| openstackgerrit | Takashi NATSUME proposed openstack/ironic master: Fix misuse of assertTrue https://review.openstack.org/643208 | 00:37 |
|---|---|---|
| *** andrein has quit IRC | 00:38 | |
| *** gyee has quit IRC | 00:39 | |
| openstackgerrit | Takashi NATSUME proposed openstack/ironic master: Fix misuse of assertTrue https://review.openstack.org/643208 | 00:39 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/ironic master: prevent serial console lockout https://review.openstack.org/642734 | 01:39 |
| *** whoami-rajat has joined #openstack-ironic | 02:02 | |
| *** hoonetorg has quit IRC | 02:59 | |
| openstackgerrit | QianBiao Ng proposed openstack/ironic master: Add Huawei iBMC driver support https://review.openstack.org/639288 | 03:06 |
| *** hoonetorg has joined #openstack-ironic | 03:12 | |
| *** MattMan_ has quit IRC | 03:24 | |
| *** MattMan_ has joined #openstack-ironic | 03:25 | |
| openstackgerrit | Nguyen Van Trung proposed openstack/ironic master: Support firmware upgrade for irmc driver https://review.openstack.org/599992 | 04:04 |
| *** zhangfei has joined #openstack-ironic | 04:20 | |
| openstackgerrit | Merged openstack/sushy stable/stein: Fix wrong default JsonDataReader() argument https://review.openstack.org/642837 | 05:35 |
| openstackgerrit | Rachit Kapadia proposed openstack/ironic master: Set boot_mode in node properties during OOB Introspection https://review.openstack.org/639698 | 05:43 |
| openstackgerrit | Kaifeng Wang proposed openstack/ironic master: TEST lower-constraint job https://review.openstack.org/643231 | 06:01 |
| *** jtomasek has joined #openstack-ironic | 06:15 | |
| *** w14161_1_ has joined #openstack-ironic | 07:02 | |
| *** w14161_1 has quit IRC | 07:05 | |
| openstackgerrit | Nguyen Van Trung proposed openstack/ironic master: Support firmware upgrade for irmc driver https://review.openstack.org/599992 | 07:07 |
| openstackgerrit | Varsha Verma proposed openstack/ironic master: Deprecates `hash_distribution_replicas` config option https://review.openstack.org/643002 | 07:14 |
| *** Qianbiao has joined #openstack-ironic | 07:17 | |
| Qianbiao | Hello, everyone. | 07:17 |
| *** mkrai has joined #openstack-ironic | 07:22 | |
| *** verma-varsha has joined #openstack-ironic | 07:22 | |
| Qianbiao | Hello, may someone help to identify why failed to active node | 07:28 |
| Qianbiao | logs: http://paste.openstack.org/show/747767/ | 07:28 |
| mkrai | Hi, is the cleaning API exposed via Nova? | 07:28 |
| mkrai | I mean the "openstack baremetal node clean" API? | 07:29 |
| Qianbiao | I did not use clean API | 07:29 |
| Qianbiao | openstack baremetal maintenance set $NODE && openstack baremetal delete $NODE | 07:30 |
| Qianbiao | then recreate a node. | 07:30 |
| mkrai | Qianbiao: From your log, it seems node is busy with some task | 07:30 |
| mkrai | Try the steps after the current task is completed | 07:30 |
| Qianbiao | What kind of task? where i can get the task info. | 07:31 |
| openstackgerrit | Kaifeng Wang proposed openstack/ironic-inspector master: Follow up to api-ref revision https://review.openstack.org/643238 | 07:31 |
| Qianbiao | because it's a fresh node. | 07:31 |
| Qianbiao | i delete it and then recreate | 07:31 |
| mkrai | I would suggest to wait for the completion of one API before running the second one | 07:32 |
| Qianbiao | Hmmm. | 07:32 |
| Qianbiao | the cmd i run has an option wait | 07:32 |
| mkrai | Your node is deleted? | 07:32 |
| Qianbiao | openstack baremetal node manage $NODE && openstack baremetal node provide $Node | 07:33 |
| Qianbiao | --wait [<time-out>] Wait for a node to reach the desired state, manageable. | 07:33 |
| Qianbiao | Optionally takes a timeout value (in seconds). The | 07:33 |
| Qianbiao | default value is 0, meaning it will wait indefinitely. | 07:33 |
| Qianbiao | by default it will wait indefinitely. | 07:33 |
| Qianbiao | Yes, I delete the node first, and then create it again. | 07:34 |
| mkrai | so you faced this error at which API? | 07:34 |
| Qianbiao | I thought it's node provide API | 07:34 |
| mkrai | What is the state of your node now? | 07:36 |
| Qianbiao | manageable | 07:36 |
| mkrai | can you try to run the provide API now? | 07:36 |
| Qianbiao | sure | 07:37 |
| Qianbiao | weird, it's ok now. | 07:37 |
| Qianbiao | available | 07:37 |
| Qianbiao | it turns to available | 07:37 |
| mkrai | Yes so the issue was you ran the two apis one after another before the manageable API could complete | 07:38 |
| Qianbiao | <mkrai> check the help info of "openstack baremetal node manage" | 07:39 |
| Qianbiao | it should wait indefinitely. | 07:39 |
| Qianbiao | I use the cmd before, it never failed. | 07:39 |
| mkrai | It waits indefinitely until it completes and during that time the node is locked by Ironic | 07:39 |
| Qianbiao | yes, than it should be ok, right? | 07:40 |
| mkrai | Yes it's ok | 07:40 |
| Qianbiao | it will wait manage finished, and then run provide. | 07:40 |
| Qianbiao | so, what should i do to fix this. because i need to use them in ansible playbook. i donot know how long it will take. | 07:41 |
| Qianbiao | If the API works as it desc, it should wait and no error occurs. | 07:42 |
| arne_wiebalck | good morning, ironic! | 07:42 |
| mkrai | arne_wiebalck: GM | 07:43 |
| arne_wiebalck | mkrai: o/ | 07:43 |
| *** e0ne has joined #openstack-ironic | 07:45 | |
| *** yaawang has quit IRC | 07:45 | |
| *** yaawang has joined #openstack-ironic | 07:47 | |
| Qianbiao | mkrai should i upgrade openstack client | 07:47 |
| Qianbiao | openstack baremetal --version | 07:47 |
| Qianbiao | openstack 3.16.1 | 07:47 |
| Qianbiao | It's weird, the script has been used for half a year, and this is the first time it happens. | 07:47 |
| *** gkadam_ has joined #openstack-ironic | 07:50 | |
| *** pcaruana has joined #openstack-ironic | 08:07 | |
| *** pcaruana has quit IRC | 08:11 | |
| *** tssurya has joined #openstack-ironic | 08:12 | |
| *** pcaruana has joined #openstack-ironic | 08:23 | |
| openstackgerrit | Mark Goddard proposed openstack/ironic master: Fix idrac Job.state renamed to Job.status https://review.openstack.org/643142 | 08:24 |
| *** trungnv has quit IRC | 08:25 | |
| *** hoonetorg has quit IRC | 08:32 | |
| openstackgerrit | Kaifeng Wang proposed openstack/ironic master: TEST lower-constraints job https://review.openstack.org/643231 | 08:34 |
