*** jaybeale has quit IRC | 00:07 | |
*** thiagop has quit IRC | 00:09 | |
*** PollyZ has joined #openstack-ironic | 00:18 | |
*** hoangcx has quit IRC | 00:18 | |
*** bapalm has quit IRC | 00:22 | |
*** PollyZ has quit IRC | 00:22 | |
*** bapalm has joined #openstack-ironic | 00:25 | |
*** yibo has joined #openstack-ironic | 00:26 | |
*** mbound has joined #openstack-ironic | 00:41 | |
*** fragatina has quit IRC | 00:44 | |
*** mbound has quit IRC | 00:46 | |
*** ijw has quit IRC | 00:50 | |
*** bvandewa has joined #openstack-ironic | 00:52 | |
*** bvandewa has quit IRC | 00:54 | |
*** PollyZ has joined #openstack-ironic | 00:54 | |
*** bvandewa has joined #openstack-ironic | 00:58 | |
*** PollyZ has quit IRC | 01:00 | |
*** ijw has joined #openstack-ironic | 01:09 | |
*** garthb has quit IRC | 01:17 | |
*** mtanin___ has joined #openstack-ironic | 01:21 | |
*** mtanino has quit IRC | 01:22 | |
*** rbudden has joined #openstack-ironic | 01:41 | |
openstackgerrit | Merged openstack/ironic: Centralize config options - [glance] https://review.openstack.org/304829 | 01:42 |
---|---|---|
*** piet has quit IRC | 01:52 | |
*** _milan_ has joined #openstack-ironic | 01:55 | |
*** PixieBoots has quit IRC | 01:55 | |
*** PixieBoots has joined #openstack-ironic | 01:57 | |
openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: [WIP] Nova-compatible serial console: socat console_utils https://review.openstack.org/328168 | 01:57 |
openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: Mock global variables https://review.openstack.org/338665 | 01:57 |
openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: IPMITool: add IPMISocatConsole class https://review.openstack.org/293873 | 01:57 |
*** milan has quit IRC | 01:57 | |
*** PollyZ has joined #openstack-ironic | 01:59 | |
*** PollyZ has quit IRC | 02:03 | |
*** joprovost has joined #openstack-ironic | 02:06 | |
*** rama_y has quit IRC | 02:12 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Add a doc about appending kernel parameters to boot instances https://review.openstack.org/320848 | 02:12 |
*** Sukhdev_ has quit IRC | 02:15 | |
openstackgerrit | yuyafei proposed openstack/ironic-lib: Correct reraising of exception https://review.openstack.org/338797 | 02:16 |
openstackgerrit | Tan Lin proposed openstack/ironic: Improve _check_depolying_status to handle nodes in maintenance https://review.openstack.org/324269 | 02:18 |
*** yuikotakadamori has joined #openstack-ironic | 02:26 | |
*** amotoki has joined #openstack-ironic | 02:36 | |
*** jaybeale has joined #openstack-ironic | 02:40 | |
*** piet has joined #openstack-ironic | 02:43 | |
*** mtanino has joined #openstack-ironic | 02:45 | |
*** mtanin___ has quit IRC | 02:47 | |
*** jaybeale has quit IRC | 02:50 | |
*** mtanino has quit IRC | 02:50 | |
*** krtaylor has joined #openstack-ironic | 02:51 | |
*** piet has quit IRC | 02:54 | |
*** bvandewa has quit IRC | 02:54 | |
*** ijw has quit IRC | 02:54 | |
*** piet has joined #openstack-ironic | 02:54 | |
*** Nisha1 has joined #openstack-ironic | 02:55 | |
*** joprovost has quit IRC | 02:58 | |
*** joprovost has joined #openstack-ironic | 02:58 | |
*** amotoki has quit IRC | 02:59 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic: Add support for building ISO for deploy ramdisk https://review.openstack.org/264590 | 03:07 |
*** joprovost has quit IRC | 03:16 | |
*** piet has quit IRC | 03:16 | |
*** amotoki has joined #openstack-ironic | 03:21 | |
*** rbudden has quit IRC | 03:21 | |
*** jaybeale has joined #openstack-ironic | 03:22 | |
*** amotoki has quit IRC | 03:26 | |
*** amotoki has joined #openstack-ironic | 03:27 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic: Config variable to configure [glance] section https://review.openstack.org/266803 | 03:35 |
*** jaybeale has quit IRC | 03:36 | |
*** ijw has joined #openstack-ironic | 03:45 | |
*** PollyZ has joined #openstack-ironic | 03:45 | |
*** amotoki has quit IRC | 03:49 | |
*** ijw has quit IRC | 03:50 | |
*** mtanino has joined #openstack-ironic | 03:51 | |
*** rloo has quit IRC | 03:59 | |
*** amotoki has joined #openstack-ironic | 03:59 | |
*** links has joined #openstack-ironic | 04:02 | |
*** Sukhdev has joined #openstack-ironic | 04:14 | |
*** garthb has joined #openstack-ironic | 04:22 | |
*** davidlenwell has quit IRC | 04:28 | |
*** bvandewa has joined #openstack-ironic | 04:32 | |
*** davidlenwell has joined #openstack-ironic | 04:38 | |
*** d0ugal has quit IRC | 05:01 | |
openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add nodes tagging support - objs https://review.openstack.org/233357 | 05:02 |
*** bvandewa has quit IRC | 05:08 | |
*** moshele has joined #openstack-ironic | 05:11 | |
*** rama_y has joined #openstack-ironic | 05:19 | |
openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: IPMITool: add IPMISocatConsole class https://review.openstack.org/293873 | 05:20 |
moshele | dtantsur|afk: can you remove your -2 from https://review.openstack.org/#/c/264263/ and https://review.openstack.org/#/c/264257/. the spec in now approved | 05:22 |
*** garthb has quit IRC | 05:23 | |
*** sdake_ has joined #openstack-ironic | 05:25 | |
openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 05:27 |
*** PollyZ has quit IRC | 05:46 | |
*** jvgrant_ has joined #openstack-ironic | 05:46 | |
*** jvgrant has quit IRC | 05:49 | |
*** Sukhdev has quit IRC | 05:53 | |
*** ChubYann has quit IRC | 05:55 | |
*** rama_y has quit IRC | 05:56 | |
*** rcernin has joined #openstack-ironic | 05:58 | |
*** edand has joined #openstack-ironic | 06:03 | |
*** Nisha1 has quit IRC | 06:03 | |
*** amotoki has quit IRC | 06:03 | |
*** d0ugal has joined #openstack-ironic | 06:04 | |
openstackgerrit | Vasyl Saienko proposed openstack/ironic: Switch to Neutron Flat network. https://review.openstack.org/340695 | 06:07 |
*** bvandewa has joined #openstack-ironic | 06:08 | |
*** mjura has joined #openstack-ironic | 06:11 | |
*** bvandewa_ has joined #openstack-ironic | 06:12 | |
*** bvandewa has quit IRC | 06:12 | |
*** lennyb has quit IRC | 06:12 | |
*** rajinir has quit IRC | 06:15 | |
*** amotoki has joined #openstack-ironic | 06:17 | |
*** mjura has quit IRC | 06:18 | |
*** lennyb has joined #openstack-ironic | 06:19 | |
*** mtanino has quit IRC | 06:24 | |
*** Fdaisuke has quit IRC | 06:29 | |
*** mjura has joined #openstack-ironic | 06:31 | |
*** hoangcx has joined #openstack-ironic | 06:34 | |
*** d0ugal has quit IRC | 06:43 | |
*** Fdaisuke has joined #openstack-ironic | 06:47 | |
*** sdake_ has quit IRC | 06:49 | |
*** amotoki has quit IRC | 06:52 | |
*** mkovacik__ has joined #openstack-ironic | 06:56 | |
*** _milan_ has quit IRC | 06:57 | |
*** _milan_ has joined #openstack-ironic | 06:59 | |
*** tesseract- has joined #openstack-ironic | 07:00 | |
*** pcaruana has joined #openstack-ironic | 07:00 | |
*** mkovacik__ has quit IRC | 07:02 | |
*** zer0c00l_ has quit IRC | 07:04 | |
*** ljjjustin has joined #openstack-ironic | 07:08 | |
openstackgerrit | Dao Cong Tien proposed openstack/ironic: Nova-compatible serial console: socat console_utils https://review.openstack.org/328168 | 07:12 |
*** zer0c00l has joined #openstack-ironic | 07:13 | |
*** amotoki has joined #openstack-ironic | 07:14 | |
*** fragatina has joined #openstack-ironic | 07:17 | |
*** fragatina has quit IRC | 07:18 | |
*** fragatina has joined #openstack-ironic | 07:19 | |
*** amotoki has quit IRC | 07:19 | |
*** UForgotten has joined #openstack-ironic | 07:22 | |
*** ifarkas has joined #openstack-ironic | 07:23 | |
*** UForgotten has quit IRC | 07:26 | |
*** UForgotten has joined #openstack-ironic | 07:26 | |
openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add DB layer support for node tags filter https://review.openstack.org/272624 | 07:27 |
*** daemontool has joined #openstack-ironic | 07:53 | |
*** Romanenko_K has joined #openstack-ironic | 07:54 | |
*** jrist has joined #openstack-ironic | 07:55 | |
_milan_ | #pixiesay morning Ironic! -m flexing | 07:57 |
PixieBoots | ᕙʕ⇀ᴥ⇀ʔᕗ: morning Ironic! | 07:57 |
yuikotakadamori | good morning, _milan_ and PixieBoots! | 07:58 |
_milan_ | yuikotakadamori, good morning :) | 07:58 |
*** zzzeek has quit IRC | 08:00 | |
*** zzzeek has joined #openstack-ironic | 08:00 | |
*** athomas has joined #openstack-ironic | 08:03 | |
vsaienk0 | Morning _milan_, yuikotakadamori, dtantsur and all Ironic'ers | 08:03 |
_milan_ | vsaienk0, good morning :) | 08:03 |
yuikotakadamori | vsaienk0, morning :) | 08:04 |
*** amotoki has joined #openstack-ironic | 08:09 | |
*** amotoki has quit IRC | 08:13 | |
*** jtomasek_ has joined #openstack-ironic | 08:13 | |
*** lucas|afk is now known as lucasagomes | 08:15 | |
lucasagomes | morning all :-) | 08:15 |
*** yuikotakadamori has quit IRC | 08:15 | |
*** jpich has joined #openstack-ironic | 08:15 | |
Romanenko_K | Good morning! | 08:17 |
*** MattMan has quit IRC | 08:19 | |
*** MattMan has joined #openstack-ironic | 08:19 | |
*** derekh has joined #openstack-ironic | 08:20 | |
*** jistr has joined #openstack-ironic | 08:22 | |
Romanenko_K | <lucasagomes>, could you repeat your +2 on this patch? I was fixing Ruby's comments after your previous review there: https://review.openstack.org/#/c/322788/ | 08:32 |
*** yogi_ has joined #openstack-ironic | 08:33 | |
alineb | morning all! | 08:35 |
ifarkas | morning all! | 08:36 |
*** moshele has quit IRC | 08:38 | |
*** edand has quit IRC | 08:39 | |
*** Fdaisuke has quit IRC | 08:39 | |
*** moshele has joined #openstack-ironic | 08:40 | |
*** mgould|afk is now known as mgould | 08:49 | |
mgould | morning Ironic | 08:49 |
mgould | and in particular _milan_ dtantsur|afk alineb lucasagomes Romanenko_K ifarkas PixieBoots | 08:50 |
