| tchaypo | II take it back, there is a bug | 00:01 |
|---|---|---|
| tchaypo | https://bugs.launchpad.net/openstack-ci/+bug/1270382 | 00:01 |
| uvirtbot | Launchpad bug 1270382 in openstack-ci "Name resolution failure on gate job" [Medium,Confirmed] | 00:01 |
| StevenK | Bleh, looking at the wrong console log | 00:03 |
| StevenK | 2014-03-25 23:42:46.216 | + '[' overcloud = overcloud ']' | 00:04 |
| StevenK | 2014-03-25 23:42:46.216 | + setup-baremetal --service-host seed --nodes /dev/fd/63 | 00:04 |
| StevenK | 2014-03-25 23:42:46.252 | ++ jq '.nodes - [.nodes[0]]' /tmp/tmp5JOyMb | 00:04 |
| StevenK | lifeless: ^ | 00:04 |
| StevenK | 2014-03-25 23:42:46.252 | new | 00:04 |
| tchaypo | uvirtbot: oh, you're playing nice today, well done! | 00:04 |
| uvirtbot | tchaypo: Error: "oh," is not a valid command. | 00:04 |
| StevenK | Haha | 00:04 |
| StevenK | uvirtbot: botsnack | 00:04 |
| uvirtbot | StevenK: Error: "botsnack" is not a valid command. | 00:04 |
| StevenK | Heh | 00:04 |
| lifeless | StevenK: so its good ? | 00:22 |
| *** mestery has quit IRC | 00:23 | |
| *** newell has quit IRC | 00:26 | |
| *** matsuhashi has joined #tripleo | 00:27 | |
| *** morganfainberg is now known as morganfainberg_Z | 00:33 | |
| *** yamahata has quit IRC | 00:34 | |
| StevenK | lifeless: Hm, not sure why undercloud-ironic broke | 00:36 |
| * tchaypo heads into the office for vaccination | 00:47 | |
| *** d0ugal has quit IRC | 00:53 | |
| *** d0ugal has joined #tripleo | 01:07 | |
| *** akrivoka has quit IRC | 01:09 | |
| *** matsuhashi has quit IRC | 01:12 | |
| *** matsuhashi has joined #tripleo | 01:14 | |
| *** sdake_ has quit IRC | 01:15 | |
| *** sdake has quit IRC | 01:16 | |
| *** eguz has quit IRC | 01:18 | |
| *** weshay has quit IRC | 01:25 | |
| *** nosnos has joined #tripleo | 01:27 | |
| *** sseago has joined #tripleo | 01:28 | |
| *** ramishra has joined #tripleo | 01:30 | |
| *** sdake has joined #tripleo | 01:35 | |
| *** sdake_ has joined #tripleo | 01:35 | |
| lifeless | SpamapS: slagle: https://review.openstack.org/#/c/82066/ | 01:45 |
| lifeless | StevenK: the deploy failed | 01:46 |
| lifeless | StevenK: pull the log tarball | 01:46 |
| lifeless | 2014-03-26 00:07:49.063 4480 ERROR nova.compute.manager [req-5c546f0f-c3ab-4531-a963-746c51791f99 d6ceae50d0d74b1abc080f1bf99e7761 7094b5e353a3441fabe3242425ffcadf] [instance: 3630003f-52dd-45e1-a004-0815618b83da] Instance failed to spawn | 01:48 |
| lifeless | 2014-03-26 00:07:49.063 4480 TRACE nova.compute.manager [instance: 3630003f-52dd-45e1-a004-0815618b83da] Traceback (most recent call last): | 01:48 |
| lifeless | 2014-03-26 00:07:49.063 4480 TRACE nova.compute.manager [instance: 3630003f-52dd-45e1-a004-0815618b83da] File "/opt/stack/venvs/nova/local/lib/python2.7/site-packages/nova/compute/manager.py", line 1718, in _spawn | 01:48 |
| lifeless | 2014-03-26 00:07:49.063 4480 TRACE nova.compute.manager [instance: 3630003f-52dd-45e1-a004-0815618b83da] block_device_info) | 01:48 |
| lifeless | then check nova-baremetal-deploy-helper | 01:48 |
| lifeless | 2014-03-26 00:07:39.615 4429 ERROR nova.virt.baremetal.deploy_helper [req-cf9f97eb-583b-48f9-b2eb-1d8771942699 None None] Cmd : sudo nova-rootwrap /etc/nova/rootwrap.conf dd if=/mnt/state/var/lib/nova/instances/instance-00000001/disk of=/dev/disk/by-path/ip-192.0.2.3:3260-iscsi-iqn-3630003f-52dd-45e1-a004-0815618b83da-lun-1-part1 bs=1M oflag=direct | 01:49 |
| lifeless | 2014-03-26 00:07:39.622 4429 ERROR nova.virt.baremetal.deploy_helper [req-cf9f97eb-583b-48f9-b2eb-1d8771942699 None None] StdOut : '' | 01:49 |
