| *** sdake has joined #openstack-ironic | 00:04 | |
| *** ijw has joined #openstack-ironic | 00:47 | |
| *** sdake has quit IRC | 01:08 | |
| *** ijw has quit IRC | 01:15 | |
| *** betherly has joined #openstack-ironic | 01:17 | |
| *** ijw has joined #openstack-ironic | 01:17 | |
| *** betherly has quit IRC | 01:22 | |
| *** _fragatina has quit IRC | 01:22 | |
| *** hwoarang has quit IRC | 01:23 | |
| *** hwoarang has joined #openstack-ironic | 01:29 | |
| *** gyee has quit IRC | 01:38 | |
| *** bfournie has joined #openstack-ironic | 01:41 | |
| *** dsneddon has quit IRC | 01:57 | |
| * tonyb wonders if TheJulia is around and meeting free? | 02:00 | |
| * mcarden points tonyb to the backscroll at about 00:35 (~13hrs ago) today | 02:07 | |
| tonyb | mcarden: Thanks and #rats | 02:09 |
|---|---|---|
| mcarden | :) | 02:09 |
| *** sdake has joined #openstack-ironic | 02:09 | |
| *** betherly has joined #openstack-ironic | 02:10 | |
| *** betherly has quit IRC | 02:14 | |
| TheJulia | tonyb: boot_interface as of steinnsuppoets ipxe instead of pxe | 02:17 |
| TheJulia | Sorry, was put eating dinner | 02:18 |
| tonyb | TheJulia: Thanks no need to appologise | 02:18 |
| *** rloo has quit IRC | 02:18 | |
| tonyb | TheJulia: I think I need to do ... nde set --boot-interface pxe <node> to force it to pxe now that both are enbaled right? | 02:19 |
| tonyb | ... where both enbaled means: enabled_boot_interfaces=ilo-pxe,ipxe,pxe | 02:20 |
| TheJulia | Is ipxe_enabled set in the config? | 02:21 |
| tonyb | ironic/ironic.conf:ipxe_enabled=False | 02:21 |
| *** dsneddon has joined #openstack-ironic | 02:22 | |
| * tonyb suspects there is work to be done in puppet-ironic | 02:23 | |
| TheJulia | Nope | 02:24 |
| tonyb | Nope? | 02:25 |
| TheJulia | Ipxe_enabledneffectively forces pxe to be ipxe | 02:25 |
| TheJulia | Sorry, on my phone at the moment | 02:25 |
| tonyb | Okay don't worry enjoy your evening | 02:25 |
| tonyb | doing that --boot-interface has fixed my issue so I can make forward progress | 02:26 |
| tonyb | TheJulia: we can worry about how the issue came up and where the bug is later ;P | 02:26 |
| *** dsneddon has quit IRC | 02:27 | |
| TheJulia | tonyb: we didn't change behavior at all... | 02:29 |
| TheJulia | So... I'm curious about bug | 02:29 |
| TheJulia | Anyway, I need to get some sleep, I am two time zones ahead of my normal time zone and need to get up early. | 02:30 |
| tonyb | Go sleep I'll send an email ;p | 02:30 |
| TheJulia | Okay | 02:36 |
| *** dsneddon has joined #openstack-ironic | 02:38 | |
| *** dsneddon has quit IRC | 02:44 | |
| *** sdake has quit IRC | 02:46 | |
| *** dsneddon has joined #openstack-ironic | 03:07 | |
| *** andrein has quit IRC | 03:10 | |
| *** ijw has quit IRC | 03:13 | |
| *** ijw_ has joined #openstack-ironic | 03:13 | |
| *** dsneddon has quit IRC | 03:14 | |
| *** dsneddon has joined #openstack-ironic | 03:14 | |
| *** sdake has joined #openstack-ironic | 03:16 | |
| *** sdake has quit IRC | 03:17 | |
| *** hjensas has quit IRC | 03:18 | |
| *** dsneddon has quit IRC | 03:19 | |
| *** sdake has joined #openstack-ironic | 03:22 | |
| *** dsneddon has joined #openstack-ironic | 03:22 | |
| *** sdake has quit IRC | 03:23 | |
| *** Nisha_Agarwal has joined #openstack-ironic | 03:25 | |
| Nisha_Agarwal | TheJulia, ping | 03:25 |
| *** sdake has joined #openstack-ironic | 03:27 | |
| *** dsneddon has quit IRC | 03:27 | |
| *** ijw_ has quit IRC | 03:38 | |
| *** sdake has quit IRC | 03:40 | |
| *** dsneddon has joined #openstack-ironic | 03:48 | |
| *** dsneddon has quit IRC | 04:04 | |
| *** hjensas has joined #openstack-ironic | 04:11 | |
| *** mbeierl has quit IRC | 04:24 | |
| *** Nisha_Agarwal has quit IRC | 04:26 | |
| *** betherly has joined #openstack-ironic | 04:26 | |
| *** betherly has quit IRC | 04:31 | |
| openstackgerrit | paresh sao proposed openstack/ironic master: [docs] IPv6 support for iLO https://review.openstack.org/638590 | 04:31 |
| *** dnuka has joined #openstack-ironic | 04:42 | |
| dnuka | good morning ironic | 04:43 |
| *** _fragatina has joined #openstack-ironic | 04:54 | |
| *** gkadam has joined #openstack-ironic | 04:55 | |
| *** dsneddon has joined #openstack-ironic | 05:00 | |
| *** mbeierl has joined #openstack-ironic | 05:17 | |
| *** sdake has joined #openstack-ironic | 05:30 | |
| *** _fragatina has quit IRC | 05:41 | |
| *** sdake has quit IRC | 06:02 | |
| *** sdake has joined #openstack-ironic | 06:21 | |
| *** moshele has joined #openstack-ironic | 06:24 | |
| *** sdake has quit IRC | 06:26 | |
| *** gkadam has quit IRC | 06:27 | |
| *** sdake has joined #openstack-ironic | 06:28 | |
| *** sdake has quit IRC | 06:28 | |
| *** dnuka has quit IRC | 06:31 | |
| *** ijw has joined #openstack-ironic | 06:32 | |
| *** dnuka has joined #openstack-ironic | 06:38 | |
| *** sdake has joined #openstack-ironic | 06:42 | |
| etingof | dnuka, o/ | 06:45 |