*** e0ne has joined #openstack-ironic | 08:51 | |
_milan_ | morning mgould lucasagomes ifarkas Romanenko_K :) | 08:51 |
*** electrofelix has joined #openstack-ironic | 08:56 | |
*** amotoki has joined #openstack-ironic | 09:03 | |
*** ohamada has joined #openstack-ironic | 09:04 | |
*** dtantsur|afk is now known as dtantsur | 09:08 | |
dtantsur | morning _milan_, mgould, alineb, ifarkas, lucasagomes, vsaienk0 and everyone else :) | 09:08 |
_milan_ | dtantsur, good morning :) | 09:09 |
lucasagomes | Romanenko_K, looking | 09:09 |
*** amotoki has quit IRC | 09:09 | |
lucasagomes | mgoddard, dtantsur _milan_ morning :-) | 09:09 |
_milan_ | lucasagomes, good morning :) | 09:09 |
mgoddard | morning lucasgomes! | 09:09 |
dtantsur | moshele, done | 09:10 |
_milan_ | mgoddard, good morning :) | 09:11 |
mgoddard | morning _milan_ | 09:12 |
mgould | mordred: mgoddard | 09:13 |
mgould | bah | 09:13 |
mgould | morning mgoddard | 09:13 |
mgould | my client is not smart enough to autocomplete "morning" :-) | 09:13 |
mgoddard | mordred mgould :-p | 09:13 |
_milan_ | :) | 09:15 |
mordred | zomg. I'm becomming a greeting | 09:17 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-python-agent: Small refactor in the root device loop matching logic https://review.openstack.org/338763 | 09:17 |
*** moshele has quit IRC | 09:19 | |
*** mbound has joined #openstack-ironic | 09:21 | |
*** moshele has joined #openstack-ironic | 09:22 | |
Romanenko_K | lucasagomes, thanks! | 09:22 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-specs: Collect system logs specs: Change the file name template https://review.openstack.org/336108 | 09:23 |
lucasagomes | Romanenko_K, np :-) | 09:24 |
moshele | dtantsur: thanks | 09:24 |
*** ccamacho has joined #openstack-ironic | 09:28 | |
mgould | mordred: sorry! | 09:31 |
mgould | also, good morning :-) | 09:31 |
mordred | morning mgould ! and no worries ;) | 09:36 |
openstackgerrit | Merged openstack/python-ironicclient: Negative tests for testing chassis-create command https://review.openstack.org/322788 | 09:44 |
*** mbound has quit IRC | 09:56 | |
*** amotoki has joined #openstack-ironic | 09:57 | |
*** athomas has quit IRC | 10:01 | |
*** d0ugal has joined #openstack-ironic | 10:02 | |
*** amotoki has quit IRC | 10:02 | |
*** ohamada has quit IRC | 10:03 | |
*** ohamada has joined #openstack-ironic | 10:03 | |
vdrok | morning ironic! | 10:09 |
*** athomas has joined #openstack-ironic | 10:10 | |
*** hoangcx has quit IRC | 10:12 | |
mgould | morning vdrok! | 10:16 |
vdrok | good morning mgould :) | 10:18 |
*** ljjjustin has left #openstack-ironic | 10:21 | |
*** ljjjustin has joined #openstack-ironic | 10:21 | |
* mgould uses "Perl" and "principled" in the same comment on https://review.openstack.org/#/c/339596; I hope they don't take my Pythonista badge away | 10:28 | |
*** ohamada has quit IRC | 10:32 | |
*** ohamada has joined #openstack-ironic | 10:32 | |
lucasagomes | mgould, and haskell :-) | 10:33 |
mgould | muahahaha | 10:42 |
*** sborkows has joined #openstack-ironic | 10:58 | |
*** sdake has joined #openstack-ironic | 11:07 | |
*** sdake_ has joined #openstack-ironic | 11:08 | |
*** sdake has quit IRC | 11:13 | |
*** bvandewa_ has quit IRC | 11:20 | |
*** amotoki has joined #openstack-ironic | 11:25 | |
*** yibo has quit IRC | 11:43 | |
*** sambetts|afk is now known as sambetts | 11:44 | |
sambetts | Morning all | 11:44 |
vsaienk0 | morning sambetts | 11:48 |
dtantsur | morning sambetts | 11:48 |
vsaienk0 | sambetts, have you seen https://github.com/openstack/neutron/commit/b672c26cb42ad3d9a17ed049b506b5622601e891 ? | 11:48 |
sambetts | vsaienk0: I hadn't but thats awesome, and we should hook into it some how :D | 11:50 |
sambetts | I've just read your email | 11:50 |
vsaienk0 | sambetts, we can always do like nova :) | 11:51 |
sambetts | yeah, I think that might be pretty cool, I still think its a shame we have to do it by API, instead of a message bus publish subscribe msg queue that we can subscribe too for notifications... | 11:54 |
*** Nisha1 has joined #openstack-ironic | 12:00 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Collect deployment logs from IPA https://review.openstack.org/336102 | 12:04 |
*** lucasagomes is now known as lucas-hungry | 12:04 | |
*** Nisha1 has quit IRC | 12:04 | |
openstackgerrit | Yatin Kumbhare proposed openstack/ironic: Add Python 3.5 classifier and venv https://review.openstack.org/340902 | 12:05 |
sambetts | dtantsur: Morning, did you see my response to deva's write up? | 12:10 |
dtantsur | not yet, just moving to it | 12:11 |
dtantsur | and this it's more or less in agreement with what he wrote, no? | 12:12 |
dtantsur | s/and this/I think/ | 12:12 |
* dtantsur has no clue what he's typing | 12:12 | |
TheJulia | dtantsur: coffee? | 12:12 |
sambetts | there are 2 major corrections to his write up in my email | 12:12 |
dtantsur | yeah | 12:12 |
dtantsur | TheJulia, oh yeah, I definitely need more (even though it's 2pm already) | 12:13 |
TheJulia | heh | 12:13 |
sambetts | just wanted to make sure you saw them because your email touched on the same parts | 12:13 |
*** amotoki_ has joined #openstack-ironic | 12:13 | |
*** amotoki has quit IRC | 12:13 | |
dtantsur | yeah, I think we're in agreement here too | 12:13 |
*** ccamacho is now known as ccamacho|lunch | 12:15 | |
*** sdake_ has quit IRC | 12:16 | |
sborkows | Hi, I have a problem with testing my local change to ironic python agent. I followed the instructions in https://github.com/openstack/ironic-python-agent/tree/master/imagebuild/coreos and built cpio.gz and vmlinuz files. Now I do not know how to substitute it with actual image to my node. In general I want to check that some code from agent is called by ironic inspector during inspection process. How can I check/debug such thing? | 12:16 |
*** rbudden has joined #openstack-ironic | 12:17 | |
sambetts | sborkows: you need to add it to glance like specified in 3. here: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#image-requirements | 12:18 |
sambetts | sborkows: then update the node with the new UUIDs as specified in 4. here: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#flavor-creation | 12:18 |
*** mbound has joined #openstack-ironic | 12:21 | |
*** piet has joined #openstack-ironic | 12:22 | |
sborkows | sambetts: Thanks, I will try it out.' | 12:23 |
*** caiobo has joined #openstack-ironic | 12:26 | |
*** mbound has quit IRC | 12:26 | |
*** trown|outtypewww is now known as trown | 12:26 | |
*** dprince has joined #openstack-ironic | 12:27 | |
caiobo | good morning Ironic :) | 12:29 |
sambetts | morning caiobo | 12:29 |
caiobo | morning sambetts! | 12:30 |
*** jrist has quit IRC | 12:31 | |
mgould | morning caiobo sambetts | 12:32 |
caiobo | morning mgould | 12:32 |
* mgould wonders if he can make coffee without scalding himself this time | 12:33 | |
mgould | gotta be worth a try | 12:33 |
openstackgerrit | Vladyslav Drok proposed openstack/ironic: Expose node's network_interface field in API https://review.openstack.org/317392 | 12:36 |
*** baoli has joined #openstack-ironic | 12:39 | |
*** piet has quit IRC | 12:40 | |
*** piet has joined #openstack-ironic | 12:40 | |
*** amotoki_ has quit IRC | 12:41 | |
*** Goneri has joined #openstack-ironic | 12:46 | |
*** ccamacho|lunch is now known as ccamacho | 12:46 | |
*** thiagop has joined #openstack-ironic | 12:47 | |
*** baoli_ has joined #openstack-ironic | 12:47 | |
thiagop | Good morning folks | 12:48 |
sambetts | Morning thiagop | 12:49 |
*** baoli has quit IRC | 12:49 | |
*** amotoki has joined #openstack-ironic | 12:50 | |
*** jrist has joined #openstack-ironic | 12:51 | |
*** moshele has quit IRC | 12:53 | |
*** edand has joined #openstack-ironic | 12:53 | |
mjturek1 | morning thiagop, you were looking for me yesterday | 12:54 |
mjturek1 | ? | 12:54 |
sborkows | sambetts: I have another question: am I right that the python agent image is still used during the hardware inspection? I mean, if I start inspection on a given node with modified ramdisk, will the inspector call the agent on that ramdisk? | 12:55 |
*** rloo has joined #openstack-ironic | 12:56 | |
*** lucas-hungry is now known as lucasagomes | 12:56 | |
sambetts | sborkows: no, the ironic inspector even though it uses IPA uses a separate ramdisk to the one configured in Ironic, to change the one configured for inspector you need to copy your ramdisk and kernel into your TFTP location, under the names ironic-agent.kernel and ironic-agent.initramfs | 12:57 |
* mgould has now booked his transport to the summit! | 12:58 | |
*** caiobo is now known as caiobo-afk | 12:58 | |
sambetts | mgould: \o/ | 12:58 |
sborkows | sambetts: Do you mean something like 'tftpboot' or 'httpboot' directory? | 12:58 |
mgould | looking forward to putting some faces to names :-) | 12:58 |
sambetts | sborkows: whichever location is configured in the ironic-inspecotor dnsmasq.conf as the tftp-root | 12:59 |
*** makowals has joined #openstack-ironic | 13:00 | |
*** mbound has joined #openstack-ironic | 13:02 | |
*** amotoki has quit IRC | 13:03 | |
*** Goneri has quit IRC | 13:07 | |
*** loki_ has joined #openstack-ironic | 13:07 | |
thiagop | mjturek1: Hi! Yes, I was. krtaylor said that you or him would like to participate on tha panel about 3rd party CI | 13:08 |
thiagop | mjturek1: we are drafting the resume right now and I would like to check if you are still interested in joining | 13:09 |
sborkows | sambetts: I found that there are ramdisk and kernel inside httpboot, not tftpboot. The node's driver is agent-ssh. Also, the whole dnsmasq.conf file is commented. Does it change anything? | 13:10 |
sambetts | sborkows: how are you deploying inspector devstack or is this a production system? | 13:12 |