| lifeless | 2014-03-26 00:07:39.623 4429 ERROR nova.virt.baremetal.deploy_helper [req-cf9f97eb-583b-48f9-b2eb-1d8771942699 None None] StdErr : '/bin/dd: writing \xe2\x80\x98/dev/disk/by-path/ip-192.0.2.3:3260-iscsi-iqn-3630003f-52dd-45e1-a004-0815618b83da-lun-1-part1\xe2\x80\x99: Input/output error\n1+0 records in\n0+0 records out\n0 bytes (0 B) copied, 120.693 s, 0.0 kB/s\n' | 01:49 |
| *** yamahata has joined #tripleo | 01:58 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Remove backwards compatability in setup-baremetal https://review.openstack.org/78467 | 02:00 |
| *** matsuhashi has quit IRC | 02:00 | |
| *** matsuhashi has joined #tripleo | 02:08 | |
| lifeless | greghaynes: so how goes HA? With Ironic sorted (pending reviews I have more bw for HA and perf) | 02:09 |
| StevenK | Blah, that remove backwards compat branch will need a recheck no bug when the toci change lands | 02:10 |
| tchaypo | https://bugs.launchpad.net/tripleo/+bug/1297324 is doing my head in | 02:14 |
| uvirtbot | Launchpad bug 1297324 in tripleo "iptables element scripts don't work on Ubuntu" [Undecided,New] | 02:14 |
| tchaypo | can't figure out if there's actually a problem, or if the requestor is just seeing rules and assuming they need to be applied | 02:14 |
| tchaypo | it looks like the add-rules script very deliberately targets only the redhat builds, which suggests this is deliberate... but why would redhat need iptables rules and debian-based distros not require them? but in any case if the lack of firewall rules was a problem I'd expect to see CI fail | 02:15 |
| clarkb | tchaypo: perhaps because rhel images enforce rules by default whereas debian/ubuntu allow all traffic to flow? | 02:18 |
| tchaypo | ah. https://review.openstack.org/#/c/68075/ has a comment saying that as of F20 iptables doesn't run on fedora anyway so these probably aren't even needed there any more | 02:18 |
| tchaypo | clarkb: this is the impression I'm getting. | 02:19 |
| tchaypo | I'm going to mark the bug as incomplete with a suggestion that unless there are actual problems observed with the traffic flow this should probably be closed | 02:19 |
| lifeless | so we may want to firewall things up | 02:20 |
| lifeless | but yes the rules exist to unbreak fedora | 02:21 |
| tchaypo | we may want to, yes; but it seems that everything we have so far has been designed to unbreak fedora | 02:25 |
| tchaypo | it seems to me that switching to "here's a set of iptables rules that will work generically" is a bigger goal | 02:25 |
| tchaypo | also, it bugs me that launchpad lets me triage/update priority/comment on a bug and still doesn't think I want to get emails about changes. I wish I could find a setting that lets me get auto-subscribed to everything I touch | 02:25 |
| *** noslzzp has quit IRC | 02:26 | |
| *** vkozhukalov has joined #tripleo | 02:31 | |
| *** saurabhs1 has quit IRC | 02:32 | |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Kill commands that outlive test env https://review.openstack.org/75402 | 02:33 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Set the worker timeout https://review.openstack.org/76172 | 02:33 |
| lifeless | tchaypo: you can subscribe to the whole feed | 02:34 |
| *** fandi has quit IRC | 02:34 | |
| lifeless | tchaypo: or per bug basis, I don't *think* there is 'if I commented', but I may be wrong - try putting a custom feed subscription in place | 02:35 |
| StevenK | There's also a setting for e-mailing you about things you changed | 02:36 |
| lifeless | StevenK: https://review.openstack.org/#/c/79051 needs tweaks | 02:36 |
| StevenK | Yeah, let me fix that up. | 02:38 |
| *** BadCub01 has quit IRC | 02:39 | |
| *** lazy_prince has quit IRC | 02:39 | |
| *** killer_prince has joined #tripleo | 02:39 | |
| StevenK | lifeless: I was also trying to work out how to store the route-dev in the object | 02:39 |
