| *** matsuhashi has joined #openstack-ironic | 00:18 | |
| *** eguz has quit IRC | 00:28 | |
| *** eghobo has joined #openstack-ironic | 00:39 | |
| *** nosnos has joined #openstack-ironic | 01:30 | |
| *** radsy has quit IRC | 02:41 | |
| *** matsuhashi has quit IRC | 02:47 | |
| *** radsy has joined #openstack-ironic | 02:54 | |
| *** coolsvap has joined #openstack-ironic | 02:56 | |
| *** radsy has quit IRC | 02:56 | |
| *** radsy has joined #openstack-ironic | 02:57 | |
| *** nosnos has quit IRC | 03:02 | |
| *** matsuhashi has joined #openstack-ironic | 03:04 | |
| *** eghobo has joined #openstack-ironic | 03:15 | |
| *** nosnos has joined #openstack-ironic | 03:20 | |
| *** matsuhashi has quit IRC | 03:29 | |
| *** eghobo has quit IRC | 03:49 | |
| *** eghobo has joined #openstack-ironic | 03:49 | |
| *** nosnos has quit IRC | 03:51 | |
| *** radsy has quit IRC | 03:52 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 04:00 | |
| *** radsy has joined #openstack-ironic | 04:22 | |
| *** eghobo has quit IRC | 04:38 | |
| *** eghobo has joined #openstack-ironic | 04:39 | |
| *** eguz has joined #openstack-ironic | 04:50 | |
| *** eghobo has quit IRC | 04:50 | |
| *** saju_m has joined #openstack-ironic | 05:13 | |
| *** sabah has joined #openstack-ironic | 05:16 | |
| *** sabah has quit IRC | 05:20 | |
| *** radsy has quit IRC | 05:21 | |
| *** mrda has quit IRC | 05:29 | |
| *** adam_g has quit IRC | 05:29 | |
| *** wendar_ has joined #openstack-ironic | 05:29 | |
| *** lifeless has quit IRC | 05:30 | |
| *** wendar has quit IRC | 05:30 | |
| *** lifeless has joined #openstack-ironic | 05:31 | |
| *** mrda has joined #openstack-ironic | 05:32 | |
| *** adam_g has joined #openstack-ironic | 05:33 | |
| *** adam_g has quit IRC | 05:33 | |
| *** adam_g has joined #openstack-ironic | 05:33 | |
| *** Mikhail_D_ltp has quit IRC | 05:34 | |
| dtantsur | morning Ironic \o/ | 06:07 |
|---|---|---|
| openstackgerrit | Jenkins proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/83956 | 06:07 |
| *** pradipta_away is now known as pradipta | 06:28 | |
| *** rameshg87 has joined #openstack-ironic | 06:30 | |
| *** rwsu has joined #openstack-ironic | 06:35 | |
| *** saju_m has quit IRC | 06:41 | |
| *** saju_m has joined #openstack-ironic | 06:54 | |
| *** sabah has joined #openstack-ironic | 06:56 | |
| *** max_lobur has joined #openstack-ironic | 07:01 | |
| *** sabah has quit IRC | 07:03 | |
| *** datajerk1 has joined #openstack-ironic | 07:09 | |
| *** datajerk has quit IRC | 07:09 | |
| *** max_lobur has quit IRC | 07:10 | |
| *** lsmola_ has joined #openstack-ironic | 07:16 | |
| *** eguz has quit IRC | 07:22 | |
| *** killer_prince has joined #openstack-ironic | 07:37 | |
| *** derekh has joined #openstack-ironic | 07:40 | |
| *** mrda has quit IRC | 07:45 | |
| *** mrda has joined #openstack-ironic | 07:46 | |
| *** derekh has quit IRC | 07:46 | |
| *** mrda has left #openstack-ironic | 07:46 | |
| *** mrda has joined #openstack-ironic | 07:49 | |
| *** mrda is now known as mrda_away | 07:50 | |
| *** mrda_away is now known as mrda | 07:57 | |
| *** mrda is now known as mrda_away | 07:58 | |
| *** pbrooko has joined #openstack-ironic | 08:02 | |
| *** yuriyz has joined #openstack-ironic | 08:06 | |
| *** jistr has joined #openstack-ironic | 08:08 | |
| *** martyntaylor has joined #openstack-ironic | 08:08 | |
| *** max_lobur has joined #openstack-ironic | 08:08 | |
| *** max_lobur1 has joined #openstack-ironic | 08:09 | |
| *** max_lobur has quit IRC | 08:13 | |
| *** mkerrin has joined #openstack-ironic | 08:14 | |
| *** lucasagomes has joined #openstack-ironic | 08:16 | |
| *** dshulyak has joined #openstack-ironic | 08:16 | |
| *** Mikhail_D_wk has left #openstack-ironic | 08:17 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 08:18 | |
| *** ndipanov has joined #openstack-ironic | 08:21 | |
| *** eloib_ has joined #openstack-ironic | 08:27 | |
| *** romcheg has quit IRC | 08:33 | |
| *** Mikhail_D_ltp has quit IRC | 08:34 | |
| *** matsuhashi has joined #openstack-ironic | 08:35 | |
| *** nosnos has joined #openstack-ironic | 08:40 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 08:48 | |
| agordeev | morning Ironic! | 08:56 |
| agordeev | dtantsur: morning ! | 08:56 |
| *** athomas has joined #openstack-ironic | 08:59 | |
| Mikhail_D_ltp | Good morning folks! :) | 09:22 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Supports filtering port by address https://review.openstack.org/82773 | 09:25 |
| *** martyntaylor1 has joined #openstack-ironic | 09:26 | |
| *** saju_m has quit IRC | 09:27 | |
| agordeev | Mikhail_D_ltp: morning | 09:28 |
| openstackgerrit | Vladimir Kozhukalov proposed a change to openstack/ironic-python-agent: Added lvm partitioner https://review.openstack.org/85672 | 09:28 |
| *** martyntaylor has quit IRC | 09:30 | |
| *** martyntaylor1 has quit IRC | 09:32 | |
| *** romcheg has joined #openstack-ironic | 09:35 | |
| openstackgerrit | Rohan Kanade proposed a change to openstack/ironic: Fix seamicro power.validate() method definition https://review.openstack.org/85675 | 09:38 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add DiskPartitioner https://review.openstack.org/83396 | 09:55 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use DiskPartitioner https://review.openstack.org/83399 | 09:56 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Get rid of the swap partition https://review.openstack.org/83726 | 09:56 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use GB instead of MB for swap https://review.openstack.org/83788 | 09:56 |
| lucasagomes | morning agordeev Mikhail_D_ltp :) | 09:56 |
| Mikhail_D_ltp | Morning lucasagomes :) | 09:56 |
| lucasagomes | Shrews, http://logs.openstack.org/59/84859/1/check/gate-ironic-python26/2734491/console.html that's the prob you have seem in ur tests as well right> | 09:58 |
| lucasagomes | ?* | 09:58 |
| *** martyntaylor has joined #openstack-ironic | 10:01 | |
| *** romsin has quit IRC | 10:03 | |
| lucasagomes | romcheg, ping | 10:05 |
| lucasagomes | romcheg, re https://review.openstack.org/#/c/85135/ | 10:05 |
| lucasagomes | did you found the problem of running it in OSX? | 10:05 |
| lucasagomes | cause otherwise that patch lgtm | 10:05 |
| lucasagomes | it works on linux | 10:06 |
| romcheg | Morning lucasagomes | 10:06 |
| romcheg | *is looking* | 10:06 |
| lucasagomes | romcheg, morning there :) | 10:06 |
| romcheg | lucasagomes: I found the problem but it is not in the tox.ini, but in generate-config.sh | 10:07 |
| romcheg | should we merge those patches separately? | 10:08 |
| lucasagomes | romcheg, ah right, yes, so that patch doesn't touch the generate-config.sh at all | 10:08 |
| lucasagomes | so I think it's fair to merge it separately | 10:08 |
| romcheg | let's land this one then | 10:08 |
| lucasagomes | ack | 10:08 |
| lucasagomes | romcheg, thanks for verifying :) | 10:09 |
| * lucasagomes has no os x handy here | 10:09 | |
| romcheg | Oh, U laready put a +2 | 10:12 |
| romcheg | I'm going to approve it | 10:12 |
| lucasagomes | :) | 10:13 |
| *** matsuhashi has quit IRC | 10:16 | |
| *** matsuhashi has joined #openstack-ironic | 10:16 | |