sambetts | lucasagomes: do you know if we use ironic-specs for IPA??? | 13:14 |
lucasagomes | sambetts, hi there, I don't think so if the change is only in IPA | 13:15 |
sborkows | sambetts: It is a devstack deployment | 13:15 |
*** Goneri has joined #openstack-ironic | 13:16 | |
sambetts | sborkows: in a devstack deployment normally the tftp root location is in /opt/stack/data/ironic | 13:16 |
sambetts | lucasagomes: hmm, I have a TODO to write a spec for doing IPA versioning stuff, and I don't know where to put it, I guess that touches Ironic too | 13:17 |
sborkows | sambetts: yes, these directories I have under that location | 13:17 |
lucasagomes | sambetts, oh ironic-specs I think, because it will require changes to ironic no? | 13:17 |
sambetts | lucasagomes: yeah I think so | 13:17 |
lucasagomes | sambetts, wasn't the idea to query IPA about the version it has and have the ironic code to adapt to it ? | 13:18 |
jroll | morning y'all, I'm actually here today | 13:18 |
lucasagomes | or something like that, so ironic-specs seems fine since the change affects both projects | 13:18 |
lucasagomes | jroll, morning | 13:18 |
sambetts | lucasagomes: yeah I think that makes sense | 13:18 |
lucasagomes | jroll, feeling better? | 13:18 |
jroll | sambetts: I'd use ironic specs for IPA API versioning | 13:18 |
jroll | lucasagomes: yeah, all good today, thanks :) | 13:19 |
sambetts | jroll: cool :) | 13:19 |
lucasagomes | good good :-) | 13:19 |
mgould | morning jroll, glad to hear you're feeling better | 13:19 |
jroll | :) | 13:19 |
dtantsur | morning jroll | 13:19 |
sborkows | sambetts: So inspector will load ramdisk image located in /opt/stack/data/ironic/httpboot/. So I cannot test my changes in python agent with inspection? | 13:22 |
sambetts | sborkows: yup, it should do | 13:23 |
*** amotoki has joined #openstack-ironic | 13:23 | |
sborkows | sambetts: So is it any possibility to test on my environment changes in python agent? | 13:26 |
sambetts | you just need copy your generated ramdisk into httpboot and replace the existing ones | 13:26 |
sambetts | and it'll serve that ramdisk for inspection | 13:27 |
sborkows | sambetts: even if it is an agent ramdisk? | 13:27 |
*** jcoufal has joined #openstack-ironic | 13:28 | |
sambetts | inspector uses an agent ramdisk, just not the one uploaded to glance | 13:28 |
sborkows | sambetts: Oh, I understand now :) thanks | 13:28 |
*** amotoki has quit IRC | 13:28 | |
*** jpich_ has joined #openstack-ironic | 13:29 | |
*** jrist_ has joined #openstack-ironic | 13:29 | |
*** jtomasek__ has joined #openstack-ironic | 13:29 | |
*** d0ugal_ has joined #openstack-ironic | 13:30 | |
*** jrist has quit IRC | 13:32 | |
*** jpich has quit IRC | 13:32 | |
*** d0ugal has quit IRC | 13:33 | |
*** jtomasek_ has quit IRC | 13:33 | |
*** ChrisAusten has joined #openstack-ironic | 13:33 | |
*** baoli_ has quit IRC | 13:34 | |
openstackgerrit | Vasyl Saienko proposed openstack/python-ironicclient: Fix Quick-start example syntax. https://review.openstack.org/340944 | 13:34 |
*** jrist_ has quit IRC | 13:34 | |
*** d0ugal_ is now known as d0ugal | 13:35 | |
*** mtanino has joined #openstack-ironic | 13:36 | |
*** d0ugal is now known as Guest65715 | 13:36 | |
*** Guest65715 has quit IRC | 13:36 | |
*** d0ugal_ has joined #openstack-ironic | 13:36 | |
*** baoli has joined #openstack-ironic | 13:37 | |
*** xhku has quit IRC | 13:38 | |
*** xhku has joined #openstack-ironic | 13:39 | |
*** baoli_ has joined #openstack-ironic | 13:39 | |
*** sdake_ has joined #openstack-ironic | 13:41 | |
*** baoli has quit IRC | 13:42 | |
*** mtanino has quit IRC | 13:43 | |
*** mgoddard_ has joined #openstack-ironic | 13:44 | |
*** d0ugal_ has quit IRC | 13:44 | |
vsaienk0 | Morning jroll, lucasagomes, thiagop and all! | 13:44 |
lucasagomes | vsaienk0, hey there morning | 13:44 |
*** d0ugal has joined #openstack-ironic | 13:44 | |
thiagop | morning vsaienk0 | 13:44 |
*** amotoki has joined #openstack-ironic | 13:45 | |
*** jrist_ has joined #openstack-ironic | 13:45 | |
vsaienk0 | folks could you please review: https://review.openstack.org/#/c/334382, the patch adds flat network support to our devstack plugin | 13:46 |
sborkows | sambetts: I tried to do that and received kernel panic -> I think replacing initramfs image with cpio.gz was not correct. When building image I just ran 'make' inside ironic-python-agent/imagebuild/coreos location | 13:47 |
*** mgoddard has quit IRC | 13:47 | |
*** clenimar has quit IRC | 13:48 | |
*** e0ne has quit IRC | 13:49 | |
*** Goneri has quit IRC | 13:49 | |
sambetts | sborkows: I think the reason that you are seeing a kernel panic is most likely because of not having enough ram in the BM VMs to boot coreos | 13:50 |
*** PollyZ has joined #openstack-ironic | 13:50 | |
sborkows | sambetts: Actually it says that it cannot open device "(null)" or unknown-block(0,0) | 13:51 |
thiagop | sambetts: sborkows don't know about your problem exactly, but I think I found a bug on the coreos ramdisk on master. I'm confirming it now, but appears that it can't detect disks on my hardware | 13:51 |
sambetts | :/ | 13:52 |
thiagop | stable-mitaka is fine though | 13:52 |
thiagop | You can try to download and try it | 13:52 |
*** e0ne has joined #openstack-ironic | 13:53 | |
thiagop | sborkows: on devstack, you can just use the property IPA_DOWNLOAD_BRANCH=stable/mitaka on local.conf | 13:53 |
mgould | morning vsaienk0 thiagop | 13:54 |
thiagop | morning mgould | 13:54 |
sborkows | thiagop: I actually need to check my modifications to python agent and wanted to di this with inspection | 13:55 |
thiagop | sborkows: own, so you're building it | 13:55 |
thiagop | :/ | 13:55 |
jroll | thiagop: "stable/mitaka is fine", I'd bet money on the coreos version. hrm. any special kernel mods needed for your hardware or is it just plane sata | 13:56 |
vsaienk0 | o/ mgould | 13:56 |
sborkows | thiagop: yes, I'm building it locally | 13:56 |
sambetts | sborkows: have you tried building a tinyipa version instead of coreos/ | 13:57 |
sambetts | ? | 13:57 |
sborkows | sambetts: Now I'm trying with disk-image-builder | 13:57 |
*** Goneri has joined #openstack-ironic | 13:57 | |
loki_ | review please https://review.openstack.org/#/c/272658/ | 13:58 |
sborkows | If this won't help, I will switch to tinyipa | 13:58 |
sambetts | if its an up to date master IPA then it requires more RAM to run coreos than specified in the Ironic devstack script | 13:59 |
sambetts | the default RAM amout is 1024 and it now requires 1280 at least | 14:00 |
* sambetts is going to create a bug and change that in our devstack script I think | 14:01 | |
*** piet has quit IRC | 14:04 | |
*** Nakato has quit IRC | 14:15 | |
*** edand has quit IRC | 14:15 | |
*** links has quit IRC | 14:15 | |
*** Nakato has joined #openstack-ironic | 14:16 | |
*** wajdi has joined #openstack-ironic | 14:24 | |
openstackgerrit | Merged openstack/ironic: Add a doc about appending kernel parameters to boot instances https://review.openstack.org/320848 | 14:27 |
vgadiraj | Morning everyone, wanted to introduce myself on this channel. My name is Varun Gadiraju, I am a new OSIC member working on Ironic and I'm very excited to get started and work with all of you :) | 14:27 |
jroll | vgadiraj: welcome to the party :D | 14:28 |
rloo | welcome vgadiraj! | 14:28 |
*** piet has joined #openstack-ironic | 14:28 | |
TheJulia | it's a piet! :) Good morning | 14:29 |
piet | Hi! | 14:29 |
TheJulia | :) | 14:29 |
thiagop | vgadiraj: Be welcome to the Ironic community, the most prone to memes and sci-fi quotes on openstack | 14:29 |
vgadiraj | thanks all :) I have looked through the summary of the midcycle that jroll provided me and am going through the notes from midcycle that is on etherpad to try to get up to speed | 14:29 |
vgadiraj | hahah i look forward to it | 14:30 |
TheJulia | thiagop: I love that description! | 14:30 |
sambetts | hehe | 14:30 |
* rloo wonders if what thiagop said is true... | 14:30 | |
thiagop | #pixiesay -m happy welcome vgadiraj | 14:30 |
PixieBoots | ʕ^ᴥ^ʔ: welcome vgadiraj | 14:30 |
*** mtanino has joined #openstack-ironic | 14:32 | |
*** rama_y has joined #openstack-ironic | 14:32 | |
openstackgerrit | Carol Bouchard proposed openstack/ironic: Increase devstack BM VM RAM for coreos to boot https://review.openstack.org/340972 | 14:35 |
thiagop | Who is Carol Bouchard here on the channel? | 14:36 |
*** sabeen has joined #openstack-ironic | 14:37 | |
jroll | cisco, sambetts might know :) | 14:37 |
TheJulia | Her launchpad profile doesn't list an irc handle :( | 14:38 |
thiagop | sambetts: Maybe you (and/or her) should look at this too: https://github.com/openstack-infra/devstack-gate/blob/3ef48cec70c771242d743467768f22d1ecb55ecb/devstack-vm-gate.sh#L359 | 14:38 |
mgould | morning rloo TheJulia vgadiraj | 14:39 |
sambetts | thiagop: woah?! We already have an override for that I thoguht! | 14:39 |
*** mgoddard has joined #openstack-ironic | 14:40 | |
sambetts | She's one of my co-workers, I handed this bug off to her as she's not had much time to work on upstream this cycle but still wanted to get involved :) | 14:40 |
*** mgoddard_ has quit IRC | 14:40 | |
rloo | hi mgould, sambetts, thiagop. Good morning TheJulia, jroll, and everyone else. | 14:40 |
thiagop | rloo: morning | 14:40 |
sambetts | thiagop: we inject it into devstack local config in our jobs which I guess overrides anything set by devstack-gate https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/devstack-gate.yaml | 14:41 |
sambetts | https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/devstack-gate.yaml#L703 | 14:42 |
sambetts | including line number ^ | 14:42 |
TheJulia | good morning mgould, rloo | 14:42 |
mariojv | morning Ironic | 14:42 |