| StevenK | Since we currently use a specified value in toci, and the fallback for devtest. | 02:40 |
| *** nati_ueno has quit IRC | 02:42 | |
| lifeless | StevenK: so that patch isn't about toci using testenv; as such you could just document it. | 02:43 |
| *** ramishra has quit IRC | 02:45 | |
| *** ramishra has joined #tripleo | 02:45 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Update documentation. https://review.openstack.org/82741 | 02:49 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Default to 30G of disk per bm VM. https://review.openstack.org/82753 | 02:49 |
| *** ramishra has quit IRC | 02:50 | |
| *** cwolferh has quit IRC | 02:54 | |
| *** ramishra has joined #tripleo | 02:54 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Update documentation. https://review.openstack.org/82741 | 02:57 |
| *** ramishra has quit IRC | 03:02 | |
| *** ramishra has joined #tripleo | 03:02 | |
| *** cwolferh has joined #tripleo | 03:06 | |
| *** ramishra has quit IRC | 03:07 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Adding image update policy to undercloud template https://review.openstack.org/82066 | 03:11 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Create pip manifest files of installed packages https://review.openstack.org/80644 | 03:12 |
| *** matsuhashi has quit IRC | 03:15 | |
| greghaynes | lifeless: not a lot of progress, got distracted by trying to actually learn how heat works. Got the percona config to be templated out by a hosts list from heat this afternoon so hoping to work on the bootstrapping logic this eve/tomorrow morning. | 03:17 |
| lifeless | awesome | 03:17 |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Update documentation. https://review.openstack.org/82741 | 03:17 |
| lifeless | [2014-03-26 16:17:15] Total runtime: 1396 s | 03:18 |
| lifeless | [2014-03-26 16:17:15] ramdisk : 0 s | 03:18 |
| lifeless | [2014-03-26 16:17:15] seed : 213 s | 03:18 |
| lifeless | [2014-03-26 16:17:15] undercloud : 353 s | 03:18 |
| lifeless | [2014-03-26 16:17:15] overcloud : 820 s | 03:18 |
| lifeless | real 23m16.304s | 03:18 |
| lifeless | user 0m57.494s | 03:18 |
| lifeless | sys 0m16.543s | 03:18 |
| lifeless | time with -c | 03:18 |
| greghaynes | wat | 03:18 |
| tchaypo | lifeless: ^^^ seemed easier than just telling you about the nit | 03:18 |
| lifeless | tchaypo: thanks :) | 03:18 |
| tchaypo | greghaynes: oh, there you are. i was just wondering where your in-progress change is because i want to start learning from it | 03:18 |
| SpamapS | lifeless: hey, so heat-admin is no longer necessary... | 03:18 |
| SpamapS | lifeless: if we stop setting it in heat.conf, we'll get whatever the default from the image is. | 03:19 |
| lifeless | SpamapS: youknowwhattodo | 03:19 |
| greghaynes | tchaypo: not sure what change you are referring to? | 03:19 |
| lifeless | SpamapS: where do we set it? Also how will we know what user to ssh in to setup keystone with ? | 03:19 |
| tchaypo | greghaynes: the work you're doing on percona. I was hoping there was a change | 03:20 |
| lifeless | bbiab, C time | 03:20 |
| * tchaypo should have looked at trello | 03:20 | |
| greghaynes | ah. The switch to xtradb landed, just getting started on actual clustering. Ill try to push a WIP soon and hl you | 03:21 |
| SpamapS | lifeless: 1: heat.conf, admin_user=heat-admin. 2: I know.. trade one problem for another one. ;) | 03:21 |
| greghaynes | oh, trello. Forgot about that :) | 03:21 |
| *** Matt1 has quit IRC | 03:21 | |
| SpamapS | lifeless: I was thinking we could bake a locked down user into the images for it. | 03:22 |
| tchaypo | thanks :) | 03:23 |
