interrupting during DEPLOYWAIT is possible today | 18:26 |
| lucasagomes | yeah it wasn't via nova | 18:26 |
| lucasagomes | now it is | 18:26 |
| devananda | ahhh cool! | 18:26 |
| lucasagomes | due that patch ^ | 18:26 |
| lucasagomes | but what I mean is, interface-wise we should somehow allow to do it in DEPLOYING as well, since it's allowed for other drivers in nova | 18:27 |
| lucasagomes | but ironic | 18:27 |
| devananda | lucasagomes: question -- this exception will cause self.spawn() to call self.destroy from its exception handler | 18:30 |
| devananda | that's great -- but | 18:30 |
| *** openstackgerrit has quit IRC | 18:30 | |
| *** Sukhdev has quit IRC | 18:30 | |
| devananda | look at destroy, at L827 | 18:30 |
| devananda | what if the node is in state DEPLOYING ? | 18:30 |
| lucasagomes | devananda, no it doesn't, it just breaks the loop | 18:30 |
| * lucasagomes looks | 18:30 | |
| *** ijw has quit IRC | 18:30 | |
| *** openstackgerrit has joined #openstack-ironic | 18:30 | |
| devananda | lucasagomes: L744 starts a timer on _wait_for_active | 18:31 |
| devananda | if that raises an exception, it calls destroy on L751 | 18:31 |
| devananda | in destroy, it checks the node state | 18:31 |
| *** romcheg has quit IRC | 18:31 | |
| devananda | if the node state is DEPLOYING, it *does not call* self._unprovision | 18:31 |
| devananda | and instead, just calls _cleanup_deploy | 18:32 |
| devananda | which is going to fail, because a PATCH request to remove instance_info will be refused while the node is in DEPLOYING state | 18:32 |
| lucasagomes | right yeah | 18:33 |
| lucasagomes | even if we allow yeah, _unprovision() for DEPLOYING it will fail because DEPLOYING doesn't allow "deleted" | 18:34 |
| devananda | right | 18:34 |
| lucasagomes | with the "abort" idea it would work | 18:34 |
| devananda | lucasagomes: so, while your change works for nodes in DEPLOYWAIT state, I think this is going to create orphaned nodes, if it happens to get called on a node thta is DEPLOYING | 18:35 |
| *** achanda has quit IRC | 18:35 | |
| lucasagomes | devananda, I will give it a go, I think I've tested it | 18:35 |
| *** achanda has joined #openstack-ironic | 18:36 | |
| JoshNang | i've gotta run, bbiab | 18:39 |
| NobodyCam | :) | 18:39 |
| JoshNang | devananda: i'll review the state machine changes later today | 18:39 |
| *** Marga_ has quit IRC | 18:40 | |
| *** Marga_ has joined #openstack-ironic | 18:40 | |
| *** achanda has quit IRC | 18:41 | |
| lucasagomes | devananda, JoshNang yeah can we continue this tomorrow? It's a bit late here | 18:41 |
| lucasagomes | devananda, I will take a look at the spec machine review tomorrow morning too | 18:41 |
| * lucasagomes marks him on the spec | 18:42 | |
| lucasagomes | where is it? | 18:42 |
| devananda | thx | 18:42 |
| devananda | https://review.openstack.org/#/c/196320/1/specs/devref/new-ironic-state-machine.rst,cm | 18:42 |
| lucasagomes | ah was looking for state machine in the title | 18:43 |
| lucasagomes | devananda, thanks | 18:43 |
| lucasagomes | lol "Topic: versions-are-hard" | 18:43 |
| *** romcheg has joined #openstack-ironic | 18:44 | |
| lucasagomes | ok have a good night all! | 18:44 |
| *** krtaylor has quit IRC | 18:44 | |
| lucasagomes | see you later | 18:44 |
| *** lucasagomes is now known as lucas-dinner | 18:44 | |
| NobodyCam | night lucas-dinner | 18:44 |
| NobodyCam | devananda: looking at 196320.. line # 343 ... can I ask why 415 vs something like 406? | 18:48 |
| devananda | The resource identified by the request is only capable of generating response entities which have content characteristics not acceptable according to the accept headers sent in the request. | 18:49 |
| devananda | The server is refusing to service the request because the entity of the request is in a format not supported by the requested resource for the requested method. | 18:49 |
| devananda | NobodyCam: AIUI, 406 is specific to the Content-Type header | 18:50 |
| devananda | eg, xml vs json | 18:50 |
| NobodyCam | ahh ok :) | 18:50 |
| devananda | sorry, i mean the "Accept: Content-Type" header | 18:50 |
| NobodyCam | Unsupported Media Type just seemd strange but I think I get it now | 18:50 |
| NobodyCam | TY | 18:51 |
| *** Sukhdev has joined #openstack-ironic | 18:51 | |
| devananda | oooh | 18:51 |
| devananda | I've been reading from the old list | 18:51 |
| devananda | http://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml is actualy the reference we want to use | 18:52 |
| NobodyCam | ya I like 426 but its kinda a miss use | 18:52 |
| *** Marga_ has quit IRC | 18:53 | |
| devananda | omg. there's an RFC for HTTP-based resource locking | 18:53 |