| dnuka | good morning etingof o/ | 06:45 |
| zer0c00l | Any one had problems using the discount code for the denver summit? | 06:50 |
| arne_wiebalck | good morning, ironic! dnuka etingof o/ | 07:12 |
| dnuka | good morning arne_wiebalck o/ | 07:12 |
| *** sdake has quit IRC | 07:15 | |
| *** sdake has joined #openstack-ironic | 07:18 | |
| etingof | hey-hey, good news - winter is officially ending today! o/ | 07:18 |
| *** sdake has quit IRC | 07:19 | |
| *** ijw has quit IRC | 07:25 | |
| *** ijw has joined #openstack-ironic | 07:25 | |
| *** ijw has quit IRC | 07:34 | |
| openstackgerrit | Merged openstack/sushy master: Add support for ilo Virtual Media https://review.openstack.org/636845 | 08:00 |
| *** jtomasek has joined #openstack-ironic | 08:01 | |
| *** gkadam has joined #openstack-ironic | 08:02 | |
| rpittau|sardegna | good morning ironic! o/ | 08:07 |
| dnuka | good morning rpittau|sardegna | 08:12 |
| rpittau|sardegna | hi dnuka :) | 08:13 |
| dnuka | :) | 08:17 |
| *** tssurya has joined #openstack-ironic | 08:19 | |
| *** sburette has joined #openstack-ironic | 08:28 | |
| *** amoralej|off is now known as amoralej | 08:32 | |
| *** pcaruana has joined #openstack-ironic | 08:32 | |
| *** andrein has joined #openstack-ironic | 08:35 | |
| iurygregory_mtg | good morning everyone | 09:02 |
| dnuka | morning iurygregory_mtg | 09:06 |
| *** dougsz has joined #openstack-ironic | 09:06 | |
| *** iurygregory_mtg is now known as iurygregory | 09:06 | |
| *** dnuka is now known as dnuka|brb | 09:07 | |
| *** sdake has joined #openstack-ironic | 09:14 | |
| *** andrein has quit IRC | 09:16 | |
| openstackgerrit | Digambar proposed openstack/ironic master: Modify the iDRAC driver to use realtime RAID creation https://review.openstack.org/634903 | 09:18 |
| *** olivierbourdon38 has joined #openstack-ironic | 09:20 | |
| *** olivierbourdon38 has quit IRC | 09:22 | |
| *** iurygregory has quit IRC | 09:24 | |
| *** iurygregory has joined #openstack-ironic | 09:25 | |
| *** e0ne has joined #openstack-ironic | 09:32 | |
| *** diga has joined #openstack-ironic | 09:33 | |
| *** derekh has joined #openstack-ironic | 09:35 | |
| *** andrein has joined #openstack-ironic | 09:37 | |
| openstackgerrit | Digambar proposed openstack/ironic master: Modify the iDRAC driver to use realtime RAID creation https://review.openstack.org/634903 | 09:45 |
| diga | arne_wiebalck: Hi | 09:46 |
| arne_wiebalck | hey diga o/ | 09:46 |
| diga | Updated the patch, can you go through it | 09:46 |
| arne_wiebalck | diga: sure | 09:46 |
| diga | :) | 09:46 |
| *** sdake has quit IRC | 09:46 | |
| *** rcernin has quit IRC | 09:48 | |
| *** dtantsur|afk is now known as dtantsur | 09:49 | |
| dtantsur | morning ironic | 09:49 |
| arne_wiebalck | diga: shouldn’t the reboot param be set to ‘False’ in the else branch? | 09:51 |
| arne_wiebalck | diga: my point is still the same as on the review | 09:51 |
| arne_wiebalck | diga: only the last controller should be rebooted (if anyone at all), the else branch does not handle the last controller, so it should always be ‘False’ | 09:52 |
| openstackgerrit | Mark Goddard proposed openstack/ironic master: Fix pysendfile requirement marker https://review.openstack.org/639967 | 09:56 |
| *** dnuka|brb is now known as dnuka | 09:56 | |
| dnuka | good morning dtantsur o/ | 09:57 |
| *** andrein has quit IRC | 09:57 | |
| diga | arne_wiebalck: okay | 10:01 |
| diga | arne_wiebalck: Let me think over this | 10:02 |
| diga | arne_wiebalck: one question, do you think we should not add reboot=True in else clause | 10:03 |
| diga | ? | 10:03 |
| *** iurygregory has quit IRC | 10:03 | |
| arne_wiebalck | diga: no :) | 10:04 |
| arne_wiebalck | diga: why would you think we need reboot=True ? | 10:04 |
| arne_wiebalck | diga: I’m comparing with the original logic. | 10:05 |
| diga | arne_wiebalck: okay | 10:06 |
| diga | Let's go one by one scenario | 10:07 |
| diga | If node has three controller | 10:07 |
| diga | 1. supports realtime, 2. not supports realtime 3 supports realtime | 10:07 |
| diga | in else clause, first two controllers come into picture | 10:08 |
| arne_wiebalck | diga: right | 10:08 |
| diga | arne_wiebalck: so first will set to True, second will be set to False | 10:08 |
| diga | all_realtime = False now | 10:09 |
| arne_wiebalck | diga: correct | 10:09 |
| diga | arne_wiebalck: when we reach to last controller | 10:09 |
| *** iurygregory has joined #openstack-ironic | 10:09 | |
| arne_wiebalck | diga: wait | 10:09 |
| arne_wiebalck | diga: stay with the 2. controller for a sec | 10:09 |
| diga | arne_wiebalck: ok | 10:09 |
| * iurygregory my wifi connection in the office is a pain today =( | 10:09 | |
| diga | arne_wiebalck: okay | 10:09 |
| arne_wiebalck | diga: what do you expect to happen in the else branch for the reboot parameter for controllers 1 and 2? | 10:10 |
| arne_wiebalck | diga: not what your code does, but you want to happen | 10:10 |