| SpamapS | otherwise we have to map distro -> cloud-init default. | 03:23 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Default to 30G of disk per bm VM. https://review.openstack.org/82753 | 03:25 |
| * SpamapS releases os-collect-config 0.1.13 with cfn deployments support | 03:27 | |
| *** nosnos has quit IRC | 03:28 | |
| StevenK | Heh, I guess you can't create RHEL images without a RHN account | 03:29 |
| SpamapS | AFAIK, yes | 03:32 |
| SpamapS | It's one of the reasons I've never logged into a RHEL server. :) | 03:33 |
| mordred | SpamapS: those are a thing? I thought they were only a myth | 03:33 |
| StevenK | Heh | 03:33 |
| SpamapS | mordred: A story told to the stock holders to help them sleep at night? ;) | 03:34 |
| mordred | SpamapS: :) | 03:34 |
| StevenK | I'm trying to work out if we can use 'ip route replace' | 03:34 |
| StevenK | Fedora 20 is fine, at least | 03:35 |
| tchaypo | "control plane node" == "notCompute", right? | 03:36 |
| tchaypo | ls | 03:36 |
| greghaynes | . .. | 03:38 |
| * tchaypo cuddles tmux and its quirky some-times keystroke eating | 03:38 | |
| * tchaypo is sitting in an office right near a myITpc | 03:39 | |
| * tchaypo learns from the advertising on the screen that he will soon have 2 whole Gb of storage space in outlook | 03:39 | |
| StevenK | tchaypo: From like Aug 2013 | 03:39 |
| tchaypo | StevenK: tmux, or noComputer? | 03:40 |
| tchaypo | s/noComputer/notCompute/ | 03:40 |
| StevenK | tchaypo: The 2 whole Gb of storage in Outlook | 03:40 |
| tchaypo | oh right. according to the screen it's "COMING SOON" | 03:40 |
| * StevenK considers connecting to the VPN to find docs to back up his comment. | 03:41 | |
| tchaypo | oh, i believe you | 03:41 |
| StevenK | I do wonder just how many machines we have dedicated to Exchange | 03:44 |
| tchaypo | one. | 03:44 |
| tchaypo | it might be spread across multiple chassis but effectively it's one computer *nods* | 03:44 |
| StevenK | tchaypo: Exchange supports clustering natively | 03:44 |
| greghaynes | Im sorry you know that | 03:45 |
| StevenK | greghaynes: I discussed it vaguely with a friend of mine | 03:45 |
| clarkb | StevenK: you need a lot | 03:45 |
| StevenK | He knows Exchange, the poor bastard, not me. | 03:45 |
| clarkb | we had a ton at Intel and only a quarter the employees | 03:46 |
| StevenK | clarkb: Yeah, my rough calculations were on the order of like 4-6 racks per region | 03:46 |
| *** ramishra has joined #tripleo | 03:46 | |
| StevenK | (With 50 machines per rack) | 03:46 |
| *** blamar has quit IRC | 03:46 | |
| * tchaypo gives dirty looks to th eman with the Escorted Visitor badge and the macbook air | 03:49 | |
| StevenK | Haha | 03:49 |
| clarkb | tchaypo: huh everyone in the seattle office has a macbook | 03:49 |
| StevenK | tchaypo: That was you in Moffet? :-P | 03:50 |
| clarkb | everyone but the tiny corner I sit in but slowly we are losing out | 03:50 |
| StevenK | clarkb: http://www.youtube.com/watch?v=Jln3mi0vfJU | 03:50 |
| *** mestery has joined #tripleo | 03:51 | |
| *** killer_prince is now known as lazy_prince | 03:53 | |
| tchaypo | StevenK: i didn't have an "escorted visitor" badge | 03:55 |
| tchaypo | clarkb: I've got an 8570w.I need to glare at people with macbook airs to take my mind of the awful crushing weight I carry with me. | 03:55 |
| *** blamar has joined #tripleo | 03:56 | |
| StevenK | Haha | 03:56 |
| clarkb | tchaypo: we must resist | 03:56 |
| clarkb | though the bricks are pretty terrible | 03:56 |
| tchaypo | the office muzak is curerntly "i wanna dance with somebody" | 03:56 |
| StevenK | I'm happy enough with the 9470m | 03:57 |
| StevenK | I would like to know where my slice battery is, though | 03:57 |
| clarkb | StevenK: its a decent machine. The touchpad and display could be better though | 03:57 |