| *** hjensas has quit IRC | 08:35 | |
| *** hoonetorg has joined #openstack-ironic | 08:45 | |
| iurygregory | good morning everyone o/ | 08:52 |
| *** sburette has joined #openstack-ironic | 08:53 | |
| *** dtantsur|afk is now known as dtantsur | 08:59 | |
| *** dougsz has joined #openstack-ironic | 08:59 | |
| dtantsur | morning ironic | 08:59 |
| iurygregory | morning dtantsur | 08:59 |
| openstackgerrit | Kaifeng Wang proposed openstack/ironic master: TEST lower-constraints job https://review.openstack.org/643231 | 09:00 |
| mkrai | Hi, is the cleaning API exposed via Nova? | 09:02 |
| mkrai | I mean the "openstack baremetal node clean" API? | 09:02 |
| dtantsur | mkrai: hi, no. | 09:05 |
| dtantsur | folks, is the lower-constraints job broken? is anybody already looking into it? | 09:05 |
| * dtantsur asks infra | 09:08 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Add libpq-dev to bindep https://review.openstack.org/643251 | 09:12 |
| dtantsur | maybe this will help ^^ | 09:13 |
| openstackgerrit | Kaifeng Wang proposed openstack/ironic master: TEST lower-constraints job https://review.openstack.org/643231 | 09:26 |
| *** andrein has joined #openstack-ironic | 09:27 | |
| *** mkrai has quit IRC | 09:30 | |
| openstackgerrit | Kaifeng Wang proposed openstack/ironic master: Fix lower-constraints job https://review.openstack.org/643231 | 09:33 |
| dtantsur | rpittau|afk: morning! do you think that pinning the postgres job to xenial can unbreak us for now? | 09:41 |
| iurygregory | if CI seems broken is probably because is on Bionic (for legacy jobs) =) | 09:42 |
| *** e0ne has quit IRC | 09:43 | |
| dtantsur | iurygregory: well, the lower-constraints job is fixed with the patch above. the postgres job is non-voting, but it fails because of bionic, and we need it green before the final release | 09:44 |
| dtantsur | which may mean sticking with xenial for stable/stein | 09:44 |
| iurygregory | dtantsur, gotcha | 09:44 |
| *** e0ne has joined #openstack-ironic | 09:46 | |
| iurygregory | wow =O | 09:46 |
| iurygregory | thats sad =( | 09:46 |
| *** pcaruana has quit IRC | 09:48 | |
| *** hjensas has joined #openstack-ironic | 09:57 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-tempest-plugin master: discovery: optional support for testing changing driver https://review.openstack.org/643264 | 09:58 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector master: Pass reset_interfaces when updating a driver from the rules https://review.openstack.org/641340 | 10:03 |
| rpittau|afk | dtantsur: we can give it a try, I can convert the job in the test I have | 10:05 |
| rpittau|afk | Coming back from Paris, had to do an errand this morning and trains were late | 10:06 |
| rpittau|afk | Instead of 1:30 took me more than 2 hours :/ | 10:06 |
| dtantsur | trains.. | 10:06 |
| dtantsur | okay, let's try reverting to xenial, then propose a patch with moving it to bionic + fixes | 10:07 |
| rpittau|afk | dtantsur: ack | 10:07 |
| *** rpittau|afk is now known as rpittau | 10:13 | |
| rpittau | I made it back \o/ | 10:13 |
| rpittau | good morning ironic! o/ | 10:13 |
| rpittau | dtantsur, I think we're good on the lower constraints fix patch, going to +2 that asap | 10:18 |
| dtantsur | yep, it's good, feel free to approve it | 10:21 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] changes to fix postgres job https://review.openstack.org/642430 | 10:25 |
| mgoddard | I have to say, recent experience of using the idrac drivers rather than ipmitool has not been too pleasant | 10:26 |
| mgoddard | things keep getting wedged | 10:27 |
| iurygregory | morning rpittau o/ | 10:32 |
| rpittau | hi iurygregory :) | 10:32 |
| *** w14161_1_ has quit IRC | 10:32 | |
| *** w14161_1 has joined #openstack-ironic | 10:39 | |
| hjensas | good morning. | 11:00 |
| mgoddard | so close to deploy time RAID. After 30 minutes of watching various reboots, and even applying the RAID config, the idrac comes back with 'Configuration job already created, cannot create another config job on specified target until existing job is completed or is cancelled' | 11:00 |
| hjensas | rpittau: the pxelinux.0 load issue still relevant? | 11:00 |
| rpittau | hey hjensas | 11:00 |
| rpittau | hjensas, it's relevant for bionic | 11:00 |
| rpittau | the update to the dist filter is necessary or bionic won't find pxelinux.0 | 11:01 |
| *** yolanda has quit IRC | 11:02 | |
| *** yolanda has joined #openstack-ironic | 11:03 | |
| hjensas | rpittau: this one right? https://review.openstack.org/642507 | 11:07 |
| patchbot | patch 642507 - ironic - Update dist filter for devstack ubuntu - 1 patch set | 11:07 |
| rpittau | hjensas, correct | 11:07 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: [WIP] pinning postgres job to xenial https://review.openstack.org/642430 | 11:12 |
| rpittau | oh the irony, it rains like hell but there's no water in my building | 11:15 |
| *** zhangfei has quit IRC | 11:18 | |
| dtantsur | heh | 11:22 |
| *** derekh has joined #openstack-ironic | 11:29 | |
| *** pcaruana has joined #openstack-ironic | 11:32 | |
| openstackgerrit | Ilya Etingof proposed openstack/sushy-tools master: Add libvirt-based virtual media emulation https://review.openstack.org/640452 | 11:35 |
| etingof | raining cats and dogs on rpittau o/ | 11:35 |
| *** sburette has quit IRC | 11:37 | |
| rpittau | hi etingof :D | 11:37 |
| openstackgerrit | Ilya Etingof proposed openstack/sushy-tools master: Render System->Managers linkage https://review.openstack.org/643094 | 11:38 |
| openstackgerrit | Ilya Etingof proposed openstack/sushy-tools master: Add docs on libvirt virtual media emulation https://review.openstack.org/643027 | 11:38 |
| *** bfournie has quit IRC | 11:42 | |
| rpittau | dtantsur, it seems like the issue with postgres is indeed related to bionic, pinning to xenial makes it pass https://review.openstack.org/642430 | 12:06 |
| patchbot | patch 642430 - ironic - [WIP] pinning postgres job to xenial - 7 patch sets | 12:06 |
| dtantsur | rpittau: cool, let's pin it for now and prepare a proper patch with migration | 12:06 |
| rpittau | dtantsur, ack | 12:07 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Pinning postgres job to xenial https://review.openstack.org/642430 | 12:13 |