| arne_wiebalck | diga: not what your code does, but what you want to happen | 10:10 |
| diga | for 1st controller, reboot sets to True, and 2nd controller, reboot should set to False | 10:11 |
| arne_wiebalck | diga: why should controller 1 be rebooted? it supports realtime | 10:12 |
| diga | sorry | 10:12 |
| diga | 1st should reboot to False, 2nd to True | 10:12 |
| arne_wiebalck | diga: why should the 2nd be rebooted? we will reboot the last one | 10:12 |
| arne_wiebalck | diga: b/c of the all_realtime flag | 10:13 |
| diga | arne_wiebalck: okay | 10:13 |
| diga | because else clause should server the request depends on controller supports realtime or not | 10:14 |
| diga | arne_wiebalck: what you think is, if 2nd supports not realtime, then in else clause we should not set to True as it will be rebooted in the last iteration anyway | 10:15 |
| arne_wiebalck | diga: that’s what I understood, yes | 10:15 |
| arne_wiebalck | diga: so, in the end we will only reboot one | 10:16 |
| diga | arne_wiebalck: yes | 10:16 |
| arne_wiebalck | diga: is my understanding not correct? | 10:16 |
| diga | arne_wiebalck: you are right | 10:17 |
| arne_wiebalck | diga: as mentioned on the review as well, it’d be good to add a brief description of the logic, as it may get confusing :) | 10:18 |
| diga | arne_wiebalck: in that case, we don't need if-else loop here | 10:18 |
| diga | arne_wiebalck: okay I will add description | 10:19 |
| arne_wiebalck | diga: I think we do as we need to know when we’re handling the last controller, no? | 10:19 |
| diga | arne_wiebalck: yeah, but what I think is we should respect 1st/2nd controller execution as per the supported params | 10:21 |
| diga | arne_wiebalck: and anyway, in the end we are handling the scenario for last controller | 10:21 |
| diga | arne_wiebalck: I will discuss this with Chris also, then will update the patch | 10:22 |
| arne_wiebalck | diga: sounds good | 10:22 |
| diga | arne_wiebalck: Thank you | 10:23 |
| arne_wiebalck | diga: np | 10:23 |
| diga | arne_wiebalck: thank for reviewing the patch and your comments are really helpful also :) | 10:23 |
| arne_wiebalck | diga: :) | 10:23 |
| dnuka | dtantsur: thanks for the review :) | 10:25 |
| dtantsur | np | 10:25 |
| rpittau|sardegna | welcome back dtantsur :) | 10:27 |
| *** dnuka is now known as dnuka|brb | 10:31 | |
| dtantsur | thanks :) | 10:31 |
| *** sdake has joined #openstack-ironic | 10:37 | |
| *** iurygregory has quit IRC | 10:40 | |
| *** iurygregory has joined #openstack-ironic | 10:41 | |
| *** iurygregory has quit IRC | 10:42 | |
| *** iurygregory has joined #openstack-ironic | 10:42 | |
| etingof | if iurygregory trusts in networkmanager, he could probably use wire connection as well | 10:54 |
| *** dnuka|brb is now known as dnuka | 10:57 | |
| *** diga has quit IRC | 11:04 | |
| openstackgerrit | Merged openstack/sushy master: Add foundation for supporting Redfish OEMs https://review.openstack.org/463829 | 11:07 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Update the log message for ilo drivers https://review.openstack.org/639989 | 11:07 |
| openstackgerrit | Merged openstack/sushy master: Add support for the `CompositionService` resource https://review.openstack.org/608563 | 11:07 |
| dnuka | \o/ | 11:07 |
| arne_wiebalck | dnuka: :) | 11:09 |
| dnuka | :) | 11:10 |
| *** sdake has quit IRC | 11:18 | |
| *** sdake has joined #openstack-ironic | 11:21 | |
| openstackgerrit | Dhanuka Warusadura proposed openstack/sushy master: Add unit test case to `get_allowed_transfer_protocol_values()` https://review.openstack.org/637799 | 11:25 |
| *** andrein has joined #openstack-ironic | 11:26 | |
| *** dnuka is now known as dnuka|brb | 11:29 | |
| *** e0ne has quit IRC | 11:30 | |
| *** e0ne has joined #openstack-ironic | 11:36 | |
| *** _fragatina has joined #openstack-ironic | 11:46 | |
| *** dnuka|brb is now known as dnuka | 11:47 | |
| *** sdake has quit IRC | 11:55 | |
| *** amoralej is now known as amoralej|lunch | 12:05 | |
| iurygregory | etingof, i dont have ethernet cable in my desk =( | 12:09 |
| iurygregory | probably i will request one now XD | 12:10 |
| etingof | iurygregory, should be easy | 12:10 |
| rpioso | zer0c00l: I heard the expiration date was entered wrong. Just email summitreg@openstack.org and/or kendall@openstack.org and they will sort it out. | 12:11 |
| iurygregory | etingof, truth =) | 12:11 |
| openstackgerrit | Mark Goddard proposed openstack/ironic master: Deploy templates: fix updating steps in Python 3 https://review.openstack.org/640014 | 12:13 |
| * dnuka waves at rpioso :) | 12:23 | |
| * rpioso return’s dnuka’s morning greeting | 12:26 | |
| dnuka | :) | 12:26 |
| *** rh-jelabarre has joined #openstack-ironic | 12:52 | |
| *** trown|outtypewww is now known as trown | 12:58 | |
| dnuka | o/ | 13:25 |
| *** dnuka has quit IRC | 13:25 | |
| *** rh-jelabarre has quit IRC | 13:31 | |