| *** matsuhashi has quit IRC | 10:20 | |
| *** romcheg1 has joined #openstack-ironic | 10:23 | |
| *** romcheg has quit IRC | 10:24 | |
| *** coolsvap has quit IRC | 10:27 | |
| *** saju_m has joined #openstack-ironic | 10:32 | |
| *** saju_m has quit IRC | 10:49 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Fix Nova rescheduling tear down problem https://review.openstack.org/84497 | 10:52 |
| romcheg1 | lucasagomes: did I miss any discussion about parted vs sfdisk? | 10:54 |
| romcheg1 | I've seen lifeless' comment here https://review.openstack.org/#/c/84396/ | 10:55 |
| lucasagomes | romcheg1, right not that I know, I saw his comments as well | 10:55 |
| lucasagomes | gotta answer that and point him to the other patch as well | 10:56 |
| NobodyCam | good morning Ironic | 11:18 |
| NobodyCam | so we are going to be hitting the road in just a few minutes | 11:19 |
| lucasagomes | morning NobodyCam | 11:20 |
| agordeev | lucasagomes, NobodyCam morning! | 11:20 |
| lucasagomes | drive-safe | 11:20 |
| lucasagomes | NobodyCam, were you heading to? | 11:20 |
| agordeev | romcheg1: also morning! | 11:20 |
| NobodyCam | if all goes well we shouldbe docked by 9 / 9:30 locl time (8 - 8:30 pst time) | 11:21 |
| lucasagomes | :) | 11:21 |
| NobodyCam | morning lucasagomes agordeev | 11:21 |
| romcheg1 | Morning everyone :) | 11:21 |
| openstackgerrit | A change was merged to openstack/ironic: Use xargs -0 instead of --null https://review.openstack.org/85135 | 11:21 |
| romcheg1 | NobodyCam: You are here so early today | 11:21 |
| NobodyCam | we are current in lodrsburg New mexico, and we are off to Almogordo New Mexico | 11:21 |
| NobodyCam | morning romcheg1 :) | 11:22 |
| *** romcheg1 is now known as romcheg | 11:22 | |
| NobodyCam | we're heding to atlanta :) | 11:22 |
| NobodyCam | and need I want ro docked before the meeting | 11:22 |
| *** saju_m has joined #openstack-ironic | 11:24 | |
| lucasagomes | nice :D | 11:26 |
| *** matsuhashi has joined #openstack-ironic | 11:26 | |
| NobodyCam | ya we're having a blast so far... only been on the road for two days now. | 11:29 |
| NobodyCam | romcheg: how did your meeting go? | 11:30 |
| NobodyCam | well I hope :) | 11:30 |
| romcheg | NobodyCam: Which one? ^) | 11:30 |
| NobodyCam | for visa? | 11:31 |
| romcheg | Oh, you missed my message. The visa was approved | 11:31 |
| NobodyCam | w000t !!!! | 11:31 |
| romcheg | So I'm going to be present at the summit | 11:31 |
| NobodyCam | ya | 11:31 |
| NobodyCam | awesome :) | 11:32 |
| romcheg | After dealing with some of the corporate red tape | 11:32 |
| NobodyCam | :-p | 11:32 |
| *** pradipta is now known as pradipta_away | 11:33 | |
| Shrews | lucasagomes: yes | 11:34 |
| lucasagomes | Shrews, :) ack commented on the bug | 11:34 |
| lucasagomes | and clicked on affect me as well | 11:34 |
| romcheg | lucasagomes: Did anyone consider using PyParted? https://fedorahosted.org/pyparted/ | 11:37 |
| lucasagomes | romcheg, yes | 11:38 |
| lucasagomes | romcheg, 2 problems: the process need to run as root | 11:38 |
| lucasagomes | can't be installed via pip | 11:38 |
| lucasagomes | but I like the python binds, have used it in another project | 11:38 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Fix hosts mapping for conductor's periodic tasks https://review.openstack.org/84142 | 11:42 |
| *** rameshg87 has left #openstack-ironic | 11:44 | |
| romcheg | lucasagomes: Ah, I remembered that :) | 11:48 |
| lucasagomes | :D | 11:49 |
| *** matsuhashi has quit IRC | 11:49 | |
| *** matsuhashi has joined #openstack-ironic | 11:50 | |
| NobodyCam | on the road... back a few hours | 11:50 |
| *** matsuhas_ has joined #openstack-ironic | 11:53 | |
| *** matsuhashi has quit IRC | 11:54 | |
| *** nosnos has quit IRC | 11:59 | |
| *** lucasagomes is now known as lucas-hungry | 12:00 | |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Implement caching for master images https://review.openstack.org/85387 | 12:00 |
| *** sabah has joined #openstack-ironic | 12:14 | |
| *** sabah has quit IRC | 12:24 | |
| *** zigo has quit IRC | 12:26 | |
| *** jdob has joined #openstack-ironic | 12:37 | |
| *** rloo has joined #openstack-ironic | 12:49 | |
| *** martyntaylor has quit IRC | 12:59 | |
| *** lucas-hungry is now known as lucasagomes | 12:59 | |
| openstackgerrit | David Shrewsbury proposed a change to openstack/ironic: Encapsulate Ironic client retry logic https://review.openstack.org/83105 | 13:00 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Remove usage of Glance from PXE clean_up() https://review.openstack.org/85703 | 13:01 |
| Shrews | lucasagomes: when you get the chance, can you take a look at 83105 again? I just rebased it to get rid of some merge conflicts. | 13:01 |
| lucasagomes | ack :) thanks | 13:02 |
| *** Mikhail_D_ltp has quit IRC | 13:03 | |
| *** rustlebee is now known as russellb | 13:08 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 13:09 | |
| *** ndipanov has quit IRC | 13:15 | |
| *** derekh has joined #openstack-ironic | 13:15 | |
| *** martyntaylor has joined #openstack-ironic | 13:17 | |
| *** matty_dubs|gone is now known as matty_dubs | 13:18 | |
| *** ndipanov has joined #openstack-ironic | 13:19 | |
| *** linggao has joined #openstack-ironic | 13:21 | |
| *** matsuhas_ has quit IRC | 13:27 | |
| *** matsuhashi has joined #openstack-ironic | 13:28 | |
| *** matsuhas_ has joined #openstack-ironic | 13:31 | |
| *** jbjohnso_ has joined #openstack-ironic | 13:32 | |
| *** matsuhashi has quit IRC | 13:32 | |
| *** matsuhas_ has quit IRC | 13:34 | |
| *** overlayer has joined #openstack-ironic | 13:42 | |
| openstackgerrit | Vladimir Kozhukalov proposed a change to openstack/ironic-python-agent: Added lvm partitioner https://review.openstack.org/85672 | 13:51 |
| Shrews | dtantsur: thanks for your review. i addressed your comments. let me know if you have further questions. :) | 13:52 |
| openstackgerrit | Vladimir Kozhukalov proposed a change to openstack/ironic-python-agent: Added lvm partitioner https://review.openstack.org/85672 | 14:01 |
| dtantsur | Shrews, you're welcome :) | 14:10 |
| *** mdickson has quit IRC | 14:15 | |
| *** mdickson has joined #openstack-ironic | 14:17 | |
| *** jgrimm has joined #openstack-ironic | 14:18 | |
| *** romcheg has quit IRC | 14:29 | |
| *** romcheg has joined #openstack-ironic | 14:29 | |
| *** ndipanov has quit IRC | 14:39 | |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Implement caching for master images https://review.openstack.org/85387 | 14:43 |
| *** Mikhail_D_ltp has quit IRC | 14:48 | |
| *** saju_m has quit IRC | 14:50 | |
| *** ndipanov has joined #openstack-ironic | 14:54 | |
| *** mgagne1 has quit IRC | 14:54 | |
| *** romcheg1 has joined #openstack-ironic | 15:01 | |
| *** romcheg has quit IRC | 15:01 | |
| *** blamar has joined #openstack-ironic | 15:01 | |
| *** romcheg has joined #openstack-ironic | 15:06 | |
| *** romcheg1 has quit IRC | 15:06 | |
| romcheg | lucasagomes: I'm trying to deal with https://bugs.launchpad.net/ironic/+bug/1286244 | 15:08 |
| lucasagomes | right | 15:08 |
| romcheg | I was going to try 2 approaches: | 15:08 |
