| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Additional services to support multiple swift storage nodes https://review.openstack.org/76391 | 00:02 |
|---|---|---|
| *** hewbrocca has quit IRC | 00:03 | |
| tchaypo | bnemec: hear, let me sharpen your razor before you start on the next yak | 00:03 |
| bnemec | tchaypo: Indeed. :-/ | 00:05 |
| tchaypo | *here | 00:07 |
| saurabhs | Hi I am trying to understand how tripleo-image-elements are layed out to help write elements for trove. I can see that there is nova element and a separate element for nova-api. Inside nova-api there is element-deps on nova, so when we include nova-api, nova will get included by default. So this is pretty clear here. | 00:09 |
| saurabhs | But when I look at nova elements I can see that its installing nova through git, but nova-api elements are installing it through install-packages. why is it different? | 00:09 |
| *** michchap has quit IRC | 00:10 | |
| *** rbrady-afk is now known as rbrady | 00:10 | |
| *** edmund has quit IRC | 00:10 | |
| rbrady | saurabhs: the nova-api element was recently refactored to support package installs too | 00:11 |
| SpamapS | saurabhs: in the git/source based install, we don't split things out. We just install the software in 'nova' | 00:12 |
| saurabhs | rbrady: so I should use nova-api sort of a reference point then? also when you say it it now support package installs too, do you mean it also support source based install in nova-api elements. in nova-api/install.d/nova-source-install/80-nova-api, is that installing from source? I am not clear what its doing there | 00:14 |
| SpamapS | saurabhs: Heat will be more like Trove than nova. | 00:15 |
| SpamapS | saurabhs: the *-source-install dirs do the git based/virtualenv approach. the *-package-install dirs do a package based install. | 00:16 |
| saurabhs | Spamaps: so when I include nova-api which also includes nova elements, they will overwrite nova install. correct? | 00:16 |
| SpamapS | saurabhs: no they will not overwrite nova install. They will be combined. | 00:16 |
| saurabhs | SpamapS: so how do we configure which approach to use, whether to use source-install or package-install? | 00:17 |
| SpamapS | saurabhs: I don't know. Default is source. | 00:18 |
| SpamapS | saurabhs: ignore package based. ;) | 00:18 |
| SpamapS | that's for the package based people to figure out IMO :-P | 00:18 |
| *** michchap has joined #tripleo | 00:18 | |
| SpamapS | and if trove isn't even in Fedora/RHEL/RDO ... then well it will just be a noop. :-P | 00:18 |
| saurabhs | Spamaps : when we say installing from source, is it checking it out somewhere and then installing from that checkout stuff? I think it is correct? | 00:21 |
| *** michchap has quit IRC | 00:21 | |
| SpamapS | saurabhs: correct | 00:22 |
| saurabhs | if thats true then what this line is doing, is it installing nova from source or only creating daemon/upstart kind of thing | 00:22 |
| saurabhs | os-svc-daemon nova-api nova nova-api "--config-dir /etc/nova" | 00:22 |
| SpamapS | saurabhs: that is installing a startup script and creating users to run as. | 00:22 |
| *** michchap has joined #tripleo | 00:22 | |
| saurabhs | Spamaps: so os-svc-daemon really nows where is the nova-api executable? I don't see that we specified the /opt/stack/nova path here, where it was checked out or /opt/stack/venv where it was installed. | 00:25 |
| saurabhs | is that a fixed path where things are supposed to be? | 00:25 |
| openstackgerrit | Ryan Brady proposed a change to openstack/diskimage-builder: Add NFS package mapping https://review.openstack.org/76398 | 00:28 |
| SpamapS | saurabhs: https://git.openstack.org/cgit/openstack/tripleo-image-elements/tree/elements/os-svc-install/bin/os-svc-daemon#n108 | 00:28 |
| SpamapS | saurabhs: yes | 00:28 |
| lifeless | ok wow much meeting | 00:29 |
| lifeless | so many | 00:29 |
| lifeless | now to book my hotel room | 00:29 |
| saurabhs | Spamaps: thanks this is now more clear | 00:29 |
| StevenK | lifeless: The hotel still doesn't know about me | 00:29 |
| *** matsuhashi has joined #tripleo | 00:30 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Adds nfs-server element https://review.openstack.org/74712 | 00:30 |
| SpamapS | lifeless: doge much? | 00:31 |
| *** CaptTofu has joined #tripleo | 00:34 | |
| lifeless | StevenK: did you use the group code? | 00:35 |
| lifeless | StevenK: use the group code and add yourself | 00:35 |
| greghaynes | lifeless: Did your commit c1a97145 close bug 1283921? I dont seem to get that error anymore | 00:37 |
| StevenK | lifeless: Except I'm in the list that CWT were apparently adding | 00:38 |
| greghaynes | that commit is in t-i-e btw | 00:38 |
| lifeless | StevenK: epic fail do it anyhow :) | 00:38 |
| lifeless | greghaynes: I think we can close that yes | 00:38 |
| lifeless | greghaynes: I think I know why we got it | 00:39 |
| SpamapS | Worst case they'll have two rooms for you and you can throw your after parties in one, and sleep in the other. | 00:39 |
| greghaynes | Just wasnt running migrations? | 00:39 |
| SpamapS | You can probably even request adjoining rooms with yourself. | 00:39 |
| lifeless | greghaynes: we use the ovs plugin *and* the ml2 plugin. | 00:39 |
| lifeless | greghaynes: this isn't supported. | 00:39 |
| greghaynes | ah | 00:39 |
| lifeless | greghaynes: we have to migrate the ovs agent to use the ml2 plugin with ovs link driver. | 00:39 |
| lifeless | greghaynes: so I think it ran the ovs plugin migrations when the ovs agent connected, and then the ml2 migrations didn't run | 00:40 |
| StevenK | SpamapS: I bought that up, their system will automatically cancel duplicate reservations for the same person | 00:40 |
| lifeless | because OMG YOU DON"T WANT TO KNOW | 00:40 |
| greghaynes | haha, wierd | 00:40 |
| *** ramishra has quit IRC | 00:40 | |
| tchaypo | I'll use the group code as soon as I have flights | 00:43 |
| tchaypo | which, if everything goes according to plan, should happen this afternoon after I get my account activated in the office at rhodes | 00:43 |
| StevenK | I'm hoping internal webmail doesn't require a class B | 00:44 |
| StevenK | But we shall see | 00:45 |
| *** rbrady is now known as rbrady-afk | 00:46 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Store node details in the JSON datafile https://review.openstack.org/73145 | 00:49 |
| greghaynes | StevenK: It doesnt | 00:49 |
| lifeless | ok omg maybe I get to code today | 00:50 |
| StevenK | greghaynes: \o/ | 00:50 |
| lifeless | meetings done | 00:50 |
| lifeless | email done | 00:50 |
| StevenK | lifeless: No. No coding for you. | 00:50 |
| lifeless | breakfast done; need a sanity break. | 00:50 |
| lifeless | StevenK: wwaaahh | 00:50 |
| StevenK | Haha | 00:50 |
| *** UtahDave has quit IRC | 00:51 | |
| *** ccrouch has quit IRC | 00:52 | |
| lifeless | StevenK: reviewed https://review.openstack.org/#/c/73145/ | 01:02 |
| *** bauzas has quit IRC | 01:03 | |
| saurabhs | SpamapS : nova/os-refresh-config/configure.d/10-nova-state says : 'install -D -d -o nova -g nova -m 0775 /mnt/state/var/lib/nova' is it just creating that empty directory with correct permissions there or its doing something more than that? | 01:04 |
| StevenK | lifeless: arch is already present | 01:04 |
| StevenK | power-manager comes in a later branch | 01:05 |
| StevenK | lifeless: So, I was considering pulling out MACs in setup-baremetal, but I got stuck on trying to work out array indexs | 01:07 |
| lifeless | StevenK: want me to have a noodle on it? | 01:07 |
| StevenK | lifeless: Sure | 01:07 |
| *** killer_prince has quit IRC | 01:08 | |
| SpamapS | saurabhs: install -D just creates dirs yes | 01:09 |
| lifeless | StevenK: ok will in a little | 01:10 |
| *** bauzas has joined #tripleo | 01:12 | |
| *** killer_prince has joined #tripleo | 01:15 | |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-incubator: Enable horizon on overcloud control node https://review.openstack.org/74603 | 01:16 |
| StevenK | lifeless: Re: jq, precise->saucy has 1.2, which has -r | 01:19 |
| *** bauzas has quit IRC | 01:27 | |
| lifeless | StevenK: only in backports | 01:29 |
| *** saurabhs has quit IRC | 01:29 | |
| *** xuhanp has joined #tripleo | 01:36 | |
| *** cd-undercloud has joined #tripleo | 01:37 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 01:37 |
| *** cd-undercloud has quit IRC | 01:37 | |
| *** nosnos has joined #tripleo | 01:41 | |
| StevenK | lifeless: Huh. I should learn to read. precise and quantal, only in backports, not in raring at all, saucy and trusty universe | 01:55 |
| *** cd-undercloud has joined #tripleo | 02:07 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 02:07 |
| *** cd-undercloud has quit IRC | 02:07 | |
| lifeless | StevenK: ok, I have the lock on your patch | 02:31 |
| StevenK | lifeless: I am just about to meet up with tchaypo so I can spin him up, so have at it | 02:32 |
| lifeless | heh | 02:32 |
| lifeless | /home/robertc/work/tripleo-incubator/scripts/create-nodes: line 44: scripts/get-vm-mac: No such file or directory | 02:32 |
| lifeless | first issue :P | 02:32 |
| StevenK | So I had to do that to get it work in a sub-shell | 02:33 |
| lifeless | its in the path | 02:33 |
| lifeless | removing the scripts/ prefix fixes for me, did that not work for you? | 02:34 |
| StevenK | Without scripts/ it gave no such file or directory, I think | 02:35 |
| StevenK | lifeless: Let me push up my rebase, which reverts part of undercloud that you questioned? | 02:36 |
| lifeless | StevenK: sure | 02:36 |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Store node details in the JSON datafile https://review.openstack.org/73145 | 02:37 |
| *** cd-undercloud has joined #tripleo | 02:41 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 02:41 |
| *** cd-undercloud has quit IRC | 02:41 | |
| *** cd-undercloud has joined #tripleo | 02:44 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 02:44 |
| *** cd-undercloud has quit IRC | 02:44 | |
| *** sseago_ has quit IRC | 03:01 | |
| *** sseago_ has joined #tripleo | 03:02 | |
| *** jcooley_ has quit IRC | 03:23 | |
| *** jcooley_ has joined #tripleo | 03:27 | |
| *** ramishra_ has joined #tripleo | 03:27 | |
| *** cody-somerville has joined #tripleo | 03:29 | |
| *** cd-undercloud has joined #tripleo | 03:30 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 03:30 |
| *** cd-undercloud has quit IRC | 03:30 | |
| *** CaptTofu has quit IRC | 03:35 | |
| *** newell has quit IRC | 03:35 | |
| *** matsuhashi has quit IRC | 03:37 | |
| *** ccrouch has joined #tripleo | 03:56 | |
| *** cd-undercloud has joined #tripleo | 04:05 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 04:05 |
| *** cd-undercloud has quit IRC | 04:05 | |
| lifeless | NobodyCam: around ? | 04:10 |
| lifeless | NobodyCam: what does Ironic want for the virtual power driver - path to a key, or the keys contents? | 04:18 |
| lifeless | NobodyCam: I'm going to guess at key contents | 04:18 |
| *** rha has quit IRC | 04:20 | |
| *** rha has joined #tripleo | 04:21 | |
| *** cd-undercloud has joined #tripleo | 04:27 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 04:27 |
| *** cd-undercloud has quit IRC | 04:27 | |
| *** jcooley_ has quit IRC | 04:27 | |
| *** akuznetsov has joined #tripleo | 04:28 | |
| *** xuhanp has quit IRC | 04:32 | |
| *** matsuhashi has joined #tripleo | 04:32 | |
| *** xuhanp has joined #tripleo | 04:34 | |
| *** xuhanp has quit IRC | 04:39 | |
| *** xuhanp has joined #tripleo | 04:50 | |
| *** jcooley_ has joined #tripleo | 04:55 | |
| *** xuhanp has quit IRC | 04:55 | |
| *** cd-undercloud has joined #tripleo | 04:56 | |
| cd-undercloud | ************** overcloud complete status=6 ************ | 04:56 |
| *** cd-undercloud has quit IRC | 04:56 | |
| *** boris-42 has quit IRC | 04:57 | |
| *** boris-42 has joined #tripleo | 04:59 | |
| *** jcooley_ has quit IRC | 05:00 | |
| *** jcooley_ has joined #tripleo | 05:01 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Store node details in the JSON datafile https://review.openstack.org/73145 | 05:07 |
| lifeless | StevenK: ^thats my concept | 05:07 |
| lifeless | StevenK: stir to taste | 05:07 |