| *** verma-varsha has quit IRC | 12:17 | |
| *** bfournie has joined #openstack-ironic | 12:18 | |
| *** trown|outtypewww is now known as trown | 12:19 | |
| *** e0ne has quit IRC | 12:31 | |
| Qianbiao | <rpittau> what is ack stand for? | 12:38 |
| Qianbiao | the same as TCP ack ? | 12:38 |
| iurygregory | Qianbiao, ack = acknowledged | 12:44 |
| rpittau | Qianbiao: kind of :) | 12:44 |
| Qianbiao | i see. | 12:45 |
| iurygregory | Qianbiao, did you have lucky with your env to see the tests with standalone queens? =) | 12:45 |
| Qianbiao | kind like "hi, received" | 12:45 |
| Qianbiao | It works in my ENV. | 12:46 |
| iurygregory | the tempests tests run fine? | 12:46 |
| Qianbiao | I didnot use tempest :) | 12:46 |
| iurygregory | you have a conf that i could check what is different in the job? | 12:46 |
| Qianbiao | I code a ansible playbook myself. | 12:46 |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic master: Update doc for UEFI first https://review.openstack.org/631225 | 12:47 |
| Qianbiao | I checked the log, but no lucky. | 12:47 |
| iurygregory | well the devstack works the problem is run the tempest tests that get many different errors XD | 12:48 |
| *** Qianbiao has quit IRC | 12:49 | |
| openstackgerrit | Merged openstack/ironic master: Fix lower-constraints job https://review.openstack.org/643231 | 12:50 |
| *** Qianbiao has joined #openstack-ironic | 12:50 | |
| *** hwoarang has quit IRC | 12:56 | |
| *** hwoarang has joined #openstack-ironic | 12:57 | |
| dtantsur | okay, the CI should be back to norm. recheck the hell out of it! | 12:59 |
| iurygregory | recheck all the things! | 13:01 |
| *** andrein has quit IRC | 13:11 | |
| *** sburette has joined #openstack-ironic | 13:21 | |
| *** e0ne has joined #openstack-ironic | 13:32 | |
| *** baha has joined #openstack-ironic | 13:34 | |
| openstackgerrit | Ilya Etingof proposed openstack/sushy-tools master: Add HTTP image download for virtual media emulation https://review.openstack.org/643330 | 13:34 |
| *** mjturek has joined #openstack-ironic | 13:37 | |
| *** mjturek has quit IRC | 13:38 | |
| *** efried has joined #openstack-ironic | 13:38 | |
| *** Wryhder has joined #openstack-ironic | 13:40 | |
| *** Lucas_Gray has joined #openstack-ironic | 13:44 | |
| *** mjturek has joined #openstack-ironic | 13:46 | |
| *** hwoarang has quit IRC | 13:53 | |
| *** hwoarang has joined #openstack-ironic | 13:55 | |
| openstackgerrit | Nikolay Fedotov proposed openstack/ironic master: Kg key for IPMIv2 authentication https://review.openstack.org/641658 | 14:06 |
| *** hjensas has quit IRC | 14:06 | |
| *** Wryhder has quit IRC | 14:12 | |
| *** rh-jelabarre has joined #openstack-ironic | 14:13 | |
| *** andrein has joined #openstack-ironic | 14:14 | |
| *** rcernin has quit IRC | 14:17 | |
| *** hjensas has joined #openstack-ironic | 14:20 | |
| Qianbiao | dtantsur TheJulia huawei ironic ci is ready now | 14:20 |
| Qianbiao | check the newest comment https://review.openstack.org/#/c/639288/ | 14:20 |
| patchbot | patch 639288 - ironic - Add Huawei iBMC driver support - 16 patch sets | 14:20 |
| dtantsur | thanks, will check when I finish my meetings | 14:20 |
| Qianbiao | ok thanks | 14:20 |
| openstackgerrit | Ilya Etingof proposed openstack/sushy-tools master: Add HTTP image download for virtual media emulation https://review.openstack.org/643330 | 14:24 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: Removing two deprecated values from json responses https://review.openstack.org/643350 | 14:24 |
| dtantsur | Qianbiao: I'm getting "The server at 112.93.129.99 is taking too long to respond" :( | 14:24 |
| dtantsur | can there be some kind of firewall or IP filtering? | 14:24 |
| dtantsur | Qianbiao: honestly, I cannot even ping that IP | 14:28 |
| Qianbiao | why | 14:34 |
| Qianbiao | http://112.93.129.99/builds/Ironic-3rd-party-CI/146/ | 14:34 |
| Qianbiao | log is here | 14:34 |
| *** hwoarang has quit IRC | 14:35 | |
| *** hwoarang has joined #openstack-ironic | 14:36 | |
| rpittau | dtantsur, I can access the server | 14:36 |
| dtantsur | interesting | 14:38 |
| dtantsur | may be some global routes from my place.. | 14:38 |
| iurygregory | i cant access too =( | 14:39 |
| *** dougsz has quit IRC | 14:45 | |
| *** hwoarang has quit IRC | 14:46 | |
| *** yaawang has quit IRC | 14:47 | |
| Qianbiao | :( it's a normal server in china zone. | 14:47 |
| *** yaawang has joined #openstack-ironic | 14:48 | |
| *** hwoarang has joined #openstack-ironic | 14:52 | |
| iurygregory | i will try to access from home since I'm at the office and I'll see if it works | 14:56 |
| *** yolanda has quit IRC | 14:57 | |
| Qianbiao | ok thanks. | 14:57 |
| Qianbiao | i will re-trigger CI | 14:57 |
| Qianbiao | again. because the log contains some sensitive info... i need to hide them | 14:57 |
| *** baha has quit IRC | 14:58 | |
| dtantsur | rpittau: do you think we can split out the jobs that do work from https://review.openstack.org/#/c/636169/ ? | 14:58 |
| patchbot | patch 636169 - ironic-tempest-plugin - Run all defined jobs defined in check and gate - 15 patch sets | 14:58 |
| iurygregory | dtantsur, i was thinking about this | 14:59 |
| iurygregory | if you want i can push the patch without wholedisk and ironic-standalone queens | 15:00 |
| rpittau | dtantsur, yeah | 15:01 |
| dtantsur | iurygregory: oh, it's your patch, sorry :) | 15:01 |
| dtantsur | yeah, let's do small steps. | 15:01 |
| iurygregory | sure =) | 15:02 |
| iurygregory | after meeting im going to update | 15:02 |
| rpittau | I wonder if we should split by branch also ? | 15:02 |
| iurygregory | rpittau, for the ones that arent working you mean? | 15:04 |
| *** Qianbiao has quit IRC | 15:04 | |
| rpittau | iurygregory, yes :) | 15:09 |
| iurygregory | rpittau, sure makes sense =) to not be blocking if we get another working | 15:10 |
| *** hjensas has quit IRC | 15:12 | |
| *** yajushiSri has joined #openstack-ironic | 15:18 | |
| *** dougsz has joined #openstack-ironic | 15:31 | |
| *** baha has joined #openstack-ironic | 15:34 | |
| openstackgerrit | Nikolay Fedotov proposed openstack/ironic master: Kg key for IPMIv2 authentication https://review.openstack.org/641658 | 15:35 |
| *** rachit7_ has joined #openstack-ironic | 15:38 | |