| romcheg | One is to check whether the file named "mkfs.filesystem" exists, the other one is to check the error message of the mkfs | 15:09 |
| lucasagomes | right, the #1 sounds better for me | 15:09 |
| lucasagomes | mkfs should have a mkfs.<fs> command for each fs it supports | 15:09 |
| lucasagomes | afaict | 15:09 |
| romcheg | Apparently both are wrong: the list of the paths where mkfs looks for the specific builder is defined on compile time so it might be different on different distributions | 15:10 |
| romcheg | the #2 won't work because on different locales the output is different | 15:12 |
| romcheg | http://paste.openstack.org/show/75217/ | 15:12 |
| lucasagomes | yeah hmmm :/ | 15:12 |
| * lucasagomes thinking | 15:13 | |
| romcheg | We can assume some of the directories for the #1 but that solution is not very portable | 15:13 |
| dtantsur | romcheg, is there some mkfs subcommand that is harmless and always succeedes (like --version?) | 15:13 |
| romcheg | There are also only two return codes: 0 and -1 | 15:13 |
| dtantsur | so you can just trey to call and error exit code would clearly mean "no such command"? | 15:13 |
| romcheg | *is looking* | 15:13 |
| dtantsur | romcheg, $ mkfs.ext2 -V mke2fs 1.42.8 (20-Jun-2013) | 15:14 |
| dtantsur | $ mkfs.vfat -V <-- invalid option >_< | 15:14 |
| romcheg | dtantsur: we use mkfs because it is a unified frontend for mkfs.<FS> | 15:15 |
| dtantsur | yeah, I understand | 15:15 |
| dtantsur | romcheg, maybe use whatever bash uses to provide autocompletion for mkfs.<TAB> ? | 15:15 |
| romcheg | dtantsur: won't work | 15:16 |
| romcheg | dtantsur: AFAIR some of the builders start with something different that "mkfs" | 15:16 |
| dtantsur | on my Fedora there's always an alias mkfs.* | 15:17 |
| lucasagomes | yeah idk hw we can get a list of supported filesystem w/o relying on the mkfs.<fs> subcommands | 15:17 |
| agordeev | romcheg: #2, enforcing LANG to C. What is the problem to perform checking prior the call and with LANG=C set? | 15:18 |
| romcheg | agordeev: that's an Idea | 15:19 |
| dtantsur | that makes sense | 15:19 |
| lucasagomes | yeah sounds reasonable | 15:20 |
| *** romcheg has quit IRC | 15:20 | |
| *** romcheg1 has joined #openstack-ironic | 15:20 | |
| *** max_lobur1 has quit IRC | 15:21 | |
| * dtantsur relocates again to be ready for Ironic meeting | 15:24 | |
| *** jgrimm has quit IRC | 15:24 | |
| *** jgrimm has joined #openstack-ironic | 15:25 | |
| openstackgerrit | Vladimir Kozhukalov proposed a change to openstack/ironic-python-agent: Added lvm partitioner https://review.openstack.org/85672 | 15:26 |
| *** mgagne has joined #openstack-ironic | 15:26 | |
| *** zul has quit IRC | 15:28 | |
| *** mgagne1 has joined #openstack-ironic | 15:29 | |
| agordeev | hah, processutils from oslo doesn't support passing additional env variables to subprocess.Popen | 15:29 |
| lucasagomes | :( | 15:31 |
| *** mgagne has quit IRC | 15:31 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: IPMINative set_boot_device persistent https://review.openstack.org/85742 | 15:31 |
| matty_dubs | Say, Ironic is the only component I'm ATC on that I didn't get a voting email for. That's just because devananda is running uncontested, right? | 15:31 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: IPMINative set_boot_device persistent https://review.openstack.org/85742 | 15:32 |
| *** zul has joined #openstack-ironic | 15:33 | |
| lucasagomes | matty_dubs, I think so | 15:34 |
| lucasagomes | there's only one candidate | 15:34 |
| * matty_dubs just wanted to make sure he wasn't missing anythign | 15:34 | |
| matty_dubs | *anything | 15:35 |
| romcheg1 | matty_dubs, lucasagomes: That reminded me elections in Soviet Union, when only one candidate was in ballots :) | 15:38 |
| *** yuriyz has quit IRC | 15:40 | |
| lucasagomes | romcheg1, heh :( | 15:44 |
| *** romcheg1 has quit IRC | 15:45 | |
| matty_dubs | romcheg1: Heh, that's not quite the image I had here, but I guess it's historically-accurate ;) | 15:46 |
| *** eghobo has joined #openstack-ironic | 15:54 | |
| *** romcheg has joined #openstack-ironic | 15:55 | |
| *** bearhands is now known as comstud | 15:56 | |
| openstackgerrit | Aleksandr Gordeev proposed a change to openstack/ironic-python-agent: Move split_command to utils https://review.openstack.org/85748 | 15:57 |
| openstackgerrit | Aleksandr Gordeev proposed a change to openstack/ironic-python-agent: Add FlowExtension https://review.openstack.org/85412 | 15:57 |
| *** matty_dubs is now known as matty_dubs|lunch | 15:57 | |
| *** martyntaylor has quit IRC | 15:59 | |
| linggao | Good morning lucasagomes. | 16:04 |
| lucasagomes | morinig linggao | 16:04 |
| lucasagomes | morning* | 16:04 |
| linggao | lucasagomes, I saw your comment on the console patch. | 16:04 |
| linggao | You said a periodic task to check if console is running or not. | 16:05 |
| lucasagomes | linggao, right, yeah, processutils.execute() won't work for background tasks (even passing --background to the shellinabox) | 16:05 |
| lucasagomes | linggao, that's one way to do it | 16:05 |
| jroll | happy monday, ironic | 16:06 |
| lucasagomes | but the idea is to have a way to verify whether the shellinabox command is running (console enabled) or not | 16:06 |
| lucasagomes | since we have a loose control of that process | 16:06 |
| lucasagomes | jroll, morning | 16:06 |
| linggao | so we still need to fix the problem of processutils.execute() hung for background process, right? | 16:07 |
| lucasagomes | linggao, yeah, or use subprocess.Popen for our case | 16:08 |
| lucasagomes | idk how they will see it in oslo | 16:08 |
| lucasagomes | to make processutils.execute() to not check for returned code or stderr | 16:08 |
| jroll | devananda: care to take another look at this today? https://review.openstack.org/#/c/81919/ | 16:09 |
| jroll | I'd really like to get that or something similar landed | 16:10 |
| jroll | we can talk about it at the meeting, too, if you'd rather | 16:10 |
| lucasagomes | jroll, +1 to talk on the meeting, yeah idk hw it look for u guys but it looks a bit overcomplicated to have 2 vendor_passthru's | 16:10 |
| linggao | lucasagomes, the periodic task thing is in additon to the fix for processutils.execute() or use subprocess.Popen, correct? | 16:11 |
| lucasagomes | jroll, maybe it's better to have only 1 vendor_passthru on the top level | 16:11 |
| lucasagomes | and pass the node_uuid as parameter | 16:11 |
| jroll | lucasagomes: I was just going to suggest that :) | 16:11 |
| lucasagomes | if node_uuid is passed then you generate a topic for it, if not u generate a random topic | 16:11 |
| jroll | makes sense | 16:11 |
| lucasagomes | jroll, at least for me (I haven't thought deeply about it) | 16:11 |
| lucasagomes | having only one seems more reasonable | 16:11 |
| jroll | I'll throw it on the agenda | 16:11 |
| russell_h | lucasagomes: I'm not sure it makes sense to overload that call though | 16:11 |
| lucasagomes | jroll, ack | 16:12 |
| *** saju_m has joined #openstack-ironic | 16:12 | |
| russell_h | lucasagomes: like, its not really the same call just with or without a node UUID, they're completely different calls | 16:12 |
