| *** rbudden has quit IRC | 00:02 | |
| *** wajdi has joined #openstack-ironic | 00:10 | |
| *** garthb has quit IRC | 00:10 | |
| *** fragatin_ has joined #openstack-ironic | 00:11 | |
| *** wajdi has quit IRC | 00:14 | |
| *** fragatin_ has quit IRC | 00:14 | |
| *** fragatina has quit IRC | 00:14 | |
| *** fragatina has joined #openstack-ironic | 00:15 | |
| *** rloo has joined #openstack-ironic | 00:20 | |
| *** Sukhdev has joined #openstack-ironic | 00:26 | |
| *** mtanino has quit IRC | 00:28 | |
| *** baoli has joined #openstack-ironic | 00:34 | |
| *** baoli has quit IRC | 00:36 | |
| *** baoli has joined #openstack-ironic | 00:36 | |
| *** praneshp has quit IRC | 00:46 | |
| *** keedya has quit IRC | 00:50 | |
| jaypipes | jroll: "sounds like a major win for the team" <--- heh, nice. :) | 00:51 |
|---|---|---|
| *** Sukhdev has quit IRC | 00:56 | |
| *** baoli has quit IRC | 00:59 | |
| *** hoangcx has joined #openstack-ironic | 01:03 | |
| *** baoli has joined #openstack-ironic | 01:04 | |
| *** fragatin_ has joined #openstack-ironic | 01:12 | |
| *** fragatin_ has quit IRC | 01:14 | |
| *** caiobo has joined #openstack-ironic | 01:15 | |
| *** fragatina has quit IRC | 01:16 | |
| *** Sukhdev has joined #openstack-ironic | 01:19 | |
| openstackgerrit | Merged openstack/ironic: Fixed nits in the new inspection doc page https://review.openstack.org/308862 | 01:22 |
| *** Sukhdev has quit IRC | 01:27 | |
| *** baoli has quit IRC | 01:27 | |
| *** rloo has quit IRC | 01:32 | |
| *** Keedya_ has joined #openstack-ironic | 01:43 | |
| *** thrash is now known as thrash|g0ne | 01:45 | |
| *** vishwanathj has quit IRC | 01:48 | |
| *** Keedya_ has quit IRC | 01:49 | |
| *** Keedya_ has joined #openstack-ironic | 01:52 | |
| *** Keedya_ has quit IRC | 01:59 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/311535 | 02:00 |
| openstackgerrit | Dan Bode proposed openstack/bifrost: Add Ubuntu 16.04 defaults for ironic role https://review.openstack.org/312303 | 02:01 |
| *** keedya_ has joined #openstack-ironic | 02:05 | |
| *** ijw has quit IRC | 02:06 | |
| *** jaybeale has quit IRC | 02:22 | |
| *** hemna_ has joined #openstack-ironic | 02:23 | |
| *** fragatina has joined #openstack-ironic | 02:31 | |
| *** caiobo has quit IRC | 02:35 | |
| *** jaypipes has quit IRC | 02:40 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Make use of oslo-config-generator https://review.openstack.org/247331 | 03:17 |
| *** baoli has joined #openstack-ironic | 03:31 | |
| *** hemna_ has quit IRC | 03:36 | |
| *** baoli has quit IRC | 03:37 | |
| *** Nisha has joined #openstack-ironic | 03:54 | |
| *** jaybeale has joined #openstack-ironic | 03:54 | |
| *** fragatina has quit IRC | 03:56 | |
| *** fragatina has joined #openstack-ironic | 03:56 | |
| *** e0ne has joined #openstack-ironic | 03:59 | |
| *** e0ne has quit IRC | 04:01 | |
| *** e0ne has joined #openstack-ironic | 04:02 | |
| *** e0ne has quit IRC | 04:05 | |
| *** vmud213 has joined #openstack-ironic | 04:06 | |
| *** links has joined #openstack-ironic | 04:25 | |
| *** amit213 has quit IRC | 04:44 | |
| *** harlowja_at_home has joined #openstack-ironic | 04:46 | |
| *** JoshNang has quit IRC | 04:50 | |
| *** harlowja_at_home has quit IRC | 05:02 | |
| *** ChrisAusten has joined #openstack-ironic | 05:05 | |
| *** vmud213 has quit IRC | 05:06 | |
| *** ChubYann has quit IRC | 05:11 | |
| *** fragatina has quit IRC | 05:15 | |
| *** yolanda has joined #openstack-ironic | 05:16 | |
| *** bnemec has quit IRC | 05:17 | |
| *** rbrud has quit IRC | 05:28 | |
| *** ayoung has quit IRC | 05:31 | |
| *** hoangcx has quit IRC | 05:38 | |
| *** hoangcx has joined #openstack-ironic | 05:38 | |
| *** vmud213 has joined #openstack-ironic | 05:45 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic-specs: Baremetal provisioning in UEFI secure boot mode https://review.openstack.org/230274 | 05:49 |
| *** Sukhdev has joined #openstack-ironic | 05:49 | |
| *** david-lyle has quit IRC | 06:05 | |
| *** jrist has quit IRC | 06:06 | |
| *** openstackgerrit has quit IRC | 06:17 | |
| *** openstackgerrit has joined #openstack-ironic | 06:18 | |
| *** Sukhdev has quit IRC | 06:19 | |
| *** e0ne has joined #openstack-ironic | 06:41 | |
| *** tesseract has joined #openstack-ironic | 07:01 | |
| *** tesseract is now known as Guest23319 | 07:01 | |
| *** sacharya has quit IRC | 07:05 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Clear target_power_state on conductor startup https://review.openstack.org/310090 | 07:13 |
| *** jaybeale has quit IRC | 07:19 | |
| *** jrist has joined #openstack-ironic | 07:20 | |
| *** ifarkas has joined #openstack-ironic | 07:22 | |
| openstackgerrit | Debayan Ray proposed openstack/ironic: File download fails with swift pseudo folder https://review.openstack.org/306929 | 07:23 |
| *** yolanda has quit IRC | 07:23 | |
| *** deray has joined #openstack-ironic | 07:25 | |
| *** yolanda has joined #openstack-ironic | 07:32 | |
| *** jrist has quit IRC | 07:38 | |
| *** gugl has quit IRC | 07:42 | |
| *** ayoung has joined #openstack-ironic | 07:49 | |
| *** dmk0202 has joined #openstack-ironic | 07:54 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:00 | |
| *** stendulker has joined #openstack-ironic | 08:05 | |
| *** sacharya has joined #openstack-ironic | 08:06 | |
| lucasagomes | morning all | 08:10 |
| *** sacharya has quit IRC | 08:11 | |
| *** mbound has joined #openstack-ironic | 08:11 | |
| stendulker | morning lucasagomes | 08:20 |
| *** ohamada has joined #openstack-ironic | 08:21 | |
| alineb | morning all! | 08:21 |
| *** vmud213 has quit IRC | 08:23 | |
| *** vmud213 has joined #openstack-ironic | 08:23 | |
| *** jistr has joined #openstack-ironic | 08:24 | |
| *** lennyb has quit IRC | 08:24 | |
| lucasagomes | stendulker, alineb morning | 08:28 |
| *** keedya_ has quit IRC | 08:28 | |
| openstackgerrit | Merged openstack/proliantutils: Add `.flash` extn to supported raw firmware files https://review.openstack.org/304028 | 08:33 |
| openstackgerrit | Merged openstack/proliantutils: Add proper error message for extraction failure https://review.openstack.org/305331 | 08:33 |
| *** jrist has joined #openstack-ironic | 08:33 | |
| alineb | o/ lucasagomes | 08:35 |
| deray | lucasagomes, stendulker alineb and all others .. g'morning | 08:37 |
| stendulker | o/ alineb deray | 08:37 |
| deray | stendulker, o/ | 08:38 |
| alineb | stendulker, deray o/ | 08:38 |
| deray | alineb, o/ | 08:39 |
| *** mgould has joined #openstack-ironic | 08:40 | |
| Nisha | morning Ironic | 08:47 |
| mkovacik | morning all! | 08:48 |
| deray | Nisha, g'afternoon :) | 08:48 |
| deray | mkovacik, o/ | 08:48 |
| Nisha | :) | 08:48 |
| mgould | morning Nisha mkovacik deray and everyone else! | 08:48 |
| mkovacik | deray, o/ :) | 08:48 |
| Nisha | mgould, o/ | 08:49 |
| mkovacik | mgould, morning! :) | 08:49 |
| *** divya has joined #openstack-ironic | 08:49 | |
| deray | mgould, o/ | 08:49 |
| divya | Hi Ironicers | 08:50 |
| Nisha | lucasagomes, i was going through the comments on https://review.openstack.org/#/c/302403/6/devstack/lib/ironic | 08:50 |
| mkovacik | divya, o/ | 08:50 |
| vmud213 | morning all | 08:52 |
| divya | mgould : yesterday i enabled log for dnsmasq and here is the logs http://paste.openstack.org/show/495956/ | 08:53 |
| mgould | divya, excellent | 08:54 |
| *** derekh has joined #openstack-ironic | 08:56 | |
| divya | mgould : From this line, May 3 18:16:54 dnsmasq-dhcp[17537]: 0 DHCPRELEASE(tapd6d47e79-1b) 100.100.100.23 98:be:94:2b:0e:f4 | 08:57 |
| mgould | do you know what has the MAC address 98:be:94:2b:0e:f4? | 08:57 |
| mgould | [also, how did you enable logging in the end?] | 08:58 |
| divya | neutron-dhcp-agent is trying to assign this ip 100.100.100.23 to the baremetal server's eth mac 98:be:94:2b:0e:f4 | 08:58 |
| divya | This is my current config in dhcp_agent.ini file dnsmasq_base_log_dir = /tmp/dhcp_dns_log | 09:00 |
| divya | if it is /tmp log files are not created under /tmp | 09:00 |
| mgould | divya, how annoying of it! | 09:01 |
| *** electrofelix has joined #openstack-ironic | 09:01 | |
| divya | mgould : as mentined in the line DHCPRELEASE(tapd6d47e79-1b) tap is also created. | 09:02 |
| mgould | divya, that's less logging than I'd hoped to see: can you edit dnsmasq.conf directly, add the line `log-dhcp`, and try restarting dnsmasq? | 09:03 |
| mgould | [this may not work] | 09:04 |
| mgould | brb | 09:04 |
| divya | i am using devstack. i don't find dnsmasq.conf file. do i need to edit dnsmasq-ironic.conf? | 09:06 |
| *** sacharya has joined #openstack-ironic | 09:07 | |
| *** sacharya has quit IRC | 09:12 | |
| sambetts | Morning all | 09:20 |
| mgould | divya, yes, looks like that will do it: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#networking-service-configuration | 09:20 |
| mgould | morning sambetts | 09:20 |
| mgould | divya, or you could run tcpdump on the tapd6d47e79-1b interface | 09:21 |
| dtantsur | Morning Ironic | 09:21 |
| mgould | morning dtantsur, welcome back :-) | 09:21 |
| dtantsur | :) | 09:21 |
| dtantsur | sambetts, lucasagomes, how was your trip back? | 09:21 |
| lucasagomes | dtantsur, hi there, not bad. I'm glad I didn't miss my connection flight because of the delay in Austin | 09:22 |
| mgould | divya, I *think* Neutron is creating that tap interface to connect the DHCP server to the provisioning network | 09:22 |
| dtantsur | alineb, deray, divya, stendulker, morning | 09:22 |
| mgould | so presumably Neutron will somehow wire it up to the physical NIC | 09:22 |
| lucasagomes | got to the gate at boarding time | 09:22 |
| dtantsur | lucasagomes, yeah, that's great | 09:22 |
| mgould | lucasagomes, always nerve-wracking when that happens | 09:22 |
| dtantsur | we've jumped on an earlier bus with mkovacik, so got to Brno even earlier than expected | 09:22 |
| * mgould googles, and sees that it should be "nerve-racking" | 09:22 | |
| mgould | English is hard, kids | 09:22 |
| dtantsur | omg it is | 09:23 |
| lucasagomes | mgould, yup | 09:23 |
| lucasagomes | dtantsur, oh nice | 09:23 |
| deray | vmud213 divya sambetts dtantsur g'morning | 09:23 |
| mgould | http://grammarist.com/usage/rack-wrack/ if anyone's interested - it looks like "wrack" is basically obsolete now | 09:23 |
| vmud213 | deray o/ | 09:24 |
| stendulker | o/ mgould, sambetts, divya, Nisha, vmud213, mkovacik | 09:25 |
| mgould | morning stendulker | 09:25 |
| sambetts | dtantsur: not bad :) converence flu is still eating away at me though | 09:26 |
| vmud213 | morning stendulker | 09:26 |
| divya | deray, stendulker : o/ | 09:26 |
| vmud213 | mgould: i have seen your comments on https://review.openstack.org/#/c/302143/ | 09:27 |
| Nisha | stendulker, o/ | 09:27 |
| dtantsur | sambetts, oh, get better :( I somehow escaped it seemingly | 09:28 |
| deray | mgould, I cannot (but) (help) agree(ing).. with your "English is hard". I always get confused with the correct one/s | 09:28 |
| dtantsur | sambetts, ifarkas, I could use some reviews on https://review.openstack.org/304185 please | 09:28 |
| sambetts | dtantsur: thank :) it took until monday to appear, and hit me real bad | 09:28 |
| vmud213 | Unfortunately i could not test it with redhat | 09:29 |
| deray | divya, o/ | 09:29 |
| vmud213 | mgould: | 09:29 |
| ifarkas | dtantsur, looking | 09:29 |
| mgould | vmud213, oh I see :-( | 09:31 |
| mgould | you can test on CentOS without a license, FWIW | 09:31 |
| mgould | [or rather, without having to pay for a license] | 09:32 |
| mgould | deray, hah, nice example :-) | 09:33 |
| deray | mgould, :-) | 09:33 |
| vmud213 | mgould:Which DIB element should i use? centos, centos7 or centos-minimal? | 09:34 |
| *** daemontool has joined #openstack-ironic | 09:35 | |
| vmud213 | mgould:Hope anything works | 09:35 |
| *** mbound has quit IRC | 09:35 | |
| mgould | vmud213, I'm not sure, but try centos7 | 09:35 |
| mgould | are Ubuntu/RHEL/CentOS/Fedora the only distros we support? | 09:35 |
| *** yolanda has quit IRC | 09:36 | |
| mgould | if the problem is "Fedora is using a bleeding-edge GRUB", then I'd expect to see problems with other bleeding-edge distros like Arch or Gentoo | 09:36 |
| vmud213 | mgould:tbh, no idea about other distros :( | 09:37 |
| vmud213 | But can confirm the issue with latest Fedora... Ubuntu doesn't have this issue | 09:37 |