| openstackgerrit | Chris St. Pierre proposed openstack/ironic: Replace is_hostname_safe with a better check https://review.openstack.org/193587 | 18:53 |
| NobodyCam | brb | 18:56 |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Mute ipmi debug log output https://review.openstack.org/196267 | 18:59 |
| *** ijw has joined #openstack-ironic | 19:05 | |
| *** Marga_ has joined #openstack-ironic | 19:10 | |
| *** boris-42 has quit IRC | 19:12 | |
| *** dguerri` is now known as dguerri | 19:18 | |
| *** dguerri is now known as dguerri` | 19:20 | |
| *** achanda has joined #openstack-ironic | 19:22 | |
| *** maurosr has quit IRC | 19:28 | |
| *** maurosr has joined #openstack-ironic | 19:32 | |
| *** maurosr has quit IRC | 19:39 | |
| *** maurosr has joined #openstack-ironic | 19:40 | |
| *** romcheg has quit IRC | 19:46 | |
| *** romcheg has joined #openstack-ironic | 19:47 | |
| *** dguerri` is now known as dguerri | 19:47 | |
| *** achanda has quit IRC | 19:48 | |
| *** gridinv_ has quit IRC | 19:49 | |
| *** gridinv_ has joined #openstack-ironic | 19:54 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: WIP: Ansible 2.0 compatability **DO NOT MERGE** https://review.openstack.org/196832 | 19:58 |
| *** achanda has joined #openstack-ironic | 20:00 | |
| *** ijw has quit IRC | 20:05 | |
| *** gridinv_ has quit IRC | 20:22 | |
| *** achanda has quit IRC | 20:22 | |
| *** gridinv_ has joined #openstack-ironic | 20:22 | |
| *** dontalton has quit IRC | 20:26 | |
| *** cdearborn has quit IRC | 20:30 | |
| *** maurosr has quit IRC | 20:30 | |
| *** maurosr has joined #openstack-ironic | 20:30 | |
| *** boris-42 has joined #openstack-ironic | 20:31 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/196849 | 20:33 |
| *** zz_jgrimm has quit IRC | 20:37 | |
| *** maurosr has quit IRC | 20:37 | |
| *** gridinv_ has quit IRC | 20:37 | |
| *** krtaylor has joined #openstack-ironic | 20:38 | |
| *** krtaylor has quit IRC | 20:39 | |
| *** krtaylor has joined #openstack-ironic | 20:40 | |
| *** marzif_ has joined #openstack-ironic | 20:43 | |
| *** jgrimm has joined #openstack-ironic | 20:45 | |
| *** maurosr has joined #openstack-ironic | 20:46 | |
| *** maurosr has quit IRC | 20:51 | |
| *** Sukhdev has quit IRC | 20:51 | |
| *** afaranha has quit IRC | 20:52 | |
| *** jgrimm has quit IRC | 20:52 | |
| *** jgrimm has joined #openstack-ironic | 20:55 | |
| *** maurosr has joined #openstack-ironic | 20:56 | |
| *** marzif_ has quit IRC | 20:58 | |
| *** Marga_ has quit IRC | 20:59 | |
| *** ijw has joined #openstack-ironic | 21:04 | |
| *** trown is now known as trown|outttypeww | 21:05 | |
| *** thiagop has quit IRC | 21:06 | |
| *** maurosr has quit IRC | 21:11 | |
| *** jgrimm has quit IRC | 21:13 | |
| *** krtaylor has quit IRC | 21:13 | |
| *** radek_ has quit IRC | 21:13 | |
| *** ijw has quit IRC | 21:19 | |
| *** dprince has quit IRC | 21:20 | |
| *** r-daneel has quit IRC | 21:23 | |
| *** Sukhdev has joined #openstack-ironic | 21:27 | |
| *** kevinbenton has quit IRC | 21:31 | |
| *** natorious is now known as zz_natorious | 21:37 | |
| *** maurosr has joined #openstack-ironic | 21:51 | |
| *** krtaylor has joined #openstack-ironic | 21:52 | |
| *** zz_jgrimm has joined #openstack-ironic | 21:52 | |
| *** ijw has joined #openstack-ironic | 22:02 | |
| *** ijw has quit IRC | 22:03 | |
| *** ijw has joined #openstack-ironic | 22:03 | |
| *** rloo has quit IRC | 22:08 | |
| mrda | Morning Ironic | 22:18 |
| *** absubram has quit IRC | 22:19 | |
| NobodyCam | good UGT mornig mrda | 22:19 |
| mrda | NobodyCam: umm, ok :) | 22:19 |
| NobodyCam | hehehe | 22:20 |
| BadCub | howdy mrda :) | 22:25 |
| mrda | hey BadCub | 22:25 |
| *** lucas-dinner has quit IRC | 22:39 | |
| *** sinval has joined #openstack-ironic | 23:00 | |
| *** romcheg has quit IRC | 23:05 | |
| *** ijw has quit IRC | 23:05 | |
| *** ijw has joined #openstack-ironic | 23:05 | |
| *** ijw has quit IRC | 23:07 | |
| *** ijw has joined #openstack-ironic | 23:07 | |
| *** ijw has quit IRC | 23:10 | |
| jlvillal | Good morning mrda | 23:10 |
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Add port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 23:11 |
| *** dguerri is now known as dguerri` | 23:17 | |
| *** ijw has joined #openstack-ironic | 23:18 | |
| mrda | hey jlvillal, welcome back | 23:21 |
| *** Sukhdev has quit IRC | 23:27 | |
| *** yuanying has joined #openstack-ironic | 23:30 | |
| *** boris-42 has quit IRC | 23:32 | |
| *** smoriya has joined #openstack-ironic | 23:54 | |
| jlvillal | mrda, Thanks | 23:57 |
| *** Sukhdev has joined #openstack-ironic | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!