| lucasagomes | linggao, yes, it's something else that we might want to have in that patch... because once we execute shellinabox with no pipe to the stderr | 16:12 |
| russell_h | there is some thinking to be done there... the vendor interface was a little bit of a weird spot to put that | 16:13 |
| lucasagomes | linggao, we won't know if it worked or not | 16:13 |
| lucasagomes | russell_h, yeah... maybe might suggestion is a bit more generic... like if we are going to have a vendor_passthru, maybe it's better to have only one URI for all the cases | 16:14 |
| lucasagomes | node agnostic or not | 16:14 |
| lucasagomes | russell_h, I'm more worried in having to mainting 2 vendorpassthru controllers | 16:14 |
| russell_h | gotcha | 16:14 |
| linggao | lucasagomes, devananda's comment says that "shell=True" is a security risk because the arguments are user-supplied. How do we resolve this? | 16:14 |
| *** dwalleck has joined #openstack-ironic | 16:15 | |
| linggao | lucasagomes, remving "shell=True" got an error. | 16:15 |
| lucasagomes | linggao, yeah if you remove the shell=True you will have to pass a list of arguments instead of one string (' '.join(x)) | 16:17 |
| lucasagomes | linggao, but I didn't quite see the problem with the shell=True there to be honest | 16:17 |
| lucasagomes | lemme take another look | 16:17 |
| *** martyntaylor has joined #openstack-ironic | 16:17 | |
| NobodyCam | Good mornig Again Ironic | 16:20 |
| lucasagomes | linggao, so I think that the problem could be things like: args.append(CONF.ipmi.terminal_cert_dir) | 16:21 |
| linggao | Good morning NobodyCam | 16:21 |
| lucasagomes | where the user could change the terminal_cert_dir option to inject more options to the process | 16:21 |
| jroll | morning NobodyCam | 16:21 |
| lucasagomes | terminal_cert_dir="/path/test -p blah blah -c blah blah" | 16:21 |
| lucasagomes | NobodyCam, morning | 16:22 |
| linggao | lucasagomes, so the problem is that user defined terminal_cert_dir, not the shell=True, right? | 16:25 |
| NobodyCam | :) | 16:25 |
| NobodyCam | morning jroll linggao lucasagomes | 16:26 |
| NobodyCam | oh and russell_h | 16:26 |
| lucasagomes | linggao, yeah basically shell=True will prevent users from passing more options in the argument like | 16:27 |
| lucasagomes | linggao, lemme see if I can show it better with code | 16:27 |
| lucasagomes | 1 min | 16:27 |
| * jroll pokes cores about checking this out, too https://review.openstack.org/#/c/79466/ | 16:28 | |
| *** zigo has joined #openstack-ironic | 16:28 | |
| russell_h | NobodyCam: morning | 16:28 |
| lucasagomes | linggao, http://paste.openstack.org/show/75225/ | 16:29 |
| lucasagomes | linggao, so the arguments should be "./" (localdir) for e.g | 16:29 |
| lucasagomes | linggao, and then user change it to be "./ -a" to pass an extra option | 16:30 |
| lucasagomes | if shell=True it will work | 16:30 |
| lucasagomes | if shell=False it won't, because every option should be an parameter to the execute() | 16:30 |
| lucasagomes | I think that's what devananda meant as a security flaw of having shell=True there | 16:30 |
| *** zigo has quit IRC | 16:33 | |
| NobodyCam | jroll: lgtm +a's | 16:33 |
| NobodyCam | s/s/d/ | 16:33 |
| jroll | nice, thanks :) | 16:33 |
| *** zigo has joined #openstack-ironic | 16:34 | |
| linggao | lucasagomes, I see what you mean. thanks for the help. | 16:34 |
| *** saju_m has quit IRC | 16:35 | |
| lucasagomes | linggao, ack... I think that's the problem, yeah it's a problem | 16:35 |
| lucasagomes | linggao, one thing I spotted right now in that patch also | 16:35 |
| lucasagomes | linggao, is that _parse_driver_info doesn't validate if port is an integer value | 16:35 |
| lucasagomes | anyhoo, just a nit not really related to that patch | 16:36 |
| lucasagomes | but if worth fixing within the changes maybe :) | 16:36 |
| linggao | can you write a comment on and I can fix it while am modifying the file? | 16:36 |
| lucasagomes | linggao, sure | 16:37 |
| linggao | lucasagomes, do you still think we need to bring the console issue (execute) up in today's meeting? | 16:37 |
| lucasagomes | linggao, hmm if there's time in the open discussion section and nobody is talking about anything relevant I'd say yes | 16:38 |
| lucasagomes | if not I would just talk here in the normal channel about it | 16:38 |
| linggao | ok. thanks. | 16:38 |
| lucasagomes | the console patch is pretty much done, just need some adjustments to workaround few problems | 16:39 |
| *** jistr has quit IRC | 16:39 | |
| linggao | yes | 16:39 |
| lucasagomes | lifeless, ping re: 84396, what you mean by the cross product of 0, 1, 5 (in whatever unit) for each of (swap, root, ephemeral) ? | 16:40 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 16:40 | |
| *** zigo has quit IRC | 16:41 | |
| JayF | lucasagomes: linggao: Or even worse; CONF.ipmi.terminal_cert_dir = ";rm -rf /" | 16:42 |
| lucasagomes | JayF, yeah, thanks | 16:43 |
| *** dwalleck has quit IRC | 16:43 | |
| lifeless | no swap, 1g root, 0g ephemeral; no swap 1g root, 1g ephemeral; no swap 1g root 5g ephemeral; no swap, 5g root, 0g ephemeral etc | 16:43 |
| lucasagomes | lifeless, 0g ephemeral == no ephemeral | 16:45 |
| lucasagomes | right? | 16:45 |
| *** zigo has joined #openstack-ironic | 16:46 | |
| *** max_lobur has joined #openstack-ironic | 16:47 | |
| lucasagomes | lifeless, btw there's other patches you might be interesting in looking at as well : https://review.openstack.org/#/c/83726 ( get rid of the swap part), https://review.openstack.org/#/c/83788/ (convert swap MB to GB) | 16:48 |
| NobodyCam | brb... | 16:51 |
| *** harlowja has joined #openstack-ironic | 16:56 | |
| lifeless | lucasagomes: yes | 16:58 |
| *** pbrooko has quit IRC | 16:58 | |
| NobodyCam | gah supper laggy internet :( | 16:58 |
| *** lucasagomes is now known as lucas-afk | 17:00 | |
| openstackgerrit | A change was merged to openstack/ironic: Add Node.instance_info field https://review.openstack.org/79466 | 17:09 |
| *** derekh has quit IRC | 17:10 | |
| linggao | JayF, yes. I'll change to shell=False. Thanks. | 17:13 |
| *** matty_dubs|lunch is now known as matty_dubs | 17:14 | |
| *** athomas has quit IRC | 17:16 | |
| openstackgerrit | Jay Faulkner proposed a change to openstack/ironic-python-agent: Compatibility fixes for Python 3.3 https://review.openstack.org/85481 | 17:20 |
| *** blamar has quit IRC | 17:20 | |
| *** jdob_ has joined #openstack-ironic | 17:20 | |
| *** mgagne1 is now known as mgagne | 17:22 | |
| openstackgerrit | Jenkins proposed a change to openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/85233 | 17:30 |
| NobodyCam | romcheg: happy BirthDay | 17:31 |
| romcheg | NobodyCam: Thanks! | 17:32 |
| *** martyntaylor has left #openstack-ironic | 17:33 | |
| *** max_lobur has left #openstack-ironic | 17:39 | |
| *** max_lobur has joined #openstack-ironic | 17:39 | |
| *** jdob_ has quit IRC | 17:40 | |
| devananda | morning, all | 17:40 |
| NobodyCam | good morning devananda :) | 17:41 |
| jroll | morning devananda | 17:44 |
| dtantsur | devananda, morning | 17:44 |
| * dtantsur still cannot get used to saying "Good morning" at 7pm | 17:45 | |
| NobodyCam | lol | 17:45 |
| *** dwalleck has joined #openstack-ironic | 17:45 | |