| *** amoralej|lunch is now known as amoralej | 13:36 | |
| *** moshele has quit IRC | 13:40 | |
| openstackgerrit | Mark Goddard proposed openstack/ironic-tempest-plugin master: Deploy templates: add API tests https://review.openstack.org/637187 | 13:43 |
| dtantsur | etingof: could you check https://review.openstack.org/#/c/637799/ please? we're about to get the final release | 13:44 |
| patchbot | patch 637799 - sushy - Add unit test case to `get_allowed_transfer_protoc... - 3 patch sets | 13:44 |
| dtantsur | unfortunately, the message registry seems to fall out because of merge conflicts :( | 13:44 |
| dtantsur | TheJulia, jroll, I've extended https://storyboard.openstack.org/#!/story/2005083 with network_data, since it seems really useful indeed (e.g. to set up bonding) and because the relevant SDK patch was approved and will be released soon | 13:49 |
| *** rloo has joined #openstack-ironic | 13:51 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Update the log message for ilo drivers https://review.openstack.org/639989 | 13:56 |
| openstackgerrit | Mark Goddard proposed openstack/python-ironicclient master: Deploy templates: client support https://review.openstack.org/636931 | 13:57 |
| *** baha has joined #openstack-ironic | 13:58 | |
| *** betherly has joined #openstack-ironic | 14:02 | |
| *** moshele has joined #openstack-ironic | 14:03 | |
| *** mmethot has joined #openstack-ironic | 14:06 | |
| jroll | dtantsur: awesome, thanks | 14:09 |
| jroll | oh, I missed your last comment, agree | 14:10 |
| *** jistr is now known as jistr|call | 14:11 | |
| iurygregory | morning jroll o/ | 14:11 |
| jroll | morning :) | 14:12 |
| dtantsur | jroll: g'morning | 14:13 |
| jroll | \o | 14:14 |
| rpittau|sardegna | hi jroll :) | 14:16 |
| jroll | morning rpittau|sardegna | 14:16 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Allow building configdrive from JSON in the API https://review.openstack.org/639050 | 14:19 |
| dtantsur | ^^ WIP until openstacksdk release (and to avoid stealing microversion from deploy templates) | 14:20 |
| *** ianychoi has joined #openstack-ironic | 14:20 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic master: Update the log message for ilo drivers https://review.openstack.org/639989 | 14:23 |
| openstackgerrit | Rachit Kapadia proposed openstack/ironic master: Set boot_mode in node properties during OOB Introspection https://review.openstack.org/639698 | 14:28 |
| *** mjturek has joined #openstack-ironic | 14:39 | |
| mgoddard | dtantsur: you're very kind :) | 14:49 |
| mgoddard | [hands off my microversion :p] | 14:49 |
| dtantsur | lol | 14:51 |
| dtantsur | mgoddard: I'm even kinder: I'm reviewing the API patch right now :) | 14:51 |
| w14161_1 | Hi All, I deploy ironic to a baremetal node by devstack with neutron setting like http://paste.openstack.org/show/746505/, before run stack.sh, the baremetal node installed a ubuntu server 18.04 with physical network interface eno1, internet is ok, ping is ok. But during deploy, the internet and ping did not work so I only can access the node by IPMI console, by checking, found the eno1 already | 14:51 |
| w14161_1 | did not have original IP address, instead, at a interface br-ex, contains the original IP address, it seems the neutron network used br-ex so stack.sh use br-ex to connect to external network, that was I expected, but why internet and ping was broken during stack.sh? Any more settings to do at local.conf? | 14:52 |
| mgoddard | dtantsur: how far through? have fixes for ruby's comments I could push if you like? | 14:53 |
| dtantsur | mgoddard: please hold on, I'll throw a few nits there | 14:53 |
| dtantsur | reviewing tests right now | 14:53 |
| mgoddard | dtantsur: sure | 14:53 |
| rloo | Good morning ironic'ers | 14:53 |
| dtantsur | mgoddard: also let's rebase on top of the gate-fixing patch | 14:53 |
| dtantsur | morning rloo | 14:53 |
| *** rh-jelabarre has joined #openstack-ironic | 14:54 | |
| mgoddard | dtantsur: I added a depends-on (not pushed yet) | 14:54 |
| rloo | hjensas: want to check with you wrt the network event stuff; is there a PR that you want reviewed? | 14:54 |
| mgoddard | morning rloo | 14:54 |
| rloo | morning dtantsur, mgoddard :) | 14:54 |
| dtantsur | mgoddard: within the same repository it's better to just stack things on top of each other | 14:54 |
| hjensas | rloo: not right now, I hope to push 3 changes later today. | 14:55 |
| mgoddard | dtantsur: I guess. Either works | 14:55 |
| dtantsur | mgoddard: done! a few weird places in the tests, everything else looks okay | 15:00 |
| mgoddard | dtantsur: great, thanks. Conductor piece is pretty much ready, just finishing testing | 15:00 |
| *** arne_wiebalck_ has joined #openstack-ironic | 15:00 | |
| dtantsur | rloo: the gate fixing patch if you have a minute: https://review.openstack.org/#/c/639967/ | 15:00 |