| *** xuhanp has joined #tripleo | 05:07 | |
| lifeless | StevenK: I really did just mean to address the range offset stuff, but PM was in the way, so I moved it in, and that lead to another thing etc;) | 05:08 |
| lifeless | StevenK: there's a little sci-fi in the help for setup-baremetal now | 05:08 |
| *** killer_prince has quit IRC | 05:09 | |
| *** coolsvap has joined #tripleo | 05:11 | |
| *** jcooley_ has quit IRC | 05:12 | |
| *** killer_prince has joined #tripleo | 05:14 | |
| *** jcooley_ has joined #tripleo | 05:18 | |
| *** killer_prince has quit IRC | 05:19 | |
| *** ramishra_ has quit IRC | 05:19 | |
| *** jcooley_ has quit IRC | 05:21 | |
| *** rpodolyaka1 has joined #tripleo | 05:22 | |
| *** xuhanp has quit IRC | 05:27 | |
| *** killer_prince has joined #tripleo | 05:28 | |
| *** ramishra has joined #tripleo | 05:29 | |
| *** cd-undercloud has joined #tripleo | 05:32 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 05:32 |
| *** cd-undercloud has quit IRC | 05:32 | |
| *** CaptTofu has joined #tripleo | 05:36 | |
| *** killer_prince2 has joined #tripleo | 05:37 | |
| *** xuhanp has joined #tripleo | 05:39 | |
| *** CaptTofu has quit IRC | 05:41 | |
| *** sdake_ has quit IRC | 05:47 | |
| *** saurabhs has joined #tripleo | 05:50 | |
| *** xuhanp has quit IRC | 05:52 | |
| *** xuhanp has joined #tripleo | 05:53 | |
| *** killer_prince has quit IRC | 05:57 | |
| *** killer_prince has joined #tripleo | 05:58 | |
| *** ramishra has quit IRC | 06:07 | |
| *** cd-undercloud has joined #tripleo | 06:09 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 06:09 |
| *** cd-undercloud has quit IRC | 06:09 | |
| *** vkozhukalov_ has joined #tripleo | 06:10 | |
| lifeless | greghaynes: so if I want to dogpile on HA db | 06:13 |
| lifeless | greghaynes: where should I start? open reviews - which ones? + what needs doing / testing / etc? | 06:13 |
| greghaynes | https://review.openstack.org/#/c/75540/ | 06:13 |
| StevenK | lifeless: You have some trailing whitespace around the setup-baremetal calls, but it looks really good. | 06:14 |
| greghaynes | theres a todo in https://review.openstack.org/#/c/75540/5/elements/mysql/install.d/10-mysql which im not sure if we should address | 06:14 |
| StevenK | lifeless: The .nodes[] - [.nodes[0]] is a bit scary | 06:14 |
| greghaynes | Then theres also the issue of - we currently install mariadb on fedora and this installs by tgz | 06:15 |
| *** vkozhukalov_ has quit IRC | 06:16 | |
| greghaynes | All this is just for one node setup, https://review.openstack.org/#/c/74860/ is needed along with other things to let us actually do the multi-node | 06:19 |
| *** ramishra has joined #tripleo | 06:19 | |
| *** jcooley_ has joined #tripleo | 06:19 | |
| *** rwsu has quit IRC | 06:27 | |
| *** cd-undercloud has joined #tripleo | 06:31 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 06:31 |
| *** cd-undercloud has quit IRC | 06:31 | |
| StevenK | lifeless: Your fallback seems to be a bad plan | 06:31 |
| *** rlandy has joined #tripleo | 06:32 | |
| StevenK | It might even work locally with the new style | 06:32 |
| *** jcooley_ has quit IRC | 06:37 | |
| *** pblaho has joined #tripleo | 06:47 | |
| *** rcarrillocruz has joined #tripleo | 06:48 | |
| *** rcarrillocruz1 has quit IRC | 06:50 | |
| *** saju_m has joined #tripleo | 06:50 | |
| *** cd-undercloud has joined #tripleo | 06:52 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 06:52 |
| *** cd-undercloud has quit IRC | 06:52 | |
| *** xuhanp has quit IRC | 06:58 | |
| *** xuhanp has joined #tripleo | 07:03 | |
| lifeless | StevenK: how so ? | 07:03 |
| lifeless | greghaynes: looking | 07:04 |
| *** rcarrillocruz1 has joined #tripleo | 07:06 | |
| *** sdake_ has joined #tripleo | 07:06 | |
| * SpamapS thinking that the hot-software-config stuff may actually get us to within < 100 lines of code change to get graceful updates | 07:07 | |
| lifeless | cool | 07:07 |
| *** rcarrillocruz has quit IRC | 07:07 | |
| lifeless | greghaynes: so, the t-h-t thing, I've -1'd, explanation in the review | 07:07 |
| SpamapS | and the best part is we don't actually have to switch to HOT to use it | 07:07 |
| lifeless | greghaynes: let me know if it scans, or if I've missed something | 07:07 |
| lifeless | SpamapS: fuckyeah, $nation | 07:07 |
| SpamapS | since it is just new types | 07:08 |
| lifeless | is percona in debian *and* Ubuntu now ? | 07:08 |
| lifeless | hmm, we're going to run into user id issues in /mnt/state data at some point | 07:09 |
| lifeless | I see a user-assertion thing being needed | 07:09 |
| lifeless | sadface | 07:09 |
| *** rcarrillocruz has joined #tripleo | 07:10 | |
| greghaynes | lifeless: agree with that -1 | 07:12 |
| SpamapS | lifeless: Percona no. | 07:12 |
| lifeless | greghaynes: ok so ^ per spamaps. | 07:12 |
| *** rcarrillocruz1 has quit IRC | 07:12 | |
| SpamapS | lifeless: Percona has just been uploaded to NEW in Debian about 2 weeks ago. | 07:12 |
| lifeless | greghaynes: I think we should just use the tarball thing to install always | 07:12 |
| greghaynes | sounds good | 07:13 |
| SpamapS | lifeless: regarding the user-id thing.. Could we symlink /etc/passwd -> /mnt/state/etc/passwd ? | 07:13 |
| lifeless | greghaynes: since an apt package from a third-party repo isn't actually any better | 07:13 |
| lifeless | SpamapS: no (think it through) | 07:13 |
| SpamapS | lifeless: or maybe nss lets us specify multiple passwd/shadow files | 07:13 |
| greghaynes | Yes, I need to do some digging into how to actually prevent installing mariadb in fedora (and likely suse) | 07:14 |
| *** jcooley_ has joined #tripleo | 07:15 | |
| lifeless | greghaynes: prevent it? what triggers it? | 07:17 |
| lifeless | SpamapS: so the issue is goingto be skew either way | 07:17 |
| lifeless | SpamapS: we eithe rskew in /mnt/* or in / | 07:17 |
| lifeless | SpamapS: and one we can fixup :) | 07:17 |
| greghaynes | I just saw it flash by while watching the image build | 07:17 |
| greghaynes | Not sure why | 07:17 |
| clarkb | mariadb is the db in fedora | 07:18 |
| clarkb | you ahve to do extra special community contrib stuff to get mysql in fedora now | 07:18 |
| *** mrunge has joined #tripleo | 07:19 | |
| clarkb | s/ db / mysql derivative db / | 07:19 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Fixing error in readme https://review.openstack.org/75276 | 07:19 |
| *** ifarkas has joined #tripleo | 07:20 | |
| SpamapS | lifeless: I do see what you mean. Would it be too simple if we just select static userids in a given range in os-svc-daemon? Something like 800 - 900, and just pass it in the call to os-svc-daemon? | 07:21 |
| lifeless | SpamapS: its very debian-90's :P | 07:21 |
| *** jcooley_ has quit IRC | 07:21 | |
| *** ramishra has quit IRC | 07:21 | |
| SpamapS | lifeless: aye, but we have a need for them to be stable. | 07:22 |
| greghaynes | lifeless: left a comment on https://review.openstack.org/#/c/75540/5/elements/mysql/install.d/10-mysql | 07:22 |
| greghaynes | woo for -1ing my own code | 07:22 |
| greghaynes | the adding vars to source-repositories doesnt 'just work' either because of amd64 vs x86_64 :( | 07:23 |
| lifeless | greghaynes: oh fugly | 07:24 |
| lifeless | erm | 07:24 |
| lifeless | you can do a conditional thingy | 07:24 |
| lifeless | let me look in a sec | 07:24 |
| greghaynes | Yea, in that install script I can... kind of baloons up complexity for modifying source-repositories | 07:24 |
| *** cd-undercloud has joined #tripleo | 07:27 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 07:27 |
| *** cd-undercloud has quit IRC | 07:27 | |
| *** ramishra has joined #tripleo | 07:29 | |
| lifeless | greghaynes: so when do we want amd64 vs x86_64 ? | 07:29 |
| greghaynes | the percona download url wants x86_64, but DIB defines $ARCH as amd64 | 07:30 |
| lifeless | so, define FEDORA_ARCH simulataneously or something in evnironment.d ? | 07:32 |
| greghaynes | aha! | 07:32 |
| * greghaynes is new to the whole DIB thing | 07:32 | |
| lifeless | it may not be all glued up for that yet, but it should be close | 07:32 |
| greghaynes | ok, well late here | 07:33 |
| lifeless | night! | 07:33 |
| greghaynes | gn | 07:33 |
| lifeless | I'm not around much tomorrow | 07:33 |
| *** rpodolyaka1 has quit IRC | 07:34 | |
| greghaynes | good timing catching up now, then :) | 07:34 |
| *** CaptTofu has joined #tripleo | 07:36 | |
| *** jcoufal has joined #tripleo | 07:38 | |
| *** CaptTofu has quit IRC | 07:41 | |
| *** jang1 has quit IRC | 07:46 | |
| *** vkozhukalov_ has joined #tripleo | 07:48 | |
| *** jprovazn has joined #tripleo | 07:51 | |
| lifeless | StevenK: I don't know what you mean | 07:58 |
| *** akuznetsov has quit IRC | 07:58 | |
| *** e0ne has joined #tripleo | 07:58 | |
| *** ramishra has quit IRC | 08:00 | |
| *** akuznetsov has joined #tripleo | 08:00 | |
| *** cd-undercloud has joined #tripleo | 08:06 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 08:06 |
| *** cd-undercloud has quit IRC | 08:06 | |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-heat-templates: Add rabbit.cookie attribute https://review.openstack.org/70238 | 08:08 |
| rpodolyaka | morning all | 08:09 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Update openstack config files to support rabbitmq cluster https://review.openstack.org/68393 | 08:10 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Fix rabbitmq env file name and log path https://review.openstack.org/70236 | 08:10 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Add cluster support for rabbitmq element https://review.openstack.org/68392 | 08:10 |
| jprovazn | Hi tripleo folks, if you have a minute, could you please review at the above patches above? | 08:18 |
| jprovazn | s/review at the above/review/ | 08:18 |
| rpodolyaka | jprovazn: sure! | 08:20 |
| jprovazn | rpodolyaka: thanks | 08:20 |
| rpodolyaka | np | 08:20 |
| *** bauzas has joined #tripleo | 08:32 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Enable openvswitch https://review.openstack.org/75036 | 08:33 |
| *** cd-undercloud has joined #tripleo | 08:34 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 08:34 |
| *** cd-undercloud has quit IRC | 08:34 | |
| *** andreaf has joined #tripleo | 08:39 | |
| *** d0ugal has joined #tripleo | 08:40 | |
| *** saurabhs has left #tripleo | 08:43 | |
| *** jistr has joined #tripleo | 08:49 | |
| openstackgerrit | A change was merged to openstack/tuskar: Handle overcloud-block-storage in heat template https://review.openstack.org/74969 | 08:50 |
| *** killer_prince has quit IRC | 08:51 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Update pypi element to suggest pypi-mirror https://review.openstack.org/74679 | 08:52 |
| *** rsacharya_ is now known as rsacharya | 08:52 | |
| *** xuhanp has quit IRC | 08:53 | |
| *** ramishra has joined #tripleo | 09:00 | |
| *** killer_pr has joined #tripleo | 09:03 | |
| *** killer_pr is now known as killer_prince | 09:03 | |
| *** jcooley_ has joined #tripleo | 09:03 | |
| *** cd-undercloud has joined #tripleo | 09:09 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 09:09 |
| *** cd-undercloud has quit IRC | 09:09 | |
| *** jcooley_ has quit IRC | 09:09 | |
| *** rdopieralski has joined #tripleo | 09:11 | |
| *** derekh has joined #tripleo | 09:14 | |
| *** marun has quit IRC | 09:15 | |
| *** killer_prince has quit IRC | 09:18 | |
| *** lazy_prince has joined #tripleo | 09:18 | |
| *** lazy_prince is now known as killer_prince | 09:18 | |
| *** jcooley_ has joined #tripleo | 09:20 | |
| *** pbelanyi has joined #tripleo | 09:21 | |
| *** athomas has quit IRC | 09:22 | |
| *** hashar has joined #tripleo | 09:22 | |
| *** hashar has quit IRC | 09:22 | |
| *** xuhanp has joined #tripleo | 09:23 | |
| *** jcooley_ has quit IRC | 09:25 | |
| *** rcarrillocruz1 has joined #tripleo | 09:25 | |
| *** hashar has joined #tripleo | 09:25 | |
| *** rcarrillocruz has quit IRC | 09:27 | |
| *** athomas has joined #tripleo | 09:37 | |
| *** martyntaylor has joined #tripleo | 09:37 | |
| *** CaptTofu has joined #tripleo | 09:37 | |
| *** CaptTofu has quit IRC | 09:42 | |