| dtantsur | gentoo is pretty much supported in openstack since recently | 09:38 |
| mgould | vmud213, is Ubuntu using GRUB 2? | 09:40 |
| vmud213 | mgould:yes, it is | 09:40 |
| openstackgerrit | Merged openstack/ironic: File download fails with swift pseudo folder https://review.openstack.org/306929 | 09:41 |
| vmud213 | mgould: wait? Don't take my words.It's not on top of my head right know. | 09:41 |
| *** yolanda has joined #openstack-ironic | 09:41 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-inspector: Use PortOpt type for port options https://review.openstack.org/312408 | 09:42 |
| vmud213 | dtantsur: https://review.openstack.org/#/c/303503/ is pending for a long time :(. Would you mind taking some time to review? | 09:43 |
| dtantsur | vmud213, it's not a big surprise, everyone was on the summit and just starting to catch up.. could you resolve the pending -1 there? | 09:44 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-inspector: Use PortOpt type for port options https://review.openstack.org/312408 | 09:45 |
| mkovacik | sambetts, hi, could you please paste the pull request to Neutron to support creating ports without MAC addresses? I can't see it in https://etherpad.openstack.org/p/newton-baremetal-networking | 09:45 |
| sambetts | mkovacik: Do you want a link to the spec added? | 09:46 |
| sambetts | because the spec for that isn't merged and the code isn't written for that part yet | 09:47 |
| mkovacik | sambetts, IRC here will do, thanks :) | 09:47 |
| * mkovacik needs to mention in the HA inspector spec as an alternative to our own firewall | 09:47 | |
| sambetts | mkovacik: https://review.openstack.org/#/q/status:open+project:openstack/neutron-specs+branch:master+topic:bug/1512666 | 09:48 |
| sambetts | those are the 2 specs related to making it work ^ | 09:48 |
| mkovacik | sambetts, thx! | 09:48 |
| sambetts | we of course need neutron guys to review our sanity on the serve unknown host one ;) | 09:49 |
| mgould | vmud213, reading that fedoraproject page, it looks like it probably is just a problem with fedora >=18 | 09:49 |
| mkovacik | sambetts, yeah :) sanity | 09:49 |
| openstackgerrit | Merged openstack/ironic: Fix API node name updates https://review.openstack.org/300983 | 09:50 |
| openstackgerrit | Merged openstack/ironic: Updating dev-quickstart.rst file links https://review.openstack.org/308821 | 09:50 |
| vmud213 | mgould:Looking at it | 09:51 |
| *** mbound has joined #openstack-ironic | 09:51 | |
| mgould | vmud213, dtantsur: which distros do we support in IPA? {Ubuntu, RHEL, CentOS, recent Fedora}, anything else? | 09:52 |
| mgould | because if that's all, and it works with CentOS, then checking for Fedora is fine | 09:52 |
| * mgould wonders how hard it would be to write a MirageOS unikernel that does IPA's job | 09:53 | |
| dtantsur | mgould, did someone confirm centos? we also have tinycore | 09:53 |
| mgould | less hard than getting OCaml code merged into OpenStack, I'm guessing :-) | 09:53 |
| mgould | dtantsur, I think vmud213 is checking centos now | 09:53 |
| dtantsur | got it | 09:53 |
| * sambetts worrys if this is something we need to worry about in the future, if it to do with a newer grub2 | 09:54 | |
| * vmud213 building the image | 09:54 | |
| sambetts | can we confirm grub2 versions? | 09:54 |
| mgould | is it Ubuntu 14.04 we're using, or something newer? | 09:55 |
| vmud213 | sambetts,mgould,dtantsur: correct me if i am wrong. it's actually grub-mkinstall that is breaking.. | 09:55 |
| *** athomas has quit IRC | 09:55 | |
| *** vdrok_ is now known as vdrok | 09:55 | |
| mgould | vmud213, grub2-install according to the Fedora wiki | 09:55 |
| mgould | oh, MirageOS only runs on Xen | 09:56 |
| sambetts | yeah: grub2-install shouldn't be used on EFI systems. The grub2-efi package installs a prebaked grubx64.efi on the EFI System partition, which looks for grub.cfg on the ESP in /EFI/fedora/ whereas the grub2-install command creates a custom grubx64.efi, deletes the original installed one, and looks for grub.cfg in /boot/grub2/. | 09:56 |
| * mgould 's beautiful dream collapses | 09:56 | |
| sambetts | :/ seems like a bug to me | 09:56 |
| mgould | yep | 09:57 |
| vmud213 | mgould: i agree, it's grub2-install | 09:57 |
| mgould | but is it a bug in fedora or in grub2-install? | 09:57 |
| *** fragatina has joined #openstack-ironic | 09:59 | |
| sambetts | Sounds like it might/could have been a configure misconfiguration when the person who packaged it built the source | 09:59 |
| openstackgerrit | vinay kumar muddu proposed openstack/ironic-python-agent: Wait for all interfaces to be up before node lookup https://review.openstack.org/303503 | 09:59 |
| *** dmk0202 has quit IRC | 09:59 | |
| sambetts | e.g. specfied the wrong directory in a option or something | 09:59 |
| vmud213 | sambetts: apparently it is | 10:01 |
| vdrok | morning ironic! | 10:01 |
| vmud213 | i also tried to move the existing grub to different directories, and the config file as well. but no luck. | 10:01 |
| vmud213 | morning vdrok | 10:02 |
| vdrok | morning vmud213 | 10:02 |
| *** jrist has quit IRC | 10:02 | |
| *** athomas has joined #openstack-ironic | 10:02 | |
| mgould | vmud213, I've withdrawn my -1, but please check whether the bug affects CentOS | 10:03 |
| mgould | vdrok, morning! | 10:03 |
| vdrok | good morning mgould | 10:03 |
| mgould | vmud213, if it's common to fedora and centos, then just check whether /etc/redhat-release exists | 10:04 |
| vmud213 | mgould: sure..will check it | 10:04 |
| vmud213 | mgould: got it. | 10:05 |
| *** e0ne has quit IRC | 10:06 | |
| mgould | vmud213, awesome | 10:06 |
| *** e0ne has joined #openstack-ironic | 10:06 | |
| mgould | sambetts, did you still want to pair on grenade testing? | 10:07 |
| sambetts | mgould: Sure, I've just been trying out running things like tox -e smoke etc, confused as hell on how to get actual granade stuff working so that would be cool | 10:09 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector-specs: High Availability for Ironic Inspector https://review.openstack.org/253675 | 10:09 |
| mgould | sambetts, cool, let me grab a coffee | 10:11 |
| dtantsur | on an unrelated note: I guess the US-only nature of the new format design summits is already a confirmed thing? | 10:13 |
| sambetts | I kinda bummed out I missed that talk at the summit, is there somewhere I can catch up? was it a recorded one? | 10:13 |
| dtantsur | yep, can't find a link, but it was like 'oh, lets to the first 2 in US, as it's easier' | 10:14 |
| sambetts | -_- | 10:14 |
| dtantsur | and as we all know, there's nothing more permanent than temporary things, so... | 10:14 |
| sambetts | dtantsur: oh it doesn't look like it was a recorded one because it was a cross project workshop... | 10:16 |
| openstackgerrit | Merged openstack/ironic: Migrate api-ref into our tree https://review.openstack.org/312181 | 10:17 |
| dtantsur | mkovacik, is the HA spec ready for final reviews now? | 10:26 |
| *** yolanda has quit IRC | 10:33 | |
| *** lennyb has joined #openstack-ironic | 10:34 | |
| vdrok | sambetts: seems like that's it - https://etherpad.openstack.org/p/newton-design-summit-format | 10:35 |
| *** hoangcx has quit IRC | 10:37 | |
| *** yolanda has joined #openstack-ironic | 10:38 | |
| *** MattMan has quit IRC | 10:39 | |
| *** MattMan has joined #openstack-ironic | 10:39 | |
| dtantsur | aha | 10:39 |
| dtantsur | hopefully they *do* start rotating it in 2018 | 10:40 |
| openstackgerrit | Merged openstack/ironic-inspector: Refactor test_process https://review.openstack.org/304185 | 10:42 |
| *** links has quit IRC | 10:45 | |
| *** jrist has joined #openstack-ironic | 10:45 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: API: Check for reserved words when naming a node https://review.openstack.org/308965 | 10:49 |
| lucasagomes | dtantsur, sambetts AFAIK it seems like a confirmed things, devananda can probably clarify it better | 10:49 |
| lucasagomes | and I believe it's the 3 firsts in US | 10:50 |
| lucasagomes | the first 3* | 10:50 |
| dtantsur | meh... | 10:52 |
| *** stendulker has quit IRC | 10:52 | |
| lucasagomes | dtantsur, sambetts thierry's blog has some posts about it | 10:54 |
| lucasagomes | http://ttx.re/ | 10:54 |
| * lucasagomes goes grab some food | 10:56 | |
| *** lucasagomes is now known as lucas-hungry | 10:56 | |
| *** baoli has joined #openstack-ironic | 10:57 | |
| *** baoli has quit IRC | 10:57 | |
| *** baoli has joined #openstack-ironic | 10:57 | |
| *** david-lyle has joined #openstack-ironic | 10:58 | |
| jroll | sup y'all | 10:58 |
| dtantsur | hey jroll | 10:58 |
| jroll | mgould: I totally thought it was "nerve-wracking" | 10:58 |
| *** baoli has quit IRC | 10:58 | |
| *** oberezovskyi has quit IRC | 10:59 | |
| *** oberezovskyi has joined #openstack-ironic | 11:00 | |
| *** oberezovskyi has joined #openstack-ironic | 11:02 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Store ramdisk logs on all processing failures, not only reported by the ramdisk https://review.openstack.org/300011 | 11:05 |
| *** deray has quit IRC | 11:07 | |
| mgould | morning jroll | 11:08 |
| TheJulia | good morning everyone | 11:08 |
| mgould | morning TheJulia | 11:08 |
| *** sacharya has joined #openstack-ironic | 11:08 | |
| * TheJulia tries to wake up with delicious coffee | 11:09 | |
| *** sacharya has quit IRC | 11:13 | |
| *** wajdi has joined #openstack-ironic | 11:17 | |
| dtantsur | TheJulia, morning | 11:18 |
| jroll | http://magic.io/blog/uvloop-make-python-networking-great-again/ | 11:19 |
| *** jjohnson2 has joined #openstack-ironic | 11:19 | |
| TheJulia | interesting | 11:20 |
| dtantsur | cool | 11:20 |
| *** thrash|g0ne is now known as thrash | 11:22 | |
| TheJulia | I would think http request processing and handling would be the real bottle neck unless it is kept wafter thin, but still awesome | 11:23 |
| jroll | TheJulia: yeah, they actually get into that at some point | 11:24 |
| TheJulia | yeah, 2nd to last paragraph :) | 11:25 |
| *** daemontool has quit IRC | 11:27 | |
| *** jjohnson2 has quit IRC | 11:29 | |
| *** wajdi_ has joined #openstack-ironic | 11:29 | |
| *** wajdi has quit IRC | 11:32 | |
| *** wajdi has joined #openstack-ironic | 11:33 | |
| *** wajdi_ has quit IRC | 11:33 | |
| *** wajdi_ has joined #openstack-ironic | 11:34 | |
| sambetts | lucas-hungry, dtantsur :( | 11:36 |
| sambetts | Morning jroll TheJulia | 11:36 |
| *** wajdi has quit IRC | 11:37 | |
| jroll | sambetts: morning, is your CI okay? https://review.openstack.org/#/c/312181/ | 11:37 |
| sambetts | jroll: I hope so now, it had a tantrum monday and yesterday | 11:38 |
| *** wajdi_ has quit IRC | 11:39 | |
| jroll | heh | 11:39 |
| *** mbound has quit IRC | 11:42 | |
| jroll | mgould: did you want to volunteer to be a point of contact for the upgrade testing (grenade) work? | 11:44 |
| *** irf has joined #openstack-ironic | 11:46 | |
| irf | morning Ironic !!! | 11:46 |
| dtantsur | morning irf | 11:47 |
| sambetts | Hey jroll, myself and mgould are on a hangout right now trying to run tempest locally in a VM, and discovering all the inconsistancys between our developer devstack docs and whats setup by devstack-gate | 11:48 |
| dtantsur | here come the volunteers ^^^ :D | 11:48 |
| * dtantsur is also going to do that pretty soon | 11:49 | |
| jroll | yay! welcome irf | 11:49 |
| jroll | sambetts: I'd love to see that diff :) | 11:49 |
| irf | i was configuring nova service to integrate with Ironic for bare metal provisioning ... after configuration , when i restarted the nova compute service error , i get the error "ImportError: No module named ironic.nova.compute.manager" in the log.. after analysis, i found that openstack-ironic-common package needs to be installed on the compute node... | 11:49 |
| mgould | jroll, not entirely sure what I'd be volunteering for | 11:50 |
| mgould | the person who knows what the state of the work is and what the problems are? | 11:50 |
| irf | so not getting how can i install openstack-ironic-common package ?? | 11:50 |
| jroll | mgould: help track/push that work, idk, dtantsur mentioned you | 11:50 |
| jroll | irf: so I've actually been considering removing that config from our docs, because it isn't really a supported thing | 11:51 |
| mgould | jroll, yep, I can do that | 11:51 |
| jroll | irf: but, can you not just install it if you do want to use that? | 11:51 |