| clarkb | tchaypo: you have office music!? | 03:57 |
| StevenK | clarkb: He's in the cafeteria | 03:57 |
| tchaypo | clarkb: we do in the cafeteria | 03:57 |
| clarkb | I tried using the couches in the office today but they didn't agree with my back | 03:57 |
| *** blamar has quit IRC | 03:57 | |
| StevenK | tchaypo: Oh, did you have to get another prox card stapled to the one you got in Moffet? | 03:58 |
| *** blamar has joined #tripleo | 03:58 | |
| tchaypo | no, the one i got in moffet works dandy | 03:59 |
| tchaypo | greghaynes: if i'm reading things correctly, you're working on getting percona working on 1 node first, and you're planning to attack the building of multiple nodes later | 03:59 |
| tchaypo | greghaynes: is that right? | 03:59 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Only create a tmpfs big enough for DIB_MIN_TMPFS https://review.openstack.org/79074 | 04:00 |
| greghaynes | tchaypo: works on one node right now, moving on to the >1 node setup now | 04:01 |
| tchaypo | ah, hence why you spent the day learning about heat | 04:02 |
| greghaynes | tchaypo: for that its basically template out configs so the nodes know about each other, and correctly bootstrap a first node | 04:02 |
| *** nati_ueno has joined #tripleo | 04:02 | |
| greghaynes | yep, and also I kinda have been wanting to know it :) | 04:02 |
| greghaynes | therse probably more stuff for handling updates, but not there yet | 04:03 |
| greghaynes | er, theres | 04:03 |
| lifeless | StevenK: the zbook14 slice arrived this morning | 04:03 |
| tchaypo | I was planning to attack the problem from the other direction | 04:03 |
| greghaynes | oh? | 04:04 |
| tchaypo | ie, get multiple nodes running, then figure out how to get one of the services (rabbitmq seems like a good target) to know about the other nodes | 04:04 |
| greghaynes | I think rabbitmq should be "working" | 04:04 |
| greghaynes | I havent actually tested it, but in theory its all there | 04:04 |
| tchaypo | yeah, that's why i think it's a good target | 04:04 |
| lifeless | tchaypo: ++ | 04:05 |
| greghaynes | I think jprovazn posted a heat template that might be a good place to poke at | 04:05 |
| StevenK | lifeless: Nice. It's been 7 weeks since I ordered it. | 04:05 |
| greghaynes | -> bbiab | 04:05 |
| tchaypo | greghaynes: it sounds like i'm going to be duplicating some of your effort - but i think i need to learn about heat too, so i think i'm going to go ahead anyway and assume I'll be throwing away my work | 04:06 |
| tchaypo | I got my slice about 3 days after ordering it | 04:06 |
| StevenK | *sadface* | 04:06 |
| * StevenK goes to prepare a portion of dinner | 04:06 | |
| tchaypo | It'd be more excited if my RAM had turned up | 04:07 |
| tchaypo | lifeless: as an aside - I was thinking that at some point when I'm looking for a pre-coffee task I'd like to rewrite README.md as README.rst and get it linked into the rest of the docs | 04:08 |
| tchaypo | it seems like it makes a more better landing page than the current index.html or devtest.html | 04:08 |
| tchaypo | especially with the stuff you added today. | 04:09 |
| tchaypo | Also I think I'd like to collect the notes on setting up squid, pypi-mirror etc into one place | 04:09 |
| *** blamar has quit IRC | 04:10 | |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Update selinux policies for nova's use of /mnt/state https://review.openstack.org/82976 | 04:10 |