| *** lucasagomes has joined #tripleo | 09:43 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Enable horizon on overcloud control node https://review.openstack.org/74603 | 09:46 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Use local apt and pypi mirror for tripleo-cd/ https://review.openstack.org/75746 | 09:46 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Revert "Workaround broken Neutron requirements." https://review.openstack.org/76474 | 09:47 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Use local apt and pypi mirror for tripleo-cd/ https://review.openstack.org/75746 | 09:47 |
| *** gcha has joined #tripleo | 09:48 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Allow the testenv client to dictate the worker timeout https://review.openstack.org/76171 | 09:49 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Set the worker timeout https://review.openstack.org/76172 | 09:49 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Kill commands that outlive test env https://review.openstack.org/75402 | 09:49 |
| *** bogdando has quit IRC | 09:50 | |
| *** rcarrillocruz has joined #tripleo | 09:51 | |
| *** bogdando has joined #tripleo | 09:52 | |
| *** rcarrillocruz1 has quit IRC | 09:52 | |
| *** jp_at_hp has joined #tripleo | 09:52 | |
| *** ramishra has quit IRC | 09:57 | |
| *** jcooley_ has joined #tripleo | 09:59 | |
| *** rcarrillocruz1 has joined #tripleo | 10:00 | |
| *** rcarrillocruz has quit IRC | 10:03 | |
| *** shardy is now known as shardy_afk | 10:03 | |
| *** ramishra has joined #tripleo | 10:03 | |
| *** jcooley_ has quit IRC | 10:06 | |
| *** akrivoka has joined #tripleo | 10:27 | |
| *** jang has joined #tripleo | 10:30 | |
| *** rcarrillocruz has joined #tripleo | 10:39 | |
| *** rcarrillocruz1 has quit IRC | 10:42 | |
| *** cd-undercloud has joined #tripleo | 10:45 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 10:45 |
| *** cd-undercloud has quit IRC | 10:45 | |
| *** ramishra has quit IRC | 10:55 | |
| *** xuhanp has quit IRC | 10:57 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Use local apt and pypi mirror for tripleo-cd https://review.openstack.org/75746 | 11:00 |
| killer_prince2 | hi all | 11:09 |
| killer_prince2 | I am trying to use first boot hooks for rhel. has anyone tried it before.. | 11:10 |
| killer_prince2 | I can see that the rhel element is not running anything specific to first boot hooks in finalise step... but ubuntu and fedora seems to run some script.. | 11:11 |
| *** gcha has quit IRC | 11:11 | |
| killer_prince2 | is that expected or is there any step that i am missing from my side.. | 11:12 |
| *** cd-undercloud has joined #tripleo | 11:12 | |
| cd-undercloud | ************** overcloud complete status=7 ************ | 11:12 |
| *** cd-undercloud has quit IRC | 11:12 | |
| killer_prince2 | oh.. forgot to mention that I am using dib with RHEL 6.5 | 11:12 |
| lifeless | killer_prince2: don't use first boot hooks; they are deprecated. | 11:13 |
| lifeless | killer_prince2: (see the docs which say that :)) | 11:13 |
| killer_prince2 | lifelessk.. is there any alternative that we can use..? I | 11:14 |
| killer_prince2 | I plan to run some script on first boot to perform some action on the disk. | 11:14 |
| lifeless | make it an idempotent action | 11:15 |
| lifeless | and use os-refresh-config to trigger it | 11:15 |
| lifeless | killer_prince2: what action specifically? | 11:15 |
| killer_prince2 | lifeless: lets say setting up grub on the disk.. for local boot support or even disk encryption | 11:16 |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/tuskar-ui: Manage image ID's for node profile https://review.openstack.org/75823 | 11:17 |
| lifeless | killer_prince2: so, for the bootblock case you'll want to do that everytime grub is updated - make it a configure.d script | 11:18 |
| lifeless | killer_prince2: (and an optional element to do that for baremetal machines will be quite useful - so please do send a patch :)) | 11:18 |
| lifeless | killer_prince2: for encryption, you have a chicken and egg problem there, I think you'll need to change Ironic to tackle it :) | 11:19 |
| killer_prince2 | YEah.. I plan to do so.. | 11:20 |
| killer_prince2 | lifeless: I do not see configure.d being mentioned on https://github.com/openstack/diskimage-builder | 11:21 |
| lifeless | killer_prince2: look into os-refresh-config | 11:22 |
| *** rcarrillocruz1 has joined #tripleo | 11:22 | |
| *** rcarrillocruz has quit IRC | 11:25 | |
| killer_prince2 | not sure how it will be consumed by disk-image-builder. I do not plan to use it along-with tripleo-image-elements as it builds a dependency on it. I am currently trying to build RHEL 6.5 image with apache 2 using tripleo-image-elements but its not working for me.. its failing for os-refresh-config scripts due to some lxml issue. | 11:28 |
| *** killer_prince2 has quit IRC | 11:31 | |
| *** rcarrillocruz has joined #tripleo | 11:34 | |
| *** rcarrillocruz1 has quit IRC | 11:36 | |
| *** CaptTofu has joined #tripleo | 11:38 | |
| *** CaptTofu has quit IRC | 11:43 | |
| *** matsuhashi has quit IRC | 11:46 | |
| *** cd-undercloud has joined #tripleo | 11:47 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 11:47 |
| *** cd-undercloud has quit IRC | 11:47 | |
| *** jcooley_ has joined #tripleo | 11:48 | |
| openstackgerrit | Michael Kerrin proposed a change to openstack/diskimage-builder: Add utility to mount and unmount images https://review.openstack.org/76509 | 11:48 |
| *** rbrady-afk has quit IRC | 11:50 | |
| *** jcooley_ has quit IRC | 11:54 | |
| *** killer_prince has quit IRC | 11:57 | |
| *** CaptTofu has joined #tripleo | 12:02 | |
| *** coolsvap has quit IRC | 12:02 | |
| *** pbelanyi_ has joined #tripleo | 12:05 | |
| *** pbelanyi has quit IRC | 12:05 | |
| *** pbelanyi_ is now known as pbelanyi | 12:05 | |
| *** lucasagomes has quit IRC | 12:06 | |
| *** lucasagomes has joined #tripleo | 12:07 | |
| *** rcarrillocruz1 has joined #tripleo | 12:12 | |
| *** rcarrillocruz has quit IRC | 12:14 | |
| derekh | lifeless, devananda, Ng, SpamapS_, jog0_, GheRivero, derekh, dprince, slagle: ci appears down tracking here https://etherpad.openstack.org/p/KyfhpxVgOw | 12:23 |
| *** killer_prince has joined #tripleo | 12:24 | |
| lifeless | jprovazn: http://paste.openstack.org/show/69307/ looks sane to me | 12:24 |
| *** cody-somerville has quit IRC | 12:24 | |
| *** hashar has quit IRC | 12:24 | |
| *** cd-undercloud has joined #tripleo | 12:24 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 12:24 |
| *** cd-undercloud has quit IRC | 12:24 | |
| lifeless | derekh: looks up to me | 12:24 |
| lifeless | derekh: nova list works | 12:25 |
| lifeless | derekh: lots of live slaves | 12:25 |
| derekh | lifeless: yes, cloud is up, geard doesn't seem to be doing its job | 12:25 |
| lifeless | derekh: ah, righto | 12:25 |
| jprovazn | lifeless: cool, could you please take a look at implementation of it? https://review.openstack.org/#/c/68392/ - I'd really love to finally get rabbitmq HA in | 12:25 |
| lifeless | I can ssh into the broker | 12:25 |
| *** lucasagomes is now known as lucas-hungry | 12:25 | |
| derekh | lifeless: I don't seem to have the right key to log onto broker, can you? | 12:26 |
| lifeless | derekh: yes, have done | 12:26 |
| jprovazn | lifeless: it's markes is WIP, but just because of cross repo dep | 12:26 |
| lifeless | derekh: your key must not be in the cd admins keylist | 12:26 |
| lifeless | nope | 12:27 |
| lifeless | I must have failed at registerng keys | 12:27 |
| lifeless | anyhow, a) fixing | 12:27 |
| lifeless | derekh: try logging in now | 12:29 |
| derekh | lifeless: im in, thanks | 12:29 |
| lifeless | telnet localhost 4730 | 12:29 |
| lifeless | Trying 127.0.0.1... | 12:29 |
| lifeless | Connected to localhost. | 12:29 |
| lifeless | Escape character is '^]'. | 12:29 |
| lifeless | status | 12:29 |
| lifeless | Connection closed by foreign host. | 12:29 |
| lifeless | it is status right, to debug? | 12:29 |
| derekh | lifeless: error: [Errno 24] Too many open files | 12:30 |
| lifeless | restarting geard | 12:30 |
| lifeless | ok, thats happier | 12:30 |
| lifeless | so lets file a bug on geard | 12:30 |
| derekh | lifeless: yup, restart and I'll see if I can find the leak | 12:30 |
| lifeless | clarkb: where do such bugs go ? | 12:30 |
| lifeless | we may have a few bad jobs go through | 12:31 |
| lifeless | because state will be confused | 12:31 |
| lifeless | dunno, we'll find out I guess ;) | 12:31 |
| lifeless | derekh: thanks for noting! | 12:31 |
| derekh | lifeless: I guess, we should rotate that log file | 12:32 |
| derekh | -rw-r----- 1 root root 15G Feb 26 12:30 geard.log | 12:32 |
| lifeless | lol | 12:32 |
| derekh | lifeless: hmm, it is being rotated, I guess once it hit the error it kept spitting out logs | 12:33 |
| lifeless | jprovazn: yes, my friday thugh - am off tomorrow/today:P | 12:35 |
| lifeless | jprovazn: I'm keen too | 12:35 |
| jprovazn | lifeless: yea - I was thinking that it's extremely early morning for you now :) | 12:36 |
| lifeless | jprovazn: btw - https://pypi.python.org/pypi/concoord looks quite interesting | 12:38 |
| * jprovazn looks | 12:39 | |
| lifeless | and/or https://pypi.python.org/pypi/essential-paxos | 12:39 |
| lifeless | looking at building blocks for doing a correct 'build a cluster from zero knowledge other than the intended peers in the cluster' | 12:40 |
| lifeless | jprovazn: I found it from http://openreplica.org which is cool too, in aneven more way out way :) | 12:41 |
| lifeless | jprovazn: sanlock would do what we need... but only if one has a san :) | 12:42 |
| *** CaptTofu has quit IRC | 12:43 | |
| *** CaptTofu has joined #tripleo | 12:43 | |
| openstackgerrit | Fengqian.gao proposed a change to openstack/tuskar: Python 3: replace "im_self" by "__self__" https://review.openstack.org/76522 | 12:44 |
| *** jdob has joined #tripleo | 12:44 | |
| *** cd-undercloud has joined #tripleo | 12:45 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 12:45 |
| *** cd-undercloud has quit IRC | 12:45 | |
| jprovazn | lifeless: TBH, after reading basic info about openreplica and concoord, I'm still not sure how this fits into any cluster setup in TripleO. Most probably because I'm not sure, what are consequences/requirements of "build a cluster from zero knowledge other than the intended peers in the cluster". | 12:52 |
| lifeless | jprovazn: solving the 'and we use the node with 0 in its name to be the seed' ugliness | 12:56 |
| lifeless | jprovazn: aka master election | 12:56 |
| lifeless | jprovazn: ditto for coordinating 'db sync' execution. | 13:00 |
| jprovazn | lifeless: ah, I see - so you would create a simple replicated python object which would keep info about nodes in cluster. If a node starts, it would save itself into a list of nodes in cluster (saved as the replicated object) | 13:00 |
| lifeless | handwaving-level, but yes. | 13:00 |
| lifeless | not sure if we'd store the node list in the objet | 13:00 |
| lifeless | or state | 13:01 |
| lifeless | like | 13:01 |
| lifeless | 'we are ready to intiialize the clsuter' | 13:01 |
| lifeless | 'we have initialized the cluster and folk can use it now' | 13:01 |
| lifeless | 'we are ready to run nova db sync' | 13:01 |
| lifeless | and the master node is the one that would react to the message and actually do the work | 13:02 |
| lifeless | the master would move around as needed | 13:02 |
| *** killer_prince is now known as lazy_prince | 13:05 | |
| jprovazn | lifeless: yes, interesting idea | 13:08 |
| *** marun has joined #tripleo | 13:08 | |
| *** killer_p- has joined #tripleo | 13:08 | |
| *** killer_p- is now known as killer_prince | 13:09 | |
| *** lazy_prince has quit IRC | 13:09 | |
| *** bauzas has quit IRC | 13:12 | |
| *** imbrandon has joined #tripleo | 13:14 | |
| *** imbrandon has left #tripleo | 13:14 | |
| *** e0ne has quit IRC | 13:21 | |
| *** e0ne has joined #tripleo | 13:21 | |
| *** cd-undercloud has joined #tripleo | 13:24 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 13:24 |
| *** cd-undercloud has quit IRC | 13:24 | |
| openstackgerrit | A change was merged to openstack/tuskar: Update context from Oslo https://review.openstack.org/76182 | 13:26 |
| *** sdake has quit IRC | 13:28 | |