| dtantsur | mgould, I just remembered you were working with jlvillal on grenade, hence I've mentioned you | 11:51 |
| mgould | dtantsur, I was working on it a month or so ago, but other things got in the way | 11:52 |
| mgould | but yes! I would like to get back to it | 11:52 |
| jroll | cool, ty mgould | 11:52 |
| irf | i am using ubuntu box and i just fired the command : apt-get install openstack-ironic-common | 11:53 |
| irf | thats does not solve the issue | 11:53 |
| irf | do i need to import any repo so that apt-get can understand ? | 11:53 |
| dtantsur | hmm, I don't think ubuntu packages has "openstack-" prefix.. or do they? | 11:54 |
| jroll | irf: I'd recommend just removing the compute_manager config, and let it be default | 11:54 |
| * jroll proposes a docs change | 11:54 | |
| *** lucas-hungry is now known as lucasagomes | 11:55 | |
| lucasagomes | jroll, yo hey there | 11:55 |
| lucasagomes | sambetts, what is the ":(" about? design summit split? | 11:56 |
| dtantsur | sambetts, where are you putting your notes? nothing in https://etherpad.openstack.org/p/ironic-grenade-tests | 11:56 |
| *** wajdi has joined #openstack-ironic | 11:56 | |
| jroll | lucasagomes: mornin' | 11:57 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Remove ClusteredComputeManager docs https://review.openstack.org/312466 | 11:57 |
| jroll | dtantsur: https://etherpad.openstack.org/p/ironic-newton-summit-grenade-worksession is the main pad | 11:57 |
| dtantsur | thnx | 11:58 |
| *** links has joined #openstack-ironic | 11:59 | |
| irf | ok let me | 12:00 |
| dtantsur | jroll, should we also deprecate the code itself? | 12:03 |
| jroll | dtantsur: it's still useful until we have a better thing | 12:04 |
| *** e0ne has quit IRC | 12:05 | |
| lucasagomes | ilo folks, the inband RAID configuration is already merged, right? | 12:05 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Add newton priorities doc https://review.openstack.org/311530 | 12:05 |
| jroll | ^ I think that covers everything so far | 12:06 |
| dtantsur | lucasagomes, I think it is, I'm aware of one DIB patch though | 12:06 |
| irf | ok .. that issue is resolved .. but when now i am creating instances from horizon dashboard .. i am getting error like : No valid host was found. There are not enough hosts available. | 12:08 |
| lucasagomes | dtantsur, perfect, thanks | 12:09 |
| jroll | irf: http://docs.openstack.org/developer/ironic/deploy/troubleshooting.html#nova-returns-no-valid-host-was-found-error | 12:10 |
| *** jaypipes has joined #openstack-ironic | 12:11 | |
| *** raildo-afk is now known as raildo | 12:13 | |
| *** davidlenwell has quit IRC | 12:13 | |
| *** davidlenwell has joined #openstack-ironic | 12:15 | |
| *** mbound has joined #openstack-ironic | 12:16 | |
| *** dprince has joined #openstack-ironic | 12:17 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Support Ironic node names in our API https://review.openstack.org/276331 | 12:18 |
| irf | before working on this error , i would like to mentions that , the driver i am using is : pxe_wol | 12:21 |
| irf | i assume nova does not have any issue with this driver .. i mean no any compatibility issues | 12:22 |
| *** e0ne has joined #openstack-ironic | 12:22 | |
| jroll | irf: that error is kind of a catch all error in nova, so it's hard to say the problem | 12:22 |
| jroll | but no, nova works fine with all drivers afaik | 12:22 |
| irf | ok | 12:22 |
| irf | fine let me proceed and solve this issue first ... | 12:23 |
| * TheJulia sifts through grenade logs and ponders | 12:23 | |
| *** oberezovskyi has quit IRC | 12:24 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: is-empty conditions should accept missing values https://review.openstack.org/310031 | 12:27 |
| dtantsur | spotted this when working on my slides ^^^ :) | 12:27 |
| *** wajdi has quit IRC | 12:27 | |
| *** keedya_ has joined #openstack-ironic | 12:31 | |
| *** jjohnson2 has joined #openstack-ironic | 12:33 | |
| *** keedya_ has quit IRC | 12:35 | |
| *** baoli has joined #openstack-ironic | 12:38 | |
| *** hshiina has joined #openstack-ironic | 12:40 | |
| *** baoli_ has joined #openstack-ironic | 12:40 | |
| *** baoli has quit IRC | 12:43 | |
| *** trown|outtypewww is now known as trown | 12:46 | |
| *** links has quit IRC | 12:46 | |
| *** rloo has joined #openstack-ironic | 12:47 | |
| vdrok | so in theory, does the grenade runs with smoke regex? | 12:50 |
| vdrok | or is it smth like smoke(?!slow) | 12:51 |
| lucasagomes | jroll, since there's no votes yet on the priorities do you mind if I submit a new patch set adding my name to the boot from volume? This is something I want to work on because RH is interested on it | 12:51 |
| TheJulia | vdrok: on the VM I'm working in, the tests time out since the initial deploy times out | 12:51 |
| TheJulia | vdrok: 600 seconds, and basically CoreOS is only getting an IP address at minute 8.... mid-way through unpacking IPA when the timeout hits | 12:51 |
| *** divya has quit IRC | 12:52 | |
| vdrok | TheJulia: hm, that's strange. I think I'll start with trying to run smoke serially | 12:52 |
| vdrok | btw, we might want this - https://review.openstack.org/#/c/311225/1 | 12:53 |
| TheJulia | vdrok: not really.. that strange, I'm retrying with ipxe | 12:53 |
| *** caiobo has joined #openstack-ironic | 12:53 | |
| jroll | lucasagomes: yeah, go for it | 12:53 |
| lucasagomes | cool, ty | 12:53 |
| caiobo | good morning Ironic :) | 12:53 |
| vdrok | also, should we do 3 vms on tinyipa instead of 1? | 12:53 |
| TheJulia | I suspect we don't have the ram to do 3 | 12:54 |
| vdrok | morning caiobo | 12:54 |
| TheJulia | at least with coreos | 12:54 |
| jroll | vdrok: I got smoke serial basically passing, see the bottom here https://etherpad.openstack.org/p/ironic-newton-summit-grenade-worksession | 12:54 |
| vdrok | TheJulia: tinyipa :) | 12:54 |
| TheJulia | yup | 12:54 |
| jroll | tinyipa at 384MB can get us 7 VMs | 12:54 |
| TheJulia | Thats not bad, that should actually work quite nicely | 12:54 |
| vdrok | jroll: IRONIC_VM_COUNT=64 # because I can :) | 12:55 |
| jroll | yep | 12:55 |
| jroll | hehe | 12:55 |
| TheJulia | only other major thing is cpu concurrency | 12:55 |
| caiobo | morning vdrok jroll TheJulia lucasagomes and everybody else | 12:55 |
| jroll | hi caiobo | 12:55 |
| vdrok | we could change the create_server cleanup to wait for the ironic node to become available | 12:55 |
| sambetts | jroll: I tried running smoke locally and got a bunch of failures that didn't match the ones you were seeing but my terminal ate half my logs :'( | 12:55 |
| vdrok | then we won't need so much vms | 12:56 |
| jroll | sambetts: without concurrency=1 it seems to get racey | 12:56 |
| jroll | the subset size config I have there should help a bit with that | 12:56 |
| *** keedya has joined #openstack-ironic | 12:56 | |
| lucasagomes | caiobo, yo hi there | 12:57 |
| *** irf has quit IRC | 12:58 | |
| lucasagomes | jroll, quick q: for the multiple compute host, the work that will be done in nova is the one (or similar) to the discussion you and deva had with jayp the day before yesterday ? | 13:00 |
| *** Goneri has joined #openstack-ironic | 13:00 | |
| *** ri0 has joined #openstack-ironic | 13:00 | |
| *** vmud213 has quit IRC | 13:01 | |
| rloo | lucasagomes: i just added comments to the priorities spec. if you submit a new patch, my comments will be lost. could you just comment please? | 13:02 |
| lucasagomes | rloo, yeah will do, I'm looking at the spec before submitting a new change :-) | 13:02 |
| rloo | morning everyone, lucasagomes, jroll, vdrok, sambetts, TheJulia, caiobo | 13:03 |
| vdrok | morning rloo :) | 13:03 |
| caiobo | morning rloo :D | 13:03 |
| dtantsur | morning rloo, caiobo | 13:03 |
| rloo | hi dtantsur! | 13:04 |
| caiobo | hello dtantsur | 13:04 |
| TheJulia | good morning rloo | 13:05 |
| sambetts | jroll: I ran with concurrency=1 :( | 13:06 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-specs: Add doc8 check on priorities/ https://review.openstack.org/312492 | 13:08 |
| lucasagomes | rloo, ^ re syntax checks on priorities ^ | 13:08 |
| * lucasagomes just saw ur comment about the ------- line | 13:08 | |
| *** sacharya has joined #openstack-ironic | 13:09 | |
| rloo | thx lucasagomes :) | 13:09 |
| *** krotscheck_dcm is now known as krotscheck | 13:10 | |
| lucasagomes | np | 13:11 |
| dtantsur | sambetts, aarefiev, could you rereview https://review.openstack.org/#/c/300011/ please? got 2x +2, but had to rebase | 13:12 |
| *** ri0 has left #openstack-ironic | 13:13 | |
| *** sacharya has quit IRC | 13:14 | |
| *** ijw has joined #openstack-ironic | 13:15 | |
| *** ifarkas has quit IRC | 13:15 | |
| * lucasagomes *holy cows* receives a pull request that contains 4 years of work in one of his open source projects https://github.com/umago/carbono/pull/2 | 13:16 | |
| dtantsur | lucasagomes, wow, that's a thing :D | 13:17 |
| lucasagomes | yeah, someone has been using it a lot I guess | 13:17 |
| lucasagomes | heh | 13:17 |
| lucasagomes | will take me a while to check that code | 13:17 |
| rloo | lucasagomes: wow :) :( | 13:18 |
| lucasagomes | problem is, many of the commits are in portuguese | 13:19 |
| openstackgerrit | Merged openstack/ironic-specs: Add doc8 check on priorities/ https://review.openstack.org/312492 | 13:20 |
| jroll | lucasagomes: so we still need to evaluate the viability there of the new nova work, but yeah it will probably go in that direction | 13:20 |
| jroll | gotta run an errand, will fix up priorities doc after | 13:22 |
| openstackgerrit | Merged openstack/ironic-inspector: Use PortOpt type for port options https://review.openstack.org/312408 | 13:23 |
| lucasagomes | jroll, right on, yeah well that's a big change in nova I guess | 13:24 |
| lucasagomes | I don't think it's likely to merge in newton, but ofc we can start working/helping on it | 13:24 |
| lucasagomes | jroll, btw, another quick q, I'm about to put a patch to project-config to make the ipmitool jobs voting (as part of the gate effort) | 13:25 |
| lucasagomes | I might need ur ack there soon, will give you the link once I put it up | 13:25 |
| *** electrofelix has quit IRC | 13:28 | |
| keedya | Good Morning All o/ | 13:28 |
| * sambetts and mgould are just trying to benchmark tinyipa vs coreos IPA pxe times | 13:29 | |
| sambetts | jroll: ^ | 13:29 |
| *** ifarkas has joined #openstack-ironic | 13:29 | |
| lucasagomes | sambetts, startup time? network? | 13:30 |
| sambetts | we're working out the time from power on to kernel [0.0000] | 13:30 |
| mgould | keedya, morning | 13:30 |
| rloo | morning keedya | 13:30 |
| * lucasagomes laughs with https://github.com/umago/carbono/pull/2/commits/740608464110c7041eb19ad909fb893e4a6056c1 (portuguese only) | 13:30 | |
| sambetts | so as to try to narrow it down to just the kernel time | 13:30 |
| sambetts | pxe boot time * | 13:30 |
| *** electrofelix has joined #openstack-ironic | 13:31 | |
| mgould | lucasagomes, 143 commits, wow | 13:31 |
| lucasagomes | mgould, yeah someone just made a pull request with 4 years of work improving the code in a open source project I've started | 13:32 |
| lucasagomes | mgould, apparently they have been using it a lot and since it's GPL licensed they now had to publish it or something | 13:32 |
| lucasagomes | they have really cool stuff in there | 13:33 |
| * lucasagomes is skimming though the commits | 13:33 | |
| mgould | hurrah for open source | 13:33 |
| vdrok | lucasagomes: google translate only says something about the cake :) | 13:34 |
| lucasagomes | yeah things that worked poorly such as dealing with GPT partition tables seems to be much better now | 13:34 |
| lucasagomes | vdrok, exactly the comment was somthing like "ask him if we should put it under or over this line. We are going bring cake" (doesn't makes much sense even in portuguese) | 13:35 |
| vdrok | :) | 13:35 |
| lucasagomes | vdrok, and the commit message was "cake is good. but not in the code" | 13:35 |
| lucasagomes | heh | 13:35 |
| lucasagomes | removing that line | 13:35 |
| keedya | I am running tempest on my devstack setup, I keep failing test_baremetal_server_ops in BaremetalBasicOps with error "timeout waiting to get ironic node by instance" | 13:37 |
| *** rbudden has joined #openstack-ironic | 13:37 | |