thiagop | sambetts: oww, yeah. Right. | 14:42 |
mariojv | lucasagomes: I had two questions about L620 here if you have a second: https://review.openstack.org/#/c/338763/5/ironic_python_agent/hardware.py | 14:42 |
mariojv | 1. will hint_value implicitly be converted to a string while logging? | 14:43 |
mariojv | 2. does IPA use _("log message) for i18n? | 14:43 |
*** clenimar has joined #openstack-ironic | 14:43 | |
sambetts | thiagop: /me hates all the extra logic like that devstack gate... its a nightmare for working out whats actually going on | 14:43 |
*** clenimar_ has joined #openstack-ironic | 14:43 | |
*** clenimar has quit IRC | 14:43 | |
*** clenimar has joined #openstack-ironic | 14:44 | |
sambetts | thiagop: don't actually bother with it in my CI, I just have a local.conf file I define that I inject some extra values into | 14:44 |
thiagop | sambetts: you tell me... And some options are hardcoded, I had a problem to discover how to increase deploywait_timeout on tempest | 14:44 |
thiagop | sambetts: and how to influence active_timeout through other completely different variable | 14:45 |
mgould | morning mariojv | 14:45 |
sambetts | thiagop: its these sort of things that I'd love to cover in the panel | 14:46 |
thiagop | sambetts: we will :) | 14:46 |
sambetts | thiagop: like a "put your hand up if you use devstack-gate" | 14:46 |
thiagop | thiagop: I use because we have a full infra here (zuul, jjb, jenkins, nodepool) and I think it's kind of easier from future newcomers on our project to have a similar structure to manage here and in the community | 14:48 |
*** edand has joined #openstack-ironic | 14:48 | |
thiagop | (and it kind of scales, but shhhh...) | 14:48 |
*** rama_y has quit IRC | 14:48 | |
mariojv | lucasagomes: i commented on the line in question, took away the -1 | 14:49 |
lucasagomes | mariojv, thanks a lot will take a look | 14:50 |
mariojv | np thank you | 14:50 |
sambetts | thiagop: I have a slightly modfied infra ;-P zuul + jenkins + nodepool + manually defined jobs + custom test code to run devstack and tempest with my settings | 14:50 |
lucasagomes | mariojv, sorry for the delay was looking at some downstream stuff | 14:51 |
lucasagomes | mariojv, I will reply inline on that patch | 14:51 |
mariojv | np | 14:51 |
mariojv | thanks | 14:51 |
sambetts | but my custom test code is literally just a python test utils wrapper around setting up the local.conf + starting devstack + running tempest | 14:51 |
thiagop | sambetts: jjb will start to get useful in the next month, we have brand new mitaka cloud being put in place and I plan to move the CI there :) | 14:51 |
thiagop | sambetts: YEEEAHH, tempest tests running! | 14:52 |
thiagop | o/\o/\o/\o/\o\/ | 14:52 |
thiagop | TheJulia: your guess was right. Apparently, the problem is on the last coreos build | 14:53 |
sambetts | thiagop: whats the issue? missing drivers or something/ | 14:53 |
sambetts | ? | 14:53 |
thiagop | sambetts: Don't know exactly. lsblk wasn't discovering the disks correctly | 14:54 |
thiagop | sambetts: I have a screenshot, let me discover how to paste it somewhere | 14:54 |
TheJulia | thiagop: awesome /o\ | 14:54 |
sambetts | hmm, what did you change to make it work? roll back the version or something? | 14:55 |
jroll | heads up, v2 api meeting in 5 minutes | 14:55 |
TheJulia | yes, screen shot would be awesome | 14:55 |
jroll | https://wiki.openstack.org/wiki/Meetings/Ironic_v2_API | 14:55 |
jroll | #openstack-meeting-3 | 14:55 |
thiagop | sambetts: I just set the IPA_DOWNLOAD_BRANCH to stable/mitaka | 14:55 |
lucasagomes | mariojv, replied, thanks for the review | 14:55 |
*** loki_ has quit IRC | 14:56 | |
sambetts | thiagop: hmm, I wonder if it because of this: https://review.openstack.org/#/c/327891/ | 14:56 |
sambetts | TheJulia: ^ | 14:56 |
*** edand has quit IRC | 14:57 | |
mariojv | ah, i did not catch the s after the %(...) | 14:57 |
TheJulia | sambetts: that is the link I sent thiagop yesterday :) | 14:57 |
thiagop | sambetts TheJulia I don't speak IPA enough to confirm anything, but... | 14:58 |
sambetts | TheJulia, thiagop: could you try cloning IPA, roll back to before that change merged and try manually building a coreos image ?? | 14:58 |
sambetts | thiagop: ^ | 14:58 |
*** jaybeale has joined #openstack-ironic | 14:59 | |
TheJulia | thiagop: I would be curious to see what the kernel and manual command output is between your stable/mitaka ramdisk and your master branch ramdisk | 15:00 |
jroll | api v2 meeting now in -meeting-3 | 15:00 |
TheJulia | and yes, sorry, that means booting the image up with an ssh key defined and collecting data :( | 15:00 |
dtantsur | jroll, now? Oo | 15:00 |
jroll | dtantsur: yes, that's what the email said, right? | 15:00 |
dtantsur | then I seriously misunderstood it | 15:00 |
dtantsur | jroll, 1800 UTC is now? | 15:01 |
jroll | dtantsur: yeah, I'm wrong | 15:01 |
jroll | somehow my local calendar was wrong | 15:01 |
sambetts | :-P | 15:01 |
jroll | sorry :( | 15:01 |
sambetts | I was excited for a meeting that was not in the middle of me eating dinner :-P | 15:02 |
TheJulia | heh | 15:02 |
dtantsur | that's pity, I already got glad I can attend it actually :) | 15:02 |
*** mjura has quit IRC | 15:02 | |
TheJulia | any meeting not in the middle of lunch is awesome to me | 15:02 |
*** jaybeale has quit IRC | 15:03 | |
jroll | yeah, finding a slot was hard :( | 15:04 |
TheJulia | yeah :( | 15:04 |
thiagop | TheJulia sambetts the output during cleaning was this -> https://owncloud.lsd.ufcg.edu.br/index.php/s/G024Uuq5eBsU93I | 15:05 |
thiagop | sorry, during deployment | 15:05 |
*** jaybeale has joined #openstack-ironic | 15:05 | |
sambetts | I can only think that the new CoreOS version is missing some drivers from the kernel | 15:06 |
sambetts | so its no longer finding your disks | 15:07 |
thiagop | sambetts: I'll try to rollback the change to see what I got | 15:07 |
*** jaybeale has quit IRC | 15:09 | |
*** vishwanathj has joined #openstack-ironic | 15:19 | |
*** ametts has joined #openstack-ironic | 15:21 | |
JayF | sambetts: thiagop: Like, I don't think CoreOS dropping drivers is a valid reason to rollback that change. | 15:23 |
JayF | sambetts: thiagop: We can't ship out of date software; it may just mean a CoreOS ramdisk isn't well-suited for that use case. | 15:23 |
thiagop | JayF: I'll rollback locally just to check if the problem exists without the change | 15:24 |
thiagop | JayF: Do you have another suggestion of ramdisk to use? TinyIPA doesn't have network drivers for our blades either... | 15:24 |
* thiagop feels like in a deadlock | 15:25 | |
JayF | There are DIB ramdisks that supposedly work but are not gated | 15:25 |
JayF | and those allow for more customization | 15:25 |
JayF | I'm not an expert on those but there are folks in here who know things about it | 15:25 |
JayF | also, CoreOS uses stock kernels, so you can build custom modules and inject them into your image, just by building them out of vanilla kernels of the same version | 15:26 |
*** e0ne has quit IRC | 15:26 | |
*** harlowja_at_home has joined #openstack-ironic | 15:27 | |
thiagop | JayF: the setup here is for CI, I rather not use out of date software too... | 15:28 |
JayF | oh man :( | 15:28 |
JayF | I would suggest going down the DIB path then | 15:28 |
JayF | I think cinerama and maybe TheJulia know things about that ^ but it's not well documented yet | 15:28 |
TheJulia | cinerama was beginning to look at what would be required to move the ironic-agent element into the IPA repo so we could build a gate job around it that we control, and then deprecate the old element. I don't know the status of that because it raised a bug in diskimage-builder | 15:30 |
JayF | TheJulia: I think what would help thiagop would just be docs on how to build an ironic-agent image using DIB and/or customize it if needed (althoug h I suspect a DIB ramdisk may work out of the box) | 15:31 |
TheJulia | thiagop: So the ilo CI team had to do the same, so the devstack variables allow definition of other pre-built images | 15:33 |
TheJulia | thiagop: I feel like we've discussed this before though. | 15:34 |
sean-k-mooney | TheJulia: hi. quick question | 15:35 |
TheJulia | thiagop: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#buildingdibbaseddeployramdisk | 15:35 |
TheJulia | sean-k-mooney: sure | 15:35 |
sean-k-mooney | TheJulia: does ironic observer the normal milestone 3 code freeze and milestone 2 spec approval deadline as other project such as nova? | 15:35 |
*** rajinir has joined #openstack-ironic | 15:35 | |
JayF | sean-k-mooney: no. | 15:35 |
sean-k-mooney | oh ok | 15:36 |
*** sdake_ has quit IRC | 15:36 | |
TheJulia | thiagop: https://github.com/openstack/ironic/blob/master/devstack/lib/ironic#L145 | 15:36 |
JayF | sean-k-mooney: We do indepedent releases; we sorta view the milestones as a "good spot" to think of doing a release | 15:36 |
JayF | sean-k-mooney: we generally release an X.0.0 around final release freeze, which ends up being our stable branch | 15:36 |
TheJulia | JayF: Thank you :) | 15:36 |
*** sdake_ has joined #openstack-ironic | 15:36 | |
dtantsur | leaving earlier today, see you tomorrow | 15:36 |
JayF | TheJulia: I honestly didn't even see that was pointed at you. lol | 15:36 |
TheJulia | sean-k-mooney: are you at the midcycle in Durham? | 15:36 |
sean-k-mooney | JayF: oh ok i was looking at http://releases.openstack.org/newton/schedule.html and noticed ironic follows http://governance.openstack.org/reference/tags/release_cycle-with-intermediary.html | 15:37 |
sean-k-mooney | TheJulia: am no sorry. | 15:37 |
thiagop | Thanks TheJulia | 15:37 |
thiagop | night dtantsur | 15:37 |
TheJulia | sean-k-mooney: okay, I was going to suggest going and having a beer if you were :) | 15:37 |