| *** sdake has joined #tripleo | 13:28 | |
| *** gcha has joined #tripleo | 13:29 | |
| Ng | morning | 13:30 |
| * Ng imbibes a great deal of plague fighting remedies | 13:30 | |
| *** hashar has joined #tripleo | 13:31 | |
| *** jcooley_ has joined #tripleo | 13:36 | |
| rpodolyaka | morning Ng, get well! | 13:36 |
| Ng | rpodolyaka: thanks :) | 13:37 |
| *** dprince has joined #tripleo | 13:37 | |
| *** gcha has quit IRC | 13:37 | |
| *** nosnos has quit IRC | 13:41 | |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-incubator: WIP - Move wait_for for checking exit status to checking for condition to be true https://review.openstack.org/76537 | 13:44 |
| *** jistr is now known as jistr|english | 13:44 | |
| *** lucas-hungry is now known as lucasagomes | 13:48 | |
| *** rbrady has joined #tripleo | 13:49 | |
| *** gcha has joined #tripleo | 13:51 | |
| *** shardy_afk is now known as shardy | 13:52 | |
| openstackgerrit | Jon-Paul Sullivan proposed a change to openstack/diskimage-builder: Enable simple modification of git repo location https://review.openstack.org/74791 | 13:52 |
| *** mestery has quit IRC | 13:54 | |
| *** weshay has joined #tripleo | 13:59 | |
| *** cd-undercloud has joined #tripleo | 14:00 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 14:00 |
| *** cd-undercloud has quit IRC | 14:00 | |
| *** lblanchard has joined #tripleo | 14:02 | |
| *** bauzas has joined #tripleo | 14:02 | |
| *** mestery has joined #tripleo | 14:14 | |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Add waitcondition on test environments https://review.openstack.org/76264 | 14:16 |
| *** jcooley_ has quit IRC | 14:16 | |
| *** jprovazn has quit IRC | 14:17 | |
| *** jprovazn has joined #tripleo | 14:17 | |
| *** julim has joined #tripleo | 14:19 | |
| *** jprovazn has quit IRC | 14:21 | |
| *** matty_dubs|gone is now known as matty_dubs | 14:23 | |
| *** jprovazn has joined #tripleo | 14:24 | |
| *** e0ne_ has joined #tripleo | 14:32 | |
| *** e0ne has quit IRC | 14:36 | |
| *** cd-undercloud has joined #tripleo | 14:36 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 14:36 |
| *** cd-undercloud has quit IRC | 14:36 | |
| *** jistr|english is now known as jistr | 14:40 | |
| *** jpeeler has quit IRC | 14:40 | |
| *** hewbrocca has joined #tripleo | 14:41 | |
| *** gcha has quit IRC | 14:47 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Shutdown the worker callback client https://review.openstack.org/76546 | 14:50 |
| derekh | lifeless, devananda, Ng, SpamapS_, jog0_, GheRivero, derekh, dprince, slagle: ^^ the cause of this mornings geard outage | 14:51 |
| *** jcoufal has quit IRC | 14:52 | |
| dprince | derekh: seems fine (I +2d). Does it belong in a finally block though? | 14:57 |
| slagle | i was going to ask the same :) | 14:58 |
| slagle | all the code in run_callback could be in a try: finally: | 14:58 |
| *** e0ne has joined #tripleo | 15:01 | |
| *** saju_m has quit IRC | 15:02 | |
| *** hewbrocc` has joined #tripleo | 15:02 | |
| *** hewbrocca has quit IRC | 15:04 | |
| *** e0ne_ has quit IRC | 15:05 | |
| *** dkehn_ has left #tripleo | 15:05 | |
| derekh | dprince: slagle yup, it would be better in a try/finally , the 3 cases I tried (command success, command error and command times out), run the shutdown as needed, but yes the whole thing should be wrapped | 15:05 |
| *** sballe_ has joined #tripleo | 15:07 | |
| openstackgerrit | Jon-Paul Sullivan proposed a change to openstack/diskimage-builder: Enable custom apt.conf in apt-conf element https://review.openstack.org/74992 | 15:09 |
| *** sballe has quit IRC | 15:10 | |
| *** bauzas has quit IRC | 15:11 | |
| *** cd-undercloud has joined #tripleo | 15:14 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 15:14 |
| *** cd-undercloud has quit IRC | 15:14 | |
| *** pblaho` has joined #tripleo | 15:15 | |
| *** pblaho has quit IRC | 15:16 | |
| *** jrist_ has quit IRC | 15:17 | |
| *** jrist_ has joined #tripleo | 15:17 | |
| *** rwsu has joined #tripleo | 15:18 | |
| *** jpeeler has joined #tripleo | 15:19 | |
| NobodyCam | lifeless: also virtual power type. currently this is stored in seed-stack-config elements config.jason, would be great if it was in the $TE_DATAFILE. | 15:23 |
| tchaypo | surely sleepless isn't still online? | 15:24 |
| openstackgerrit | Stuart McLaren proposed a change to openstack/tripleo-image-elements: Glance: Allow logging to be more configurable https://review.openstack.org/76557 | 15:25 |
| NobodyCam | tchaypo: I hope not... Juse answering questions he posted to me lastnight | 15:27 |
| NobodyCam | while I was asleep :) | 15:27 |
| *** pblaho` is now known as pblaho | 15:27 | |
| rbrady | derekh: ping | 15:27 |
| derekh | rbrady: ack | 15:28 |
| *** hewbrocc` is now known as hewbrocca | 15:28 | |
| rbrady | derekh: I keep getting random errors on check-tripleo-overcloud-precise | 15:29 |
| derekh | rbrady: which review are you looking at ? | 15:30 |
| rbrady | derekh: latest log from changing the map-packages in the fedora element: http://logs.openstack.org/98/76398/1/check-tripleo/check-tripleo-overcloud-precise/f91d882/console.html | 15:30 |
| rbrady | https://review.openstack.org/#/c/76398/ | 15:30 |
| *** hewbrocca has quit IRC | 15:36 | |
| *** jprovazn_ has joined #tripleo | 15:36 | |
| derekh | rbrady: ya, I've seen them two, havn't looked into tracking it down yet though, it seems to happen on some disk image builds, but not consistently reproducible the once you pointed at is an error building a seed which is weird because the seed built ok in the other two jobs in the review | 15:37 |
| rbrady | derekh: I just wanted to make sure this was noted to someone closer to the CI stuff. I can keep commenting "recheck no bug" for now | 15:38 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add missing +x https://review.openstack.org/74468 | 15:38 |
| *** jprovazn_ has quit IRC | 15:38 | |
| *** jprovazn has quit IRC | 15:39 | |
| openstackgerrit | Jon-Paul Sullivan proposed a change to openstack/tripleo-incubator: Add source-repository element var setting script https://review.openstack.org/74393 | 15:39 |
| derekh | rbrady: ok, probably the best thing todo for now, in the meantime if anybody can reproduce locally it would be great to try and track it down http://logs.openstack.org/98/76398/1/check-tripleo/check-tripleo-overcloud-precise/f91d882/console.html | 15:40 |
| derekh | rbrady: can you file it in launchpad aswell so it gets some attention | 15:41 |
| slagle | pblaho: hi, we're holding off on +A'ing reviews until we get successful CI runs | 15:41 |
| derekh | rbrady: looks like I lied, the error was during the build of the overcloud notcompute image | 15:44 |
| rbrady | derekh: I will dig up the failures from the patches I have now and post them into a bug in launchpad | 15:44 |
| derekh | rbrady: cool, thanks | 15:45 |
| derekh | rbrady: I got another error I'm trying the track down, then will see if I can reproduce | 15:45 |
| *** e0ne has quit IRC | 15:50 | |
| *** e0ne has joined #tripleo | 15:50 | |
| *** cd-undercloud has joined #tripleo | 15:50 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 15:50 |
| *** cd-undercloud has quit IRC | 15:50 | |
| *** rbrady has quit IRC | 15:50 | |
| *** cd-undercloud has joined #tripleo | 15:51 | |
| cd-undercloud | ************** overcloud complete status=6 ************ | 15:51 |
| *** cd-undercloud has quit IRC | 15:51 | |
| *** dprince has quit IRC | 15:53 | |
| *** rbrady has joined #tripleo | 15:53 | |
| *** dprince has joined #tripleo | 15:53 | |
| NobodyCam | StevenK: did you get your work email? I'm getting a bounce from stevenk@mangled.wedontsleep.org | 15:54 |
| *** jrist_ is now known as jrist | 15:55 | |
| *** rlandy has quit IRC | 15:58 | |
| *** julim_ has joined #tripleo | 16:05 | |
| *** julim has quit IRC | 16:05 | |
| tchaypo | NobodyCam: I can confirm that he got access to it - he took me into the office today and helped me set mine up | 16:07 |
| *** jprovazn has joined #tripleo | 16:07 | |
| *** pbelanyi has quit IRC | 16:08 | |
| NobodyCam | tchaypo: TY :) will search in the directory | 16:08 |
| *** e0ne_ has joined #tripleo | 16:12 | |
| NobodyCam | tchaypo: Ty Found him!! :) | 16:13 |
| *** e0ne has quit IRC | 16:16 | |
| *** stevebaker has quit IRC | 16:18 | |
| *** cody-somerville has joined #tripleo | 16:24 | |
| *** cody-somerville has joined #tripleo | 16:24 | |
| openstackgerrit | Stuart McLaren proposed a change to openstack/tripleo-image-elements: Glance: Allow logging to be more configurable https://review.openstack.org/76557 | 16:25 |
| *** cd-undercloud has joined #tripleo | 16:28 | |
| cd-undercloud | ************** overcloud complete status=7 ************ | 16:28 |
| *** cd-undercloud has quit IRC | 16:28 | |
| *** bauzas has joined #tripleo | 16:28 | |
| *** cody-somerville has quit IRC | 16:29 | |
| *** jcooley_ has joined #tripleo | 16:31 | |
| *** ifarkas has quit IRC | 16:32 | |
| *** rsacharya has left #tripleo | 16:37 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Link db sync utilities to /usr/local/bin... https://review.openstack.org/76579 | 16:40 |
| openstackgerrit | Stuart McLaren proposed a change to openstack/tripleo-image-elements: Glance: Make API versions configurable https://review.openstack.org/76580 | 16:40 |
| *** rsacharya has joined #tripleo | 16:41 | |
| openstackgerrit | A change was merged to openstack/tuskar: Python 3: replace "im_self" by "__self__" https://review.openstack.org/76522 | 16:41 |
| *** shardy is now known as shardy_afk | 16:43 | |
| *** CaptTofu has quit IRC | 16:45 | |
| *** rdopieralski has quit IRC | 16:48 | |
| *** CaptTofu has joined #tripleo | 16:51 | |
| *** hashar has quit IRC | 16:56 | |
| *** cadenzajon has joined #tripleo | 16:56 | |
| *** coolsvap has joined #tripleo | 16:56 | |
| *** jistr has quit IRC | 16:57 | |
| *** killer_prince is now known as lazy_prince | 17:01 | |
| *** rpodolyaka has left #tripleo | 17:03 | |
| *** mrunge has quit IRC | 17:03 | |
| *** rpodolyaka has joined #tripleo | 17:03 | |
| *** ftcjeff has joined #tripleo | 17:03 | |
| *** e0ne_ has quit IRC | 17:05 | |
| *** cd-undercloud has joined #tripleo | 17:06 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 17:06 |
| *** cd-undercloud has quit IRC | 17:06 | |
| *** saju_m has joined #tripleo | 17:06 | |
| *** lazy_prince is now known as killer_prince | 17:08 | |
| *** sseago_ has quit IRC | 17:23 | |
| *** sseago_ has joined #tripleo | 17:23 | |
| *** coolsvap has quit IRC | 17:25 | |
| *** coolsvap has joined #tripleo | 17:25 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Link db sync utilities to /usr/local/bin... https://review.openstack.org/76579 | 17:26 |
| *** cody-somerville has joined #tripleo | 17:31 | |
| *** cody-somerville has quit IRC | 17:31 | |
| *** cody-somerville has joined #tripleo | 17:31 | |
| *** vkozhukalov_ has quit IRC | 17:32 | |
| *** coolsvap has quit IRC | 17:34 | |
| *** coolsvap has joined #tripleo | 17:35 | |
| *** cd-undercloud has joined #tripleo | 17:39 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 17:39 |
| *** cd-undercloud has quit IRC | 17:39 | |
| *** cd-undercloud has joined #tripleo | 17:40 | |
| cd-undercloud | ************** overcloud complete status=6 ************ | 17:40 |
| *** cd-undercloud has quit IRC | 17:40 | |
| *** matty_dubs is now known as matty_dubs|lunch | 17:49 | |
| *** jcooley_ has quit IRC | 17:52 | |
| *** jcooley_ has joined #tripleo | 17:56 | |
| *** jcooley_ has quit IRC | 18:01 | |
| *** bcrochet has quit IRC | 18:01 | |
| *** bcrochet has joined #tripleo | 18:03 | |
| *** killer_prince has quit IRC | 18:06 | |
| *** jeremydei has quit IRC | 18:08 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Stop using the os-svc-install -n option. https://review.openstack.org/76604 | 18:08 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Drop the os-svc-install -n -c options. https://review.openstack.org/76605 | 18:08 |
| *** andreaf has quit IRC | 18:08 | |
| SpamapS | derekh: https://review.openstack.org/#/c/76546/ all good to +A now? it looks sane to me. | 18:09 |
| *** andreaf has joined #tripleo | 18:10 | |
| slagle | the CI jobs are running for it | 18:10 |