| keedya | any idea what I m missing here? | 13:37 |
| TheJulia | keedya: so provisioning ironic nodes takes time... depending on hardware/ram it might timeout in any number of places | 13:40 |
| TheJulia | keedya: I would start with the baremetal vm logs, and go from there | 13:40 |
| jlvillal | lucasagomes, Not sure if you saw I added a shell script that did what you had written up on the etherpad | 13:40 |
| keedya | thanks TheJulia | 13:41 |
| lucasagomes | jlvillal, oh I haven't no, where's it? | 13:42 |
| jlvillal | In the devstack-gate-test project | 13:43 |
| jlvillal | lucasagomes, vm-setup.sh | 13:43 |
| lucasagomes | jlvillal, cool that will facilitate yes | 13:43 |
| jlvillal | lucasagomes, The error I am seeing is this: http://paste.openstack.org/show/495993/ | 13:44 |
| jlvillal | Well it shows where I am failing. Not the underlying error. | 13:44 |
| jlvillal | Is that the same as you? | 13:44 |
| lucasagomes | yesterday I was actually able to run it through but it failed on the smoking tests later on | 13:45 |
| lucasagomes | lemme try again | 13:45 |
| openstackgerrit | Merged openstack/ironic: Remove ClusteredComputeManager docs https://review.openstack.org/312466 | 13:46 |
| lucasagomes | jlvillal, it seems to be somehting in the grenade.sh script itself right? Is it a fresh env? | 13:46 |
| *** d0ugal has quit IRC | 13:46 | |
| *** d0ugal has joined #openstack-ironic | 13:47 | |
| jlvillal | lucasagomes, My test was a fresh environment. Yes grenade | 13:47 |
| jlvillal | lucasagomes, I wanted to make sure you got to the resource phase create step | 13:47 |
| lucasagomes | AFAIR yes, I iwll start fresh again | 13:49 |
| lucasagomes | let you know in some ~hour | 13:49 |
| lucasagomes | in a ~hour* | 13:49 |
| jlvillal | lucasagomes, Thanks! | 13:50 |
| *** electrofelix has quit IRC | 13:52 | |
| *** electrofelix has joined #openstack-ironic | 13:52 | |
| *** mbound has quit IRC | 13:54 | |
| *** ametts has joined #openstack-ironic | 13:55 | |
| *** mbound has joined #openstack-ironic | 13:55 | |
| jroll | lucasagomes: yeah cool, let me know when you get the project-config thing up | 13:56 |
| mgould | keedya, sambetts and I found that the timeout for that is higher in the gate | 13:56 |
| mgould | set BUILD_TIMEOUT=600 in local.conf and try again | 13:56 |
| *** electrofelix has quit IRC | 13:57 | |
| * mgould -> fresh air... | 13:57 | |
| *** electrofelix has joined #openstack-ironic | 13:57 | |
| mkovacik | sambetts, aarefiev, dtantsur, mgould and all fellow ironicers interested, the Inspector HA spec received the summit updates and is ready for your review! Thanks a lot! https://review.openstack.org/#/c/253675/ | 13:57 |
| * jlvillal makes note to self to not upgrade his home network hardware at 9:45pm when he usually goes to bed at 10:00pm... | 13:58 | |
| rloo | jlvillal and anyone else with an opinion about files for functional tests: https://github.com/openstack/python-ironicclient/tree/master/ironicclient/tests/functional | 13:58 |
| keedya | mgould: it is failing node validation on my end; because one or more parameters are missing from its instance_info.. Missing are: ['ramdisk', 'kernel', 'image_source'] | 13:58 |
| rloo | jlvillal: what do you think about separating node testing into different files? | 13:59 |
| jlvillal | rloo, I think vdrok might have an opinion or know someone who does. | 13:59 |
| jlvillal | looking... | 13:59 |
| vdrok | morning jlvillal | 13:59 |
| rloo | jlvillal: ok. i was just looking at a patch and haven't been following functional tests. vdrok ^^? | 14:00 |
| rloo | jlvillal: and I see that another test_node_delete.py file is being proposed. | 14:00 |
| rloo | jlvillal: so wanted to check if we want an explosion of files or not. | 14:00 |
| vdrok | rloo: that would make sense, right now there are not many of them, but more to come | 14:00 |
| jlvillal | rloo, I don't have a strong opinion one way or another on different files. I could see a benefit to a split. | 14:01 |
| vdrok | rloo: but we need some clear separation | 14:01 |
| rloo | jlvillal, vdrok: for unit tests, we tend to have a test file corresponding to a code file. | 14:01 |
| rloo | vdrok: right. what do you suggest? | 14:02 |
| vdrok | currently, in test_node there are create/update tests, and maintenance too | 14:02 |
| jlvillal | rloo, Right, but functional testing doesn't really fall into that category. | 14:02 |
| vdrok | I should think about it :) | 14:02 |
| *** jtomasek has quit IRC | 14:02 | |
| jlvillal | As far as matching up a test file to a code file in a 1:1 match | 14:02 |
| rloo | jlvillal: yes, functional testing is different. but we should be consistent with whatever we decide :) | 14:02 |
| rloo | jlvillal: i was wondering if there had been any decision. will see what vdrok thinks :) | 14:03 |
| jlvillal | Sounds good. | 14:03 |
| sambetts | keedya: thats expected, because nova sets those parameters | 14:03 |
| vdrok | rloo: maybe split them by controller method? like states controller, boot device controller, console controller, and basic operations? | 14:04 |
| *** jtomasek has joined #openstack-ironic | 14:04 | |
| vdrok | maintenance and vendor passthru controllers are there too | 14:05 |
| vdrok | I mean these https://github.com/openstack/ironic/blob/master/ironic/api/controllers/v1/node.py | 14:05 |
| keedya | sambetts: So how can I get a clean run? Do I need to skip this test? | 14:06 |
| rloo | vdrok: hmmm. the tests are in the client. should the client know about the controllers? | 14:06 |
| sambetts | there is a test for if the node validates? | 14:07 |
| rloo | vdrok: I suppose it isn't a big deal. maybe things will become clearer as more tests are added. we can always rearrange them later i guess. | 14:07 |
| *** mbound has quit IRC | 14:08 | |
| mgould | rloo, I think trying to come up with the One True File Layout at this point is premature | 14:08 |
| vdrok | rloo: but still, the commands are tied to specific controller underneath - node-get-boot-device, node-get-supported-boot-devices, node-set-boot-device. yep, can be decided afterwards | 14:08 |
| rloo | mgould: i wasn't hoping for the one true file layout. just hoping for someone to give me a reason for what is there/where it is going :) | 14:08 |
| rloo | mgould: otherwise, we could get a real mix of files/tests if more than one person is adding functional tests. | 14:09 |
| mgould | rloo, good point | 14:10 |
| dtantsur | hmm, I can't count how many times I've asked that, but does pyghmi has a bug tracker? | 14:10 |
| dtantsur | jroll, ^^^ | 14:10 |
| mgould | my instinct is "start with one file per entity type, then split once they become unwieldy" | 14:10 |
| jroll | dtantsur: uhhhhhhhhhh | 14:10 |
| dtantsur | I see no point in folks reporting bugs in our bugtracker, where they get buried forever, | 14:10 |
| rloo | vdrok: yes. i think maybe the osc commands might be more useful. cuz it is noun-verb. | 14:10 |
| dtantsur | jroll, or maybe you have contacts of the maintainer? | 14:11 |
| rloo | mgould: yes, that was my instinct too. hence my question when i saw there was more than one file per entity :) | 14:11 |
| jroll | dtantsur: jjohnson2 is the maintainer | 14:12 |
| jroll | his email should be in every commit | 14:12 |
| dtantsur | lets try irc first | 14:12 |
| jroll | dtantsur: but, don't think there's a separate tracker today | 14:12 |
| dtantsur | jjohnson2, hey, do you folks have an bug tracker for pyghmi? people tend to report bugs against ironic right now, which does not feel right. the team can't act/triage them | 14:12 |
| dtantsur | e.g. https://bugs.launchpad.net/ironic/+bug/1576811 | 14:13 |
| openstack | Launchpad bug 1576811 in Ironic "pyghmi 1.0.0 regression on OVB" [Undecided,New] | 14:13 |
| rloo | vdrok: guess i was wrong. with osc, it is baremetal node set boot-device, baremetal node get boot-device. | 14:13 |
| jjohnson2 | dtantsur, good day | 14:13 |
| *** dmk0202 has joined #openstack-ironic | 14:14 | |
| jjohnson2 | interesting | 14:14 |
| jjohnson2 | to be honest, I had assumed erroneously that when it was being set up for me a bug tracker would be part of the deal, and that the code was just without any bug | 14:15 |
| mgould | rloo, maybe the node tests got unwieldy really quickly? :-) | 14:15 |
| vdrok | rloo: I just can't come up with something other than per-controller :) everywhere in code these are single files, in our api, in client node.py and node_shell.py | 14:15 |
| jjohnson2 | how many pyghmi bugs have appeared under ironic? | 14:16 |
| mgould | test_entity_action.py seems sensible, or perhaps entity/test_action.py | 14:16 |
| dtantsur | jjohnson2, if you mean openstack infra, they only setup git. we're on our own with launchpad/whatever | 14:16 |
| rloo | mgould: maybe. I have no idea. I haven't been paying attention to the functional tests. I could just not look at the patch :) | 14:16 |
| * mgould neither, I'm afraid | 14:16 | |
| *** absubram has joined #openstack-ironic | 14:16 | |
| rloo | vdrok: that's fine for now i guess. we (or mgould!) can think about it for later :) | 14:16 |
| jroll | jjohnson2: dtantsur: I'm happy to set up a pyghmi launchpad if needed, I feel like we need one to deal with releases anyway | 14:17 |
| jjohnson2 | jroll, ok, that's cool | 14:17 |
| jroll | I think everything in our tree has to go release:managed this cycle | 14:17 |
| jjohnson2 | any way I get emailed, I'd be happy | 14:17 |
| dtantsur | jjohnson2, I'm seeing also https://bugs.launchpad.net/ironic/+bug/1352079 and https://bugs.launchpad.net/ironic/+bug/1509098 | 14:17 |
| openstack | Launchpad bug 1352079 in Ironic "pyghmi: AttributeError: 'SDREntry' object has no attribute 'tolerance'" [Medium,Fix released] | 14:17 |
| openstack | Launchpad bug 1509098 in Ironic "fakebmc broken in pyghmi" [Undecided,Invalid] | 14:17 |
| dtantsur | jroll++ lets do it | 14:17 |
| jjohnson2 | I'm kind of confused on the 1.0.0 regression output, is that really all that comes out? | 14:18 |
| jjohnson2 | I don't see it actually print out the exception... or am I missing something? | 14:18 |
| dtantsur | jjohnson2, I don't quite know, sorry. you can ask on the bug | 14:18 |
| jjohnson2 | ok, just didn't know if that was usual for debug output | 14:18 |
| * mgould -> relocate | 14:19 | |
| *** mgould has quit IRC | 14:19 | |
| *** daemontool has joined #openstack-ironic | 14:19 | |
| *** joprovost has joined #openstack-ironic | 14:21 | |
| *** bnemec has joined #openstack-ironic | 14:21 | |
| jjohnson2 | well, I have an idea, let me recreate the issue and I'll comment | 14:21 |
| lucasagomes | jjohnson2, dtantsur jroll pyghmi 1.0.1 broke the BMCs yeah, I've just tested it locally with VirtualBMC http://paste.openstack.org/show/496082/ | 14:22 |
| lucasagomes | I think mrda have seem it too | 14:22 |
| * lucasagomes should pin vbmc in pyghmi 0.9.8* | 14:22 | |
| *** mgould has joined #openstack-ironic | 14:23 | |
| jroll | fun | 14:23 |
| mat128 | lucasagomes, jjohnson2: same with our VirtualBMC equivalent | 14:23 |
| lucasagomes | jjohnson2, btw, can I add some job in the pyghmi gate to run against ironic + vbmc ? | 14:23 |
| jroll | lucasagomes: maybe rather than pinning, we run ipmitool CI against pyghmi | 14:23 |
| jroll | yeah that | 14:23 |
| lucasagomes | yes | 14:24 |
| mat128 | lucasagomes: please please :D | 14:24 |
| lucasagomes | I will need to change devstack to install pyghmi from source as well | 14:24 |
| lucasagomes | mat128, jroll will do it | 14:24 |
| jroll | should be just a libs_from_git thing but yeah | 14:24 |
| lucasagomes | yup | 14:24 |
| jroll | O_o | 14:24 |
| jjohnson2 | ok, that would be great to get going | 14:24 |
| jroll | what am I doing? | 14:24 |
| jroll | oh, misread, ignore me | 14:24 |
| jjohnson2 | jroll, make everything work perfect is all | 14:24 |
| jroll | on it! | 14:24 |
| dtantsur | jroll, do you think we need a spec for https://bugs.launchpad.net/ironic/+bug/1572511 ? the fix looks simple | 14:26 |
| openstack | Launchpad bug 1572511 in Ironic "[RFE] RAID interface to support JBOD volumes" [Wishlist,In progress] - Assigned to Imre Farkas (ifarkas) | 14:26 |
| jjohnson2 | no this won't be a fix, it's me backing up my WIP before fixing the bug | 14:26 |
| jjohnson2 | what's about to be sent in the channel | 14:26 |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add Hardware inventory from Lenovo Agentless https://review.openstack.org/312568 | 14:26 |