| *** blamar has joined #openstack-ironic | 17:50 | |
| NobodyCam | brb | 17:56 |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic-python-agent: Add BackOffLoopingCall with jitter https://review.openstack.org/84303 | 18:00 |
| *** user has joined #openstack-ironic | 18:01 | |
| *** user is now known as agordeev2 | 18:02 | |
| *** ndipanov is now known as ndipanov_gone | 18:05 | |
| *** overlayer has quit IRC | 18:09 | |
| *** jdob has quit IRC | 18:22 | |
| openstackgerrit | Jay Faulkner proposed a change to openstack/ironic-python-agent: Compatibility fixes for Python 3.3 https://review.openstack.org/85481 | 18:34 |
| NobodyCam | lucas-afk: happen to be around | 18:34 |
| openstackgerrit | David Shrewsbury proposed a change to openstack/python-ironicclient: Show maintenance mode in node list https://review.openstack.org/85785 | 18:37 |
| Shrews | NobodyCam: once lucas-afk's change that you just approved merges, that client change will be relevant ^^^^ | 18:37 |
| NobodyCam | :) heheh yes! | 18:38 |
| Shrews | dtantsur: i think you requested that, too, in the review | 18:39 |
| openstackgerrit | Jay Faulkner proposed a change to openstack/ironic-python-agent: Compatibility fixes for Python 3.3 https://review.openstack.org/85481 | 18:39 |
| dtantsur | Shrews, yes, will be great, thanks | 18:42 |
| NobodyCam | Shrews: no test changes for that one? | 18:45 |
| Shrews | NobodyCam: no | 18:47 |
| *** vkozhukalov has joined #openstack-ironic | 18:47 | |
| NobodyCam | why yes you are correct ... test node-show only check for uuid | 18:50 |
| Shrews | NobodyCam: the test fake responses already have that field, and no 'list' tests actually validate the exact fields returned | 18:50 |
| NobodyCam | :-p | 18:50 |
| NobodyCam | at some point we may want to validate what fields node-list and node-show return | 18:51 |
| *** saju_m has joined #openstack-ironic | 18:51 | |
| JayF | I have a change in to infra to disable pypy testing for ironic-python-agent until a hardlock bug in eventlet+pypy is fixed - details are in gerrit if you're curious. https://review.openstack.org/85787 | 18:51 |
| NobodyCam | s/validate/test/ | 18:51 |
| Shrews | NobodyCam: yeah, probably all of the calls are lacking in that area | 18:52 |
| Shrews | of testing | 18:52 |
| NobodyCam | wanta file a bug before the meeting | 18:52 |
| NobodyCam | hehehe :) | 18:52 |
| Shrews | NobodyCam: can do | 18:53 |
| NobodyCam | woo hoo 7 minuntes to go | 18:54 |
| NobodyCam | TY Shrews | 18:54 |
| lucas-afk | NobodyCam, hi there | 18:55 |
| lucas-afk | Shrews, https://review.openstack.org/#/c/84742/ | 18:56 |
| *** lucas-afk is now known as lucasagomes | 18:56 | |
| NobodyCam | hey hey lucas-afk I had a question... but I got the answer/ I ended up over writting your -1 on the get nova test working patch | 18:56 |
| Shrews | lucasagomes: doh! | 18:56 |
| lucasagomes | Shrews, heh it's hangin there has a while, waiting for the one in Ironic to merge | 18:57 |
| romcheg | lucasagomes: Do you think we should discuss that question about the disk partitioner on the meeting? | 18:57 |
| Shrews | lucasagomes: sorry. will abandon mine | 18:57 |
| lucasagomes | Shrews, no worries at all | 18:58 |
| lucasagomes | it's grand | 18:58 |
| lucasagomes | romcheg, yeah, if there's some time we should | 18:58 |
| lucasagomes | NobodyCam, no worries :D | 18:58 |
| NobodyCam | lucasagomes: you see my comment about client tests? | 18:58 |
| romcheg | lucasagomes: I will add that to the open discussion section | 18:58 |
| lucasagomes | NobodyCam, checking now | 18:59 |
| NobodyCam | at some point we may want to test what fields node-list and node-show return | 18:59 |
| NobodyCam | lucasagomes: ^^ | 18:59 |
| lucasagomes | NobodyCam, I had the objection because it add a new dependency in our tree and all | 18:59 |
| lucasagomes | but fine | 18:59 |
| lucasagomes | NobodyCam, +1 | 19:00 |
| NobodyCam | thwe mox or egg | 19:00 |
| lucasagomes | in the CLI u mean? | 19:00 |
| lucasagomes | NobodyCam, mox | 19:00 |
| NobodyCam | ya but it is commented nicly in the file | 19:00 |
| lucasagomes | NobodyCam, yeah, no worries we can remove it once we put things in nova | 19:01 |
| lucasagomes | romcheg, ack, thank you :) | 19:01 |
| Shrews | NobodyCam: https://bugs.launchpad.net/python-ironicclient/+bug/1303964 | 19:03 |
| NobodyCam | awesome TY Shrews | 19:03 |
| vkozhukalov | lucasagomes: romcheg: citation from man sfdisk "sfdisk doesn't understand the GUID Partition Table (GPT) format and it is | 19:04 |
| vkozhukalov | not designed for large partitions. In these cases use the more advanced | 19:04 |
| vkozhukalov | GNU parted(8)" | 19:04 |
| romcheg | vkozhukalov: there is also sgdisk which lifeless mentioned | 19:04 |
| lucasagomes | vkozhukalov, oh lol! | 19:04 |
| vkozhukalov | sgdisk is pretty convenient for wiping partitions | 19:05 |
| vkozhukalov | sgdisk --zap-all /dev/sda | 19:06 |
| *** blamar has quit IRC | 19:06 | |
| *** blamar has joined #openstack-ironic | 19:07 | |
| *** eguz has joined #openstack-ironic | 19:09 | |
| *** dwalleck has quit IRC | 19:11 | |
| *** dwalleck_ has joined #openstack-ironic | 19:11 | |
| *** eghobo has quit IRC | 19:12 | |
| *** eghobo has joined #openstack-ironic | 19:13 | |
| *** eguz has quit IRC | 19:15 | |
| *** newell_ has joined #openstack-ironic | 19:17 | |
| openstackgerrit | A change was merged to openstack/ironic: Show maintenance field in GET /nodes https://review.openstack.org/84685 | 19:27 |
| *** dwalleck has joined #openstack-ironic | 19:28 | |
| *** dwalleck_ has quit IRC | 19:31 | |
| *** Mikhail_D_ltp has quit IRC | 19:48 | |
| lifeless | NobodyCam: patch #? | 19:54 |
| lifeless | NobodyCam: Ironic should be asserting power state anyhow | 19:55 |
| NobodyCam | lifeless: https://review.openstack.org/#/c/85529 | 19:55 |
| openstackgerrit | A change was merged to openstack/ironic: Update tox.ini to also run nova tests https://review.openstack.org/84033 | 19:55 |
| NobodyCam | lifeless: ah ha Ironic is not setting the ssh port in tripleO | 19:58 |
| *** agordeev2 has quit IRC | 20:00 | |
| linggao | devananda, in current console patch, it uses ipmi_terminal_port setting for each node for the console. | 20:01 |
| lifeless | NobodyCam: and its not defaulting to 22 ? | 20:01 |
| lifeless | NobodyCam: Ironic bug :) | 20:01 |
| linggao | that require each node have different port number which is not practice if there are thousands of nodes. | 20:02 |
| *** iron1 has joined #openstack-ironic | 20:02 | |
| NobodyCam | it is using 22.. but do the gate test use 2222? | 20:02 |
| dtantsur | judging by meeting minutes, everyone can add actions :) | 20:02 |
| lifeless | NobodyCam: tripleo-ci gate? no. | 20:02 |
| devananda | dtantsur: hmm. didn't realize taht. thanks | 20:02 |
| lifeless | NobodyCam: dunno about devstack-gate | 20:02 |
| NobodyCam | ahh ok :) | 20:02 |
| linggao | devananda, should we implement a port pool so that the nodes can be assined a port at run time when console is enabled? | 20:03 |
| devananda | linggao: IIRC, the ipminative / pyghmi driver do that already | 20:03 |
| *** bigjools_ has joined #openstack-ironic | 20:03 | |
| dtantsur | enough for today, guys. Good night | 20:03 |