*** dtantsur is now known as dtantsur|afk | 15:37 | |
*** makowals has quit IRC | 15:38 | |
thiagop | btw, the manual doesn't even cite TinyIPA... | 15:38 |
JayF | sean-k-mooney: the way you're talking about before, with the explicit milestones, is cycle-with-milestones | 15:38 |
TheJulia | thiagop: :( | 15:39 |
JayF | thiagop: if you'll file a bug about that, I'm sure someone will get it in place | 15:39 |
thiagop | low-hanging-fruit | 15:39 |
JayF | thiagop: yep, exactly | 15:39 |
JayF | low-hanging-fruit == contributors magicaly drop from the ceiling and fix your bug | 15:39 |
*** rloo has quit IRC | 15:39 | |
JayF | it's glorious :P | 15:39 |
*** rloo has joined #openstack-ironic | 15:40 | |
TheJulia | That is not always true though ;) | 15:40 |
sambetts | JayF: https://review.openstack.org/#/c/340972 | 15:40 |
JayF | sambetts: should we only change that in the case where it's a coreos ramdisk? | 15:41 |
thiagop | JayF: LOL | 15:41 |
JayF | sambetts: changing that default seems like a big hammer | 15:41 |
JayF | sambetts: especially since the default, tinyipa, works in 1g | 15:41 |
sean-k-mooney | JayF: yes i just did not see any newton specific timeline on the ironic wiki so i taught i would ask as im not particalary familar with the intermediary process vs milestone approch. | 15:41 |
* thiagop would love to build a tinyipa image with the drivers he needs, but.. | 15:42 | |
JayF | sean-k-mooney: a good way to look at that would be to check our release notes | 15:42 |
JayF | sean-k-mooney: we release point releases reasonably often, at least compared to many other projects | 15:42 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add and document the "rotational" root device hint https://review.openstack.org/338264 | 15:42 |
sambetts | JayF: problem is that its hard to change once you've stacked, and people like to replace their ramdisks with thier own so I think supporting the biggest ramdisk by default is ok | 15:44 |
sean-k-mooney | JayF: thanks for the advice. ya ill check them out. just one other question then interms of accpeting new feautre is it a best effort e.g. if its merged it will be in the next dot release or is there still a deadline for inclution in a major release? | 15:44 |
*** fragatina has quit IRC | 15:44 | |
JayF | sambetts: I don't really agree :/ I think 1.25G of ram is bigger enough to make a difference | 15:45 |
sambetts | JayF: then when someone stacks they are locked to a particular ramdisk type | 15:49 |
*** fragatina has joined #openstack-ironic | 15:49 | |
*** ohamada has quit IRC | 15:49 | |
*** z- has quit IRC | 15:50 | |
*** z- has joined #openstack-ironic | 15:51 | |
JayF | sambetts: unless they specify a larger amount of ram | 15:51 |
JayF | sambetts: unstack/restack does raise the ram amount | 15:51 |
JayF | sambetts: which I usually do when changing ramdisks anyway | 15:51 |
rloo | jroll: so when is the ironic-api meeting? did it just happen, or will it be in 2 hours? | 15:51 |
*** penick has joined #openstack-ironic | 15:53 | |
openstackgerrit | Igor Kalnitsky proposed openstack/ironic: Fix fake.FakeBoot.prepare_ramdisk() signature https://review.openstack.org/341056 | 15:54 |
*** sborkows has quit IRC | 15:54 | |
vdrok | rloo: 1800 utc :) | 15:55 |
rloo | vdrok: ok, as scheduled then. thx! | 15:56 |
TheJulia | sean-k-mooney: We make a final release at the end of the cycle, so for that in a sense there are some deadlines where we simply cannot merge major changes at the end of the cycle. Does that make sense? Does it answer your question? | 15:56 |
sambetts | JayF: then they need to know if they are going to use a different ramdisk before they stack? | 15:56 |
*** adu has joined #openstack-ironic | 15:57 | |
JayF | sambetts: I've personally always viewed a ramdisk change as needing an unstack/restack | 15:57 |
JayF | sambetts: I know jroll uses the same pattern; I'm a little surprised to hear it isn't normal | 15:58 |
JayF | but maybe we should have that discussion in the review? | 15:58 |
*** ifarkas has quit IRC | 15:58 | |
sambetts | I've often uploaded my own ramdisk build to glance and then node-updated in Ironic | 15:59 |
JayF | and you're not willing, as someone using that pattern, to set the VM size manually in your ironic.conf? | 15:59 |
thiagop | JayF: ^this way we don't have to wait 30min to re-stack | 15:59 |
JayF | I mean, if I'm wrong someone can land it over me, or convince me otherwise, I'm just not sure it's a great idea to bump that value | 16:00 |
JayF | when msot folks probably don't have more than 8g to give to a devstack | 16:00 |
JayF | you're dropping the number of instances they can run by 1 or 2 | 16:00 |
*** pcaruana has quit IRC | 16:00 | |
JayF | but I guess the same argument of "why can't you set it manually in that case" goes against that too | 16:01 |
*** rcernin has quit IRC | 16:03 | |
sambetts | as someone who has the knowledge that CoreOS has a higher ram requirement then I can make that decision but as someone who doesn't I'd be very confused to why when I uploaded a CoreOS image to my devstack install it won't boot | 16:03 |
*** ChrisAusten has quit IRC | 16:04 | |
JayF | I mean, sure, but there are other factors into if the ram is large enough | 16:04 |
JayF | like how large the image you're deploying is | 16:04 |
*** ChrisAusten has joined #openstack-ironic | 16:04 | |
JayF | (assuming it's not raw) | 16:04 |
devananda | morning, all | 16:06 |
sambetts | morning devananda | 16:06 |
mgould | morning devananda | 16:06 |
*** daemontool has quit IRC | 16:06 | |
*** dprince has quit IRC | 16:06 | |
thiagop | morning devananda | 16:06 |
*** harlowja_at_home has quit IRC | 16:06 | |
mariojv | hi deva | 16:07 |
JayF | sambetts: after reflection, going to change the vote, for this reason: we broke existing devstack configurations | 16:09 |
*** sdake_ has quit IRC | 16:09 | |
JayF | sambetts: when we pushed out the new coreos change | 16:09 |
*** tesseract- has quit IRC | 16:09 | |
*** sdake_ has joined #openstack-ironic | 16:10 | |
sambetts | JayF: +1 and we can think about adding better logic after, adding better logic would also remove a big chunk of conditional from project-config | 16:10 |
sambetts | JayF: I have no idea what a sane default is for a DIB based ramdisk | 16:11 |
sambetts | :/ | 16:11 |
JayF | sambetts: we have to get DIB into the gate | 16:11 |
JayF | sambetts: before we worry about it at all | 16:11 |
JayF | sambetts: AFAI am concerned, it doesn't exist until/unless we're testing on it regularly. Just too much mental load to worry about something that's not tested. | 16:11 |
openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add 'neutron' network interface https://review.openstack.org/317393 | 16:12 |
openstackgerrit | Vladyslav Drok proposed openstack/ironic: Expose node's network_interface field in API https://review.openstack.org/317392 | 16:12 |
openstackgerrit | Vladyslav Drok proposed openstack/ironic: Update the deploy drivers with network flipping logic https://review.openstack.org/213262 | 16:12 |
openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add network interface to base driver class https://review.openstack.org/285852 | 16:12 |
*** david-lyle has joined #openstack-ironic | 16:13 | |
*** joprovost has joined #openstack-ironic | 16:13 | |
sambetts | JayF: you might be interested in: https://review.openstack.org/#/c/340567/ | 16:13 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add and document the "rotational" root device hint https://review.openstack.org/338264 | 16:14 |
jroll | sambetts: JayF: the bummer is that someone not re-stacking can't just update a config file to bump VM ram amount, you actually have to edit or re-create the VMs in qemu | 16:16 |
sambetts | yeah that sucks :( | 16:16 |
openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Tests for testing node-create command https://review.openstack.org/262055 | 16:17 |
Romanenko_K | Reduced test code size in half | 16:17 |
thiagop | Romanenko_K: w00t | 16:18 |
*** daemontool has joined #openstack-ironic | 16:19 | |
*** jaybeale has joined #openstack-ironic | 16:19 | |
*** david-lyle has quit IRC | 16:21 | |
*** rama_y has joined #openstack-ironic | 16:22 | |
*** sdake_ has quit IRC | 16:24 | |
*** d0ugal has quit IRC | 16:24 | |
*** garthb has joined #openstack-ironic | 16:24 | |
*** jpich_ has quit IRC | 16:24 | |
TheJulia | JayF: sambetts: imho, just a base debian build + ironic-agent element in dib is fairly sane imho | 16:26 |
sambetts | TheJulia: thats what devstack does now, but whats the RAM requirement on that? | 16:26 |
*** jtomasek__ has quit IRC | 16:26 | |
TheJulia | sambetts: 1GB I think, maybe debian-minimal + ironic-agent | 16:27 |
TheJulia | sambetts: but I need to check, too much context switching at the moment | 16:27 |
sambetts | sure :) | 16:27 |
devananda | JayF: did I take that out of context, or are you suggesting we gate ironic on DIB ? | 16:28 |
*** catintheroof has joined #openstack-ironic | 16:28 | |
sambetts | devananda: we're suggesting adding a job to test a DIB based IPA ramdisk on Ironic | 16:28 |
JayF | devananda: other way around. We need to test IPA aganist DIB ramdisk if we're going to "support" it | 16:28 |
sambetts | like we have for the coreos / tinyipa ramdisks | 16:28 |
devananda | JayF: gotcha. and yes :) | 16:29 |
TheJulia | cinerama: what is the status with dib w/r/t/ the failure upon duplicate elements being defind? | 16:29 |
*** ijw has joined #openstack-ironic | 16:30 | |
*** makowals has joined #openstack-ironic | 16:30 | |
*** jrist_ has quit IRC | 16:30 | |
TheJulia | cinerama: well, duplicate elements across paths | 16:31 |