| jjohnson2 | yeah, that | 14:26 |
| jroll | dtantsur: nah, go ahead and approve it | 14:27 |
| dtantsur | done | 14:28 |
| mgould | sambetts, I'm now fed and recaffeinated, and ready to start pairing again whenever's convenient for you | 14:28 |
| xavierr | morning good Ironic and may the force be with you | 14:28 |
| dtantsur | :) | 14:28 |
| mgould | jlvillal, this morning jroll convinced me to volunteer as point of contact for the ironic-grenade work | 14:28 |
| mgould | xavierr, and also with you :-) | 14:28 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: RAID interface to support JBOD volumes https://review.openstack.org/308306 | 14:28 |
| *** mtanino has joined #openstack-ironic | 14:28 | |
| *** wajdi has joined #openstack-ironic | 14:28 | |
| vdrok | morning xavierr | 14:28 |
| sambetts | mgould: I have a meeting for 30 mins, but after that I should be free again :) | 14:29 |
| *** mbound has joined #openstack-ironic | 14:29 | |
| mgould | 1) I hope this isn't stepping on your toes, 2) what have I missed? :-) | 14:29 |
| mgould | sambetts, cool | 14:29 |
| caiobo | morning xavierr | 14:31 |
| dtantsur | jroll, lemme know when pyghmi bug tracker is up, I'll move a couple of bugs there | 14:32 |
| jroll | dtantsur: I should have it by end of week | 14:32 |
| jroll | if you want to give me the bugs I can handle that too | 14:32 |
| dtantsur | I don't have them handy, I basically did lp search with "pyghmi" | 14:33 |
| *** dmk0202 has quit IRC | 14:33 | |
| jroll | okay, cool | 14:33 |
| jroll | will do | 14:33 |
| *** mbound has quit IRC | 14:34 | |
| *** dprince has quit IRC | 14:39 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: DevStack: Support to install pyghmi from source https://review.openstack.org/312575 | 14:41 |
| * lucasagomes tests locally | 14:41 | |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Fix IPMI server functionality https://review.openstack.org/312576 | 14:41 |
| *** xhku has joined #openstack-ironic | 14:42 | |
| *** oberezovskyi has joined #openstack-ironic | 14:42 | |
| jjohnson2 | dtantsur, jroll lucasagomes, mat128 the above commit should fix it | 14:43 |
| jroll | jjohnson2: awesome, thanks | 14:43 |
| lucasagomes | jjohnson2, I will test it soon-ish just running my devstack that install pyghmi from source :-) | 14:43 |
| lucasagomes | I let you knbow | 14:43 |
| lucasagomes | know* | 14:43 |
| jjohnson2 | lucasagomes, would that be testing against the review, or from git master? | 14:44 |
| lucasagomes | jjohnson2, git master | 14:45 |
| lucasagomes | for now, but once in gate it will test against every patch | 14:45 |
| * lucasagomes is testing it locally | 14:45 | |
| lucasagomes | jjohnson2, I need to change the ironic devstack plugin AND project-config so it will test against every patch to pyghmi (once the job is added) | 14:45 |
| jjohnson2 | lucasagomes, ok, let me know anything I need to do to help/watch how it's done | 14:46 |
| lucasagomes | jjohnson2, will do, thanks a lot already for looking into fixing it :-) | 14:46 |
| jjohnson2 | lucasagomes, well, it's my shame to cause a regression... | 14:47 |
| *** irf has joined #openstack-ironic | 14:47 | |
| jjohnson2 | the change is at a low enough level, I can't imagine any consumers not getting fixed.... | 14:48 |
| lucasagomes | happens, at least now we will speed up and get tests at gate level :-D | 14:48 |
| jjohnson2 | well, one day I'll want to take another optimization pass, it takes more CPU to send than to receive at this point, so there's more mucking to be had | 14:49 |
| jjohnson2 | there's no way that makes sense | 14:49 |
| *** causten_ has joined #openstack-ironic | 14:49 | |
| jjohnson2 | also changing the crypto backend to cryptography from PyCrypto, to be consistent with paramiko | 14:49 |
| NobodyCam | good morning Ironic | 14:50 |
| *** e0ne has quit IRC | 14:50 | |
| cinerama | hi! | 14:51 |
| NobodyCam | morning cinerama | 14:51 |
| jjohnson2 | lucasagomes, ok, after thinking and thinking and thinking, I've accepted the change and it will be in master presently | 14:51 |
| jjohnson2 | if people will confirm the git master works, I'll tag a release | 14:52 |
| mgould | morning NobodyCam | 14:52 |
| vdrok | morning NobodyCam and cinerama | 14:52 |
| NobodyCam | morning mgould, vdrok | 14:52 |
| *** ChrisAusten has quit IRC | 14:53 | |
| openstackgerrit | Merged openstack/pyghmi: Fix IPMI server functionality https://review.openstack.org/312576 | 14:53 |
| lucasagomes | jjohnson2, thanks. I will let you know soon as I get my devstack to work with installing pyghmi from master | 14:53 |
| NobodyCam | morning lucasagomes | 14:53 |
| *** mbound has joined #openstack-ironic | 14:53 | |
| *** causten_ has quit IRC | 14:54 | |
| *** irf has quit IRC | 14:54 | |
| jjohnson2 | btw, I've actually submitted a talk at a conference about the pyghmi and other work we're doing | 14:54 |
| lucasagomes | NobodyCam, morning | 14:54 |
| NobodyCam | :) | 14:54 |
| jjohnson2 | finally, a conference that I felt worthy of submitting a talk for | 14:54 |
| xavierr | morning NobodyCam vdrok aarefiev mgould lucasagomes caiobo :) | 14:54 |
| NobodyCam | morning xavierr :) | 14:54 |
| jjohnson2 | it meets my particularly strict requirements | 14:54 |
| NobodyCam | jjohnson2: which one? | 14:55 |
| jjohnson2 | NobodyCam, never heard of it before: "All Things Open" | 14:55 |
| jjohnson2 | the requirement: It's being held within 7 miles of my house | 14:55 |
| NobodyCam | :) oh cool :) | 14:55 |
| NobodyCam | nice :) | 14:55 |
| jjohnson2 | hopefully will see if my talk is accepted soon | 14:56 |
| * TheJulia is still mulling the thought of submitting a talk to All Things Open | 14:56 | |
| *** garthb has joined #openstack-ironic | 14:56 | |
| jjohnson2 | TheJulia, so deadline is Friday IIRC | 14:56 |
| TheJulia | not that I'm likely to be living in North Carolina come that timeframe.... :( | 14:56 |
| TheJulia | jjohnson2: yup | 14:56 |
| NobodyCam | TheJulia: :/ | 14:57 |
| * NobodyCam joins his first call of the day. | 14:57 | |
| mgould | xavierr, morning | 14:58 |
| jjohnson2 | also, at the rate the objections are going, I wouldn't be surprised if the conference moves to another state next year | 14:59 |
| jjohnson2 | so hopefully I get a talk before it moves inconveniently away | 14:59 |
| jjohnson2 | NC legislature, making us an international joke.... | 15:00 |
| *** yolanda has quit IRC | 15:00 | |
| TheJulia | jjohnson2: and putting lives at risk :( | 15:01 |
| *** yolanda has joined #openstack-ironic | 15:03 | |
| jjohnson2 | TheJulia, hmm, that allude to something specific I didn't know about? | 15:03 |
| sambetts | mgould: I'm free to join a call again if you are | 15:03 |
| lucasagomes | xavierr, hi there :-) | 15:04 |
| jjohnson2 | TheJulia, I know it's bad for discrimination (even beyond the blatant LGBT part), bad for business, bad for blocking lawsuits... To hear there's even worse stuff would just sadden me more | 15:05 |
| TheJulia | jjohnson2: The games played in the public eye has spurred lots of baseless hate tl;dr I no longer feel safe in public | 15:06 |
| mgould | TheJulia, that really sucks :-( | 15:06 |
| jjohnson2 | TheJulia, ah, ok, the stirring up the hatred part... not part of the bill, but making the crazy just that much worse | 15:07 |
| *** Nisha has quit IRC | 15:08 | |
| * TheJulia sighs, and goes back to the tempest test that is oddly exploding after only 33 seconds :( | 15:09 | |
| *** sacharya has joined #openstack-ironic | 15:10 | |
| *** sacharya has quit IRC | 15:14 | |
| *** yolanda has quit IRC | 15:15 | |
| lucasagomes | jjohnson2, your change works http://paste.openstack.org/show/496094/ | 15:16 |
| *** yolanda has joined #openstack-ironic | 15:17 | |
| *** hshiina has quit IRC | 15:17 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Add newton priorities doc https://review.openstack.org/311530 | 15:17 |
| jroll | rloo: lucasagomes: ^ | 15:17 |
| *** absubram has quit IRC | 15:17 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: DevStack: Support to install pyghmi from source https://review.openstack.org/312575 | 15:17 |
| lucasagomes | tested locally ^ works | 15:17 |
| lucasagomes | jroll, will take a look | 15:17 |
| jroll | thanks | 15:17 |
| *** absubram has joined #openstack-ironic | 15:18 | |
| jjohnson2 | lucasagomes, still so weird to see virsh and ipmi doing the same thing... | 15:19 |
| *** absubram has quit IRC | 15:19 | |
| lucasagomes | jjohnson2, heh that's how we test ipmitool on gate | 15:19 |
| lucasagomes | since we don't have real baremetal | 15:19 |
| *** ChrisAusten has joined #openstack-ironic | 15:20 | |
| *** Nisha has joined #openstack-ironic | 15:20 | |
| jlvillal | mgould: Awesome :) | 15:20 |
| jjohnson2 | I wonder about the relative uptake of ipmitool v ipminative | 15:21 |
| jlvillal | mgould: Have you tried out the latest devstack-gate-test? | 15:21 |
| lucasagomes | jjohnson2, I believe it would be a good thing to have a job in the pyghmi gate that uses ironic with ipminative | 15:21 |
| jjohnson2 | my dream was to have a library that would auto-sense and branch when vendor-specific needs came along, rather than being told who the vendor was upfront... | 15:22 |
| jjohnson2 | lucasagomes, yeah, that would be a welcome change | 15:22 |
| * lucasagomes adds to his todo | 15:23 | |
| jjohnson2 | lucasagomes, see, that's what you get for suggesting things | 15:23 |
| lucasagomes | let's get the ipmitool one first to execise the pyghmi's BMC | 15:23 |
| *** Teknologeek has joined #openstack-ironic | 15:23 | |
| lucasagomes | then we can work on that | 15:23 |
| Teknologeek | Hello :) | 15:23 |
| lucasagomes | cause that will require another change in devstack ironic (right now it checks for *_ipmitool :-/) | 15:23 |
| jjohnson2 | well, I'm going to minimize again and work on some other stuff for now | 15:24 |
| lucasagomes | jjohnson2, mind releasing pyghmi with that fix you merged ? | 15:24 |
| jjohnson2 | lucasagomes, already done | 15:24 |
| jjohnson2 | 1.0.2 is on pypi | 15:24 |
| lucasagomes | wow that was quick, ack! | 15:25 |
| lucasagomes | thanks a lot | 15:25 |
| jroll | thanks jjohnson2 ! | 15:25 |
| jjohnson2 | release early, release often | 15:25 |
| jroll | +2 | 15:25 |
| lucasagomes | jroll, https://review.openstack.org/#/c/312596/ to make it voting, tho I set workflow -1 until we get the job against the pyghmi's gate | 15:26 |
| jroll | rloo: fwiw, I want to put the "small things" somewhere else | 15:26 |
| Teknologeek | hey guys, i have a little issue with ironic, may i ask a question ? | 15:26 |
| jroll | lucasagomes: cool, thanks | 15:26 |
| jroll | Teknologeek: don't ask to ask, just ask :) | 15:26 |
| Teknologeek | My ironic installation deploys the image on the bare metal node, but when the node reboots, i don't have any ip address | 15:27 |
| * jroll grabs lunch before a meeting | 15:27 | |
| rloo | jroll: ok, although i like the visibility (and 'ack') of putting them in the priorities page. | 15:27 |
| Teknologeek | after some reboot in single mode i found out that my centos official image is looking for an interface named eth0 | 15:27 |
| Teknologeek | but the interface of my server actually is eno49 | 15:28 |
| jroll | rloo: maybe once we have networking done, we can add them there :) | 15:28 |
| Teknologeek | is there a way to specify the interface name in ironic or add biosdevname=1 to my node's kernel parameters before it boots ? | 15:28 |
| rloo | jroll: I don't know if it is 'fair' to change the priorities page mid-stream. Maybe it is. Dunno :) | 15:29 |
| *** yolanda has quit IRC | 15:29 | |
| rloo | jroll: I guess I'd like that page to reflect what we discussed at the summit wrt priorities. | 15:29 |
| jroll | rloo: /shrug, idk, I don't want folks running off and working on those things until we finish some stuff | 15:29 |
| *** Nisha has quit IRC | 15:29 | |
| TheJulia | Teknologeek: are you using a cloud image with cloud-init pre-installed? | 15:30 |
| jroll | rloo: right, though I said there "only if we get to them", we have too much work here already as it is | 15:30 |
| rloo | jroll: true but... they are lower priority in that list of priorities. | 15:30 |
| rloo | jroll: some of those are easy, just need reviews I think. Anyway, not going to push too hard about that. | 15:30 |