| openstackgerrit | Merged openstack/ironic-inspector stable/rocky: Use getaddrinfo instead of gethostbyname while resolving BMC address https://review.openstack.org/642560 | 15:39 |
| *** hamzy has quit IRC | 15:43 | |
| openstackgerrit | Merged openstack/ironic-inspector stable/rocky: Fix lookup when ipmi_address is a hostname https://review.openstack.org/642760 | 15:44 |
| openstackgerrit | Merged openstack/ironic master: Fix misuse of assertTrue https://review.openstack.org/643208 | 15:44 |
| *** yolanda has joined #openstack-ironic | 15:45 | |
| openstackgerrit | Merged openstack/ironic master: Adding ansible python interpreter as driver_info https://review.openstack.org/642402 | 16:00 |
| *** Lucas_Gray has quit IRC | 16:04 | |
| rpioso | Good morning, ironicers. | 16:16 |
| rpioso | mgoddard: Sorry to hear about those challenges. | 16:16 |
| mgoddard | rpioso: not your fault | 16:16 |
| rpioso | mgoddard: Have you been able to get beyond it? | 16:16 |
| mgoddard | I ended up modifying the idrac driver to use ipmitool for power & management | 16:16 |
| rpioso | mgoddard: Ugh! | 16:17 |
| mgoddard | rpioso: we always use ipmitool drivers | 16:17 |
| mgoddard | rpioso: can I recommend that you add them to the supported power & management interfaces? | 16:18 |
| * rpioso hasn't tried idrac RAID on master; however, our downstream solution based on queens, the last LTS version, uses it, and it works. | 16:18 | |
| mgoddard | I kept getting various errors from the idrac | 16:20 |
| rpioso | mgoddard: We've batted that idea around recently, in the context of the story I'm working on to add Redfish interfaces to it. I'll discuss it some more with my team. | 16:20 |
| mgoddard | this one in particular: | 16:20 |
| mgoddard | Configuration job already created, cannot create another config job on specified target until existing job | 16:20 |
| mgoddard | is completed or is cancelled | 16:20 |
| mgoddard | redfish might help. Is it supported on idrac 8? | 16:20 |
| rpioso | mgoddard: Was the job queue clear before you started? | 16:20 |
| mgoddard | rpioso: job queue was clear | 16:21 |
| rpioso | mgoddard: 13G, right? | 16:21 |
| *** rachit7_ has quit IRC | 16:21 | |
| mgoddard | R630 | 16:21 |
| rpioso | mgoddard: That's a yes :-) | 16:21 |
| mgoddard | idrac firmware is just over a year old | 16:22 |
| rpioso | mgoddard: It is, however, I believe the Redfish support on 14G is ahead of 13G. | 16:22 |
| rpioso | mgoddard: Redfish power and management should be fine on either. | 16:22 |
| mgoddard | rpioso: can it set the power state and boot mode? | 16:22 |
| mgoddard | ok | 16:23 |
| rpioso | mgoddard: Did the modification fix the issue? | 16:26 |
| mgoddard | rpioso: the RAID one? | 16:26 |
| mgoddard | Yes, it's looking better now | 16:26 |
| mgoddard | Managed to get a RAID deploy step working, I think. Need to validate the disk is as expected | 16:27 |
| rpioso | mgoddard: This one -- "Configuration job already created, cannot create another config job on specified target until existing job is completed or is cancelled" | 16:27 |
| mgoddard | rpioso: oh, no - switched to ipmitool at that point - couldn't get past that error | 16:28 |
| * rpioso is lost | 16:28 | |
| rpioso | mgoddard: I thought you modified the idrac h/w type. | 16:28 |
| mgoddard | rpioso: I did. The error was when setting the boot device, ipmitool management interface avoided it | 16:29 |
| rpioso | mgoddard: So it was a BIOS configuration job that was already created, right? | 16:29 |
| mgoddard | rpioso: I couldn't see any job in the queue. Hadn't created any BIOS jobs. I think it was a job to set the boot device | 16:30 |
| rpioso | mgoddard: Which boot mode is the server configured to use? | 16:30 |
| mgoddard | preece: seems it couldn't set the boot mode or power state: http://paste.openstack.org/show/747797/ | 16:31 |
| mgoddard | sorry preece, not you. | 16:32 |
| mgoddard | rpioso: ^ | 16:32 |
| rpioso | mgoddard: You may have run into a known race condition. Let me dig up the artifacts. | 16:35 |
| rpioso | mgoddard: Which boot mode is the server configured to use? | 16:35 |
| mgoddard | rpioso: local | 16:35 |
| *** jtomasek has quit IRC | 16:35 | |
| rpioso | mgoddard legacy BIOS or UEFI? | 16:36 |
| mgoddard | rpioso: oops, BIOS | 16:36 |
| openstackgerrit | Merged openstack/ironic master: Deploy templates: conductor and API nits https://review.openstack.org/640446 | 16:36 |
| rpioso | mgodddard: We're working on a fix to it -- https://review.openstack.org/#/c/633833/ | 16:37 |
| patchbot | patch 633833 - ironic - Fix potential race condition on node power on and ... - 4 patch sets | 16:37 |
| rpioso | mgoddard: Btw, boot device configuration is a BIOS thing. | 16:38 |
| *** e0ne has quit IRC | 16:38 | |
| rpioso | mgoddard: Internally, I'm reviewing an alternative approach mbeierl and I are batting around. | 16:40 |
| rpioso | mbeierl: ^^^^^ | 16:40 |
| rpioso | mgoddard: Are you using master? | 16:40 |
| mgoddard | rpioso: yes | 16:41 |
| iurygregory | dtantsur, rpittau just fyi im going to remove the following jobs from the patch just to unblock http://paste.openstack.org/show/747796/ my decision was based on the last patch i saw the job running and the results are differents between the patches (except for wholedisk-pike that was broken in both) make sense? | 16:43 |
| rpioso | mgoddard: For the time being, if you're game for trying idrac power and management again, you could add a short sleep immediately after https://github.com/openstack/ironic/blob/9625b8aa71896bbf626971f8bbabe48d0d34b39e/ironic/drivers/modules/drac/management.py#L249 | 16:44 |
| rpioso | mgoddard: Five seconds should do the trick. | 16:44 |
| rpittau | iurygregory, looks good, also I don't know how much we want to invest in pike tests :/ | 16:44 |
| mgoddard | rpioso: thanks, I'll stick with ipmitool so I can focus on other things | 16:44 |
| iurygregory | rpittau, yeah XD | 16:45 |
| *** tssurya has quit IRC | 16:45 | |
| *** emine__ has quit IRC | 16:46 | |
| dtantsur | iurygregory: let's make them non-voting rather than removing? otherwise yes | 16:46 |
| rpioso | mgoddard: Understood. Once we settle on the fix, I'd be glad to let you know. | 16:46 |
| mgoddard | rpioso: thanks :) | 16:46 |
| iurygregory | dtantsur, ack o/ | 16:46 |