| derekh | SpamapS: yup, although I think we're supposed to be waiting on 3 successful jobs, a log of the overcloud jobs seem to be failing, trying to figure out why at the moment | 18:10 |
| SpamapS | derekh: is tripleo-ci in the ci? I only see a noop job. | 18:11 |
| *** newell has joined #tripleo | 18:12 | |
| *** killer_prince has joined #tripleo | 18:12 | |
| derekh | SpamapS: yup, its a separate queue so returns separately, although the ci jobs will go nowhere near testing this change | 18:13 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-image-elements: Install latest version of pip and virtualenv https://review.openstack.org/75071 | 18:13 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Testing a hunch https://review.openstack.org/76607 | 18:16 |
| * derekh will check back later to see if that sheds any light to the overcloud ci issues | 18:16 | |
| *** e0ne has joined #tripleo | 18:17 | |
| SpamapS | derekh: Ok.. I really don't understand at all how this works so I'll just leave my +2 :-P | 18:18 |
| derekh | SpamapS: no prob, thanks :-), I think I'll put something together to try and explain it all at the sprint | 18:19 |
| *** cody-somerville has quit IRC | 18:20 | |
| *** jcooley_ has joined #tripleo | 18:20 | |
| *** derekh has quit IRC | 18:21 | |
| jp_at_hp | Can any core reviewers take a look at https://review.openstack.org/#/c/74992/ please? | 18:21 |
| jp_at_hp | slagle: SpamapS ^^^ ?? | 18:21 |
| dprince | derekh: the most recent set of failures for me were due to Failed to fetch http://security.ubuntu.com.... errors. | 18:21 |
| dprince | derekh: undercloud and overcloud, is that what you are seeing as well? | 18:22 |
| *** andreaf has quit IRC | 18:22 | |
| dprince | derekh: send this at will sir https://review.openstack.org/#/c/76546/ | 18:22 |
| *** e0ne has quit IRC | 18:22 | |
| *** e0ne has joined #tripleo | 18:22 | |
| SpamapS | jp_at_hp: as you wish | 18:26 |
| * SpamapS will now become the dread pirate roberts | 18:26 | |
| *** cd-undercloud has joined #tripleo | 18:26 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 18:26 |
| *** cd-undercloud has quit IRC | 18:26 | |
| SpamapS | dprince: heh, time for local apt-mirrors :) | 18:27 |
| dprince | SpamapS: yeah. we spoke of that yesterday as well. | 18:27 |
| dprince | SpamapS: we'll need Fedora too... so may as well do that too | 18:27 |
| *** killer_prince has quit IRC | 18:27 | |
| *** lazy_prince has joined #tripleo | 18:28 | |
| *** lazy_prince is now known as killer_prince | 18:28 | |
| *** saju_m has quit IRC | 18:35 | |
| SpamapS | dprince: we have debian-mirror and pypi-mirror elements already.. so probably not much work to get them up and running and then start feeding that into the process. | 18:35 |
| SpamapS | dprince: I presume mirroring fedora is also fairly straight forward | 18:36 |
| dprince | SpamapS: shouldn't be too bad. Although I haven't tested the pypi element on Fedora so there may be work there too | 18:36 |
| *** matty_dubs|lunch is now known as matty_dubs | 18:38 | |
| SpamapS | dprince: Both the 'pypi' element (consumes local mirror) and the pypi-mirror element (makes local mirror) are git/source/pip based and shouldn't have any ubuntu/debian-isms. As long as Fedora has cron.d I think it will be fine. | 18:38 |
| *** saurabhs has joined #tripleo | 18:39 | |
| SpamapS | looks like apt-mirror was orphaned out of fedora... hrm | 18:39 |
| dprince | SpamapS: yep. cron.d is there | 18:39 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-image-elements: Install latest version of pip and virtualenv https://review.openstack.org/75071 | 18:40 |
| *** lucasagomes has quit IRC | 18:40 | |
| *** akrivoka has quit IRC | 18:40 | |
| *** athomas has quit IRC | 18:40 | |
| *** jpeeler has quit IRC | 18:40 | |
| *** jpeeler has joined #tripleo | 18:41 | |
| dprince | slagle: could we use environment variables in the source-repository scripts (source-repository-nova for example) | 18:45 |
| *** cd-undercloud has joined #tripleo | 18:47 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 18:47 |
| *** cd-undercloud has quit IRC | 18:47 | |
| *** jang1 has joined #tripleo | 18:48 | |
| slagle | dprince: to do what? i don't think we can as it is today | 18:52 |
| *** julim_ has quit IRC | 18:53 | |
| dprince | slagle: just checking out the mechanism a bit today. I don't need to touch it ATM... | 18:53 |
| slagle | dprince: we do a "read line" now to get lines from the flie now | 18:53 |
| slagle | don't think that is going to interpret environment variables | 18:53 |
| slagle | maybe if we just eval it after that? | 18:54 |
| jp_at_hp | SpamapS: I take it the changes to the apt-conf element are being mirrored in the apt-sources element from whioch apt-conf was suggested? | 18:54 |
| SpamapS | slagle: greghaynes pointed out that envsubst would do nicely in that situation. | 18:55 |
| SpamapS | jp_at_hp: I know you're just cargo culting, but what a great opportunity to improve both. :) | 18:55 |
| *** pblaho has quit IRC | 18:56 | |
| SpamapS | jp_at_hp: if the other one is really full of all that fail, we should go back and improve on it. For now, we don't want to repeat that mistake for the third thing that cargo cults it and now we have _three_ bits of debt. | 18:56 |
| greghaynes | Sometimes its nice, sometimes eval echo "$foo" is nice | 18:56 |
| jp_at_hp | SpamapS: Can I discuss something that I disagree with in your comments? | 18:56 |
| jp_at_hp | I exit if the apt-conf element is included, but there is nothing specified for source to copy in. | 18:57 |
| greghaynes | jprovazn: Have a question about https://review.openstack.org/#/c/68392/6/elements/rabbitmq-server/os-refresh-config/post-configure.d/40-rabbitmq | 18:57 |
| *** CaptTofu has quit IRC | 18:57 | |
| jp_at_hp | I would argue this is correct behaviour, because this element requires that an apt.conf be specified to copy in. | 18:57 |
| *** CaptTofu has joined #tripleo | 18:58 | |
| jp_at_hp | There is already a mechanism for adding optional elements, so this prevents this element being added without the parameters it needs to funciton. | 18:58 |
| SpamapS | jp_at_hp: Right, that is one approach to take. Another is to have it in a static list of elements that are always used, and if the environment variable is not specified, then we do nothing. | 18:58 |
| jprovazn | greghaynes, sure | 18:58 |
| jp_at_hp | Why have optional elements if elements need to be able to work without the parameters that make them do anything? | 18:59 |
| greghaynes | jprovazn: The logic for that loop seems odd to me, seems like it should be checking if I (the node of the current host) is not in cluster status | 18:59 |
| greghaynes | rather than the node in $NODES | 18:59 |
| jp_at_hp | It seems to me it is promoting 2 different ways of achieving the same thing, which can result in user error causing the wrong thing to happen. | 18:59 |
| greghaynes | on line 19 | 19:00 |
| SpamapS | jp_at_hp: I think it is just a different philosophy. I prefer to include apt-conf in any build where we might want to change the apt.conf .. and you're saying you're only going to include it if you're going to change the conf. | 19:00 |
| SpamapS | jp_at_hp: It is entirely philosophical. I err toward flexibility and resilience, you're suggesting we err toward rigidity and fail faster. | 19:01 |
| jprovazn | greghaynes, I think you are looking at older version of the patchset (the link you pasted above), check patchset 8 | 19:01 |
| greghaynes | aha! | 19:01 |
| SpamapS | Both have merrits, so were that the only thing, I would certainly defer to overall reviewer sentiment, and not block a patch purely on that basis. | 19:01 |
| * greghaynes rereads | 19:02 | |
| SpamapS | jp_at_hp: -1 is "I would prefer". There are plenty of disagreements between cores... you just need two of us to +2. :) | 19:02 |
| *** CaptTofu has quit IRC | 19:02 | |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/diskimage-builder: Add rsync to the package list used by debootstrap https://review.openstack.org/76013 | 19:02 |
| jp_at_hp | I'm saying I'm preventing user-error by specifying that I want to change the apt.conf without specifying what I want to change it to. | 19:03 |
| jp_at_hp | If someone wants to change it there is a defined mechanism for adding the optional element, and use of that element requires a parameter to be set. | 19:03 |
| jp_at_hp | I think it's a clearer user experience, with reduced scope for user error. | 19:03 |
| greghaynes | jprovazn: Makes a lot more sense now :) | 19:03 |
| jprovazn | greghaynes, great :) | 19:04 |
| SpamapS | jp_at_hp: Yes, you should express that in the review. I disagree in that I think it is extra code that will lead to as many frustrating situations as it will solve. | 19:05 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Nova: don't hard code the source install path https://review.openstack.org/76619 | 19:06 |
| dprince | slagle: do you buy this approach https://review.openstack.org/#/c/76619/ | 19:06 |
| SpamapS | jp_at_hp: consider that if it only hinges on the environment variable, we can include it in the static default list of extra elements, and then users can choose whether or not to override apt-conf only with their .profile which sets DIB_APT_CONF to the thing that makes it better locally... | 19:06 |
| SpamapS | jp_at_hp: and it will always change the apt-conf in all of the places that include the element by default. "Such win" | 19:07 |
| dprince | slagle: if you buy that for Nova then I'll do it everywhere... and once we have those in place the global venv thing is cake | 19:07 |
| * dprince untested cake of course | 19:07 | |
| SpamapS | jp_at_hp: I really do understand and respect the situation you're trying to prevent, I just don't find the user interface ramifications attractive. I also do not feel so strongly about it that I want to push very hard on it. So as I said, I will gladly defer to the other reviewers. :) | 19:08 |
| slagle | dprince: i'll check it out | 19:11 |
| greghaynes | jprovazn: left you some comments, seems pretty close though | 19:11 |
| dprince | slagle: it is just the first step... but no sense replicating it if I'm on the wrong path... | 19:11 |
| openstackgerrit | Jon-Paul Sullivan proposed a change to openstack/diskimage-builder: Enable custom apt.conf in apt-conf element https://review.openstack.org/74992 | 19:11 |
| jprovazn | greghaynes, thanks, going to look | 19:12 |
| jp_at_hp | SpamapS: I think the problem is a lack of clarity in standard approach to elements in dib. What is meant to be the primary method for selecting what elements do. The easiest way for a developer to try and understand is use existing elements as a template for new elements, but it appears that is not a reliable way of determining the way things *should* be done :( | 19:12 |
| lifeless | jp_at_hp: thats a great point. | 19:14 |
| SpamapS | jp_at_hp: We miss things. I did not review the element you cribbed from, or I'd have given the same feedback. Sorry that you're stuck in a confusing position. | 19:14 |
| lifeless | jp_at_hp: 2 quirks with it though - firstly, https://etherpad.openstack.org/p/tripleo-incubator-rationalise-ui | 19:14 |
| lifeless | jp_at_hp: which describes an architectural problem closely related to the question you raise | 19:14 |
| lifeless | jp_at_hp: and a proposed solution- so would love your thoughts on that | 19:14 |
| lifeless | jp_at_hp: secondly, we have learnt over time and not always gone back and fixed 'everything' - partly because we've only just gotten full CI up and have not yet gotten full coverage. | 19:15 |
| lifeless | jp_at_hp: so - sorry :) | 19:15 |
| lifeless | jp_at_hp: if this is about apt-conf, I was going to suggest in the review that folding apt-sources, apt-conf all into the apt element would make sense to me. | 19:16 |
| jp_at_hp | lifeless: SpamapS - No problem. It's an interesting discussion. | 19:16 |
| SpamapS | lifeless: +1 from me.. let's make apt/dpkg more configurable rather than require adding more elements to the list. | 19:16 |
| *** andreaf has joined #tripleo | 19:16 | |
| jp_at_hp | lifeless: that is not the first time I heard that. I believe SpamapS said exactly the same thing | 19:16 |