| jroll | rloo: I like to think of this as our priority efforts, not priority of all the things | 15:30 |
| * rloo going to review them anyway, probably. | 15:30 | |
| *** rloo has quit IRC | 15:31 | |
| *** rloo has joined #openstack-ironic | 15:31 | |
| *** yolanda has joined #openstack-ironic | 15:32 | |
| jroll | rloo: right, and that's okay. fwiw, I'm going to work on a better dashboard that reviewers can use to decide what's important | 15:32 |
| rloo | jroll: thx. | 15:32 |
| *** absubram has joined #openstack-ironic | 15:34 | |
| *** jistr has quit IRC | 15:37 | |
| *** fragatina has quit IRC | 15:37 | |
| mgould | jlvillal, not recently | 15:39 |
| jlvillal | mgould: Okay. We are going to do an audio session at 1700UTC today. | 15:39 |
| mgould | currently sambetts and I are trying to work out why deployment is so frigging slow | 15:39 |
| jlvillal | mgould: https://wiki.openstack.org/wiki/Meetings/Ironic-QA | 15:39 |
| mgould | jlvillal, cool | 15:39 |
| jlvillal | mgould: That would be good to know, especially if you two can fix deployments being slow :) | 15:40 |
| *** deray has joined #openstack-ironic | 15:41 | |
| jlvillal | Interesting Swift wants to move parts of their codebase to golang officially. http://lists.openstack.org/pipermail/openstack-dev/2016-May/093680.html I know they had a branch they have been working on. | 15:44 |
| NobodyCam | morning sambetts jroll rloo | 15:44 |
| devananda | morning, all | 15:44 |
| devananda | morning, NobodyCam :) | 15:44 |
| NobodyCam | and devananda too | 15:44 |
| NobodyCam | :) | 15:44 |
| NobodyCam | :p | 15:44 |
| deray | dtantsur, hi .. rgding https://review.openstack.org/#/c/311530/6/priorities/newton-priorities.rst@29 .. would like to volunteer for this. Has it already progressed significantly? | 15:44 |
| JayF | jlvillal: yeah, there's been a go implementation of some of the api stuff for a while | 15:45 |
| JayF | jlvillal: I was happy to see they unforked and got a path forward for the code to merge | 15:45 |
| dtantsur | deray, the spec is half-way through, reviews appreciated: https://review.openstack.org/188370 | 15:45 |
| jlvillal | And a proposal to add golang as an official OpenStack language I see. | 15:45 |
| jlvillal | JayF: It is interesting. | 15:46 |
| * jlvillal wants to learn Go :) | 15:46 | |
| *** davideagnello has joined #openstack-ironic | 15:46 | |
| *** Guest23319 has quit IRC | 15:46 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: [WIP] Support capabilities discovery https://review.openstack.org/308292 | 15:47 |
| deray | dtantsur, okay. so as far as implementation goes we still have the complete work left, haven't we? | 15:47 |
| *** Sukhdev has joined #openstack-ironic | 15:47 | |
| dtantsur | deray, yep, I didn't start coding | 15:47 |
| lucasagomes | jjohnson2, https://review.openstack.org/#/c/312607/ the project-config change | 15:48 |
| lucasagomes | I'm never confident if I got it right or not, let's wait for the first reviews :-) | 15:48 |
| lucasagomes | jroll, ^ you may be interested on that (sorry for the "ping-fest" against you today) | 15:49 |
| jroll | lucasagomes: thanks, and I never mind too many pings | 15:49 |
| jroll | I can always ignore you :) | 15:49 |
| NobodyCam | morning dtantsur JayF :) | 15:49 |
| lucasagomes | yes you should sometimes :-) | 15:49 |
| *** e0ne has joined #openstack-ironic | 15:49 | |
| dtantsur | morning NobodyCam | 15:49 |
| NobodyCam | :) | 15:50 |
| jroll | morning NobodyCam, devananda :) | 15:50 |
| NobodyCam | :) | 15:50 |
| NobodyCam | okay here's a off the wall question. has any one tested ironic in a multi region env? | 15:51 |
| deray | dtantsur, ok good. please consider me in then. want to pursue it. will be good learning for me, as well. Also, from my employer's side need to actively pursue the boot from cinder work.. | 15:51 |
| jroll | NobodyCam: mirantis folks have iirc, they posted some client patches | 15:51 |
| jroll | NobodyCam: I assume you mean something like one ironic per region | 15:52 |
| vdrok | NobodyCam: I did once, also IIRC mat128 runs it in production :) | 15:52 |
| jroll | oh, we also have ironic in multiple regions, heh | 15:52 |
| dtantsur | deray, cool! | 15:52 |
| jroll | but we don't use service catalog for it, it's all env variables for endpoint etc | 15:52 |
| jjohnson2 | lucasagomes, cool | 15:53 |
| NobodyCam | jroll: 3 regions .. kvm, esx, and ironic | 15:53 |
| NobodyCam | vdrok: awesome :) | 15:53 |
| jroll | NobodyCam: so one region is just ironic, one just kvm, one just esx? | 15:53 |
| NobodyCam | yep | 15:53 |
| jroll | yeah, I don't see that being a problem, as long as you know which nova you talk to, to get which tech | 15:54 |
| vdrok | there shouldn't be a problem with that, yep | 15:54 |
| NobodyCam | ya I was looking at common/keystone.py and it looks like we support regions just wanted to see if anyone had actually used / tested it :) | 15:55 |
| NobodyCam | :) | 15:55 |
| *** sacharya has joined #openstack-ironic | 15:57 | |
| dtantsur | morning devananda, jlvillal | 15:58 |
| *** jistr has joined #openstack-ironic | 15:58 | |
| *** davideagnello has quit IRC | 15:59 | |
| jlvillal | dtantsur: NobodyCam vdrok jroll lucasagomes Good morning | 16:00 |
| lucasagomes | jlvillal, morning | 16:00 |
| lucasagomes | I will call it a day a bit earlier today because I have to go to town, I see you all tomorrow | 16:00 |
| lucasagomes | have a great (ugt) evening | 16:00 |
| *** lucasagomes is now known as lucas-afk | 16:01 | |
| *** davideagnello has joined #openstack-ironic | 16:01 | |
| *** amotoki has joined #openstack-ironic | 16:02 | |
| devananda | NobodyCam: are you sharing any servives between those regions (eg, common glance) ? | 16:05 |
| *** athomas has quit IRC | 16:06 | |
| *** fragatina has joined #openstack-ironic | 16:13 | |
| *** athomas has joined #openstack-ironic | 16:14 | |
| NobodyCam | devananda: swift glance cinder shared | 16:15 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Support capabilities discovery https://review.openstack.org/308292 | 16:16 |
| NobodyCam | night lucas-afk | 16:16 |
| dtantsur | should be fine to review ^^^ | 16:16 |
| dtantsur | now I'm calling it a day as well, see you | 16:16 |
| *** dprince has joined #openstack-ironic | 16:16 | |
| *** dtantsur is now known as dtantsur|afk | 16:17 | |
| NobodyCam | night dtantsur|afk | 16:18 |
| sambetts | o/ dtantsur|afk | 16:19 |
| *** deray has quit IRC | 16:20 | |
| *** Teknologeek has quit IRC | 16:22 | |
| *** mbound has quit IRC | 16:23 | |
| *** daemontool_ has joined #openstack-ironic | 16:27 | |
| *** daemontool_ has quit IRC | 16:28 | |
| *** daemontool has quit IRC | 16:28 | |
| *** keedya has quit IRC | 16:28 | |
| *** daemontool has joined #openstack-ironic | 16:29 | |
| *** irf has joined #openstack-ironic | 16:39 | |
| *** e0ne has quit IRC | 16:46 | |
| *** e0ne has joined #openstack-ironic | 16:47 | |
| *** rbudden has quit IRC | 16:51 | |
| devananda | jlvillal: QA meeting today? | 16:53 |
| jlvillal | devananda: https://wiki.openstack.org/wiki/Meetings/Ironic-QA | 16:53 |
| jlvillal | And everyone else | 16:53 |
| jlvillal | devananda: Yes. And going to do an audio bridge this week | 16:53 |
| devananda | ooh | 16:53 |
| devananda | neat | 16:53 |
| jlvillal | :) | 16:53 |
| jlvillal | devananda: I was just about to paste that URL. Really I'm not that fast... | 16:53 |
| *** jistr has quit IRC | 16:54 | |
| jlvillal | Grenade discussion meeting in 5 minutes: https://wiki.openstack.org/wiki/Meetings/Ironic-QA | 16:54 |
| jlvillal | Using an audio bridge this week... | 16:54 |
| devananda | that's perfect. I'm going to dial in and go for a walk :) | 16:55 |
| jlvillal | :) | 16:55 |
| *** trown is now known as trown|lunch | 16:57 | |
| *** derekh has quit IRC | 16:57 | |
| *** e0ne has quit IRC | 17:01 | |
| *** e0ne has joined #openstack-ironic | 17:02 | |
| *** sivaramakrishna has joined #openstack-ironic | 17:11 | |
| *** ChubYann has joined #openstack-ironic | 17:13 | |
| *** ohamada has quit IRC | 17:15 | |
| *** irf has quit IRC | 17:30 | |
| *** thiagop has joined #openstack-ironic | 17:34 | |
| thiagop | Good morning, Ironic | 17:34 |
| thiagop | krtaylor: around? | 17:34 |
| NobodyCam | morning thiagop | 17:35 |
| thiagop | Hey NobodyCam, did you have a good travel back home? | 17:36 |
| *** electrofelix has quit IRC | 17:36 | |
| *** irf has joined #openstack-ironic | 17:36 | |
| NobodyCam | I did, :) | 17:39 |
| thiagop | Nice | 17:41 |
| *** Sukhdev has quit IRC | 17:42 | |
| thiagop | jlvillal: are you around? | 17:42 |
| NobodyCam | thiagop: I think he is on the qa meeting / call | 17:43 |
| thiagop | ow, so I'll wait | 17:43 |
| thiagop | I'm looking for the link for the molten iron tool. I think it was on the etherpad of the testing work session, but I don't have the link | 17:44 |
| thiagop | google didn't help | 17:45 |
| *** irf has quit IRC | 17:46 | |
| JayF | ooooh, I found a faster nic driver, interesting. rtl8139+ipxe booted from power on -> lookup in 3:04, compared to 3:33 for the default devstack config (virtio+ipxe) | 17:46 |
| TheJulia | nice | 17:47 |
| NobodyCam | lol | 17:48 |
| thiagop | wow | 17:49 |
| sambetts | ooo nice! | 17:49 |
| JayF | going to finish with my testing before taking action, but will benchmark a full job as my next steps | 17:49 |
| krtaylor | thiagop, sorry distracted by QA meeting :) | 17:54 |
| thiagop | NP, I need to put that on my calendar | 17:55 |
| thiagop | is it 17:00 UTC? | 17:55 |
| krtaylor | thiagop, molten iron is at: https://review.openstack.org/#/c/304683 | 17:56 |
| krtaylor | thiagop, I have some pep8 cleanup to do then I'll be begging reviews :) | 17:57 |
| thiagop | krtaylor: I found it with a very weird on UX | 17:58 |
| krtaylor | thiagop, don't understand | 17:58 |
| jlvillal | thiagop: Sorry I was in the Grenade meeting. | 17:59 |
| thiagop | krtaylor: can we use a console script for that client? Or there is a limitation due to it being inside third-party-ci-tools? | 17:59 |
| *** wajdi_ has joined #openstack-ironic | 17:59 | |
| * vdrok goes home | 17:59 | |
| vdrok | good night! | 17:59 |
| jlvillal | vdrok: Ciao! Dos ve Dos | 17:59 |
| vdrok | :) | 17:59 |
| thiagop | np jlvillal, hope you guys had a safe travel back | 17:59 |
| jlvillal | thiagop: I did, thank you. And I hope you did as well :) | 18:00 |
| *** wajdi_ has quit IRC | 18:00 | |
| krtaylor | mjturek1, ^^^ | 18:00 |
| * jlvillal heads to lunch | 18:00 | |
| *** wajdi_ has joined #openstack-ironic | 18:00 | |
| krtaylor | thiagop, do you mean how do you pull it and use it in your environment? | 18:00 |
| *** wajdi has quit IRC | 18:01 | |
| krtaylor | thiagop, ideally, I think it should be moved to a new project if others are interested | 18:01 |
| thiagop | krtaylor: yeah, it's kinda not very intuitive... but usable, though | 18:01 |
| thiagop | krtaylor: +1 | 18:01 |
| krtaylor | thiagop, that way we can add docs, testing, etc | 18:01 |
| *** wajdi_ has quit IRC | 18:01 | |
| *** keedya has joined #openstack-ironic | 18:01 | |
| *** wajdi has joined #openstack-ironic | 18:01 | |
| krtaylor | thiagop, right now I had to turn off testing for the third-party-ci-tools repo due to the fact that there are so many different types of contributions (not all python, ect) | 18:02 |
| * mgould -> home; good night! | 18:02 | |
| * sambetts -> goes home | 18:02 | |
| *** mgould has quit IRC | 18:02 | |
| *** sambetts is now known as sambetts|afk | 18:03 | |
| thiagop | night sambetts vdrok mgound | 18:03 |
| thiagop | mgould* | 18:03 |
| NobodyCam | night sambetts|afk mgould | 18:04 |
| thiagop | krtaylor: makes sense. Do you intend to use that repo as an incubator? | 18:06 |
| *** ChubYann has quit IRC | 18:09 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Optimise tinyipa boot time https://review.openstack.org/312671 | 18:10 |
| JayF | sambetts|afk: ^ what's the tl;dr of what that actually does? | 18:14 |
| JayF | sambetts|afk: actually that should probably go i nthe commit message :) | 18:14 |
| krtaylor | thiagop, it is just a place for third party CI operators to be able to share their tools they use to make CI easier | 18:15 |
| krtaylor | thiagop, MoltenIron fits that mission, although initially focused on Ironic | 18:15 |
| thiagop | krtaylor: sure. Talking to anteaya on the summit, she asked me and jroll to make the solution to this problem documented so people on the infra can mention it for people asking for help | 18:18 |