| iurygregory | also I will add comments about the jobs that we want to get back voting | 16:47 |
| rpioso | mgoddard: It's a timing issue, so it's difficult to reliably repro. It depends on the generation of the iDRAC and its firmware version. | 16:47 |
| rpioso | mgoddard: Since you've encountered it, it would be much appreciated if you could try to reproduce it when time permits and try the one line hack. And then subsequently pick up the real fix to confirm it resolves the issue. We've had trouble reproducing it in house. | 16:49 |
| *** gkadam_ has quit IRC | 16:50 | |
| mbeierl | thanks, rpiosoand mgoddard. I just came back from lunch | 16:51 |
| iurygregory | dtantsur, fow the wholedisk in the gate i should let non-voting or remove and add a note? | 16:52 |
| dtantsur | iurygregory: you mean, pike multinode? I guess nobody is going to recover it.. | 16:53 |
| mgoddard | rpioso: I'll give it a try at some point | 16:53 |
| iurygregory | dtantsur, nope ironic-tempest-ipa-wholedisk-direct-tinyipa-multinode | 16:53 |
| rpioso | mgoddard: Awesome :-) | 16:53 |
| dtantsur | iurygregory: well, master must be recovered | 16:54 |
| iurygregory | i know but since its broken and the job is on gate i should keep on gat with non-voting? | 16:54 |
| dtantsur | ah | 16:56 |
| dtantsur | iurygregory: non-voting jobs must not be in the "gate" pipeline, only in check | 16:56 |
| iurygregory | ack so i will keep a note that we want to re add to the gate | 16:57 |
| rpittau | iurygregory, weird, that job should be working just fine | 16:59 |
| *** arne_wiebalck has quit IRC | 16:59 | |
| *** arne_wiebalck has joined #openstack-ironic | 17:00 | |
| * rpioso goes to lunch | 17:00 | |
| *** rpioso is now known as rpioso|afk | 17:01 | |
| iurygregory | rpittau, yeah | 17:01 |
| *** sziviani has joined #openstack-ironic | 17:04 | |
| rpittau | good night o/ | 17:05 |
| *** rpittau is now known as rpittau|afk | 17:05 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic-tempest-plugin master: Run all defined jobs defined in check and gate https://review.openstack.org/636169 | 17:06 |
| *** trown is now known as trown|lunch | 17:06 | |
| *** gyee has joined #openstack-ironic | 17:07 | |
| *** hamzy has joined #openstack-ironic | 17:09 | |
| *** anupn has joined #openstack-ironic | 17:11 | |
| anupn | Hi Folks, any clue when Bifrost provisions the user image does it not work on the drive /dev/sda (by default) if it already had partitions on it? | 17:13 |
| anupn | My baremetal got provisioned correctly, and I see the state as active, but I don't see the OS in the BIOS although I see two partitions are created on /dev/sda | 17:14 |
| anupn | any idea if it might be OS problem or boot loader issue | 17:15 |
| *** sburette has quit IRC | 17:15 | |
| TheJulia | does it boot? | 17:15 |
| TheJulia | the bios seeing it may be it knowing how to read/grok the disk which milage may vary based on vendor and vendor focus | 17:16 |
| *** dtantsur is now known as dtantsur|afk | 17:18 | |
| dtantsur|afk | \o | 17:18 |
| anupn | The IP address that is handed out is active. So I have 4 different drives, and each one has OS in it. This OS is getting provisioned in sda. So if I boot into another OS on drive sdb I see the IP sent by DHCP being assigned to the NIC | 17:26 |
| anupn | TheJulia | 17:26 |
| anupn | It says it writes image correctly to sda | 17:26 |
| anupn | But I can't boot into sda as the OS option is not coming in BIOS | 17:27 |
| TheJulia | anupn: Ahh, so a couple different possibilities exist. Largely the first one is that you need to actually clean the drives | 17:30 |
| TheJulia | between deployments that is, we're not really going to support troubleshooting issues when cleaning is not performed because the bios nvram might have an update pointing to that other disk and there is no way for the bios to know if it is not valid if you've been writing oses across various drives | 17:31 |
| anupn | TheJulia: Here is my node show http://paste.openstack.org/show/747806/. Ah so I should erase the sda drive or whatever I chose before this deployment? | 17:32 |
| anupn | TheJulia: Also does it matter if there are multiple drives or not? | 17:33 |
| TheJulia | anupn: you should have cleaning enabled... | 17:33 |
| anupn | I am checking if cleaning was enabled by default in Bifrost | 17:34 |
| anupn | or Ironic | 17:34 |
| TheJulia | nvram, if using uefi boot, is going to choose based upon the records it has and depending on bios behavior and ultimately what your writing, it may not even be compatible. These are things that you'll need to investigate and resolve | 17:34 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: fast tracked deployment support https://review.openstack.org/635996 | 17:34 |
| anupn | TheJulia: Mine is BIOS boot. And I see cleaning is disabled by default https://github.com/openstack/bifrost/blob/734f63cdbb5458aac05a68f5fea3a3146aafec69/playbooks/roles/bifrost-ironic-install/defaults/main.yml#L4 | 17:37 |
| *** derekh has quit IRC | 17:38 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-tempest-plugin master: Add fast-track testing capability https://review.openstack.org/637039 | 17:39 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Add fast-track testing https://review.openstack.org/641104 | 17:39 |
| anupn | TheJulia: I may be wrongly understanding but it didn't complained about unable to clean earlier partitions. I see earlier there was one partition and now deployment created two partitions | 17:42 |
| *** priteau has joined #openstack-ironic | 17:45 | |
| *** e0ne has joined #openstack-ironic | 17:46 | |
| TheJulia | anupn: so that sounds like it could be the configuration drive | 17:48 |
| anupn | TheJulia: Oh so you mean one partition is configuration drive, yes that could be | 17:50 |
| openstackgerrit | Bob Fournier proposed openstack/ironic-inspector stable/queens: Use getaddrinfo instead of gethostbyname while resolving BMC address https://review.openstack.org/643409 | 17:51 |
| anupn | as I saw the contents in config drive getting attached by IPA. So what should I be trying now? A complete empty drive + only single empty drive in the system with no other drives? | 17:52 |
| anupn | TheJulia: Also will cleaning just clean the drive where image is getting written or all the drives that I have in my system? I doubt it will erase other drives where image is not getting written | 17:54 |