| lifeless | tchaypo: +1, I believe setup.cfg can cause doc/source/index.rst to be the pypi registered doc | 04:12 |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-heat-templates: WIP: Switch overcloud to software-config https://review.openstack.org/81666 | 04:12 |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Update selinux policies for glance's use of /mnt/state https://review.openstack.org/82978 | 04:12 |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Update selinux policies for neutron's use of /mnt/state and /var/run https://review.openstack.org/82980 | 04:13 |
| StevenK | lifeless, SpamapS: Do you two want to push https://review.openstack.org/#/c/82941/ along? | 04:14 |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Update selinux policies to allow openssl to read keystone ssl files https://review.openstack.org/82981 | 04:15 |
| lifeless | its 9pm for SpamapS probably still doing familty time | 04:15 |
| SpamapS | lifeless: I did family time sprinkled throughout the day actually | 04:15 |
| SpamapS | so I've got an evening pass | 04:15 |
| * tchaypo suddenly understands what "merge.py" does | 04:16 | |
| * StevenK fights with jq | 04:17 | |
| lifeless | SpamapS: .oO | 04:17 |
| lifeless | SpamapS: so land 82941 PLOX | 04:17 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Use --nodes for setup-baremetal in toci_devtest https://review.openstack.org/82941 | 04:18 |
| SpamapS | lifeless: as you wish | 04:18 |
| StevenK | SpamapS: Heh, re: your comment on 78467, it requires ^ | 04:19 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Scrub ironic config in the seed config. https://review.openstack.org/82983 | 04:23 |
| lifeless | SpamapS: \o/ | 04:23 |
| StevenK | Blah | 04:23 |
| lifeless | 82983 is needed to land the tie patch for ironic without breaking stuff | 04:23 |
| StevenK | That one liner is horrible | 04:24 |
| * tchaypo puts on One Direction in order to drown out the dreadful office muzak | 04:24 | |
| *** matsuhashi has joined #tripleo | 04:25 | |
| *** mestery has quit IRC | 04:25 | |
| StevenK | tchaypo: So you're replacing dreadful music with other dreadful music? | 04:25 |
| StevenK | Seems kinda redudant. | 04:25 |
| tchaypo | speaking of which - when is eurovision? | 04:26 |
| tchaypo | oh pooples. the final is the 10th may, which means it's going to air in Aus while I'm in the air from Aus | 04:26 |
| SpamapS | StevenK: and a rebase.. :-P | 04:27 |
| *** killer_prince has joined #tripleo | 04:27 | |
| *** eghobo has joined #tripleo | 04:27 | |
| StevenK | tchaypo: Are you going to raise a session at the summit that it must not conflict with eurovision? | 04:27 |
| SpamapS | ok.. brief switch back to family time and then back to the review treadmill | 04:27 |
| *** vkozhukalov has quit IRC | 04:28 | |
| StevenK | SpamapS: Bah :-) | 04:28 |
| tchaypo | No. What I am going to raise is a suggestion that the next mid-cycle should be ~august 4-8 and should be located in Brisbane | 04:30 |
| tchaypo | which will be great, because everyone can turn up a few days early and attend pycon-au | 04:30 |
| clarkb | that conflicts with all the summer plans :P | 04:31 |
| clarkb | we only get that one week of summer here | 04:31 |
| StevenK | Haha | 04:31 |
| tchaypo | hah | 04:31 |
| tchaypo | I'm sure brisbane mid-winter will be nicer than your summer anyway | 04:32 |
| tchaypo | wherever it i s that you are (seattle, right?) | 04:32 |
| StevenK | mid-lack-of-summer | 04:32 |
| StevenK | Er | 04:32 |
| clarkb | ya seattle | 04:32 |
| StevenK | We don't have winter in Australia, we have lack-of-summer | 04:32 |
| StevenK | And the US doesn't have summer, it has lack-of-winter | 04:33 |
| tchaypo | pfft. sydney has winter from late feb through to early december. | 04:33 |
| clarkb | no most of the US has proper summer | 04:33 |
| clarkb | hot humid days for months in massive chunks of the country | 04:34 |