| thiagop | krtaylor: so... I'll be glad to test and help :) | 18:18 |
| sambetts|afk | JayF: probably should put that in there, tl;dr installs all the system packages at build time instead the normal tinycore linux way of loading them at runtime | 18:18 |
| sambetts|afk | cuts 60+ seconds of expanding squashfs' out of the boot proceedure | 18:19 |
| JayF | nice, very very nice | 18:19 |
| JayF | if you add that blurb to the commit message I'll +2 it | 18:19 |
| JayF | that's what it looked like you were doing to me, but wanted to be sure | 18:19 |
| *** ChubYann has joined #openstack-ironic | 18:21 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Optimise tinyipa boot time https://review.openstack.org/312671 | 18:23 |
| sambetts|afk | JayF: ^ updated | 18:24 |
| JayF | TheJulia: so re: ipxe, looking at the code, it looks like it configures the dhcp server to serve ipxe rom if not already ipxe, and serve ipxe config if it's already ipxe | 18:25 |
| *** yolanda has quit IRC | 18:25 | |
| JayF | TheJulia: so I think we were both right | 18:25 |
| JayF | TheJulia: jroll ^ one of you wanna land that? | 18:26 |
| jroll | no, I want CI passing first :) | 18:26 |
| jroll | tinyipa job is non-voting, I don't want to accidentally break it | 18:26 |
| JayF | oh that's a really good point that it's -nv | 18:26 |
| JayF | was about to joke about jenkins being about to vote for itself :P | 18:27 |
| jroll | :) | 18:28 |
| jroll | sambetts|afk: what was the reason we didn't install IPA at build time before? | 18:28 |
| TheJulia | JayF: yah, that was what I was remembering :) | 18:29 |
| JayF | TheJulia: so we might want to consider having ipxe be the default for our jobs (if it's not already?) | 18:29 |
| sambetts|afk | jroll: resulted in a very slightly smaller image | 18:30 |
| jroll | sambetts|afk: right, I thought it was very smaller | 18:30 |
| *** trown|lunch is now known as trown | 18:31 | |
| sambetts|afk | jroll: but in previous patches I optimised down the dependecies, and now the difference between installing or not installing is ~3mb | 18:31 |
| jroll | anyway, +2 assuming CI passes | 18:31 |
| jroll | oh wow | 18:31 |
| jroll | neat | 18:31 |
| sambetts|afk | I've built tinyipa locally using that patch and the resulting image is 63mb | 18:32 |
| jroll | awesome | 18:32 |
| JayF | from what you said on the call | 18:32 |
| JayF | sounds like it'll save 2m per boot | 18:32 |
| JayF | compared to default config | 18:32 |
| sambetts|afk | and I've run it through tox -e all-plugin -- ironic sucessfully | 18:32 |
| sambetts|afk | manually | 18:32 |
| sambetts|afk | coreos 213mb, tinyipa 63mb so 28% the size | 18:35 |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add Hardware inventory from Lenovo Agentless https://review.openstack.org/312568 | 18:36 |
| sambetts|afk | myself and mgould also tried experimenting with un gziping the coreos initrd to see if the compression was causing any slow down, it didn't make much of a difference | 18:36 |
| sambetts|afk | uncompressing only increase the file size by 2mb | 18:36 |
| krtaylor | thiagop, thanks! I'll prob beg a review in the next day or so :) | 18:38 |
| *** xhku has quit IRC | 18:38 | |
| * sambetts|afk is really AFK now | 18:39 | |
| thiagop | ^lol | 18:42 |
| *** vishwanathj has joined #openstack-ironic | 18:50 | |
| devananda | jroll: you know if anyone else started updating the API docs yet? | 18:50 |
| jroll | devananda: not to my knowledge | 18:51 |
| devananda | cool. I'm going to do the base page and Nodes right now | 18:51 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Benchmark gate with rtl8139 NIC driver https://review.openstack.org/312690 | 18:52 |
| JayF | ^ since that was the only NIC driver that benchmarked faster, was going to try it in the gate. Honestly none of them had impressive enough numbers to be worth a change though, I think | 18:55 |
| devananda | I'm going t o unabashedly copy some parts from nova's api ref | 18:56 |
| jroll | devananda: feel free to clean up whitespace while you're in there :) | 18:56 |
| mat128 | JayF: In downstream testing, we had much better performance using virtio driver even for TFTP/PXE booting. The only backdraw was the Linux bonding module which didnt want to play nice with a NIC that has no "link speed" | 18:57 |
| JayF | mat128: I did very nieve testing against gate hardware; pretty much changed nic driver, forced the node into cleaning, and measured power on -> lookup | 18:57 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Support capabilities discovery https://review.openstack.org/308292 | 18:57 |
| JayF | mat128: that one checked out at about :30s faster than what was already in the gate | 18:58 |
| mat128 | really? interesting | 18:58 |
| JayF | mat128: https://etherpad.openstack.org/p/ironic-gate-pxe-replacement | 18:58 |
| JayF | has my results | 18:58 |
| mat128 | they all look very close, I wonder what the report will give us in https://review.openstack.org/312690 | 19:00 |
| mat128 | switching to TinyIPA sure gives the best boost as it's much smaller | 19:00 |
| JayF | yeah, I mean, I think with the exception of tinyipa the boosts we're going to get are going to be 10s here, 20s there, etc etc | 19:02 |
| JayF | only meaningful when added together | 19:02 |
| mat128 | JayF: agreed :) | 19:02 |
| *** afaranha has joined #openstack-ironic | 19:03 | |
| *** ijw has quit IRC | 19:04 | |
| *** piet has joined #openstack-ironic | 19:10 | |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add notification base classes and docs https://review.openstack.org/298461 | 19:13 |
| *** sivaramakrishna has quit IRC | 19:18 | |
| *** e0ne has quit IRC | 19:23 | |
| *** mbound has joined #openstack-ironic | 19:29 | |
| openstackgerrit | Merged openstack/ironic: RAID interface to support JBOD volumes https://review.openstack.org/308306 | 19:32 |
| jroll | mariojv: this will likely be interesting for you http://lists.openstack.org/pipermail/openstack-dev/2016-May/093734.html | 19:34 |
| mariojv | oh nice, that will be useful for making testing better | 19:37 |
| *** baoli_ has quit IRC | 19:39 | |
| *** piet has quit IRC | 19:44 | |
| *** mkovacik has quit IRC | 19:46 | |
| *** e0ne has joined #openstack-ironic | 19:46 | |
| *** openstackgerrit has quit IRC | 19:48 | |
| *** openstackgerrit has joined #openstack-ironic | 19:48 | |
| *** e0ne has quit IRC | 19:48 | |
| *** piet has joined #openstack-ironic | 19:49 | |
| *** athomas has quit IRC | 19:52 | |
| *** dprince has quit IRC | 19:58 | |
| *** e0ne has joined #openstack-ironic | 20:02 | |
| *** wajdi_ has joined #openstack-ironic | 20:03 | |
| *** wajdi has quit IRC | 20:07 | |
| *** e0ne has quit IRC | 20:07 | |
| *** piet has quit IRC | 20:08 | |
| *** harlowja has quit IRC | 20:09 | |
| *** e0ne has joined #openstack-ironic | 20:10 | |
| *** wajdi_ has quit IRC | 20:10 | |
| *** wajdi has joined #openstack-ironic | 20:11 | |
| *** athomas has joined #openstack-ironic | 20:11 | |
| *** athomas has quit IRC | 20:11 | |
| *** dmk0202 has joined #openstack-ironic | 20:14 | |
| *** rloo has quit IRC | 20:22 | |
| JayF | https://review.openstack.org/#/c/312690/ ran tests at a speed right at what we currently run at | 20:23 |
| JayF | so even if there's a benefit there, it's minimal, and probably not worth the risk of changing the driver (because we don't have reliability metrics on it) | 20:23 |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add Hardware inventory from Lenovo Agentless https://review.openstack.org/312568 | 20:26 |
| *** caiobo has quit IRC | 20:27 | |
| TheJulia | Agreed, although it might be something to keep in reserve if we really need just a few more seconds or get things working but sporattically are hitting the timeout or something a long those lines | 20:28 |
| *** rbudden has joined #openstack-ironic | 20:33 | |
| *** rloo has joined #openstack-ironic | 20:34 | |
| *** e0ne has quit IRC | 20:37 | |
| jroll | devananda: vdrok: hao<tab> :( yuriyz: TheJulia: we'd like your ack on https://review.openstack.org/#/c/311530/ when you have time please :) | 20:40 |
| *** joprovost has quit IRC | 20:40 | |
| jroll | oh and jlvillal ^ | 20:41 |
| * devananda looks | 20:42 | |
| *** Goneri has quit IRC | 20:42 | |
| *** spandhe has joined #openstack-ironic | 20:45 | |
| jlvillal | jroll: Any thought on reversing the order? Newton first? | 20:47 |
| jlvillal | Not important | 20:47 |
| jroll | jlvillal: meh, I could | 20:48 |
| jlvillal | Don't bother | 20:48 |
| jroll | I'll wait for everyone's ack and then do that if you leave a comment | 20:48 |
| jroll | that way we know everyone's on board before re-spinning | 20:48 |
| jlvillal | jroll: Looks good as is. I think that would be a follow-up patch if can figure out a way to reverse the glob ordering. | 20:52 |
| *** Sukhdev has joined #openstack-ironic | 20:52 | |
| jroll | thanks | 20:52 |
| *** Sukhdev has quit IRC | 20:54 | |
| JayF | Does devstack just work today with VMs if I set deploy driver to agent_ipmitool? | 20:55 |
| JayF | like the virtual bmcs and such are setup by default, correct? | 20:55 |
| wajdi | jroll: is "herding cats" implied to contact the people who volunteered to assist with certain priorities? | 20:56 |
| *** jaypipes has quit IRC | 20:56 | |
| *** rama_y has joined #openstack-ironic | 20:57 | |
| jroll | wajdi: yes, generally make sure people are working on the right things | 20:58 |
| jroll | JayF: I think so? there might be one more option | 20:59 |
| *** rloo has quit IRC | 20:59 | |
| JayF | yeah, it looks like that's all it needs | 20:59 |
| JayF | I'll find out soon enough :) | 20:59 |
| JayF | jroll: doing devstack on onmetal-io2 machine, with /opt on the IO cards :D | 21:00 |
| jroll | JayF: looks like it to me, yeah | 21:00 |
| jroll | nice | 21:00 |
| jroll | because reading the code is totally slow :P | 21:00 |
| wajdi | jroll: is it safe to assume that you made note of those interested from the comments made? Or will that information be gathered later on? | 21:01 |
| jroll | wajdi: I did, a bit, I can always go back and look. from what I recall internap is interested in helping with multiple compute hosts, and vdrok / sambetts|afk are interested in helping with gate things | 21:02 |
| jroll | comments are there forever | 21:02 |
| wajdi | awesome. thanks. :) | 21:03 |
| jroll | sure :) | 21:03 |
| jroll | time for me to head out, see y'all tomorrow | 21:04 |
| wajdi | cheers | 21:04 |
| JayF | o/ | 21:04 |
| JayF | sucks for you to be on est, jroll | 21:04 |
| JayF | lol | 21:04 |
| jroll | hehehe | 21:04 |
| jroll | JayF: I can always stick around if there's things people want my help with | 21:04 |
| jroll | but now I get more time with my euro friends too \o/ | 21:04 |
| JayF | I mean, just generally makes my day feel longer | 21:04 |
| JayF | I used to leave like 30-60m after you | 21:04 |
| JayF | lol | 21:04 |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add Hardware inventory from Lenovo Agentless https://review.openstack.org/312568 | 21:04 |
| jroll | JayF: think about my confusion yesterday when you took lunch at 5pm :) | 21:05 |
| JayF | jroll: heh | 21:06 |
| *** rloo has joined #openstack-ironic | 21:08 | |
| jroll | oh before I go | 21:08 |
| jroll | http://developer.openstack.org/api-ref/baremetal/ | 21:08 |
| jroll | \o/ | 21:08 |
| jroll | these come from https://github.com/openstack/ironic/tree/master/api-ref | 21:08 |
| wajdi | awesome! | 21:09 |
| *** rbrud has joined #openstack-ironic | 21:09 | |
| rloo | jroll: wrt the api-ref. are those handcrafted text/pages? | 21:10 |
| *** harlowja has joined #openstack-ironic | 21:10 | |
| jroll | rloo: define handcrafted? | 21:10 |
| rloo | jroll: it'd be great if the 'source of truth' came from the docstrings/code | 21:11 |
| jroll | they look like this: https://github.com/openstack/ironic/blob/master/api-ref/source/baremetal-api-v1-chassis.inc | 21:11 |
| jroll | well | 21:11 |
| jroll | we're bad about keeping api docs in tree up to date | 21:11 |
| *** jjohnson2 has quit IRC | 21:11 | |
| jroll | I can chat more tomorrow, or you can ask in the thread I started, but need to go for now, sorry | 21:11 |
| rloo | jroll: eg we generate this from our code base: http://docs.openstack.org/developer/ironic/webapi/v1.html#chassis | 21:11 |