| TheJulia | anupn: I really don't quite understand | 17:55 |
| TheJulia | anupn: cleaning cleans all drives | 17:55 |
| anupn | TheJulia: Ah okie. Then I would probably not use cleaning for now as I want the contents on the other drives. What I meant is should I then wipe the sda drive on my own and let that drive (sda) be alone in the system without having other drives installed? | 18:01 |
| *** trown|lunch is now known as trown | 18:01 | |
| openstackgerrit | Bob Fournier proposed openstack/ironic-inspector stable/queens: Use getaddrinfo instead of gethostbyname while resolving BMC address https://review.openstack.org/643409 | 18:23 |
| openstackgerrit | Bob Fournier proposed openstack/ironic-inspector stable/queens: Fix lookup when ipmi_address is a hostname https://review.openstack.org/643419 | 18:23 |
| *** e0ne has quit IRC | 18:33 | |
| TheJulia | anupn: worth a try | 18:36 |
| *** patchbot has quit IRC | 18:37 | |
| *** patchbot has joined #openstack-ironic | 18:39 | |
| *** e0ne has joined #openstack-ironic | 18:40 | |
| anupn | TheJulia: yup, trying that now | 18:47 |
| *** gmann is now known as gmann_afk | 18:51 | |
| *** priteau has quit IRC | 18:56 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Make metrics usable https://review.openstack.org/643436 | 19:10 |
| TheJulia | dtantsur|afk: iurygregory ^^^ | 19:10 |
| *** priteau has joined #openstack-ironic | 19:16 | |
| *** priteau has quit IRC | 19:21 | |
| *** anupn has quit IRC | 19:25 | |
| *** rpioso|afk is now known as rpioso | 19:25 | |
| *** dougsz has quit IRC | 19:27 | |
| *** absubram has joined #openstack-ironic | 19:35 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Kg key for IPMIv2 authentication https://review.openstack.org/641658 | 19:41 |
| *** yolanda has quit IRC | 19:46 | |
| efried | o/ | 20:02 |
| * efried is excited | 20:02 | |
| efried | Check this out, y'all | 20:02 |
| efried | https://review.openstack.org/#/c/642899/ | 20:02 |
| patchbot | patch 642899 - nova - WIP/PoC: Bypass ironicclient for node.get - 7 patch sets | 20:02 |
| efried | I swapped out 'node.get' with a direct API call through ksa and the job passes. | 20:03 |
| efried | (after 7 patch sets worth of debugging :) | 20:03 |
| efried | TheJulia, dtantsur|afk: ^ | 20:03 |
| dustinc | nice! | 20:09 |
| *** yajushiSri has quit IRC | 20:15 | |
| *** Lucas_Gray has joined #openstack-ironic | 20:17 | |
| mbeierl | anyone here worked with L3 routing, dhcprelay and Ironic? | 20:19 |
| mbeierl | the dnsmasq process on the ironic node is telling me: DHCPDISCOVER(br-ctlplane) 24:6e:96:22:3c:a4 ignored and I don't know why | 20:21 |
| larsks | mbeierl: there are usually a couple of dnsmasq processes. Are you sure you're looking at the right one? | 20:22 |
| mbeierl | nobody 16685 0.0 0.0 53888 1168 ? S 11:17 0:00 dnsmasq --no-hosts --no-resolv --except-interface=lo --pid-file=/var/lib/neutron/dhcp/daae77ea-f861-491d-ba4e-be44a386a9b3/pid --dhcp-hostsfile=/var/lib/neutron/dhcp/daae77ea-f861-491d-ba4e-be44a386a9b3/host --addn-hosts=/var/lib/neutron/dhcp/daae77ea-f861-491d-ba4e-be44a386a9b3/addn_hosts --dhcp-optsfile=/var/lib/neutron/dhcp/daae77ea-f861-491d-ba4e-be44a386a9b3/ | 20:23 |
| mbeierl | opts --dhcp-leasefile=/var/lib/neutron/dhcp/daae77ea-f861-491d-ba4e-be44a386a9b3/leases --dhcp-match=set:ipxe,175 --bind-interfaces --interface=tapf5dffafc-f1 --dhcp-range=set:tag0,192.168.120.0,static,255.255.255.0,86400s --dhcp-range=set:tag1,192.168.121.0,static,255.255.255.0,86400s --dhcp-option-force=option:mtu,1500 --dhcp-lease-max=512 --conf-file=/etc/dnsmasq-ironic.conf --domain=localdomain | 20:23 |
| larsks | Yeah, that's the right one. | 20:23 |
| mbeierl | larsks: thanks :) | 20:23 |
| mbeierl | larsks: oh - hang on - the log message is from the inspector PID | 20:24 |
| mbeierl | so I have no log message | 20:24 |
| larsks | Yeah, that's what I was wondering about :) | 20:24 |
| larsks | You expect to see those messages from the inspector dnsmasq instance. | 20:24 |
| mbeierl | ok | 20:25 |
| mbeierl | any way to turn up logging for ironic's dnsmasq? | 20:25 |
| mbeierl | put it into the config file? | 20:25 |
| *** priteau has joined #openstack-ironic | 20:25 | |
| larsks | I'm not sure. You can inspect the files in /var/lib/neutron/dhcp/daae77ea-f861-491d-ba4e-be44a386a9b3/ to make sure the config looks sane. | 20:26 |
| *** dsneddon has quit IRC | 20:27 | |
| larsks | Is CI healthy? I see a failure in openstack-tox-lower-constraints because some dependencies were missing (http://logs.openstack.org/34/642734/5/check/openstack-tox-lower-constraints/46a5e93/job-output.txt.gz#_2019-03-14_01_42_41_059548). | 20:29 |
| larsks | I'm going to recheck and cross my fingers. | 20:29 |
| efried | Ironic API not yet set up to handle X-OpenStack-API-Version ?? | 20:36 |
| efried | nm, must have been spelling it wrong. | 20:37 |
| mbeierl | larsks: the configs in neutron look good. I can see the entry for my MAC addr in host | 20:38 |
| mbeierl | and the set and ip addr for it all match up in the other neutron files | 20:38 |
| larsks | mbeierl: and the problem is your host is just not getting a response? | 20:39 |
| mbeierl | correct. I have wireshark running on the ctlplane network and I see the request coming in with the correct MAC addr, via the relay | 20:39 |
| *** anupn has joined #openstack-ironic | 20:40 | |
| mbeierl | if I'm looking at br-ctlplane (which is the IP address where Ironic and dnsmasq are running), what does this line in the dnsmasq ps entry mean? --interface=tapf5dffafc-f1 | 20:40 |
| mbeierl | oh, and I should mention, PXE booting from the same L2 subnet works. I get DHCP responses no problem | 20:41 |
| openstackgerrit | Merged openstack/ironic master: fast tracked deployment support https://review.openstack.org/635996 | 20:43 |
| *** priteau has quit IRC | 20:43 | |
| *** hamzy has quit IRC | 20:45 | |
| larsks | mbeierl: That seems odd. I don't have any clever ideas, sorry. | 20:46 |
| mbeierl | larsks: thanks. Just to clarify - the Ironic dnsmasq listens to the TAP interface only? | 20:47 |
| larsks | mbeierl: that should be what the --interface=tapf5dffafc-f1 option means. | 20:49 |
| mbeierl | interesting. | 20:49 |