| patchbot | patch 639967 - ironic - Fix pysendfile requirement marker - 1 patch set | 15:00 |
| rloo | hjensas: ok thx. | 15:01 |
| rloo | dtantsur: looking | 15:01 |
| *** mariojv has quit IRC | 15:08 | |
| *** mariojv has joined #openstack-ironic | 15:15 | |
| *** e0ne has quit IRC | 15:16 | |
| *** mjturek has quit IRC | 15:17 | |
| *** e0ne has joined #openstack-ironic | 15:19 | |
| rloo | mgoddard: i was thinking about the case of updating a deploy template. if we don't put a lock, and two requests come in to update it at the 'same' time, and both think they succeeded? similar thing wrt deleting template at the same time as updating. which may not matter as much. maybe we can assume for now that only one operator would actually be setting up deploy templates? | 15:21 |
| mgoddard | rloo: that's true. The only other thing I can think of that is not covered by the node lock is chassis. But no one cares about them | 15:22 |
| *** arne_wiebalck_ has quit IRC | 15:23 | |
| rloo | mgoddard: i was just about to mention the chassis. but we're not using them except as a means of grouping nodes. | 15:23 |
| dtantsur | time for etcd? :) | 15:25 |
| dtantsur | and Julia's locking interface work.. | 15:25 |
| * TheJulia whistles.... | 15:25 | |
| mgoddard | rloo: I think the main case to consider is updating the steps | 15:25 |
| rloo | mgoddard: yes. | 15:25 |
| dtantsur | oh hi TheJulia :) | 15:25 |
| TheJulia | in meeting still | 15:25 |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: Initial processing of network port events https://review.openstack.org/633729 | 15:26 |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: WiP - Implement Event Handler in driver interfaces https://review.openstack.org/637840 | 15:26 |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: WIP - Cleaning network - events https://review.openstack.org/637841 | 15:26 |
| rloo | dtantsur, TheJulia: I've been out of the loop for awhile (as you know). are there plans for this locking interface and will it solve this problem? if so, are we good with allowing the template updates w/o a lock for now, with a TODO or NOTE about it? | 15:26 |
| mgoddard | rloo: the locking interface currently only covers nodes | 15:26 |
| TheJulia | I'd have to tink about it | 15:26 |
| TheJulia | to think | 15:27 |
| mgoddard | rloo: I'd like to see it be more general than that | 15:27 |
| dtantsur | mgoddard: what's the worst case scenario of a conflict? | 15:27 |
| hjensas | rloo: ^^ if you had planned time for some reviews. The last one is still very WiP, but feedback is much appreciated. :) | 15:29 |
| mgoddard | dtantsur: two concurrent updates to template.steps, lots of possibilities | 15:29 |
| dtantsur | mgoddard: can we use transactions to at least ensure some consistent state? | 15:29 |
| rloo | hjensas: thx, i'm working on deploy templates now, so thinking maybe tomorrow... | 15:29 |
| dtantsur | i.e. serialize concurrent updates | 15:29 |
| *** baha has quit IRC | 15:30 | |
| mgoddard | dtantsur: I don't think it helps, because the read/modify/write happens outside of the DB | 15:30 |
| *** moshele has quit IRC | 15:30 | |
| dtantsur | mgoddard: can we at least make sure that the last request wins? | 15:31 |
| mgoddard | dtantsur: or rather there is a r/m/w in ironic-api, then one in the transaction | 15:31 |
| dtantsur | I'm thinking of locking the relevant deploy_templates raw with FOR UPDATE, then do everything else | 15:32 |
| mgoddard | dtantsur: I'll try to think about it. At the moment I don't think it would work like that | 15:32 |
| dtantsur | (this is how allocation API works) | 15:32 |
| mgoddard | this is how it works: | 15:32 |
| mgoddard | * query template from DB, populate object | 15:32 |
| mgoddard | * process PATCH to update object (e.g. modifying steps list) | 15:33 |
| mgoddard | * save object | 15:33 |
| mgoddard | this is outside of any transaction, in the API (or even conductor if we pushed it there) | 15:33 |
| dtantsur | okay, but can we make sure that the version from the most recent update gets saved? | 15:34 |
| dtantsur | I'm mostly worried about having some weird result in the steps | 15:34 |
| dtantsur | a whole different approach would be having a conductor_affinity (again, similar to allocations) | 15:35 |
| mgoddard | dtantsur: so probably we could make the most recent version win. What would we return to the loser though? | 15:35 |
| dtantsur | and make sure only one conductor can update (with its local log) | 15:35 |
| dtantsur | mgoddard: well, whatever version was valid at the same its update succeeded | 15:35 |
| mgoddard | dtantsur: would be a bit weird saying that the request succeeded but the result doesn't include the change | 15:37 |
| mgoddard | I probably need to sit and think about this for a little | 15:38 |
| *** sburette has quit IRC | 15:38 | |
| dtantsur | ++ | 15:39 |
| dtantsur | the conductor_affinity approach is a bit boring, but should work | 15:39 |