*** fragatina has quit IRC | 16:31 | |
cinerama | TheJulia: so we have a change that addresses that failure more generally but i'm not sure it will land soon | 16:31 |
cinerama | grabbing the link | 16:31 |
cinerama | https://review.openstack.org/#/c/334785 | 16:33 |
*** trown is now known as trown|lunch | 16:33 | |
cinerama | there's a bunch of related changes as well | 16:33 |
*** yogi_ has quit IRC | 16:34 | |
cinerama | my expectation was that we could have the discussion about moving the element into the ipa tree in parallel with trying to get that to land | 16:34 |
cinerama | since i anticipate that both will take a while | 16:34 |
*** e0ne has joined #openstack-ironic | 16:36 | |
*** edand has joined #openstack-ironic | 16:36 | |
*** Romanenko_K has quit IRC | 16:37 | |
cinerama | anyway the current failure is that svc-map dies on having duplicated services. ian's changes more specifically address override behaviour | 16:39 |
*** sdake_ has joined #openstack-ironic | 16:40 | |
*** lucasagomes is now known as lucas|afk | 16:42 | |
cinerama | for the record my change is up at https://review.openstack.org/#/c/335583/ & still pending me writing an email to the list (working on that now fwiw) | 16:43 |
devananda | so, there were a couple responses to my email about network_interface. I said we'd start talking about it at 1700UTC (15min from now) | 16:44 |
devananda | sambetts: thanks for the clarification in your response. i'm glad to learn that you were thinking the same thing I was :) | 16:45 |
devananda | and it was just a matter of typing too quickly in a meeting | 16:45 |
sambetts | yup :) | 16:45 |
sambetts | hehe | 16:46 |
devananda | dtantsur raised the same point in his reply, so we're all in agreement on that | 16:46 |
TheJulia | are we going to do it as a call on conference.openstack.org? | 16:47 |
*** athomas has quit IRC | 16:48 | |
devananda | TheJulia: yep | 16:51 |
devananda | I am just preparing my notes | 16:52 |
openstackgerrit | Sam Betts proposed openstack/ironic-specs: Add spec for IPA API versioning support https://review.openstack.org/341086 | 16:52 |
NobodyCam | good Morning Ironicers | 16:53 |
* devananda is having laptop audio problems | 16:54 | |
thiagop | morning NobodyCam | 16:56 |
*** derekh has quit IRC | 16:56 | |
*** dprince has joined #openstack-ironic | 16:57 | |
rloo | devananda, sambetts, TheJulia: I just posted a response. I can't attend the meeting. My contribution is this patch: https://review.openstack.org/#/c/341084/ | 16:57 |
devananda | rloo: thanks. I'll take a look | 16:58 |
NobodyCam | morning thiagop rloo devananda TheJulia jlvillal sambetts jroll JayF dtantsur|afk vdrok and all others not directly listed here | 16:58 |
rloo | hi NobodyCam | 16:58 |
devananda | folks who want to talk about network_interfaces, let's use +1 512-808-5750 7777# | 16:58 |
NobodyCam | :) | 16:58 |
devananda | I've also set up my notes here: https://etherpad.openstack.org/p/ironic_network_interface_discussion | 16:59 |
vdrok | ugh, no conferencing? | 16:59 |
sambetts | devananda: jumping on now | 16:59 |
JayF | jroll: wanna land https://review.openstack.org/#/c/327437/ ? | 16:59 |
JayF | NobodyCam: ^ | 16:59 |
thiagop | rajinir: thanks for editing the doc. Looks good to me | 16:59 |
devananda | vdrok: there is a SIP interface as well, details are here: https://wiki.openstack.org/wiki/Infrastructure/Conferencing | 16:59 |
vdrok | devananda: thanks! | 16:59 |
jroll | JayF: I should probably read it first | 17:00 |
rajinir | thiagop: :) | 17:00 |
NobodyCam | oh . I too need to read it first | 17:04 |
JayF | if one or both of you read it I achieved my goal :P | 17:04 |
*** ChubYann has joined #openstack-ironic | 17:10 | |
*** edand has quit IRC | 17:10 | |
*** jrist_ has joined #openstack-ironic | 17:10 | |
* mgould -> home; good night! | 17:12 | |
*** mgould is now known as mgould|afk | 17:12 | |
*** PollyZ has quit IRC | 17:13 | |
*** jtomasek__ has joined #openstack-ironic | 17:17 | |
*** sdake_ has quit IRC | 17:17 | |
*** bvandewa has joined #openstack-ironic | 17:18 | |
*** jrist_ has quit IRC | 17:21 | |
*** jtomasek__ has quit IRC | 17:22 | |
NobodyCam | night mgould|afk | 17:30 |
*** makowals has quit IRC | 17:34 | |
*** makowals has joined #openstack-ironic | 17:34 | |
*** ijw has quit IRC | 17:35 | |
*** makowals has quit IRC | 17:35 | |
*** PollyZ has joined #openstack-ironic | 17:35 | |
*** makowals has joined #openstack-ironic | 17:36 | |
*** ijw has joined #openstack-ironic | 17:38 | |
*** edand has joined #openstack-ironic | 17:39 | |
*** ijw has quit IRC | 17:39 | |
*** PollyZ has quit IRC | 17:41 | |
*** dprince has quit IRC | 17:41 | |
*** trown|lunch is now known as trown | 17:42 | |
*** dprince has joined #openstack-ironic | 17:43 | |
*** adu has quit IRC | 17:44 | |
jroll | 'v2 api meeting in 15 minutes, as a reminder | 17:44 |
*** bvandewa has quit IRC | 17:44 | |
*** jaybeale has quit IRC | 17:48 | |
*** PollyZ has joined #openstack-ironic | 17:49 | |
*** edand has quit IRC | 17:54 | |
*** PollyZ has quit IRC | 17:54 | |
*** ijw has joined #openstack-ironic | 17:56 | |
sambetts | https://review.openstack.org/#/c/341084/1/specs/approved/driver-composition-reform.rst | 17:56 |
sambetts | TheJulia: ^ | 17:56 |
*** electrofelix has quit IRC | 17:57 | |
* vdrok leaves | 17:57 | |
vdrok | good night! | 17:57 |
*** e0ne has quit IRC | 17:58 | |
rloo | bye vdrok | 17:58 |
sambetts | cya vdrok | 17:58 |
jroll | v2 api meeting in 1 minute | 17:58 |
thiagop | night vdrok | 17:59 |
devananda | jroll: what channel? | 17:59 |
TheJulia | sambetts: thank you | 17:59 |
jroll | devananda: -3 | 18:00 |
jroll | devananda: #openstack-meeting-3, that is | 18:01 |
*** PollyZ has joined #openstack-ironic | 18:07 | |
*** amotoki has quit IRC | 18:07 | |
*** sdake_ has joined #openstack-ironic | 18:11 | |
*** bvandewa has joined #openstack-ironic | 18:14 | |
rloo | devananda, TheJulia, others: sorry I couldn't attend the network meeting. Are we good to continue with the network patches or are changes needed? | 18:17 |
*** PollyZ has quit IRC | 18:17 | |
sambetts | rloo: I think we're good with contining with the network patches, I'm just leaving some comments on your patch with what we discussed in the meeting | 18:18 |
rloo | sambetts: thx. you or anyone else are welcome to update that patch; or i can do it later today. | 18:19 |
*** bvandewa has quit IRC | 18:19 | |
openstackgerrit | Merged openstack/ironic: Increase devstack BM VM RAM for coreos to boot https://review.openstack.org/340972 | 18:19 |
*** PollyZ has joined #openstack-ironic | 18:23 | |
*** PollyZ has quit IRC | 18:29 | |
*** adu has joined #openstack-ironic | 18:31 | |
*** sdake_ has quit IRC | 18:32 | |
*** Sukhdev has joined #openstack-ironic | 18:33 | |
*** sdake_ has joined #openstack-ironic | 18:33 | |
*** jaybeale has joined #openstack-ironic | 18:34 | |
*** bvandewa has joined #openstack-ironic | 18:40 | |
*** bvandewa has quit IRC | 18:41 | |
*** bvandewa has joined #openstack-ironic | 18:41 | |
*** Goneri has quit IRC | 18:46 | |
mjturek1 | thiagop: sorry I just saw your ping from earlier.. not sure what happened there. Anyways, yes I'm definitely interested in the panel (in regards to ironic 3rd party CI) | 18:48 |
*** piet has quit IRC | 18:48 | |
*** PollyZ has joined #openstack-ironic | 18:48 | |
thiagop | mjturek1: we are currently agreeing on content and finishing the resume to send | 18:48 |
thiagop | what is your e-mail? (you can send it on pvt if you like) | 18:49 |
mjturek1 | mjturek@linux.vnet.ibm.com | 18:49 |
mjturek1 | thiagop: ^ | 18:49 |
mjturek1 | thiagop: where's the discussion happening?? in an email thread? | 18:52 |
thiagop | e-mail/gdocs | 18:52 |
mjturek1 | cool cool | 18:52 |
thiagop | mjturek1: I just cc'ed you on both | 18:53 |
mjturek1 | thiagop: great, watching for it | 18:53 |
thiagop | mjturek1: please, take a look on the history of the e-mail to get up to speed on what we discussed | 18:55 |
mjturek1 | will do | 18:55 |
*** piet has joined #openstack-ironic | 18:57 | |
*** daemontool has quit IRC | 19:04 | |
sambetts | devananda: The more I dig into this the more I realise having default implementations for any interface specified in the hardware_type is pretty horrible | 19:05 |
sambetts | because you have to provide all the driver_info for your interfaces anyway, I don't see why we need defaults, node-create creates a node in enroll state, then you should be able to syncronously update all the nodes info, e.g. interfaces and driver_info etc then move it to availible via validation | 19:07 |
*** sdake_ has quit IRC | 19:08 | |
sambetts | we don't have any interfaces that just magically work out of the box without providing some driver_info, which is normally unique per node, so I have to update the node anyway, so why not just be explicit about which interfaces you want at the same time? | 19:09 |
rloo | sambetts: but why explicitly specify the interface per node? or am i missing something? | 19:11 |
sambetts | whats making me think this is that say I have a hardware_type X, and that specfies to use ipmitool power by default and also supports redfish, but in my datacenter I just want to use redfish power, I'd have to override it for every node anyway | 19:12 |
rloo | sambetts: not all interfaces require something to be set in driver_info? | 19:12 |
*** vishwanathj has quit IRC | 19:12 | |
*** vishwanathj has joined #openstack-ironic | 19:13 | |
rloo | sambetts: right, that's true. but if you have hardware_type Y and that specifies to use ipmitool power by default and also supports redfish, but in your datacenter you just want to use ipmitool, then no need to override. | 19:13 |
rloo | sambetts: maybe i don't understand, but i don't see the reason for *not* allowing for defaults. | 19:14 |