| jp_at_hp | Although I think he suggested the dpkg element | 19:17 |
| * jp_at_hp hopes this isn't the death knell for the apt-conf element | 19:18 | |
| SpamapS | Right.. dpkg is a bit of a misnomer at that point.. but.. dpkg and apt go together like steering wheels and wheels. | 19:18 |
| *** sdake_ has quit IRC | 19:18 | |
| *** cwolferh has quit IRC | 19:18 | |
| *** sdake_ has joined #tripleo | 19:18 | |
| jp_at_hp | I would ask that the apt-conf element as it stands get accepted prior to any rework. For any of dib to work within corporate firewalls that is needed, and it is painful to have to hand-patch every developers use right now... | 19:19 |
| lifeless | jp_at_hp: what config setting are you setting ? | 19:20 |
| lifeless | jp_at_hp: if its just http proxy, we should be preserving that into the environment everywhere - not apt.conf should be needed. | 19:20 |
| slagle | dprince: there's a corresponding change to os-svc-install neeeded isn't there? | 19:20 |
| jp_at_hp | lifeless: it's setting no proxy for a local apt mirror. | 19:21 |
| lifeless | jp_at_hp: in fact, see diskimage-builder/elements/local-config/install.d/61-http-proxy | 19:21 |
| dprince | slagle: that will come in time... no changes needed yet | 19:21 |
| jp_at_hp | So yes - the proxy setting is getting through fine ;-) | 19:21 |
| lifeless | jp_at_hp: weshould replicate no_proxy in as well | 19:21 |
| lifeless | jp_at_hp: and probably move much of that install.d to a pre-install.d | 19:21 |
| dprince | slagle: this is about getting in in one spot in the nova* elements. | 19:21 |
| lifeless | jp_at_hp: this is in DTRT territory | 19:21 |
| *** cd-undercloud has joined #tripleo | 19:22 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 19:22 |
| *** cd-undercloud has quit IRC | 19:22 | |
| dprince | slagle: os-svc-install will certain need a change though but that comes laster (either last... or later) | 19:22 |
| *** cwolferh has joined #tripleo | 19:22 | |
| slagle | dprince: right, ok | 19:23 |
| jprovazn | greghaynes, could you please provide more info about the issue with hostnames (firrst comment)? | 19:25 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Neutron: don't hard code the source install path https://review.openstack.org/76626 | 19:25 |
| greghaynes | Yes, if youre determining which host in the $NODES list is you, hostname -s isnt neccesarially in that list | 19:25 |
| greghaynes | on mine its overcloud-notcompute0-ijev3cgjdmpu, for example | 19:26 |
| *** CaptTofu has joined #tripleo | 19:26 | |
| *** vkozhukalov_ has joined #tripleo | 19:26 | |
| jprovazn | greghaynes, well, this is OK - thanks to lifeless - he found a way how to fetch/address hostnames of other nodes in heat template, check https://review.openstack.org/#/c/68392/8/elements/rabbitmq-server/README.md line 52 | 19:27 |
| lifeless | zomg what have I done? :) | 19:27 |
| jprovazn | greghaynes, and there is already a patch from lifless which makes sure, that /etc/hosts is updated with all hostnames in stack | 19:27 |
| greghaynes | Yes, but for line 66 in https://review.openstack.org/#/c/68392/8/elements/rabbitmq-server/os-refresh-config/post-configure.d/40-rabbitmq | 19:29 |
| greghaynes | Your hostname may not match what you put in that list exactly | 19:29 |
| greghaynes | at least in my experience it never does | 19:30 |
| slagle | lifeless: i get this isn't need anymore, but why do you feel it's not specific to installing from source? https://review.openstack.org/#/c/74717/ | 19:31 |
| jprovazn | greghaynes, well, you would have to change explicitly your hostname on the node | 19:31 |
| slagle | lifeless: we don't have git checkouts, or venvs when installing from packages. that is kind of the point | 19:31 |
| lifeless | slagle: the bug would exist in packages of trunk too. | 19:31 |
| lifeless | slagle: so we should still install the fixed client. | 19:31 |
| greghaynes | jprovazn: By default my hostnames are always things like overcloud-notcompute0-ijev3cgjdmpu, where the logical name is just a substring | 19:31 |
| jprovazn | greghaynes, sure, mine too | 19:32 |
| lifeless | slagle: sorry, someone should proably write a manifesto for what the packaging stuff is intended to do then, because that wasn't my understanding at all. | 19:32 |
| greghaynes | I dont see how that that line would not always be false then | 19:32 |
| *** martyntaylor has left #tripleo | 19:32 | |
| jprovazn | greghaynes, but these random hostnames are also fetched by method in heat template: | 19:32 |
| jprovazn | greghaynes, Fn::Select:[name, fn:getatt[node0, show]] (line 52 in README) | 19:33 |
| lifeless | slagle: my understanding is that it was about consistency in vendor products, for support, prior out of band testing and the like | 19:33 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Revert "Add cpu-{node,mem,disk} to the testenv JSON" https://review.openstack.org/76025 | 19:33 |
| greghaynes | Yes, but then using hostname -s to determing $LOCAL mixes the two naming schemes | 19:34 |
| lifeless | slagle: and someone confounded into that not using venvs (which is orthogonal IMO : we could not use venvs with git installs too) | 19:34 |
| slagle | lifeless: the packages don't use venvs. there's nothing confounded there | 19:34 |
| lifeless | slagle: source installs could choose to not use venvs too. | 19:34 |
| greghaynes | Or are you saying $NODES will be the full hostname? as I understood I would expect $NODES to be node0,node1... but hostname -s to be overcloud-node0-12345 | 19:35 |
| slagle | lifeless: distro packages install can't choose *to* use venvs | 19:35 |
| lifeless | slagle: if we use 'is not source' as equivalent to 'does not use venvs', we're confusing two different things. | 19:35 |
| lifeless | slagle: yes, packages implies !venvs | 19:35 |
| lifeless | slagle: but !venvs doesn't imply packages | 19:35 |
| slagle | but using our source install implies venvs | 19:36 |
| slagle | which makes 99-neutronclient source install specific | 19:36 |
| slagle | is there any other install type implemented that uses venvs? | 19:36 |
| jprovazn | greghaynes, NODES is list like this: overcloud-node0-12345,overcloud-node1-12346... | 19:36 |
| greghaynes | ah | 19:36 |
| lifeless | slagle: so I agree that as it stood 99-neutronclient was bugg | 19:36 |
| greghaynes | well then I retract that argument :) | 19:36 |
| *** jp_at_hp has quit IRC | 19:37 | |
| greghaynes | I do think it might be nicer if we could ask o-a-c for what my hostname is... | 19:37 |
| greghaynes | but thats pretty minor | 19:37 |
| *** DuncanT has joined #tripleo | 19:37 | |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Shutdown the worker callback client https://review.openstack.org/76546 | 19:38 |
| lifeless | slagle: I see my patch reverting it failed, :( | 19:38 |
| jprovazn | greghaynes, you mean being able to specify hostname in heat template? | 19:38 |
| *** coolsvap1 has joined #tripleo | 19:38 | |
| lifeless | slagle: but anyway, you asked why I thought it was relevant, I thought installing working code would be relevant everywhere. | 19:38 |
| greghaynes | I think it should already know it | 19:38 |
| lifeless | slagle: there are, AIUI, three different dimensions here - manner of installation, revision of code being installed, and fixing a critical bug | 19:39 |
| *** ccrouch1 has joined #tripleo | 19:39 | |
| lifeless | slagle: we have no knob at the moment (AIUI) for revision of code, other than git references. | 19:39 |
| lifeless | slagle: and the trunk client fixed a critical bug going back to H - H was broken as well. | 19:40 |
| lifeless | slagle: so stable branch package builds were broken, and trunk branch package builds were broken. | 19:40 |
| *** stevebaker has joined #tripleo | 19:40 | |
| slagle | lifeless: sure, the fix makes sense, where appicable. it's the method of installing the fix | 19:41 |
| *** dprince_ has joined #tripleo | 19:41 | |
| *** vkozhukalov_ has quit IRC | 19:41 | |
| lifeless | slagle: note that the method of installing it - install from git - is nonstandard for source installs too: we install released clients only in our servers | 19:41 |
| *** sseago__ has joined #tripleo | 19:41 | |
| *** julim has joined #tripleo | 19:41 | |
| lifeless | slagle: how do you tell yum to use an https proxy ? | 19:42 |
| lifeless | slagle: local-config doesn't propogate https proxies, only http proxies | 19:42 |
| *** dprince_ has quit IRC | 19:43 | |
| slagle | lifeless: does it not export https_proxy? | 19:45 |
| slagle | lifeless: barring that, just proxy=<url> in /etc/yum.conf | 19:45 |
| slagle | i hope that works for https proxies | 19:45 |
| lifeless | slagle: so if you http and https proxies, two line? or commas between urls ? | 19:46 |
| jprovazn | greghaynes, I've replied to your second comment - does it make sense or should I elaborate it? | 19:46 |
| greghaynes | jprovazn: Thanks for entertaining my woes - gave you the +1 | 19:46 |
| lifeless | slagle: I might leave it for a yum expert | 19:46 |
| jprovazn | greghaynes, thanks for looking at it and for valuable comments | 19:46 |
| slagle | lifeless: yea, it used to be separate options, but it got changed to just one "proxy" | 19:46 |
| *** coolsvap has quit IRC | 19:47 | |
| *** sseago_ has quit IRC | 19:47 | |
| *** bauzas has quit IRC | 19:47 | |
| *** dprince has quit IRC | 19:47 | |
| *** rbrady has quit IRC | 19:47 | |
| *** sdake has quit IRC | 19:47 | |
| *** bogdando has quit IRC | 19:47 | |
| *** ccrouch has quit IRC | 19:47 | |
| *** jtomasek has quit IRC | 19:47 | |
| *** rpodolyaka has quit IRC | 19:47 | |
| *** cd-undercloud has joined #tripleo | 19:48 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 19:48 |
| *** cd-undercloud has quit IRC | 19:48 | |
| *** vkozhukalov has joined #tripleo | 19:48 | |
| bnemec | slagle: Question about how you're handling divergent paths in package installs - rhel7 has dropped the mysql compatibility shims so it doesn't create /var/run/mysqld for the mariadb service, which is the path we configure in tripleo. | 19:50 |
| bnemec | Further, /var/run is tmpfs so I can't just create it at install-time. | 19:51 |
| bnemec | Have you run into anything like that? | 19:51 |
| *** cadenzajon_ has joined #tripleo | 19:52 | |
| *** andreaf has quit IRC | 19:53 | |
| *** sdake has joined #tripleo | 19:54 | |
| *** jtomasek has joined #tripleo | 19:54 | |
| *** rbrady has joined #tripleo | 19:55 | |
| *** cadenzajon has quit IRC | 19:56 | |
| *** sballe has joined #tripleo | 19:56 | |
| lifeless | bnemec: service unit should do that yeah? what path does rhel7 use for /var/run/$thisthing? | 19:56 |
| *** athomas has joined #tripleo | 19:57 | |
| openstackgerrit | lifeless proposed a change to openstack/diskimage-builder: Improve local-config proxy handling. https://review.openstack.org/76636 | 19:57 |
| openstackgerrit | lifeless proposed a change to openstack/diskimage-builder: Better apt-sources docs. https://review.openstack.org/73035 | 19:57 |
| bnemec | lifeless: It's /var/run/mariadb on rhel7 | 19:57 |
| *** mestery_ has joined #tripleo | 19:57 | |
| bnemec | In fedora they create both /var/run/mysld and /var/run/mariadb. | 19:57 |
| *** rpodolyaka has joined #tripleo | 19:58 | |
| *** bogdando has joined #tripleo | 19:58 | |
| slagle | bnemec: yea...os-svc-daemon creates 2 systemd services for everything: one that creates /var/run/$service, and one that creates the service itself | 19:59 |
| *** bogdando has quit IRC | 19:59 | |
| *** rpodolyaka has quit IRC | 19:59 | |
| slagle | bnemec: so for the package installs, i added the ability to call os-svc-daemon and tell it to just create the service that creates /var/run | 19:59 |
| slagle | bnemec: it's the -d switch | 19:59 |
| bnemec | slagle: Yeah, that sounds like what I need, thanks. I'll take a closer look at os-svc-daemon. | 20:01 |
| *** mestery has quit IRC | 20:01 | |
| *** sballe_ has quit IRC | 20:01 | |
| *** jdob has quit IRC | 20:01 | |
| *** jrist has quit IRC | 20:01 | |
| *** jdob_ has joined #tripleo | 20:01 | |
| *** mestery_ has quit IRC | 20:01 | |
| *** mestery_ has joined #tripleo | 20:01 | |
| *** jrist has joined #tripleo | 20:02 | |