| mgoddard | I bet this can happen in other places, we quite often apply a patch in the API, then push the result to the conductor | 15:39 |
| mgoddard | I'm not sure moving to conductor helps - we always apply a patch in the API, then push the object to the conductor | 15:40 |
| mgoddard | the problem being we did the initial read in the API | 15:40 |
| *** andrein has quit IRC | 15:42 | |
| dtantsur | you're probably right | 15:42 |
| *** mjturek has joined #openstack-ironic | 15:43 | |
| * rloo sorry she asked | 15:43 | |
| mgoddard | I spent my first few jobs working with pthreads & C, on applications that should really have been written in a higher level language :) | 15:44 |
| *** andrein has joined #openstack-ironic | 15:44 | |
| dtantsur | heh | 15:44 |
| mgoddard | gives you a nose for a concurrency problem :) | 15:44 |
| rloo | I just looked at the conductor code when updating a node and yeah, could be problems there. | 15:45 |
| *** mjturek has quit IRC | 15:45 | |
| rloo | umm... I'm good with a NOTE or TODO for now? | 15:45 |
| mgoddard | uhoh | 15:45 |
| mgoddard | sometimes best not to pull at this thread | 15:45 |
| *** baha has joined #openstack-ironic | 15:46 | |
| mgoddard | it least it's not an async API, so a single client waiting for the response would work fine | 15:46 |
| rloo | right | 15:46 |
| *** baha has quit IRC | 15:47 | |
| mgoddard | this might be a nice incentive and new use case for locking work in Train | 15:47 |
| mgoddard | would we expect all use of the locking interface to be from the conductor, or API also? that could drive the API/conductor question | 15:48 |
| dtantsur | if anyone has a second, this seems important: https://review.openstack.org/#/c/639269/ | 15:49 |
| patchbot | patch 639269 - ironic-inspector - Follow up to introspection data driver api revision - 1 patch set | 15:49 |
| *** _fragatina has quit IRC | 15:51 | |
| rloo | dtantsur: that code looks fine to me, but I'm wondering if i need to understand the context. if it doesn't accept the name anymore -- does that break any user? | 15:52 |
| dtantsur | rloo: yep, it means that /v1/introspection/<node_id>/data/unprocessed no longer works if node_id is a name | 15:53 |
| rloo | dtantsur: and we're good with that, there's a release note or something? | 15:54 |
| dtantsur | rloo: it is a temporary regression in the current release, this patch fixes it | 15:54 |
| dtantsur | no release note because the bug has never been released | 15:54 |
| *** andrein has quit IRC | 15:54 | |
| rloo | dtantsur: this patch just enforces that name is not allowed. was it never allowed? | 15:55 |
| *** baha has joined #openstack-ironic | 15:55 | |
| dtantsur | rloo: https://review.openstack.org/#/c/639269/1/ironic_inspector/main.py returns support for names back | 15:55 |
| patchbot | patch 639269 - ironic-inspector - Follow up to introspection data driver api revision - 1 patch set | 15:55 |
| dtantsur | but moves it upper to the API level | 15:56 |
| rloo | dtantsur: ah, ok, i missed that line. thx. | 15:56 |
| *** andrein has joined #openstack-ironic | 15:57 | |
| *** jistr|call is now known as jistr | 16:00 | |
| *** andrein has quit IRC | 16:01 | |
| *** andrein has joined #openstack-ironic | 16:15 | |
| *** e0ne has quit IRC | 16:16 | |
| *** SpamapS has quit IRC | 16:18 | |
| *** SpamapS has joined #openstack-ironic | 16:19 | |
| *** e0ne has joined #openstack-ironic | 16:23 | |
| mgoddard | dtantsur: rloo: conductor patch looking good, have nodes deploy stepping. Will push shortly | 16:29 |
| rloo | mgoddard: yay! | 16:29 |
| mgoddard | just printing log messages, but it's a start | 16:32 |
| mgoddard | hoping to get Dell BIOS & RAID tested | 16:33 |
| mgoddard | would like to demo that in https://www.openstack.org/summit/denver-2019/summit-schedule/events/23365/ironic-deploy-templates-bespoke-bare-metal | 16:34 |
| * dtantsur is looking forward! | 16:35 | |
| *** sthussey has joined #openstack-ironic | 16:35 | |
| mgoddard | Never work with kids or pets filming live. Nor bare metal? | 16:37 |
| dtantsur | heh | 16:37 |
| mgoddard | anyone have opinions on adding some deploy steps to the fake driver interfaces? | 16:39 |
| mgoddard | they could e.g. record in <somewhere> that they executed | 16:40 |
| *** hamzy has quit IRC | 16:40 | |
| rloo | dtantsur, TheJulia: wondering why this wasn't approved. are we waiting for something (besides someone approving it)? https://review.openstack.org/#/c/637241/ | 16:40 |
| patchbot | patch 637241 - ironic - Deprecate Cisco drivers - 1 patch set | 16:40 |
| rloo | mgoddard: if you think it is useful, go for it. it is a fake driver so it should be fine. | 16:41 |
| *** moshele has joined #openstack-ironic | 16:42 | |
| *** v12aml has quit IRC | 16:42 | |
| *** v12aml has joined #openstack-ironic | 16:43 | |
| dtantsur | rloo: I don't know. I think we have to do it.. | 16:43 |
| dtantsur | mgoddard: ++ for fake deploy steps | 16:44 |