sambetts | rloo: but I have to set ipmitool credtials an IP address on the node anyway so why not just set the interface too at that point | 19:14 |
rloo | sambetts: if you want to, go ahead and set the interface too. we allow that. | 19:15 |
sambetts | rloo: problem is in my case with hardware_type X, the ipmitool interface would be loaded and enabled whether I want to enable users to use it or not | 19:15 |
rloo | sambetts: i don't see why we should make it mandatory to specify the interface per node. that seems too restrictive. | 19:15 |
*** PollyZ has quit IRC | 19:17 | |
sambetts | I think the thing thats niggling at me is that I'm not sure I like that an operator can not disable the default_X_interface for a hardware_type | 19:19 |
rloo | sambetts: I'm not sure what you mean by that. | 19:20 |
rloo | sambetts: oh. hmm... | 19:21 |
sambetts | e.g. because ipmitool power is the default for hardware_type X I can not disable ipmitool power in my Ironic install because its the default | 19:21 |
sambetts | for that hardware_type | 19:21 |
rloo | sambetts: unless we add some new 'disable' value for a node.*_interface? | 19:21 |
rloo | sambetts: does it make sense to disable an interface, vs specifying which interface implementation to use? | 19:22 |
sambetts | thats not exactly what I mean, I mean that interfaces specified as default_x_interface in a hardware_type are automagically added to enabled_X_interfaces | 19:22 |
rloo | sambetts: i mean, eg does it make sense for an operator to want to disable the power_interface for a particular node? | 19:22 |
*** ijw has quit IRC | 19:23 | |
rloo | sambetts: oh. sorry, i misunderstood. | 19:23 |
sambetts | so as a deployer if I want to support hardware_type X but only on redfish, I can't disable IPMItool to prevent users to creating nodes with that interface | 19:24 |
rloo | sambetts: you're a mean deployer | 19:25 |
sambetts | without specified defaults on the hardware_types I'd just remove it from enabled_X_interfaces and it would be disabled and a user couldn't select it | 19:25 |
sambetts | rloo: heh | 19:25 |
rloo | sambetts: so if you only enabled your hardware type, wouldn't that work? | 19:25 |
jroll | I'm curious what the business structure looks like for a business that has deployers that have strong opinions about power management like that, and users enrolling nodes that aren't in sync with those opinions | 19:26 |
rloo | sambetts: anyway. i don't know if what you have is a corner case or something that certain hw vendors might want. worth writing up i guess. i am just glad it doesn't block the network patches :) | 19:26 |
*** sdake_ has joined #openstack-ironic | 19:26 | |
*** jaybeale has quit IRC | 19:27 | |
* sambetts needs to go and unwind, and needs to come back to this with a fresh brain and more coffee in the morning | 19:28 | |
rloo | bye sambetts, go unwind! :) | 19:29 |
jroll | night sambetts \o | 19:29 |
sambetts | o/ cya tomorrow | 19:29 |
*** sambetts is now known as sambetts|afk | 19:29 | |
*** fragatina has joined #openstack-ironic | 19:33 | |
*** PollyZ has joined #openstack-ironic | 19:38 | |
*** PollyZ has quit IRC | 19:43 | |
*** fragatina has quit IRC | 19:44 | |
*** daemontool has joined #openstack-ironic | 19:50 | |
*** PollyZ has joined #openstack-ironic | 19:52 | |
*** ijw has joined #openstack-ironic | 19:54 | |
*** PollyZ has quit IRC | 19:57 | |
*** ijw has quit IRC | 19:59 | |
mariojv | is there a reason agent_last_heartbeat should not be a datetime in driver_internal_info in ironic? https://github.com/openstack/ironic/blob/8006ec8a519fdbb34e310e785d7647ee94f68c49/ironic/drivers/modules/agent_base_vendor.py#L464 | 20:01 |
*** sdake_ has quit IRC | 20:01 | |
mariojv | other date-like fields are datetime fields it seems, not epoch seconds https://github.com/openstack/ironic/blob/master/ironic/objects/node.py#L91 | 20:02 |
*** ijw has joined #openstack-ironic | 20:03 | |
*** jcoufal has quit IRC | 20:05 | |
*** PollyZ has joined #openstack-ironic | 20:05 | |
mariojv | perhaps it has something to do with the fact that driver_internal_info is just json, but it would be nice to have that be something more readable. | 20:06 |
* TheJulia starts reading what she missed when she took her cats to the vet | 20:06 | |
*** ijw has quit IRC | 20:07 | |
rloo | mariojv: dunno, but I think those rackspace folks did that :) | 20:08 |
jroll | we would never! | 20:08 |
JayF | Oh, we totally would. | 20:09 |
rloo | mariojv, jroll: I am sure there is a way to find out. But regardless... seems a bit late to change. or if we want to, we'll have to deal with bakcwards compatibility. | 20:09 |
jroll | fwiw, that isn't actually used today | 20:10 |
rloo | mariojv: is it only an inconsistency issue, or is there another reason you ask? | 20:10 |
rloo | jroll: in that case, it can be removed. | 20:10 |
JayF | it's not used /by ironic/ today | 20:10 |
jroll | right | 20:10 |
JayF | I can attest we use it downstream in dashboard and the like. | 20:10 |
rloo | JayF: damn those downstream folks | 20:10 |
*** PollyZ has quit IRC | 20:10 | |
JayF | I mean, not like in downstream patches, just in dashboards :) | 20:10 |
*** ElCoyote_ has quit IRC | 20:11 | |
rloo | err I mean, JayF, awesome that downstream folks are so creative | 20:11 |
TheJulia | and truthfully, having it there and understanding what it represents does make it easier when troubleshooting odd issues | 20:11 |
jroll | rloo: I mean, we should keep it there because it is useful for downstreams, and also will be useful for long-running ramdisks | 20:11 |
jroll | rloo: but I wouldn't be afraid of changing it without deprecation, personally (though it doesn't hurt) | 20:11 |
JayF | jroll: It would be an API break, right? Unless we displayed it as epoch to older version-requesting clients? | 20:12 |
rloo | jroll: :) It is true that it is in driver_internal_info so that doesn't require deprecation, etc. unless it is in ironic/older versions in tree | 20:12 |
jroll | JayF: we don't promise compatibility for data within json fields, only around how a json field works | 20:12 |
JayF | ooh, okay | 20:12 |
rloo | JayF: no, not API change for driver_internal_info. | 20:12 |
jroll | but, it wouldn't be hard to deprecate right | 20:13 |
* JayF files that distinction away | 20:13 | |
jroll | move to a new name, have both for a cycle | 20:13 |
*** abashmak_ has joined #openstack-ironic | 20:13 | |
mariojv | rloo: just an inconsistency issue, and like TheJulia said, making troubleshooting easier | 20:14 |
mariojv | i always go to epochconverter.com or whatever when debugging something wrong with a node with that field | 20:14 |
mariojv | or fire up a python interpreter | 20:14 |
rloo | mariojv: ok, i like consistency. feel free to fix as suggested by JayF | 20:15 |
*** ElCoyote_ has joined #openstack-ironic | 20:15 | |
rloo | mariojv: or open a bug about it :) | 20:15 |
* mariojv opens the bug first | 20:15 | |
TheJulia | mariojv: ++ on converting the time :) | 20:15 |
mariojv | i like the move to a new name have both for a cycle idea | 20:16 |
TheJulia | but yeah, converting seconds from epoch is annoying at the same time | 20:16 |
TheJulia | Seems ideal | 20:16 |
*** rama_y has quit IRC | 20:16 | |
*** rama_y has joined #openstack-ironic | 20:16 | |
*** PollyZ has joined #openstack-ironic | 20:19 | |
*** PollyZ has quit IRC | 20:24 | |
*** Sukhdev has quit IRC | 20:27 | |
*** lucas|afk has quit IRC | 20:28 | |
mariojv | https://bugs.launchpad.net/ironic/+bug/1602410 | 20:30 |
openstack | Launchpad bug 1602410 in Ironic "agent_last_heartbeat type not consistent with other date fields" [Undecided,New] - Assigned to Mario Villaplana (mario-villaplana-j) | 20:30 |
*** PollyZ has joined #openstack-ironic | 20:30 | |
mariojv | rama_y: thank you for the review on https://review.openstack.org/#/c/337885/, good docs make me happy :) | 20:31 |
rama_y | mariojv: you are welcome :) | 20:34 |
*** lucasagomes has joined #openstack-ironic | 20:35 | |
devananda | if any spec reviewers have a few minutes, I'd be delighted if we can get another +2 on the keystone policy support spec: https://review.openstack.org/#/c/327437/ | 20:38 |
*** wajdi_ has joined #openstack-ironic | 20:40 | |
*** wajdi_ has quit IRC | 20:41 | |
NobodyCam | I'm looking now devananda :) | 20:42 |
devananda | tyvm | 20:42 |
*** ccamacho has quit IRC | 20:42 | |
devananda | I'll resume coding that once the spec is approved :) | 20:42 |
JayF | devananda: I shilled that to chris and jim this morning too :P | 20:43 |
*** wajdi has quit IRC | 20:43 | |
devananda | JayF: hehe | 20:44 |
*** priteau has joined #openstack-ironic | 20:47 | |
*** joprovost has quit IRC | 20:52 | |
*** ijw has joined #openstack-ironic | 20:54 | |
*** ijw has quit IRC | 20:55 | |
*** ijw has joined #openstack-ironic | 20:55 | |
NobodyCam | JayF: devananda: +2 with a couple of nits. perfect for a follow up. | 20:56 |
*** piet has quit IRC | 20:56 | |
JayF | NobodyCam: let me take a quick look, I might can just turn it around real quick and then land it | 20:56 |
NobodyCam | jroll: did you want to have a read over before I +a? | 20:56 |
NobodyCam | JayF: comments are all minor | 20:57 |
NobodyCam | I am happy to land as is and toss up a follow up. | 20:57 |
JayF | if you give me 5m I'll knock out your comments + miles comments | 20:57 |
JayF | and then you and deva can land? | 20:57 |
NobodyCam | either way! | 20:58 |
*** jaybeale has joined #openstack-ironic | 20:58 | |
*** ccamacho has joined #openstack-ironic | 20:58 | |
*** adu has quit IRC | 20:58 | |
* jroll +W | 20:59 | |
JayF | no wait | 21:00 |
JayF | jroll: wait | 21:00 |
jroll | before reading what JayF said | 21:00 |
devananda | heh | 21:00 |
JayF | jroll: I have the patchset like 90% done | 21:00 |