| *** lucasagomes_ has joined #openstack-ironic | 20:03 | |
| devananda | dtantsur: g'night! | 20:03 |
| *** harlowja has quit IRC | 20:03 | |
| NobodyCam | night dtantsur | 20:03 |
| NobodyCam | brb | 20:04 |
| lucasagomes_ | err my connection dropped at the very end of the meeting | 20:04 |
| linggao | devananda, I have not looked at ipminative yet. But for ipmitool implementation should we keep what it is today or should we use the port pool? | 20:04 |
| * jroll bbl | 20:04 | |
| devananda | lucasagomes_: ah, so you didn't see us #action everything to you? | 20:04 |
| devananda | :) | 20:04 |
| lucasagomes_ | devananda, lol, checking the logs now :P | 20:05 |
| lucasagomes_ | hah | 20:05 |
| linggao | lol | 20:05 |
| lucasagomes_ | anyhoo, great meeting folks, thanks for the args/suggestions | 20:05 |
| lucasagomes_ | dinner time for me | 20:06 |
| lucasagomes_ | have a g'night everybody! | 20:06 |
| NobodyCam | havea good night lucasagomes enjoy dinner | 20:06 |
| linggao | good night lucasagomes_ | 20:06 |
| *** lucasagomes_ is now known as lucas-dinner | 20:06 | |
| *** russellb_ has joined #openstack-ironic | 20:09 | |
| *** lucasagomes has quit IRC | 20:10 | |
| *** russellb has quit IRC | 20:10 | |
| *** bigjools has quit IRC | 20:10 | |
| *** russellb_ is now known as russellb | 20:10 | |
| linggao | devananda any suggestion on it? ( I am working on the console patch.) | 20:10 |
| devananda | linggao: what is involved in port pool? | 20:11 |
| devananda | linggao: and what is the limitation if we *dont* use it? | 20:11 |
| linggao | davananda, it involves setting the port range in the conf file, and store used the port number with the node in nodes table, and have a fucntion reads all the used ports from the db and come up with a free port number. | 20:13 |
| linggao | and also delete the port number from the the nodes table if the console is disabled for the node. | 20:14 |
| linggao | it is quite a lot of work. | 20:14 |
| devananda | linggao: what is the limitation if we stick with what we have now? | 20:16 |
| linggao | the limitation, hmm, what is the maximum port numbers in Linux? | 20:16 |
| JayF | like tcp/udp ports? 65535. | 20:16 |
| devananda | 64K | 20:16 |
| linggao | now the default port number for shellinabox is 4200 | 20:16 |
| linggao | 64k is pretty big :) | 20:17 |
| devananda | linggao: AIUI, terinal port is the port for the shelinaboxd instance used to proxy connections to that node, yes? | 20:18 |
| linggao | yes. | 20:18 |
| devananda | so | 20:18 |
| devananda | if there were 10 conductor services running | 20:18 |
| devananda | ** so, we get linear scaling based on number of conductor services running | 20:19 |
| *** harlowja has joined #openstack-ironic | 20:19 | |
| devananda | with eg. 10 conductors and 1K nodes, and a terminal open to each one, we'd have only 100 ports in use per conductor host | 20:19 |
| JayF | is a 100:1 ratio the usual for nodes:conductors? | 20:20 |
| linggao | devananda, are you saying each conductor "manages" a set of nodes? | 20:21 |
| linggao | but the the terminal ports cannot be overlapped. | 20:21 |
| devananda | linggao: I think we'll hit much larger scaling issues before we exhaust the number of network ports | 20:21 |
| devananda | linggao: like running that many concurrent shellinaboxd processes on the same host | 20:21 |
| linggao | this is the port when you place the url on the browser to veiw/write on console. | 20:22 |
| NobodyCam | linggao: browser? not comand line? | 20:23 |
| devananda | JayF: based on nova-baremetal, I'm guessing at a 100:1 ratio, I dont think anyone has hard data on it | 20:23 |
| linggao | for example, http://<ironic_host>:4201, that will be for one node. | 20:23 |
| devananda | JayF: I think we'd all like a much higher number eventually | 20:23 |
| devananda | linggao: have you looked at how to address that from a client? | 20:24 |
| NobodyCam | linggao: so i runn "inronic node-console uuid start" and then have to open a browser/ | 20:24 |
| linggao | NobodyCam, we use command line to eable the console. But have to use browser to watch it. | 20:24 |
| devananda | linggao: the suggested deployment topology is to keep ironic-conductor hosts unreachable by users | 20:24 |
| devananda | linggao: so we need a way for user to route the request via the API service | 20:25 |
| NobodyCam | jbjohnso_: does pyimpi need a browser for console ? | 20:25 |
| jbjohnso_ | NobodyCam, nope | 20:25 |
| NobodyCam | linggao: ^^^ | 20:26 |
| jbjohnso_ | NobodyCam, I have a console server | 20:26 |
| jbjohnso_ | NobodyCam, and said console server has http and non-http ways in | 20:26 |
| linggao | devananda, can you give more details? | 20:26 |
| JayF | From a network security standpoint, I don't want that exposure to be "hit the ironic-api on another port" | 20:26 |
| NobodyCam | jbjohnso_: TY thats what | 20:26 |
| NobodyCam | I thought | 20:26 |
| devananda | linggao: a user of nova, who has provisioned an instance on bare metal, should be abble to request a console session | 20:26 |
| devananda | linggao: however that user does not have direct network route to the ironic-conductor service | 20:27 |
| linggao | yes | 20:27 |
| jbjohnso_ | pyghmi itself knows only of bytes being flung back and fourth | 20:27 |
| devananda | linggao: they can only reach the nova-api and ironic-api (and other *-api) services | 20:27 |
| jbjohnso_ | so a wrapper script or daemon is how it makes it to real world | 20:27 |
| jbjohnso_ | I need a public server... | 20:27 |
| NobodyCam | linggao: I was envisioning something like the virsh console command | 20:28 |
| *** boris-42 has joined #openstack-ironic | 20:28 | |
| devananda | JayF: connectivity to said port should be handled with auth/access rights | 20:29 |
| linggao | devananda, I see what you mean. | 20:29 |
| devananda | linggao: i may have side tracked -- what you were working on is just the conductor side of things | 20:29 |
| devananda | linggao: in addition to that, we *also* need an API layer that does auth/access control and routing the request appropriately | 20:29 |
| jbjohnso_ | as one who generally detests using a browser and is similarly generally terrible at in-browser development | 20:30 |
| jbjohnso_ | I did the bare minimum to be satisfied that a decent widget could be provided | 20:30 |
| jbjohnso_ | javascript is one of the languages that makes me stabby to write in | 20:30 |
| devananda | jbjohnso_: how much work would it take to query pyghmi to get a text log of the all (or last N lines) of serial console output? | 20:31 |
| jbjohnso_ | it's not quite powershell stabby, but... | 20:31 |
| devananda | or is that already done? :) | 20:31 |
| jbjohnso_ | devananda, well, the log format currently has all sorts of goodies | 20:31 |
| linggao | devananda, yes. I'll work on that part after this patch. And I also plan to work on bringing ipminative implementaion of console in. | 20:31 |
| jbjohnso_ | devananda, the default behavior is to keep between 8 and 16 kilobytes in memory buffer | 20:31 |
| devananda | linggao: great | 20:31 |
| jbjohnso_ | and the on-disk log has everything, who was logged in, how many times each user was logged in, whether the remote console is alive or dead | 20:31 |