| rloo | jroll: oh, didn't know you had started a thread. will see. | 21:12 |
| jroll | rloo: yeah, just sent the same \o/ in email, feel free to reply | 21:12 |
| jroll | but that's a good question, we should work that out | 21:12 |
| devananda | rloo: these are hand-crafted, yes | 21:12 |
| jroll | or point our api docs at the new thing, and make the new thing good | 21:12 |
| devananda | I'm working on updating them right now | 21:13 |
| devananda | one of the neat things about flask, aparently, is automatic generation of WADL docs from the source tree / comments | 21:13 |
| devananda | we have different API docs generated from our source right now | 21:13 |
| *** mbound has quit IRC | 21:13 | |
| devananda | http://docs.openstack.org/developer/ironic/webapi/v1.html | 21:13 |
| jroll | we could probably figure something out | 21:13 |
| jroll | to make the two mesh | 21:14 |
| devananda | possibly | 21:14 |
| rloo | devananda: right. i'd like the stuff to mesh. | 21:14 |
| jroll | devananda: btw, feel free to change up file names / directory structure | 21:14 |
| devananda | jroll: *nod* | 21:14 |
| jroll | alright, out for real. later. | 21:14 |
| devananda | regardless of where it goes, there is going to be a lot of writing to do | 21:14 |
| rloo | devananda: it is easier to update/code review the api changes when the related code is changed, when the stuff is all in the same file. | 21:14 |
| devananda | rloo: it also makes that file huge, if we're storing sample request and response BODY inline in comments | 21:15 |
| devananda | one nice thing about the current files is, as I'm editing them, I can actually just copy/paste from a service into the file | 21:15 |
| rloo | devananda: well, let's not duplicate stuff in two places. | 21:15 |
| devananda | and we could easily store different versions of responses to things in the same directory | 21:16 |
| devananda | just with file names and links inline | 21:16 |
| rloo | devananda: so we should delete the sample stuff from our code files then. | 21:16 |
| JayF | Is there a patch for this short_source error being spammed like crazy when devstacking? | 21:16 |
| devananda | rloo: *nod* | 21:16 |
| JayF | I know unset PS4 fixes it for interactive sessions | 21:16 |
| rloo | devananda: anyway, i just want to make sure we aren't duplicating effort here and have mismatched info in two or more places. | 21:16 |
| wajdi | if you guys are interested I can share a project we have up on github that generates a pretty nice doc that is up on readthedocs. We just add our docstring to our API calls that we create, and it gets updated in the next release we push. | 21:16 |
| devananda | rloo: I agree with that sentiment whole heartedly | 21:16 |
| rloo | devananda: thanks for volunteering :D | 21:17 |
| devananda | wajdi: take a look at the link I pasted above -- that is autogenerated from our code | 21:17 |
| wajdi | missed that. thanks devananda | 21:17 |
| devananda | for comparison, take a look at Nova's API docs here: http://developer.openstack.org/api-ref-compute-v2.1.html | 21:17 |
| *** trown is now known as trown|outtypewww | 21:17 | |
| devananda | that is using the same framework that jroll just ported over to ironic/api-ref and which I'm now working on updating | 21:18 |
| rloo | wajdi, yeah, that's the discussion. I don't want us generating stuff in two ways, via-that-link-above vs the new api-ref way. | 21:18 |
| devananda | the output is much cleaner, and geared towards application developers writing things *against* our API | 21:18 |
| devananda | rather than folks working inside Ironic | 21:18 |
| jroll | devananda: that link is not using the new thing | 21:19 |
| *** Sukhdev has joined #openstack-ironic | 21:19 | |
| jroll | http://developer.openstack.org/api-ref/compute/ | 21:19 |
| jroll | is the new nova thing | 21:19 |
| devananda | oh | 21:19 |
| jroll | intended to replace what you linked, ofc | 21:19 |
| devananda | ah. thanks for the clarification | 21:19 |
| jroll | np | 21:19 |
| wajdi | oh wow there is a lot of great detail in that nova link | 21:19 |
| devananda | the format is still vastly different | 21:19 |
| jroll | the idea is /api-ref/$service/ | 21:20 |
| devananda | and the amount of detail is an order of magnitude more | 21:20 |
| jroll | and maybe version in there | 21:20 |
| wajdi | that *is* very helpful for app developers looking to communicate with the API | 21:20 |
| jroll | devananda: right, it's the same content ported to the new thing | 21:20 |
| devananda | jroll: rloo was asking if we should try to generate the same content from our inline-in-code documentation | 21:20 |
| jroll | JayF: this "keep getting distracted" thing is the worst part of EDT | 21:20 |
| devananda | my current answer is "I would rather not" | 21:21 |
| JayF | jroll: ^a^d fixes a lot of problems | 21:21 |
| jroll | devananda: right, I saw that, I need to go, asked here to reply to my "this is a thing now" email | 21:21 |
| *** piet has joined #openstack-ironic | 21:21 | |
| devananda | jroll: cool. will reply | 21:21 |
| jroll | JayF: BUT SO MANY INTERESTING THINGS IN MY TMUX SESSION | 21:21 |
| devananda | ciao! | 21:21 |
| jroll | later :) | 21:21 |
| rloo | no one talk to jroll | 21:22 |
| wajdi | (now no one is talking) | 21:23 |
| wajdi | :P | 21:23 |
| *** thiagop_ has joined #openstack-ironic | 21:23 | |
| *** thiagop has quit IRC | 21:23 | |
| *** praneshp has joined #openstack-ironic | 21:26 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Optimise tinyipa boot time https://review.openstack.org/312671 | 21:26 |
| *** thiagop_ has quit IRC | 21:29 | |
| *** thiagop has joined #openstack-ironic | 21:29 | |
| mrda | JayF: I apologise for not reviewing that thing yesterday. I will do that this morning. | 21:32 |
| * mrda slaps mrda on the wrist | 21:32 | |
| JayF | it's alright, just wanted to make sure you both were aware of it because it's a nova/ironic thing | 21:32 |
| JayF | sambetts|afk: looks like tinyipa also spends quite a bit of time on boot seeding randomness. Wonder if that's something we could at least partially speed up / disable in the gate | 21:33 |
| mrda | JayF: Or indeed at all times. I'm not sure how having good randomness helps a deploy image (but I haven't coffeed yet :) | 21:36 |
| JayF | mrda: given I run actual sshd's on my deploy images in prod | 21:37 |
| JayF | mrda: I can think of a few reasons :) | 21:37 |
| JayF | mrda: also I think the hope is that it won't be unauthenticated forever; real entropy is something we'd need to do anything cryptographically interesting in the image | 21:37 |
| JayF | just sorta one of those things I don't wanna turn off by default, because a deployer /could/ need it, and wouldn't know they were starting on shaky entropy | 21:37 |
| *** ppiela has joined #openstack-ironic | 21:42 | |
| *** chlong has quit IRC | 21:45 | |
| *** wajdi_ has joined #openstack-ironic | 21:49 | |
| *** fragatin_ has joined #openstack-ironic | 21:50 | |
| *** fragatina has quit IRC | 21:51 | |
| JayF | jroll: fyi; in onmetal+devstack, anything that uses the full disk image is failing too | 21:51 |
| JayF | jroll: because the full disk image can't get noapic passed in | 21:52 |
| *** Sukhdev has quit IRC | 21:52 | |
| *** wajdi has quit IRC | 21:52 | |
| *** Sukhdev has joined #openstack-ironic | 21:52 | |
| *** wajdi_ has quit IRC | 21:54 | |
| *** spandhe has quit IRC | 21:54 | |
| *** fragatin_ has quit IRC | 21:57 | |
| *** fragatina has joined #openstack-ironic | 21:57 | |
| mrda | JayF: I guess because the deploy image is so transitory.... of course I'd always advocate for good crypto, it's just a matter of how much randonmness we need on a isolated network segment. But I should stop arguing now :) | 21:59 |
| JayF | mrda: I just generally am on the conservative side with regards to turning something off that can have security implications + is difficult to discover without knowledge | 21:59 |
| *** dmk0202 has quit IRC | 22:01 | |
| *** Sukhdev has quit IRC | 22:04 | |
| *** rbudden has quit IRC | 22:04 | |
| *** chlong has joined #openstack-ironic | 22:05 | |
| *** Goneri has joined #openstack-ironic | 22:06 | |
| JayF | mariojv: you get a gold star for having such great docs in your notification code ✫ | 22:07 |
| mariojv | :D | 22:07 |
| *** vishwanathj has quit IRC | 22:10 | |
| *** absubram has quit IRC | 22:15 | |
| *** nick1_ has joined #openstack-ironic | 22:18 | |
| *** nick1_ has left #openstack-ironic | 22:18 | |
| *** jaybeale has joined #openstack-ironic | 22:20 | |
| *** nick1_ has joined #openstack-ironic | 22:20 | |
| nick1_ | does anyone know which dhcp server is used for the scale mentioned here? https://www.openstack.org/summit/openstack-paris-summit-2014/session-videos/presentation/scaling-ironic | 22:21 |
| JayF | That was what, Jim's? | 22:22 |
| JayF | I mean, we use dhcp_provider=None | 22:22 |
| JayF | and our static DHCP clusters are corosync+pacemaker managing services and vips, with isc-dhcp-server serving dhcp | 22:22 |
| nick1_ | OK thanks. What is the scale number today? It was 1000+ in the video | 22:23 |
| JayF | That's generally the number we give out publically | 22:23 |
| JayF | although we now have multiple environments running at that scale in different regions | 22:24 |
| *** baoli has joined #openstack-ironic | 22:24 | |
| nick1_ | Thank you | 22:24 |
| JayF | No problem, any reason you're curious? | 22:25 |
| JayF | If we were starting from scratch today, fwiw, I doubt we'd use our own external dhcp stuff vs letting neutron manage it | 22:25 |
| *** baoli_ has joined #openstack-ironic | 22:25 | |
| nick1_ | I want to play with it. I was using dnsmasq but I have scale concerns | 22:25 |
| JayF | Heh. The reason we run our own external dhcp clusters | 22:26 |
| JayF | is because I too have been scarred by dnsmasq | 22:26 |
| JayF | I'd still use neutron dhcp + dnsmasq today if I rebuilt it though, the network code that's coming down the pipe makes it even less tenable to manage than before | 22:26 |
| JayF | Glyph keeps trying to talk me into writing a twisted dhcp server and tying that into neutron too :P | 22:26 |
| nick1_ | I | 22:27 |
| nick1_ | Do you know the scale number for dnsmasq? | 22:27 |
| JayF | I don't think there's a number, but I do know we have one user of Ironic, who deployed an entire 850 node cluster with all openstack services on one box | 22:28 |
| JayF | so I think scalability issues are more around how many simultaneous builds/deletes than almost anything else | 22:28 |
| JayF | it's very cheap for nodes to be enrolled+provisioned with ironic | 22:28 |
| *** piet has quit IRC | 22:28 | |
| nick1_ | I see. Thank you. I will come up with more questions once I start playing with it | 22:29 |
| *** nick1_ has quit IRC | 22:29 | |
| *** baoli has quit IRC | 22:29 | |
| *** thiagop has quit IRC | 22:29 | |
| *** keedya has quit IRC | 22:31 | |
| *** jaybeale has quit IRC | 22:34 | |
| *** ChrisAusten has quit IRC | 22:40 | |
| *** mbound has joined #openstack-ironic | 22:46 | |
| *** ametts has quit IRC | 22:52 | |
| *** harlowja has quit IRC | 22:52 | |
| *** harlowja has joined #openstack-ironic | 22:53 | |
| *** ijw has joined #openstack-ironic | 22:57 | |
| *** rloo has quit IRC | 22:59 | |
| *** baoli_ has quit IRC | 23:05 | |
| *** baoli has joined #openstack-ironic | 23:08 | |
| *** Nisha has joined #openstack-ironic | 23:08 | |
| *** Sukhdev has joined #openstack-ironic | 23:13 | |
| *** rbrud has quit IRC | 23:14 | |
| openstackgerrit | Dan Bode proposed openstack/bifrost: Document that ssh_public_key_path must be set https://review.openstack.org/312782 | 23:21 |
| *** anteaya has joined #openstack-ironic | 23:22 | |
| *** bapalm_ has quit IRC | 23:30 | |
| *** sacharya has quit IRC | 23:33 | |
| *** fragatina has quit IRC | 23:39 | |
| *** fragatina has joined #openstack-ironic | 23:40 | |
| *** Nisha has quit IRC | 23:47 | |
| *** piet has joined #openstack-ironic | 23:48 | |
| *** mbound has quit IRC | 23:48 | |
| *** ChrisAusten has joined #openstack-ironic | 23:49 | |
| *** amit213 has joined #openstack-ironic | 23:51 | |
| *** davideagnello has quit IRC | 23:51 | |
| *** davideagnello has joined #openstack-ironic | 23:52 | |
| *** davideagnello has quit IRC | 23:56 | |
| *** ijw has quit IRC | 23:57 | |
| *** ijw has joined #openstack-ironic | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!