| mbeierl | larsks: the tap5f... in on bridge br-int inside of OVS | 20:50 |
| mbeierl | there is another br-ctlplane, but it has no tap interface | 20:50 |
| larsks | That's probably okay. That matches what my local setup looks like, at least. | 20:52 |
| mbeierl | ok | 20:52 |
| dustinc | efried: I am new to Openstack and was wondering what the benefit of using ksa over ironicclient is, are there any benefits/helpers in ironicclient that would be lost or have to be reimplemented? Is there a performance benefit to calling the API directly with ksa? | 20:52 |
| *** mjturek has quit IRC | 20:53 | |
| efried | dustinc: okay, there's a lot of questions in there... | 20:53 |
| dustinc | sorry :) | 20:54 |
| efried | dustinc: Generally speaking, OpenStack is trying to move away from having these *client layers in between service A and service B it wants to talk to. | 20:54 |
| efried | (no, no worries at all, just it's going to take some explaining) | 20:54 |
| mbeierl | larsks: ovs-tcpdump of the interface does not show the packets coming in :( | 20:55 |
| efried | The idea is that the API ought to be pretty accessible (usability-wise) and not need a client layer to massage inputs and outputs. | 20:55 |
| efried | Now what we would *really* like is to use openstacksdk - but in the case of nova-talks-to-ironic, there's some framework needed that isn't in place yet. | 20:55 |
| efried | So using direct ksa is the next best thing. | 20:55 |
| larsks | What is "ksa"? | 20:56 |
| efried | Note that ironicclient uses ksa internally anyway. So the only "benefit" is massaging inputs and outputs. | 20:56 |
| efried | keystoneauth... | 20:56 |
| larsks | Ah, thanks. | 20:56 |
| efried | https://github.com/openstack/keystoneauth | 20:56 |
| dustinc | I see. So would the intention to eventually be to completely deprecate ironicclient, or would it still exist for some other consumer? | 20:56 |
| efried | ...which is a shim layer that presents some convenient mechanisms for service A to talk to service B over a keystone-authenticated session. | 20:56 |
| *** anupn has quit IRC | 20:57 | |
| efried | dustinc: Not sure, perhaps TheJulia wants to answer that; but certainly the intention is to remove use of ironicclient completely from nova. | 20:57 |
| mbeierl | larsks: one more question - is your environment set up for L3 routed Ironic? (ie: for spine/leaf)? | 20:58 |
| efried | actually, I think ironicclient is also the CLI... | 20:58 |
| larsks | mbeierl: nope, it is not. | 20:58 |
| efried | BUT we're also trying to move all CLIs to osc with plugins, so... | 20:58 |
| dustinc | I was wondering if it was but wasn't sure | 20:58 |
| mbeierl | larsks: ok. so I might have missed something else. Thanks. I'll go digging some more. Gotta run for today! | 20:58 |
| efried | What I don't know is whether ironicclient has any other consumers. | 20:58 |
| dustinc | it seems it would make sense to have a single CLI that uses same interfaces/methods as components use internally so any massaging between is only implemented once | 20:59 |
| efried | you would think so | 20:59 |
| efried | but the client layers turn out to be more trouble than they're worth. | 20:59 |
| dustinc | (that is just a general though, I don't know enough about the project as a whole to know how it applies specifically) | 20:59 |
| dustinc | thought* | 20:59 |
| dustinc | I see | 20:59 |
| *** rh-jelabarre has quit IRC | 21:00 | |
| efried | If you dig into the ironicclient code, for example, you'll see it's like 50% overblown abstraction layers and 40% legacy code that we have to keep around for backward compatibility. | 21:00 |
| efried | The other 10% is translating input args into querystrings and calling .json() on the responses | 21:00 |
| dustinc | that sounds fun | 21:00 |
| efried | right, so the point is that all of that is very little code, and should really be done in a consumer-specific way by the consumer. | 21:01 |
| efried | so | 21:01 |
| *** baha has quit IRC | 21:01 | |
| dustinc | thanks for the info, it makes a lot more sense now. I also commented on your code with a small detail/specific question | 21:01 |
| efried | In https://review.openstack.org/#/c/642899/ I have done that, created a direct-through-ksa method for one of the otherwise-ironicclient methods, and the CI job proves that it's working, as far as it goes. | 21:01 |
| patchbot | patch 642899 - nova - WIP/PoC: Bypass ironicclient for node.get - 8 patch sets | 21:01 |
| *** anupn has joined #openstack-ironic | 21:02 | |
| dustinc | I think I am going to go dig into ksa now, thanks efried | 21:05 |
| efried | dustinc: Subsequent patches (which ima go work on now) should be smaller because the framework is already in place. | 21:05 |
| efried | dustinc: Okay, good luck with that. #openstack-keystone is your home for questions on ksa. | 21:05 |
| dustinc | as I try to ramp up on one project, my learning backlog is growing exponentially... | 21:08 |
| efried | yes, that happens | 21:12 |
| efried | dustinc: However, if you want an entry that lets you start slinging code fairly quickly without having to grok a huge amount of background, you could work on this thing... | 21:12 |
| efried | You would have both nova and ironic people chipping in to buy you a cookie | 21:13 |
| * TheJulia tries to wake up | 21:14 | |
| efried | TheJulia: Here's what you missed: That patch to get nova talking through ksa instead of ironicclient for node.get - works! | 21:15 |
| TheJulia | mbeierl: sounds like your trying to do a spine and leaf config? | 21:15 |
| TheJulia | larsks: ci was healthy at least two hours ago, haven't looked since | 21:15 |
| efried | TheJulia: so I'm going to start working on the other APIs (unless I can talk dustinc into doing it :) | 21:15 |
| TheJulia | efried: we don't handle the generic tag afaik, it has not been a priority for anyone | 21:16 |
| efried | generic tag? | 21:16 |
| *** trown is now known as trown|outtypewww | 21:16 | |
| dustinc | edfried: I can def. work on it, just going to be a bit slow to get started | 21:17 |
| TheJulia | dustinc: basically ironicclient has a lot of code that doesn't need to be carried and the overall idea that efried is heading along is to make things simpler by removing the library which should ultimately fix a lot of the headaches because the client automatically does json to object conversion magic and realistically I think everyone would be happier with just json and less hand-holding that the client library | 21:18 |