| *** moshele has quit IRC | 16:45 | |
| rloo | dtantsur: ok, I approved it. | 16:46 |
| *** bfournie has quit IRC | 16:46 | |
| openstackgerrit | Merged openstack/ironic master: Fix pysendfile requirement marker https://review.openstack.org/639967 | 16:47 |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic-python-agent master: [WIP] Software RAID: Add IPA deploy support https://review.openstack.org/639390 | 16:53 |
| *** dtantsur is now known as dtantsur|afk | 16:55 | |
| dtantsur|afk | g'night | 16:55 |
| openstackgerrit | Merged openstack/sushy-tools master: Limit instances exposure https://review.openstack.org/616516 | 16:55 |
| *** tssurya has quit IRC | 16:57 | |
| *** betherly has quit IRC | 16:59 | |
| *** e0ne has quit IRC | 17:05 | |
| openstackgerrit | OpenStack Release Bot proposed openstack/metalsmith stable/stein: Update .gitreview for stable/stein https://review.openstack.org/640104 | 17:06 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic-lib stable/stein: Update .gitreview for stable/stein https://review.openstack.org/640105 | 17:06 |
| openstackgerrit | OpenStack Release Bot proposed openstack/metalsmith stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/640106 | 17:06 |
| openstackgerrit | OpenStack Release Bot proposed openstack/ironic-lib stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/640107 | 17:06 |
| openstackgerrit | OpenStack Release Bot proposed openstack/metalsmith master: Update master for stable/stein https://review.openstack.org/640108 | 17:06 |
| TheJulia | dtantsur|afk: thanks | 17:06 |
| *** gkadam has quit IRC | 17:08 | |
| TheJulia | rloo: dtantsur|afk: fyi, some product manager is trying to setup a call with me for next week. Maybe we'll have more details later. | 17:08 |
| rloo | TheJulia: OH. should I unapprove the PR then? | 17:09 |
| TheJulia | No, lets let it mere | 17:09 |
| TheJulia | merge | 17:09 |
| rloo | TheJulia: ok. i was going to update the docs too but maybe not. yet. | 17:09 |
| *** moshele has joined #openstack-ironic | 17:10 | |
| TheJulia | rloo: yeah, I'm kind of holding off on that as well considering I've had emails going back and forth with cisco for a while now, but there is really just no forward movement so I put in at least the reno for now. I can change the docs prior to release of ironic | 17:11 |
| rloo | TheJulia: ++. thx! | 17:11 |
| rpioso | mgoddard: Re: your Summit Denver presentation, please let me know if you would like any assistance with Dell EMC h/w and/or the idrac h/w type. | 17:17 |
| *** andrein has quit IRC | 17:17 | |
| *** bfournie has joined #openstack-ironic | 17:18 | |
| mgoddard | rpioso: thanks. I just had a look actually, there's no bios interface. Is that because you're planning to use the redfish one? | 17:19 |
| * rpioso is heading to his car. | 17:22 | |
| rpioso | mgoddard: Will you be available in a couple of hours to chat more? | 17:22 |
| *** rpioso is now known as rpioso|afk | 17:23 | |
| *** munimeha1 has joined #openstack-ironic | 17:33 | |
| openstackgerrit | Mark Goddard proposed openstack/ironic master: Deploy templates: conductor https://review.openstack.org/634732 | 17:35 |
| *** _fragatina has joined #openstack-ironic | 17:36 | |
| mgoddard | rpioso|afk: probably not - on UK time. I'll be around tomorrow | 17:38 |
| *** hwoarang has quit IRC | 17:38 | |
| *** hwoarang has joined #openstack-ironic | 17:39 | |
| mgoddard | rloo: dtantsur|afk: conductor patch is ready when you are (with one todo for tomorrow): https://review.openstack.org/634732 | 17:41 |
| patchbot | patch 634732 - ironic - Deploy templates: conductor - 7 patch sets | 17:41 |
| mgoddard | night all | 17:41 |
| mgoddard | rloo: will add fixes for your and dtantsur|afk's API patch issues and push tomorrow | 17:43 |
| *** gyee has joined #openstack-ironic | 17:44 | |
| *** derekh has quit IRC | 17:46 | |
| *** moshele has quit IRC | 18:01 | |
| *** trown is now known as trown|lunch | 18:03 | |
| *** moshele has joined #openstack-ironic | 18:05 | |
| *** amoralej is now known as amoralej|off | 18:12 | |
| openstackgerrit | OpenStack Release Bot proposed openstack/sushy stable/stein: Update .gitreview for stable/stein https://review.openstack.org/640144 | 18:12 |
| openstackgerrit | OpenStack Release Bot proposed openstack/sushy stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/640146 | 18:12 |
| openstackgerrit | OpenStack Release Bot proposed openstack/sushy master: Update master for stable/stein https://review.openstack.org/640147 | 18:12 |
| *** hamzy has joined #openstack-ironic | 18:13 | |
| rloo | thx mgoddard! Will look today. | 18:16 |
| NobodyCam | Good Morning Ironic'ers | 18:30 |
| NobodyCam | happy almost Friday | 18:30 |
| *** rh-jelabarre has quit IRC | 18:40 | |
| *** rh-jelabarre has joined #openstack-ironic | 18:41 | |
| openstackgerrit | Merged openstack/ironic-inspector master: Follow up to introspection data driver api revision https://review.openstack.org/639269 | 18:44 |