| slagle | bnemec: here's an example: https://review.openstack.org/#/c/73660/5 | 20:02 |
| slagle | some other uses have already been merged | 20:02 |
| *** martyntaylor1 has joined #tripleo | 20:04 | |
| *** kiall has quit IRC | 20:04 | |
| bnemec | slagle: Okay, cool. I think this will conflict with the mysql service in Fedora, so it probably needs to be conditional on that service not already existing, right? | 20:04 |
| *** rpodolyaka has joined #tripleo | 20:05 | |
| *** bogdando has joined #tripleo | 20:05 | |
| *** kiall has joined #tripleo | 20:06 | |
| bnemec | Oh, maybe not. | 20:06 |
| * bnemec just tries it. | 20:06 | |
| slagle | yea, i'm not sure w/o seeing it :) | 20:06 |
| *** hashar has joined #tripleo | 20:07 | |
| *** cadenzajon has joined #tripleo | 20:07 | |
| *** jwalls has joined #tripleo | 20:10 | |
| *** DuncanT- has joined #tripleo | 20:10 | |
| *** sseago_ has joined #tripleo | 20:11 | |
| *** rlandy has joined #tripleo | 20:12 | |
| *** CaptTofu_ has joined #tripleo | 20:12 | |
| *** dkehn__ has joined #tripleo | 20:13 | |
| *** bauzas has joined #tripleo | 20:15 | |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Add overcloud_role method to Node https://review.openstack.org/76348 | 20:17 |
| *** cadenzajon_ has quit IRC | 20:18 | |
| *** sseago__ has quit IRC | 20:18 | |
| *** CaptTofu has quit IRC | 20:18 | |
| *** ftcjeff has quit IRC | 20:18 | |
| *** dkehn has quit IRC | 20:18 | |
| *** DuncanT has quit IRC | 20:18 | |
| *** Kim-Chi-San has quit IRC | 20:18 | |
| *** Kim-Chi-San has joined #tripleo | 20:19 | |
| SpamapS | hrm.. | 20:19 |
| SpamapS | 2014-02-26 20:19:11.456 3804 WARNING os_collect_config.cfn [-] 400 Client Error: InvalidParameterValue | 20:19 |
| *** cadenzajon has quit IRC | 20:21 | |
| *** cadenzajon has joined #tripleo | 20:21 | |
| *** jwalls has quit IRC | 20:21 | |
| *** jwalls has joined #tripleo | 20:21 | |
| *** DuncanT- has quit IRC | 20:21 | |
| *** DuncanT- has joined #tripleo | 20:21 | |
| SpamapS | whoa weird | 20:21 |
| SpamapS | we have a notCompute with no rabbitmq-server | 20:21 |
| *** cd-undercloud has joined #tripleo | 20:27 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 20:27 |
| *** cd-undercloud has quit IRC | 20:27 | |
| SpamapS | incubator/elements skew | 20:28 |
| SpamapS | lifeless: FYI, returned incubator/t-i-e/t-h-t to master branches as all checked out reviews have been merged | 20:30 |
| *** rpodolyaka1 has joined #tripleo | 20:35 | |
| *** mestery_ is now known as mestery | 20:38 | |
| *** martyntaylor1 has left #tripleo | 20:38 | |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-heat-templates: Add GlanceLogFile and GlanceNotifierStrategy https://review.openstack.org/76652 | 20:54 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Make log_file and notifier_strategy configurable https://review.openstack.org/76238 | 20:55 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add fedora-rdo-icehouse element https://review.openstack.org/73411 | 20:55 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Fix typo https://review.openstack.org/73655 | 20:55 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add missing +x https://review.openstack.org/74495 | 20:55 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add create-dir service for neutron ovs agent https://review.openstack.org/73660 | 20:56 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install lvm2 package for cinder https://review.openstack.org/73656 | 20:56 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Common cinder install code https://review.openstack.org/73657 | 20:56 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Use os-svc-restart for cinder-api https://review.openstack.org/73658 | 20:56 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Remove unneeded chown of /var/run/nova https://review.openstack.org/73659 | 20:56 |
| * Ng admits defeat and retreats to bed to sleep this silly cold off | 20:56 | |
| greghaynes | Youve got to sleep it off before the summit! | 20:56 |
| greghaynes | er meetup | 20:57 |
| * bnemec would be very concerned if a cold lasted from now until summit :-) | 20:59 | |
| greghaynes | Yes, might want to get that checked out | 20:59 |
| lifeless | SpamapS: https://review.openstack.org/76636 may help jp | 21:01 |
| lifeless | SpamapS: awesome, thanks | 21:01 |
| *** lblanchard has quit IRC | 21:08 | |
| *** CaptTofu_ has quit IRC | 21:14 | |
| *** CaptTofu has joined #tripleo | 21:14 | |
| *** CaptTofu has quit IRC | 21:14 | |
| *** rpodolyaka1 has quit IRC | 21:14 | |
| *** CaptTofu has joined #tripleo | 21:14 | |
| slagle | lifeless: did you mean to also git rm a file there? | 21:18 |
| slagle | (or maybe i'm just not understanding the change) | 21:19 |
| lifeless | oh did git utterly fail on me | 21:19 |
| slagle | i think so :) | 21:20 |
| lifeless | slagle: deleted file mode 100755 | 21:20 |
| lifeless | index 255df7e..0000000 | 21:20 |
| lifeless | --- a/elements/local-config/install.d/61-http-proxy | 21:20 |
| lifeless | slagle: gerrit is failing on you I think | 21:20 |
| *** openstackgerrit has quit IRC | 21:20 | |
| *** sseago_ has quit IRC | 21:20 | |
| *** sbadia has quit IRC | 21:20 | |
| *** shortstop has quit IRC | 21:20 | |
| *** sirushti has joined #tripleo | 21:20 | |
| *** cd-undercloud has joined #tripleo | 21:20 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 21:20 |
| *** cd-undercloud has quit IRC | 21:20 | |
| slagle | lifeless: yea, i used "git-review -d 76636" | 21:20 |
| slagle | i guess that doesn't delete the file that was deleted in the change? | 21:21 |
| slagle | locally i mean | 21:21 |
| *** jdob has joined #tripleo | 21:21 | |
| *** sbadia has joined #tripleo | 21:21 | |
| *** jdob_ has quit IRC | 21:21 | |
| *** openstackgerrit has joined #tripleo | 21:21 | |
| slagle | no, it did the right thing actually | 21:21 |
| *** rpodolyaka1 has joined #tripleo | 21:21 | |
| *** coolsvap_ has joined #tripleo | 21:23 | |
| *** mestery_ has joined #tripleo | 21:23 | |
| lifeless | slagle: I checked with log -p, it looks ok to me | 21:23 |
| slagle | so these 2 files should be extremely similar: | 21:24 |
| slagle | ./pre-install.d/02-http-proxy | 21:24 |
| slagle | ./pre-install.d/02-proxy-settings | 21:24 |
| slagle | ? | 21:24 |
| *** e0ne has quit IRC | 21:24 | |
| *** saurabhs1 has joined #tripleo | 21:24 | |
| lifeless | oh | 21:25 |
| lifeless | thats a bug | 21:25 |
| *** saurabhs1 has quit IRC | 21:26 | |
| *** saurabhs1 has joined #tripleo | 21:26 | |
| lifeless | slagle: fixed | 21:27 |
| openstackgerrit | lifeless proposed a change to openstack/diskimage-builder: Improve local-config proxy handling. https://review.openstack.org/76636 | 21:27 |
| lifeless | slagle: I changed my mind on the name and failed to cleanup properly- thanks for catching | 21:28 |
| *** coolsvap has joined #tripleo | 21:32 | |
| *** sseago_ has joined #tripleo | 21:34 | |
| *** rbrady has quit IRC | 21:34 | |
| *** mestery has quit IRC | 21:34 | |
| *** _sirushti has joined #tripleo | 21:34 | |
| *** coolsvap1 has quit IRC | 21:35 | |
| *** saurabhs has quit IRC | 21:35 | |
| *** jcooley_ has quit IRC | 21:35 | |
| *** ohadlevy has quit IRC | 21:35 | |
| *** sirushti has quit IRC | 21:35 | |
| *** _sirushti is now known as sirushti | 21:35 | |
| *** coolsvap_ has quit IRC | 21:35 | |
| *** ohadlevy has joined #tripleo | 21:35 | |
| *** ohadlevy is now known as Guest79114 | 21:36 | |
| *** rbrady1 has joined #tripleo | 21:36 | |
| *** jcooley_ has joined #tripleo | 21:36 | |
| *** Guest79114 is now known as ohadlevy_ | 21:36 | |
| rbrady1 | lifeless: ping | 21:37 |
| *** rbrady1 is now known as rbrady | 21:37 | |
| *** mestery_ is now known as mestery | 21:38 | |
| *** derekh has joined #tripleo | 21:38 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Glance/Cinder/Keystone/Swift: don't hard code venv https://review.openstack.org/76666 | 21:42 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Neutron: don't hard code the venv dir https://review.openstack.org/76626 | 21:42 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Link db sync utilities to /usr/local/bin... https://review.openstack.org/76579 | 21:42 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Nova: don't hard code the venv dir https://review.openstack.org/76619 | 21:42 |
| lifeless | rbrady: pongish | 21:43 |
| rbrady | lifeless: I was going to ask for clarification on https://review.openstack.org/#/c/74712/, but I think it just clicked for me. | 21:45 |
| *** jprovazn has quit IRC | 21:45 | |
| derekh | lifeless: that patch to close connections to the geard broker should sort out the problem from earlier, while testing it I came across another problem https://review.openstack.org/#/c/76588/ which we will probably hit at some stage | 21:46 |
| derekh | lifeless: beside those we seem to be getting fairly consitent errors building the overcloud compute image, trying to figure it out at the moment | 21:48 |
| *** shakayumi has joined #tripleo | 21:49 | |
| *** DuncanT- is now known as DuncanT | 21:50 | |
| derekh | lifeless: errors building the overcloud notcompute image | 21:50 |
| SpamapS | Updating cache of https://github.com/kanaka/noVNC.git in /root/.cache/image-create/repository-sources/eac807d7dcef19baf4 | 21:53 |
| SpamapS | dd80e1f38fc67a7720c193 with ref master | 21:53 |
| SpamapS | wonderful | 21:53 |
| SpamapS | error: The requested URL returned error: 403 Forbidden while accessing https://github.com/kanaka/noVNC.git/info/refs | 21:53 |
| *** e0ne has joined #tripleo | 21:55 | |
| SpamapS | looks like that was a relatively abnormal thing | 21:55 |
| lifeless | probably we want a cache of all repos in the CI infra | 21:55 |
| lifeless | like we need a DNS cache | 21:55 |
| lifeless | we're getting DNS issues too | 21:55 |
| lifeless | and distro etc | 21:55 |
| *** jdob has quit IRC | 21:56 | |
| SpamapS | lifeless: right, decouple downloading from image building | 21:56 |
| *** e0ne has quit IRC | 21:56 | |
| *** e0ne has joined #tripleo | 21:57 | |
| *** e0ne has quit IRC | 21:58 | |
| *** e0ne has joined #tripleo | 21:58 | |
| *** CaptTofu has quit IRC | 21:59 | |
| *** cd-undercloud has joined #tripleo | 22:00 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 22:00 |
| *** cd-undercloud has quit IRC | 22:00 | |
| *** sdake_ has quit IRC | 22:01 | |
| tchaypo | starts devtest_seed.sh running. Goes to bed. Comes back six hours later, still running. | 22:02 |
| *** julim has quit IRC | 22:02 | |
| SpamapS | tchaypo: \o/ .. no errors! | 22:02 |
| * SpamapS hums the eric idle song | 22:03 | |
| *** e0ne has quit IRC | 22:03 | |
| StevenK | NobodyCam: That's what I use to connect to IRC, not my e-mail. | 22:06 |
| *** jang1 has quit IRC | 22:06 | |
| *** CaptTofu has joined #tripleo | 22:08 | |
| NobodyCam | heheh | 22:09 |
| NobodyCam | StevenK: yep | 22:10 |
| StevenK | NobodyCam: My e-mail is in the commit logs for swift, nova, tie, dib, ... :-) | 22:11 |
| StevenK | NobodyCam: But yes, I got your e-mail to my corp address | 22:12 |
| tchaypo | well, no *fatal* errors. I might come back later and find some hidden treasure in the logs | 22:12 |
| NobodyCam | StevenK: :) | 22:12 |
| *** rpodolyaka1 has quit IRC | 22:14 | |
| SpamapS | oh interesting I may have found a bug in our heat templates | 22:15 |
| SpamapS | the CompletionConditions need to depend on the Config's | 22:15 |
| StevenK | NobodyCam: So my work is going to be a few patches - 73145 is my current WIP | 22:15 |
| SpamapS | 'ResourceNotAvailable: The Resource (NovaCompute1Config) is not available.\n'] | 22:15 |
| SpamapS | otherwise we get that | 22:16 |
| * NobodyCam looks | 22:16 | |
| SpamapS | system boot races with heat-engine | 22:16 |
| StevenK | NobodyCam: So the plan is 73145 in incubator will handle both styles (as it were), nodes in the JSON, or MACs in the JSON. | 22:17 |
| StevenK | NobodyCam: Then I'll patch toci to populate nodes in the JSON it uses, and finally we can drop the old support from incubator | 22:18 |
| tchaypo | and *just now* it got to asking me for my ssh key passphrase | 22:21 |
| NobodyCam | StevenK: are you looking at adding virsh_type to $TE_DATAFILE | 22:21 |
| greghaynes | tchaypo: I tend to regret whenever I start devtest scripts without key added to ssh-agent | 22:22 |