| TheJulia | provides. Also, an overall community desire exists to remove a lot of the heavy cross-app library dependencies for just rest APIs to reduce the overall footprint because client libraries start to pull in many dependencies | 21:18 |
| TheJulia | dustinc: since they are also geared for end user on command lines | 21:18 |
| efried | phew, well said, that's what I meant :) | 21:18 |
| efried | TheJulia: Does ironicclient have any consumers other than nova and ironic CLI? | 21:19 |
| TheJulia | efried: beyond nova, tons of consumers and the osc plugin | 21:19 |
| efried | okay, so we're not going to get rid of ironicclient after this :) | 21:19 |
| dustinc | makes sense...is the intention to continue to use them for cli, or is there a goal to replace it with something else for that as well? | 21:19 |
| *** andrein has quit IRC | 21:20 | |
| dustinc | (sorry that question is already out of date) | 21:20 |
| TheJulia | The other motivation was consumer level groking of the data, which turns out to be kind of important | 21:20 |
| TheJulia | efried: the generic api microversion request | 21:20 |
| TheJulia | oh, and dustinc... if you like beer... we buy that too | 21:21 |
| efried | TheJulia: Okay, not sure how, but sending microversion=x.y through adapter.get seems to actually work. Maybe ksa has a special case built in. (I seem to maybe kinda remember something about that.) | 21:22 |
| TheJulia | heh, yeah, irccloud is kind of laggy and I'm a bit distracted today | 21:22 |
| TheJulia | efried: yeah, there is code in ksa to handle our silliness | 21:22 |
| dustinc | waiting to find out if I go to Denver, will be nice to meet everyone I haven't yet | 21:22 |
| efried | so dustinc if you're interested in pursuing this, I can help start you off. | 21:23 |
| dustinc | TheJulia: I was just running LimeChat locally when I saw irccloud in your host string and checked it out...sold.. | 21:23 |
| TheJulia | dustinc: toatally worth the $5 a month | 21:24 |
| dustinc | efried: definitely interested | 21:24 |
| * TheJulia blames monty | 21:24 | |
| openstackgerrit | Merged openstack/ironic master: Fix idrac Job.state renamed to Job.status https://review.openstack.org/643142 | 21:24 |
| openstackgerrit | Merged openstack/ironic master: Deploy Templates: API reference https://review.openstack.org/642498 | 21:24 |
| openstackgerrit | Merged openstack/ironic master: Deploy Templates: documentation https://review.openstack.org/642531 | 21:24 |
| dustinc | TheJulia: does the mobile client stay connected reliably enough to always be within the 2-hour free window? | 21:24 |
| TheJulia | I don't know, I've paid for it from about 30 minutes after I started using it | 21:25 |
| TheJulia | I was totally sold | 21:25 |
| TheJulia | largely because I get the notifications and all | 21:25 |
| dustinc | I am still trying to figure out what is free and what is not since i am on the initial trial | 21:25 |
| TheJulia | Yeah, I've been using it for years | 21:26 |
| efried | dustinc: The pattern is going to be pretty simple: in nova/virt/ironic/driver.py search for self.ironicclient.call. The first arg is going to be like 'foo.bar'. Write a method in nova/virt/ironic/client_wrapper.py _IronicClient called foo_bar and within it duplicate the logic of the respective method in python-ironicclient/ironicclient/v1/foo/bar.py | 21:26 |
| dustinc | (I think they warned me about this in 5th grade DARE class) | 21:26 |
| *** bfournie has quit IRC | 21:26 | |
| efried | dustinc: Start off with another get-ish one to keep it simple | 21:26 |
| efried | when you need a put/post/delete you'll have to add another primitive to _IronicWrapper. | 21:27 |
| TheJulia | efried: I haven't looked recently at your patches, how much of the unit test headaches did you have to touch | 21:27 |
| TheJulia | sounds liek you kind of routed around the object craziness | 21:27 |
| efried | dustinc: Do you know how to work changes in a series? | 21:27 |
| dustinc | not exactly..was just trying to find out | 21:27 |
| efried | TheJulia: I haven't touched unit/functional test yet. I commented out all the CI except the job that does real things. | 21:27 |
| efried | dustinc: Someone recently authored an excellent article in the contributor docs on this subject. Let me go dig it up... | 21:28 |
| dustinc | I have tons of experience with Github on a single team with feature branch model but pretty mcuh nothing else | 21:28 |
| efried | dustinc: https://docs.openstack.org/contributors/code-and-documentation/patch-series-tutorial.html | 21:29 |
| dustinc | excellent..thanks | 21:30 |
| dustinc | hey look who that someone was... | 21:30 |
| *** whoami-rajat has quit IRC | 21:31 | |
| efried | I guess I could take my handle out of those screenshots. Lazy copy/paste. | 21:32 |
| hogepodge | hey everybody, wanted to make you all aware of this http://lists.openstack.org/pipermail/openstack-discuss/2019-March/003857.html | 22:20 |
| efried | dustinc: Ima revert the zuul stuff in my patch so I can start working on the unit test. If you already pulled, you'll have to pull again... | 22:21 |
| *** pcaruana has quit IRC | 22:21 | |
| dustinc | ok thanks | 22:24 |
| *** e0ne has quit IRC | 22:28 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Make metrics usable https://review.openstack.org/643436 | 22:29 |
| TheJulia | hogepodge: would it help to add to the meeting agenda? | 22:30 |
| *** dteselkin has quit IRC | 22:33 | |
| *** dteselkin has joined #openstack-ironic | 22:35 | |
| *** absubram has quit IRC | 22:36 | |
| *** gmann_afk is now known as gmann | 22:40 | |
| rpioso | TheJulia: o/ | 22:50 |
| TheJulia | o/ | 22:50 |
| rpioso | TheJulia: Do you have the info you need? | 22:50 |
| *** rcernin has joined #openstack-ironic | 23:00 | |
| TheJulia | rpioso: Thanks! | 23:02 |
| *** anupn has quit IRC | 23:13 | |
| rpioso | TheJulia: yw | 23:13 |
| *** gyee has quit IRC | 23:34 | |
| *** Lucas_Gray has quit IRC | 23:39 | |
| *** dsneddon has joined #openstack-ironic | 23:55 | |
| *** sthussey has quit IRC | 23:57 | |
| *** dsneddon has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!