| jbjohnso_ | and what second each chunk of console output came in at | 20:32 |
| jbjohnso_ | I was doing a bit of a torture test on it, so these numbers will be large (it was while :; do dmesg; done) | 20:32 |
| devananda | jbjohnso_: cool | 20:32 |
| linggao | devananda, the conductor thing. Does each ironic host has one condictor? | 20:32 |
| jbjohnso_ | -rw-rw-r-- 1 jbjohnso jbjohnso 42M Apr 7 14:38 n2 | 20:32 |
| jbjohnso_ | that's the log, and it comes with the metadata: | 20:32 |
| jbjohnso_ | -rw-rw-r-- 1 jbjohnso jbjohnso 264K Apr 7 14:38 n2.cbl | 20:32 |
| devananda | linggao: ironic services should be deployed like other openstack service | 20:32 |
| jbjohnso_ | the latter has fixed record sizes for fast seeking, trivial binary search of start and stop times, etc | 20:33 |
| jbjohnso_ | all the timestamp data is in the 'cbl' file | 20:33 |
| devananda | linggao: ironic-api service on one or more hosts, with a load balancer, and routable by both clients and other openstack services | 20:33 |
| jbjohnso_ | I need to get copyright notices in place and at least on xCAT's git repo for that daemon | 20:33 |
| linggao | davananda, got it. thanks. | 20:34 |
| devananda | linggao: ironic-conductor service on one or more hosts, without a load balancer, non-routable by clients. an RPC bus is used between the conductor and service agents. the conductor agents also need access to the IPMI network | 20:34 |
| devananda | linggao: a reasonable number of ironic-conductor service agents should be deployed, depending on how many hardware nodes are being managed and what level of concurrent deploy performance and redundancy is desired | 20:34 |
| jbjohnso_ | oh, I need to commit my 'send break' support in pyghmi... | 20:35 |
| *** mgagne has quit IRC | 20:36 | |
| *** mgagne has joined #openstack-ironic | 20:36 | |
| devananda | gotta switch cafe's... bbiab | 20:37 |
| linggao | davananda, I see. so that's why we define the host for each node when registering it? | 20:37 |
| linggao | thanks devananda for the help. | 20:39 |
| *** saju_m has quit IRC | 20:41 | |
| linggao | NobodyCam, with the current patch implementation of the API, ironic node-get-console only returns the url for accessing the console. | 20:41 |
| NobodyCam | linggao: ack :) | 20:42 |
| linggao | NobodyCam, maybe we should change that to actually return a console. | 20:42 |
| *** mgagne_ has joined #openstack-ironic | 20:43 | |
| *** mgagne has quit IRC | 20:44 | |
| jbjohnso_ | time to show the world my crimes against python.... or at least more of my crimes against python... | 20:44 |
| *** mgagne_ is now known as mgagne | 20:44 | |
| *** mgagne has quit IRC | 20:45 | |
| *** mgagne has joined #openstack-ironic | 20:45 | |
| boris-42 | devananda ping | 20:45 |
| NobodyCam | I have played much / at all with console. but as an admin I may only have command line access, ie remote locate access.. so requiring a browser may not work in all cases | 20:45 |
| NobodyCam | boris-42: devananda is moving locations atm.. he'll be back in a bit | 20:45 |
| boris-42 | NobodyCam ok thanks | 20:46 |
| NobodyCam | linggao: s/have/haven't/ | 20:46 |
| linggao | NobodyCam yes, +1 | 20:46 |
| NobodyCam | :) | 20:46 |
| jbjohnso_ | well, there it is... | 20:49 |
| jbjohnso_ | https://sourceforge.net/p/xcat/confluent/ci/master/tree/ | 20:49 |
| * NobodyCam steps awayfor a few minutes | 20:52 | |
| jbjohnso_ | may Guido have mercy upon my soul | 20:52 |
| linggao | jbjohnso_, great! I need to learn how to use it. | 20:52 |
| linggao | lol | 20:52 |
| jbjohnso_ | I need to actually make it daemonize and start consoles without a user asking | 20:52 |
| jbjohnso_ | and to be able to send break of course | 20:52 |
| NobodyCam | jbjohnso_: your going to start the console before the user requests it? | 20:53 |
| *** mrda_away is now known as mrda | 20:56 | |
| mrda | morning all. | 20:57 |
| NobodyCam | good morning mrda | 20:57 |
| mrda | Just wanted to apologise for missing the meeting today. The change is daylight savings means it's a 4:30am meeting, which I don't think is humanly possible | 20:57 |
| jbjohnso_ | NobodyCam, well, at least have the option, in HPC baremetal it's a common desire to be there waiting for kernel bad output | 20:57 |
| jbjohnso_ | NobodyCam, as an option | 20:57 |
| jbjohnso_ | I'm not very good at videos, but... | 20:58 |
| jbjohnso_ | https://www.youtube.com/watch?v=G_lDaktYnsQ | 20:58 |
| jbjohnso_ | devananda, NobodyCam that's roughly the capabilities of the daemon today... not primetime but you can get the gist | 20:59 |
| boris-42 | Btw guys are you interested in benchmarking Ironic | 20:59 |
| NobodyCam | ya, as an option, that could be cool... auto detect KP on deploy's, would also love to auto capture the console from the deploy ram disk | 20:59 |
| boris-42 | I can make some small live demo | 20:59 |
| boris-42 | of rally so you'll be able to have even gate with performance results | 20:59 |
| JoshNang | boris-42: I'd be interested to see a demo | 21:00 |
| *** dwalleck has quit IRC | 21:00 | |
| boris-42 | JoshNang nice | 21:00 |
| JayF | +1 to a demo | 21:00 |
| boris-42 | let's make some hangout call? | 21:00 |
| boris-42 | *not today* | 21:01 |
| JayF | I'd suggest something async | 21:01 |
| JayF | like either an example script we can all run against devstack | 21:01 |
| jroll | boris-42: make a video? :) | 21:01 |
| JayF | or a youtube video if that's your preferred way of doing it | 21:01 |
| boris-42 | jroll nope+) no video=) | 21:01 |
| NobodyCam | jbjohnso_: you say "with some javascript" with such distain | 21:01 |
| boris-42 | jroll only live only hardcore=) | 21:01 |
| jroll | hdh | 21:01 |
| boris-42 | jroll we are chaining to fast to make a video=) | 21:02 |
| *** dwalleck has joined #openstack-ironic | 21:02 | |
| jroll | s/hdh/heh/ | 21:02 |
| jbjohnso_ | NobodyCam, heh, compared to the alternative of 'java plugin' (or worse) I intended no disrespect to javascript | 21:02 |
| jbjohnso_ | ;) | 21:02 |
| jroll | awesome | 21:02 |
| jroll | boris-42: any results yet? | 21:02 |
| boris-42 | jroll for ironic? | 21:02 |
| jroll | yes | 21:03 |
| boris-42 | jroll nope) | 21:03 |
| jroll | ah. I'm anxious to see :) | 21:03 |
| boris-42 | jroll actually I should find somebody not from Mirnatis to make a couple of simple benchmarks lol=) | 21:04 |
| boris-42 | jroll if you are interested it won't take too much time=) | 21:04 |
| boris-42 | jroll I hope to not see something like that http://pavlovic.me/rally/glance_list.html =) | 21:04 |
| NobodyCam | good video jbjohnso_ :) | 21:05 |
| jbjohnso_ | my first ever youtube video | 21:06 |
| jbjohnso_ | I feel like those young-uns | 21:06 |
| jroll | boris-42: I'm happy to run some benchmarks if you have benchmarks to run :) | 21:06 |
| NobodyCam | :-p | 21:07 |
| boris-42 | jroll I can make it's not a big problem.. but why not just spend 5 minutes to learn how to make them?) | 21:07 |
| jroll | boris-42: I'm skeptical it will take 5 minutes, but yes, maybe one day I will do that | 21:08 |
| NobodyCam | jbjohnso_: http://www.slate.com/articles/business/the_bet/2014/03/silicon_valley_ageism_are_young_people_really_more_innovative.html | 21:08 |