| StevenK | NobodyCam: So is that going to be a global setting, or per node? | 22:22 |
| tchaypo | I'm quite surprised that I hadn't already added it. | 22:23 |
| NobodyCam | StevenK: in nova-bm it global Ironic is per node | 22:23 |
| StevenK | NobodyCam: Right, so I'm not looking at adding it, but it can probably land after 73145 does. | 22:23 |
| NobodyCam | ok. | 22:24 |
| tchaypo | + setup-baremetal 1 2048 20 i386 52:54:00:45:e7:c9 seed '' '' '' | 22:26 |
| tchaypo | ERROR: The server has either erred or is incapable of performing the requested operation. (HTTP 500) (Request-ID: req-53f53160-a6b8-48cd-91d5-dacf3f06e516) | 22:26 |
| tchaypo | that's the kind of hidden gem I'm talking about. | 22:26 |
| StevenK | Heh | 22:27 |
| * tchaypo looks forward to figuring out what exactly happened there | 22:27 | |
| greghaynes | tchaypo: I think thats missing mac addrs | 22:28 |
| greghaynes | tchaypo: so youll probably fail when it tries to heat stack-create | 22:28 |
| *** hashar has quit IRC | 22:28 | |
| greghaynes | with no valid host found | 22:28 |
| greghaynes | I might of done that a few dozen times | 22:29 |
| *** weshay has quit IRC | 22:29 | |
| *** matty_dubs is now known as matty_dubs|gone | 22:29 | |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-heat-templates: Make Wait Conditions depend on config creation https://review.openstack.org/76675 | 22:29 |
| SpamapS | greghaynes: we're going to have a "might of" and "should of" jar for times when the person should _have_ used should _have_. ;) It is just like a swear jar. Only you have to put whole bitcoins in it... | 22:30 |
| SpamapS | which may or may not mean it is free, or devastating financially. | 22:30 |
| devananda | have ya'll used the console functionailty of nova-baremetal at all? | 22:31 |
| StevenK | SpamapS: And if you don't believe in the currency at all? | 22:31 |
| greghaynes | haha, ill just have to time my poor grammar to bitcoin market conditions | 22:31 |
| *** CaptTofu has quit IRC | 22:31 | |
| devananda | StevenK: hi! at some point I'd like to bend your ear about ironic and getting support for it not only added to tripleo but *tested* too! | 22:32 |
| greghaynes | maybe I should just write a plugin for my irc client to auto change s/might of/might have | 22:32 |
| devananda | StevenK: i hear that might be a thing you might know something about or be interested in? | 22:32 |
| StevenK | devananda: At the moment, I'm refactoring devtest to support full node definitions, rather than just MACs, and have touched base with NobodyCam about ironic and virsh_type. | 22:33 |
| SpamapS | devananda: never once | 22:34 |
| devananda | StevenK: :) | 22:34 |
| SpamapS | not to say I didn't wish it was available | 22:34 |
| devananda | SpamapS: fwiw, I belieev it /was/ available all this time | 22:35 |
| devananda | SpamapS: i was goign to ask if ya'll were interested in testing console support in ironic at some point | 22:36 |
| openstackgerrit | Derek Higgins proposed a change to openstack/diskimage-builder: Replace more then just "-" in REPONAME https://review.openstack.org/76679 | 22:36 |
| SpamapS | $ nova console-log hwtest | 22:36 |
| SpamapS | ERROR: Unable to get console log, functionality not implemented (HTTP 501) (Request-ID: req-2fb702ae-b812-4e02-90ad-ca0c640eff40) | 22:36 |
| devananda | SpamapS: eg, spin up the services in isolation somewhere, enroll one of your hardware nodes, pull a patch from gerrit, and see if it gives you SOL console | 22:37 |
| devananda | SpamapS: get-vnc-console | 22:37 |
| devananda | SpamapS: you need to run shellinaboxd next to nova-compute, I think | 22:37 |
| NobodyCam | StevenK: until all the refactoring is done, what are your thoughts on adding an ENV VAR like "IRONIC_VIRT_TYPE" to https://review.openstack.org/#/c/72969 | 22:40 |
| StevenK | NobodyCam: devananda has dealt with all my concerns on that branch. I think the eventual plan is to get everything into the JSON. | 22:42 |
| devananda | that patch needs to be rebased :( | 22:43 |
| SpamapS | devananda: ETOOMUCHINVESTMENT | 22:43 |
| SpamapS | devananda: it will be amazing, and I desperately want it some day | 22:43 |
| SpamapS | but I really hope that doesn't block Ironic from being "the thing" | 22:44 |
| devananda | SpamapS: huh? | 22:44 |
| lifeless | StevenK: you said something yesterday about my refactoring not working ? | 22:44 |
| lifeless | StevenK: but I didn't understand you :) | 22:44 |
| SpamapS | devananda: vnc.. shellinaboxd ...no time to try all that ;) | 22:44 |
| lifeless | devananda: I think we (cd-admins) can definitely do it | 22:45 |
| lifeless | devananda: SpamapS can't because he's key-path on heat. | 22:45 |
| StevenK | lifeless: Sorry. From the console logs: | 22:45 |
| StevenK | 2014-02-26 06:23:14.174 | jq: error: null and array cannot be subtracted | 22:45 |
| StevenK | 2014-02-26 06:23:14.212 | new | 22:45 |
| lifeless | StevenK: how far in? | 22:45 |
| devananda | lifeless: ack - I had initially just asked whether "ya'll" had tried it, but SpamapS responded :) | 22:45 |
| lifeless | StevenK: that probably means .nodes isn't set | 22:45 |
| lifeless | devananda: haven't tried it because vnc si blech | 22:45 |
| StevenK | lifeless: Of course it isn't? | 22:45 |
| StevenK | lifeless: That's from a toci run against the branch | 22:46 |
| lifeless | StevenK: oh, crud, and jq is barfing there? | 22:46 |
| StevenK | lifeless: That's where the paste is from. | 22:46 |
| lifeless | StevenK: we should be able to do something there, let me have a look | 22:46 |
| devananda | lifeless: IIRC, pyghmi looks like it's adding a SOL text console impl in python (eg, it won't need shellinaboxd) | 22:47 |
| lifeless | StevenK: or you can:P | 22:47 |
| lifeless | devananda: ideally there would be a thing I as an admin can ssh to using my nova registered keyring | 22:47 |
| lifeless | devananda: and nova would of course know I was in a tenant that can see the box | 22:47 |
| lifeless | devananda: but honestly, shellinaboxd for now is fine | 22:48 |
| devananda | :) | 22:48 |
| StevenK | lifeless: I've been thinking about it | 22:48 |
| lifeless | devananda: the /real/ thing I'd like is to be able to get console logs etc from 'nova console-log' | 22:48 |
| lifeless | devananda: (and I mean e.g. iLO textcons output, not serialX voodoo) | 22:48 |
| lifeless | devananda: or should I say, /another/ thing I'd like | 22:49 |
| StevenK | lifeless: It looks like we don't die horribly, but I'm also not certain that $BACKWARDS_COMPAT actually gets set after the getopt fun | 22:49 |
| lifeless | devananda: that would be more valuable than on-demand interactive sessions in a lot of ways | 22:49 |
| lifeless | StevenK: I tested backwards compat 6 ways to sunday | 22:49 |
| lifeless | StevenK: just not the expression to get nodes with an old testenv json | 22:49 |
| lifeless | StevenK: I tested it with [.nodes[0]], not .nodes - [.nodes[0]] | 22:50 |
| StevenK | lifeless: Ah ha, so setup-baremetal is fine, it's just the call in _undercloud | 22:50 |
| NobodyCam | StevenK: I am looking to see if I can avoid doing somehting like line #43 in refactoring register-node script. http://paste.openstack.org/show/t6d9ZNkXqTanZpPvAOZx | 22:51 |
| devananda | lifeless: ++ | 22:51 |
| lifeless | StevenK: I expect so | 22:52 |
| lifeless | StevenK: though possibly - ah likely even - I didn't fully consider the jq expressions to select from within the nodes | 22:52 |
| lifeless | StevenK: so - thinking out loud - Rather than slicing nodes up in _seed and _undercloud, I think we should pass .nodes in verbatim | 22:53 |
| StevenK | lifeless: Right, but then we need to teach setup-baremetal how to slice it | 22:53 |
| lifeless | StevenK: and add an option to setup-baremetal to select first / rest of nodes. | 22:53 |
| StevenK | Right | 22:53 |
| lifeless | StevenK: right, and it can build a jq expression for either new or old style as needed | 22:53 |
| lifeless | StevenK: and then its all happy and light | 22:54 |
| StevenK | lifeless: --only-first and --all-but-one ? :-) | 22:54 |
| StevenK | lifeless: However, it's a pretty elegant solution, so I'd like it to return after toci changes | 22:55 |
| * greghaynes bandersnatch away! | 22:56 | |
| *** rbrady has quit IRC | 22:58 | |
| *** e0ne has joined #tripleo | 22:58 | |
| lifeless | StevenK: --first --not-first perhaps ? | 23:00 |
| *** e0ne has quit IRC | 23:03 | |
| *** cd-undercloud has joined #tripleo | 23:09 | |
| cd-undercloud | ************** overcloud complete status=1 ************ | 23:09 |
| *** cd-undercloud has quit IRC | 23:09 | |
| *** cd-undercloud has joined #tripleo | 23:11 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 23:11 |
| *** cd-undercloud has quit IRC | 23:11 | |
| SpamapS | error: Unable to find 30a7d7430f4eccec942363ac5e9f7c386ce90e98 under https://git.openstack.org/openstack/horizon | 23:20 |
| SpamapS | Cannot obtain needed object 30a7d7430f4eccec942363ac5e9f7c386ce90e98 | 23:20 |
| SpamapS | doh, we still have old broken remotes | 23:20 |
| SpamapS | with /cgit | 23:20 |
| *** lazy_prince has joined #tripleo | 23:20 | |
| *** jtomasek has quit IRC | 23:20 | |
| saurabhs1 | Hi doubt about disk image builder. the numbering of the scripts, how does that play? If two seperate elements have scripts that are of the same number what that will result it? I mean when I write my elements do I need to worry about each and every element in the tripleO image elements and then number my scripts or I should only care about the elements I directly depend upon and number my script after those? | 23:21 |
| *** cd-undercloud has joined #tripleo | 23:22 | |
| cd-undercloud | ************** overcloud complete status=128 ************ | 23:22 |
| *** cd-undercloud has quit IRC | 23:22 | |
| SpamapS | saurabhs1: they are lexically sorted | 23:22 |
| *** killer_prince has quit IRC | 23:22 | |
| *** lazy_prince is now known as killer_prince | 23:22 | |
| SpamapS | saurabhs1: so 01-a and 01-b execute 01-a then 01-b | 23:22 |
| SpamapS | saurabhs1: We have not had numbering problems thus far. | 23:22 |
| saurabhs1 | SpamapS: ok so if two scripts have same number infront of them thats fine, yeh I was just concerned about that. Thanks | 23:24 |
| SpamapS | Updating cache of https://git.openstack.org/openstack/keystone.git in /root/.cache/image-create/repository-sources/6bd7a | 23:24 |
| SpamapS | 6223fe50b4a267bc924641d331df94c833e with ref master | 23:24 |
| SpamapS | error: RPC failed; result=7, HTTP code = 0 | 23:24 |
| SpamapS | Hm looks like maybe git.o.o is having issues ATM | 23:24 |
| SpamapS | saurabhs1: I believe JP Sullivan is working on a patch to make sure elements don't use the exact same name. | 23:25 |
| lifeless | SpamapS: or possibly network glitches | 23:25 |
| SpamapS | lifeless: true | 23:26 |
| lifeless | SpamapS: or | 23:26 |
| lifeless | SpamapS: see infra they appear to be having load issues ;) | 23:26 |
| openstackgerrit | Jenkins proposed a change to openstack/diskimage-builder: Updated from global requirements https://review.openstack.org/75670 | 23:26 |
| *** vkozhukalov has quit IRC | 23:28 | |
| *** jwalls has quit IRC | 23:30 | |
| openstackgerrit | Jenkins proposed a change to openstack/os-collect-config: Updated from global requirements https://review.openstack.org/75574 | 23:33 |
| openstackgerrit | Derek Higgins proposed a change to openstack/diskimage-builder: Replace more then just "-" in REPONAME https://review.openstack.org/76679 | 23:36 |
| *** shakayumi has quit IRC | 23:38 | |
| *** rlandy has quit IRC | 23:39 | |
| derekh | pleia2: just got your mail, thanks, I'll give it a whirl tomorrow and add feedback | 23:50 |
| *** sballe has quit IRC | 23:50 | |
| *** sballe has joined #tripleo | 23:50 | |
| *** cadenzajon has quit IRC | 23:51 | |
| pleia2 | derekh: thanks :) | 23:52 |
| derekh | lifeless, devananda, Ng, SpamapS_, jog0_, GheRivero, derekh, dprince, slagle looks like we just lost the ci-overcloud again | 23:56 |
| derekh | can't ssh to ci-overcloud-notCompute0-gxezgcvv4v2q | 23:56 |
| *** e0ne has joined #tripleo | 23:57 | |
| *** sballe has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!