jroll | JayF: then upload it | 21:00 |
jroll | it's a race! | 21:00 |
devananda | JayF: uploading before the merge bot ... or upload hte diff as a follow on :p | 21:00 |
* NobodyCam is happy either way | 21:00 | |
openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Implement Keystone Policy Support https://review.openstack.org/327437 | 21:01 |
devananda | heh | 21:01 |
jroll | gg | 21:01 |
JayF | man making me sweat | 21:01 |
JayF | I always work faster on a deadline :P | 21:01 |
* JayF in b4 fails tests | 21:01 | |
jroll | super minor changes | 21:01 |
jroll | re-approved | 21:01 |
jroll | feel free to hit me next time you see me if you think that's wrong :P | 21:02 |
NobodyCam | lol | 21:02 |
* devananda re+2's | 21:02 | |
devananda | thanks. now I need to get back into that code ;) | 21:03 |
NobodyCam | :) ++++ | 21:03 |
*** piet has joined #openstack-ironic | 21:07 | |
*** fragatina has joined #openstack-ironic | 21:08 | |
rloo | jroll, devananda: wrt the network isolation patches, did we (you) decide that there should be a -2 on some patch that exposes <something?> in the API? | 21:08 |
*** trown is now known as trown|outtypewww | 21:09 | |
jroll | long ago we decided all the API stuff should land at once | 21:09 |
jroll | I care less about that now | 21:09 |
JayF | I think the concern was to not land an api that did nothing at all | 21:09 |
rloo | jroll: OK, so I'm going to ignore and if anyone cares, they can -2. | 21:09 |
jroll | rloo: okay | 21:10 |
openstackgerrit | Merged openstack/ironic-specs: Implement Keystone Policy Support https://review.openstack.org/327437 | 21:10 |
rloo | jroll: I asked, so I am not purposely sneaking around it :) | 21:10 |
jroll | :) | 21:11 |
JayF | \o/ | 21:11 |
*** adu has joined #openstack-ironic | 21:12 | |
*** mbound has quit IRC | 21:14 | |
*** fragatina has quit IRC | 21:15 | |
*** adu has quit IRC | 21:16 | |
TheJulia | vdrok: https://review.openstack.org/#/c/213262/125 appears to have merged with another patch? Is that intentional | 21:17 |
*** adu has joined #openstack-ironic | 21:17 | |
*** adu has quit IRC | 21:20 | |
*** rcernin has joined #openstack-ironic | 21:21 | |
*** PollyZ has quit IRC | 21:27 | |
TheJulia | I guess it is kind of intended looking at the dependent patch | 21:29 |
*** priteau has quit IRC | 21:32 | |
*** PollyZ has joined #openstack-ironic | 21:32 | |
*** priteau has joined #openstack-ironic | 21:33 | |
*** dprince has quit IRC | 21:36 | |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: Canery test for ironic networking changes https://review.openstack.org/330138 | 21:38 |
*** ametts has quit IRC | 21:49 | |
*** clenimar_ has joined #openstack-ironic | 21:54 | |
*** ijw has quit IRC | 21:54 | |
*** Sukhdev has joined #openstack-ironic | 21:57 | |
JayF | jroll: others; question: | 22:06 |
JayF | for https://review.openstack.org/#/c/301923/ | 22:06 |
JayF | aren't I going to have to bump the minimum version of ironic_lib in requirements/ | 22:06 |
JayF | because that code isn't even going to import on the wrong version of ironic_lib. | 22:06 |
*** clenimar_ has quit IRC | 22:06 | |
*** _milan_ has quit IRC | 22:07 | |
*** priteau has quit IRC | 22:09 | |
*** priteau has joined #openstack-ironic | 22:10 | |
openstackgerrit | Jay Faulkner proposed openstack/ironic: Metric chassis, driver, node, and port API calls https://review.openstack.org/301923 | 22:10 |
JayF | ^ put a -1 on my own patch and a question about how to handle that requirements.txt | 22:12 |
*** priteau has quit IRC | 22:13 | |
*** _milan_ has joined #openstack-ironic | 22:14 | |
*** mbound has joined #openstack-ironic | 22:14 | |
*** clenimar_ has joined #openstack-ironic | 22:19 | |
*** mbound has quit IRC | 22:20 | |
JayF | Isn't there someone in the community who is already running IPA on PPC? | 22:22 |
JayF | For some reason I wanna think some of the IBM 3rd party CI is running on power? | 22:22 |
JayF | cc: natorious ^ | 22:22 |
JayF | krtaylor: ^ do you know? | 22:23 |
*** ijw has joined #openstack-ironic | 22:26 | |
natorious | JayF: so to get it as close as our current coreos image, I could go with say Fedora or Debian and use a docker container | 22:27 |
JayF | natorious: well, our dib and tinycorelinux images don't use that pattern, so we don't neccessarily have to follow it. Honestly, I see deprecation of the CoreOS image in the future for upstream, because of some of the difficulties we've had around ... restraining systemd away from the tenant disk | 22:28 |
*** Sukhdev has quit IRC | 22:29 | |
natorious | k, good to know | 22:29 |
natorious | so I can get all the deps in the main image then and we'd run the agent service from that | 22:29 |
natorious | do you know if there was an agent user in other ppls setups? | 22:30 |
JayF | the dib image isn't officially supported/gated, but basically that's ssetup with a mix-and-match kinda deal | 22:31 |
JayF | so you can setup access with ssh, or users, by mixing in those elements | 22:31 |
JayF | the TinyIPA image is basically a minimal setup that is pretty much designed for CI, AFAIK nobody runs it in production (now that I've said that, someone will come and prove me wrong :P) | 22:31 |
*** rcernin has quit IRC | 22:35 | |
natorious | ohhh, whats the pxe deploy driver? | 22:36 |
JayF | so both drivers use IPA now to deploy | 22:37 |
JayF | but "pxe" driver, the agent exports all the local filesystems as iscsi targets, which are then mounted on the conductor and manipulated from there | 22:37 |
JayF | it tends to be better for doing a large number of parallel installs on a fast network, since the conductor acts as a cache for the image | 22:38 |
JayF | not everyone gets a free giant fast swift cluster like we do :P | 22:38 |
natorious | big data might not be so good though right? | 22:38 |
JayF | I mean, it's only during deploy phase | 22:38 |
natorious | like high io | 22:38 |
JayF | it doesn't impact anything once booted, if you're using localboot | 22:39 |
natorious | oh, so the iscsi data is used as the image data | 22:39 |
JayF | (vs non-local boot, where the server will pxe everytime, but into the OS on disk.) | 22:39 |
JayF | yes, exactly | 22:39 |
natorious | nice | 22:39 |
JayF | like think of it as the conductor just mounts it up during deploy | 22:39 |
JayF | partitions, writes image out, sets up bootloader if localboot+partition image | 22:39 |
JayF | then unmounts, reboots | 22:39 |
natorious | I'm surprised there isn't an ipxe option | 22:40 |
JayF | there is, the driver is just badly named | 22:40 |
natorious | like chain load remote kernel and initrd to automate the deploy | 22:40 |
natorious | oh, ok | 22:40 |
JayF | the "pxe" driver would be better called the "iscsi" driver | 22:40 |
JayF | and it can be booted over pxe, ipxe, virtual media | 22:40 |
JayF | I don't think there's a deploy driver that utilizes ipxe as part of the deploy mechanism, just as a means to get a ramdisk on the node to start booting, or a means to give a kernel/initrd for non-localboot after deploy | 22:41 |
openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Fix docstring warnings https://review.openstack.org/339852 | 22:41 |
natorious | JayF: yeah, maybe thats not needed right? Like what gain is there from remote booting to chain loading to copying vs just downloading and dding | 22:42 |
JayF | I don't really think there is one, yeah | 22:43 |
*** Fdaisuke has joined #openstack-ironic | 22:45 | |
natorious | JayF: apparently someone has done it. Found this under 3rd party drivers - pxe_ipmitool (ppc64) | 22:46 |
natorious | Uses the pxe_ipmitool driver on IBM OpenPOWER Servers | 22:46 |
JayF | natorious: yeah; I think that's being used by IBM third party ci | 22:46 |
JayF | you could maybe find the contact email for that third party ci and ask them about how they build their image? | 22:46 |
natorious | there is also an agent_ipmitool(ppc64) | 22:47 |
JayF | yeah. honestly don't put much credence in the driver names | 22:48 |
JayF | they're kinda badly named, and driver composition is being changed this cycle to make it more sensible | 22:48 |
natorious | np. Got some good dig starting points. Thnx for the infos :) | 22:49 |
JayF | np | 22:49 |
JayF | natorious: for bonus points; when you figure it out, you could docs patch how to build a PPC DIB ramdisk | 22:49 |
natorious | indeed | 22:50 |
*** vishwanathj has quit IRC | 22:57 | |
*** vishwanathj has joined #openstack-ironic | 22:58 | |
*** vishwanathj has quit IRC | 23:02 | |
*** vishwanathj has joined #openstack-ironic | 23:03 | |
*** Sukhdev has joined #openstack-ironic | 23:03 | |
*** vishwanathj has quit IRC | 23:07 | |
*** PollyZ has quit IRC | 23:07 | |
*** vishwanathj has joined #openstack-ironic | 23:08 | |
*** ijw has quit IRC | 23:08 | |
*** vishwanathj has quit IRC | 23:08 | |
*** vishwanathj has joined #openstack-ironic | 23:09 | |
*** catintheroof has quit IRC | 23:09 | |
*** penick has quit IRC | 23:11 | |
*** Sukhdev has quit IRC | 23:14 | |
*** jaybeale has quit IRC | 23:17 | |
*** sabeen has quit IRC | 23:20 | |
*** fragatina has joined #openstack-ironic | 23:21 | |
*** ijw has joined #openstack-ironic | 23:25 | |
*** fragatina has quit IRC | 23:28 | |
*** vishwanathj has quit IRC | 23:30 | |
*** vishwanathj has joined #openstack-ironic | 23:31 | |
*** ccamacho has quit IRC | 23:34 | |
*** baoli_ has quit IRC | 23:35 | |
*** vishwanathj has quit IRC | 23:36 | |
*** vishwanathj has joined #openstack-ironic | 23:36 | |
*** ijw has quit IRC | 23:36 | |
*** Sukhdev has joined #openstack-ironic | 23:38 | |
openstackgerrit | Xavier proposed openstack/ironic: Add Dynamic Allocation feature for the OneView drivers https://review.openstack.org/286192 | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!