| *** mgagne has quit IRC | 21:08 | |
| boris-42 | jroll okay I'll make one benchmark | 21:08 |
| openstackgerrit | Jenkins proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/85820 | 21:08 |
| boris-42 | jroll actually I will found somebody to make it lol=) | 21:09 |
| *** mgagne has joined #openstack-ironic | 21:09 | |
| *** mgagne is now known as Guest11621 | 21:09 | |
| jroll | boris-42: when I say that, I mean 5 minutes for me to learn, not 5 minutes for you to make one :P | 21:09 |
| boris-42 | jroll so what method would you like to benchmark?) | 21:09 |
| *** GheRivero_ has joined #openstack-ironic | 21:09 | |
| boris-42 | jroll it's quite simple to learn.. https://github.com/stackforge/rally/blob/master/rally/benchmark/scenarios/keystone/basic.py#L23-L26 | 21:10 |
| boris-42 | jroll ^it's benchmark lol | 21:10 |
| * jroll looks | 21:10 | |
| boris-42 | jroll 3 lines of code lol | 21:10 |
| jroll | oh, heh | 21:10 |
| jroll | but then I have to install and configure rally, no? | 21:10 |
| boris-42 | jroll yep you should run https://github.com/stackforge/rally/blob/master/install_rally.sh | 21:10 |
| jbjohnso_ | devananda, but that's probably a better explanation of what a daemon can do with pyghmi console-wise than anything I could type off hand | 21:11 |
| boris-42 | jroll something like | 21:11 |
| jbjohnso_ | and now time to call it a day | 21:11 |
| boris-42 | git clone ...rally | 21:11 |
| boris-42 | cd rally | 21:11 |
| *** Guest11621 is now known as mgagne_ | 21:11 | |
| boris-42 | ./install_rally.sh | 21:11 |
| *** mgagne_ has quit IRC | 21:11 | |
| *** mgagne_ has joined #openstack-ironic | 21:11 | |
| boris-42 | jroll not that hard ^? ) | 21:11 |
| JayF | If you wanted to help everyone, maybe toss it on an etherpad | 21:11 |
| NobodyCam | night jbjohnso_ | 21:11 |
| JayF | so since nobody seems to have time to look at it right now, maybe they could later? | 21:11 |
| lifeless | NobodyCam: so https://review.openstack.org/#/c/84361/ may still be needed | 21:12 |
| jroll | boris-42: I'll check it out when I have free time, thanks | 21:12 |
| boris-42 | jroll you can join live demo if you would like | 21:12 |
| jroll | I'd rather not :/ | 21:13 |
| jroll | sorry I just don't have time right now, boris | 21:13 |
| boris-42 | jroll it is not right now* | 21:13 |
| lifeless | NobodyCam: see commit 197eb7c7155c42d3a9c152e9bfa37af7d4261ea0 in tripleo-incubator | 21:13 |
| boris-42 | jroll I am just planing to make one | 21:13 |
| jroll | ok | 21:13 |
| NobodyCam | lifeless: ya. with my power fix patch I am getting time out waiting for ping from node. 1800 seconds | 21:14 |
| lifeless | NobodyCam: try putting them in a series | 21:14 |
| NobodyCam | the patches | 21:14 |
| lifeless | yes | 21:14 |
| NobodyCam | ack | 21:14 |
| lifeless | rebase one on the other | 21:15 |
| lifeless | then push | 21:15 |
| NobodyCam | will do | 21:15 |
| openstackgerrit | Jenkins proposed a change to openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/83488 | 21:16 |
| boris-42 | JoshNang JayF jroll so could you share your emails? and i'll send invites | 21:16 |
| boris-42 | NobodyCam lifeless btw guys would you like to take a part in Rally demo as well?) | 21:16 |
| jroll | boris-42: just ping me on IRC if you decide to do one | 21:17 |
| boris-42 | jroll i think 4/10 (in 2 days) | 21:17 |
| *** dwalleck has quit IRC | 21:19 | |
| *** dwalleck has joined #openstack-ironic | 21:21 | |
| lifeless | boris-42: sure | 21:22 |
| lifeless | boris-42: at least, I'll try :) | 21:23 |
| boris-42 | lifeless nice could you share your email | 21:23 |
| lifeless | robertc at robertcollins dot net :P | 21:23 |
| *** linggao has quit IRC | 21:23 | |
| boris-42 | lifeless sent invite =) | 21:24 |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Fix for tripleO undercloud gate tests https://review.openstack.org/85529 | 21:26 |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Workaround race with Neutron port updates https://review.openstack.org/84361 | 21:26 |
| openstackgerrit | Jenkins proposed a change to openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/83488 | 21:27 |
| NobodyCam | huh | 21:27 |
| *** jbjohnso_ has quit IRC | 21:29 | |
| *** harlowja is now known as harlowja_away | 21:31 | |
| *** harlowja_away is now known as harlowja | 21:35 | |
| *** dwalleck has quit IRC | 21:55 | |
| *** GheRivero has quit IRC | 21:55 | |
| *** GheRivero_ is now known as GheRivero | 21:55 | |
| russell_h | devananda: +2 on this? https://review.openstack.org/#/c/84814/ | 22:02 |
| *** vkozhukalov has quit IRC | 22:04 | |
| *** max_lobur has quit IRC | 22:23 | |
| *** jgrimm has quit IRC | 22:31 | |
| *** mgagne_ is now known as mgagne | 22:35 | |
| *** early has quit IRC | 22:42 | |
| openstackgerrit | A change was merged to openstack/ironic-python-agent: Compatibility fixes for Python 3.3 https://review.openstack.org/85481 | 22:42 |
| *** radsy has joined #openstack-ironic | 22:45 | |
| *** early has joined #openstack-ironic | 22:46 | |
| *** eguz has joined #openstack-ironic | 23:03 | |
| *** eghobo has quit IRC | 23:08 | |
| *** matty_dubs is now known as matty_dubs|gone | 23:10 | |
| *** rloo has quit IRC | 23:12 | |
| *** rloo has joined #openstack-ironic | 23:12 | |
| NobodyCam | lifeless: need to see why .. but still hit hitting the time out. looks like the logging fix merged so I should be able to get good logs .. TY | 23:16 |
| lifeless | \o/ | 23:17 |
| jroll | devananda: are you still around? | 23:18 |
| jroll | devananda: I want to talk about https://review.openstack.org/#/c/81919/ when you have time, since we didn't get to it in the meeting today | 23:19 |
| NobodyCam | jroll: last message I got from devananda was the the house had no internets and was rebooting things ... So I'm not 100 % sure he can be around :-p | 23:21 |
| jroll | heh | 23:21 |
| jroll | I can wait until tomorrow or whatever, hopefully he'll see those pings :) | 23:22 |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Fix for tripleO undercloud gate tests https://review.openstack.org/85529 | 23:27 |
| NobodyCam | he will | 23:27 |
| NobodyCam | :) | 23:27 |
| *** rloo has quit IRC | 23:29 | |
| *** rloo has joined #openstack-ironic | 23:29 | |
| *** rloo has quit IRC | 23:32 | |
| *** rloo has joined #openstack-ironic | 23:32 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Fix leaking DB details to API on error https://review.openstack.org/73121 | 23:35 |
| openstackgerrit | A change was merged to openstack/ironic-python-agent: Move split_command to utils https://review.openstack.org/85748 | 23:35 |
| *** lucas-dinner has quit IRC | 23:36 | |
| *** rloo has quit IRC | 23:36 | |
| *** rloo has joined #openstack-ironic | 23:37 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Fix leaking DB details to API on error https://review.openstack.org/73121 | 23:39 |
| NobodyCam | walk'n to sonic to get dinner ... will pop back in to check on the test results :) | 23:44 |
| *** bigjools_ is now known as bigjools | 23:46 | |
| *** bigjools has quit IRC | 23:46 | |
| *** bigjools has joined #openstack-ironic | 23:46 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!