| *** sdake has joined #openstack-ironic | 18:45 | |
| openstackgerrit | Merged openstack/ironic master: Deprecate Cisco drivers https://review.openstack.org/637241 | 18:49 |
| *** e0ne has joined #openstack-ironic | 18:51 | |
| *** e0ne has quit IRC | 19:03 | |
| *** ijw has joined #openstack-ironic | 19:06 | |
| *** iurygregory has quit IRC | 19:08 | |
| *** pcaruana has quit IRC | 19:14 | |
| *** jtomasek has quit IRC | 19:16 | |
| *** dougsz has quit IRC | 19:19 | |
| *** whoami-rajat has quit IRC | 19:20 | |
| *** trown|lunch is now known as trown | 19:22 | |
| *** _fragatina has quit IRC | 19:29 | |
| *** _fragatina has joined #openstack-ironic | 19:31 | |
| *** ijw has quit IRC | 19:35 | |
| *** _fragatina has quit IRC | 19:38 | |
| *** moshele has quit IRC | 19:45 | |
| openstackgerrit | Merged openstack/ironic-inspector master: Deprecates [processing]store_data_location https://review.openstack.org/639248 | 19:45 |
| *** ijw has joined #openstack-ironic | 19:51 | |
| openstackgerrit | Aija Jaunteva proposed openstack/sushy master: Add support for loading packaged standard registries https://review.openstack.org/589837 | 19:55 |
| *** ijw has quit IRC | 19:56 | |
| mbeierl | ok, I've got myself into an odd position: I am trying to clean a node, but that node's MAC address is somehow in dnsmasq blacklist, so the node never PXE boots. Any ideas? | 19:57 |
| *** sdake has quit IRC | 20:07 | |
| *** jtomasek has joined #openstack-ironic | 20:10 | |
| *** e0ne has joined #openstack-ironic | 20:16 | |
| openstackgerrit | Aija Jaunteva proposed openstack/sushy master: Add public resource loading and message parsing https://review.openstack.org/590747 | 20:29 |
| *** hamzy has quit IRC | 20:39 | |
| *** hamzy has joined #openstack-ironic | 20:43 | |
| *** sdake has joined #openstack-ironic | 21:02 | |
| *** e0ne has quit IRC | 21:19 | |
| *** sdake has quit IRC | 21:23 | |
| *** sdake has joined #openstack-ironic | 21:25 | |
| *** ijw has joined #openstack-ironic | 21:32 | |
| *** jtomasek has quit IRC | 21:42 | |
| *** e0ne has joined #openstack-ironic | 21:42 | |
| *** e0ne has quit IRC | 21:44 | |
| *** e0ne has joined #openstack-ironic | 21:45 | |
| *** hamzy has quit IRC | 21:47 | |
| *** e0ne has quit IRC | 22:01 | |
| *** _fragatina has joined #openstack-ironic | 22:07 | |
| *** MattMan_1 has quit IRC | 22:10 | |
| *** MattMan_1 has joined #openstack-ironic | 22:10 | |
| *** rh-jelabarre has quit IRC | 22:12 | |
| *** sdake has quit IRC | 22:13 | |
| *** baha has quit IRC | 22:15 | |
| *** ijw has quit IRC | 22:32 | |
| *** ijw has joined #openstack-ironic | 22:33 | |
| *** ijw has quit IRC | 22:35 | |
| *** ijw has joined #openstack-ironic | 22:35 | |
| *** bfournie has quit IRC | 23:03 | |
| *** ijw has quit IRC | 23:06 | |
| *** ijw has joined #openstack-ironic | 23:06 | |
| TheJulia | \o/ fast track works \o/ need to fix the scenario and there is concern over cleaning cleanup which requires braincells I think | 23:12 |
| TheJulia | mbeierl: remove from the blacklist? Oo | 23:12 |
| JayF | TheJulia: you have a link to that in a review or something? sounds like an interesting problem | 23:13 |
| TheJulia | sure, one moment | 23:13 |
| JayF | feel free to tldr it here for me too :P | 23:13 |
| JayF | (or instead) | 23:13 |
| TheJulia | https://review.openstack.org/#/c/635996/ | 23:14 |
| patchbot | patch 635996 - ironic - fast tracked deployment support - 19 patch sets | 23:14 |
| TheJulia | I don't think my brain can do a tl;dr, been in an all day meeting | 23:14 |
| JayF | Need to find+market to you the Yes Woman Julia Bot(tm). Sits in all day meetings and nods at appropriate times, leaving her human counterpart to go get real work done :P | 23:15 |
| *** ijw has quit IRC | 23:16 | |
| TheJulia | ++++ OMG yes | 23:16 |
| *** ijw has joined #openstack-ironic | 23:16 | |
| TheJulia | where is athomas when I need him | 23:16 |
| TheJulia | oh right... | 23:16 |
| JayF | I've been gone from ironic so long I don't even know who that is :( | 23:17 |
| TheJulia | athomas is my manager at rh | 23:17 |
| JayF | on the other hand, restaurant POS software is better than ever before! | 23:17 |
| JayF | aha | 23:17 |
| TheJulia | who happens to be sitting about 2 feet from me | 23:17 |
| JayF | the red hat will make it all the easier | 23:18 |
| JayF | just start wearing it low like Carmen Sandiego, and you can skip giving the robot a face! it's foolproof | 23:18 |
| * TheJulia wishes she took it on this trip, the weather in austin is awful right now | 23:18 | |
| *** ijw has quit IRC | 23:29 | |
| *** ijw has joined #openstack-ironic | 23:29 | |
| *** ijw has quit IRC | 23:31 | |
| *** ijw has joined #openstack-ironic | 23:32 | |
| *** hwoarang has quit IRC | 23:57 | |
| *** hwoarang has joined #openstack-ironic | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!