| StevenK | clarkb: It isn't a proper summer if it doesn't hit 104 | 04:34 |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Remove backwards compatability in setup-baremetal https://review.openstack.org/78467 | 04:35 |
| tchaypo | so the good news is, all the eurovision songs are up on the website already, so i just need to scrape them and i can take them onto the plane with me | 04:35 |
| StevenK | ... yay ? | 04:36 |
| tchaypo | oh goodie. A song about baking a cake with lyrics including "tap tap tap to-coo-coo" | 04:36 |
| *** nosnos has joined #tripleo | 04:36 | |
| *** blamar has joined #tripleo | 04:46 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Store seed details in the JSON https://review.openstack.org/79051 | 04:58 |
| *** ramishra_ has joined #tripleo | 05:04 | |
| *** ramishra has quit IRC | 05:07 | |
| *** cwolferh has quit IRC | 05:24 | |
| *** rpodolyaka1 has joined #tripleo | 05:26 | |
| *** cwolferh has joined #tripleo | 05:27 | |
| lifeless | SpamapS: https://review.openstack.org/#/c/82741/ should be ok now | 05:28 |
| *** lazy_prince has quit IRC | 05:28 | |
| *** cwolferh has quit IRC | 05:32 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: serial-console: Use udev rules to startup getty https://review.openstack.org/81616 | 05:33 |
| *** fandi has joined #tripleo | 05:44 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Teach testenv about numbers and bridges https://review.openstack.org/80874 | 05:45 |
| *** e0ne has joined #tripleo | 05:55 | |
| *** nati_ueno has quit IRC | 05:57 | |
| StevenK | 2014-03-26 05:29:04.980 | File "/opt/stack/venvs/keystone/local/lib/python2.7/site-packages/netaddr/ip/__init__.py", line 13, in <module> | 05:59 |
| StevenK | 2014-03-26 05:29:04.981 | from netaddr.strategy import ipv4 as _ipv4, ipv6 as _ipv6 | 05:59 |
| StevenK | 2014-03-26 05:29:04.981 | ValueError: bad marshal data (unknown type code) | 05:59 |
| StevenK | Blink | 05:59 |
| *** lazy_prince has joined #tripleo | 06:00 | |
| killer_prince | lifeless: regarding your comment about "Refactor code to select boot kernel (https://review.openstack.org/#/c/79873)" | 06:10 |
| *** sdake has quit IRC | 06:12 | |
| killer_prince | lifeless: first comment about syslinux, i looked at the code and i see that it is guarded by os only for kernel selection and not for syslinux configuration.. | 06:12 |
| *** sdake_ has quit IRC | 06:12 | |
| killer_prince | lifeless: this is the reason, i put a check for syslinux directory and then create config file.. | 06:13 |
| lifeless | killer_prince: hmmm, will need to think about that | 06:14 |
| killer_prince | lifeless: okay, for the 2nd comment, i have put the common code in a function as its getting reused in baremetal element and vm element.. | 06:14 |
| lifeless | killer_prince: but it seems like there would be no refactoring to do if the deprecated CLI was just removed | 06:14 |
| killer_prince | lifeless: and i intend to euse the same code later in one of the localboot element for tripleo which is pending review (https://review.openstack.org/#/c/79289/) Adds local storage Boot support in case PXE boot fails. | 06:16 |
| killer_prince | s/euse/reuse | 06:16 |
| lifeless | ok, I'll re-review tomorrow in that light | 06:16 |
| killer_prince | thanks.. | 06:16 |
| *** rpodolyaka1 has quit IRC | 06:17 | |
| killer_prince | i am fine with deleting the cli but we still have some common code shared between vm and baremetal element which needs to be consolidated (as per community feedback)... | 06:18 |
| killer_prince | I will wait for your 2nd review.. :) | 06:19 |
| *** sdake has joined #tripleo | 06:24 | |
| *** sdake has quit IRC | 06:24 | |
| *** sdake has joined #tripleo | 06:24 | |
| *** sdake_ has joined #tripleo | 06:27 | |
| *** matsuhashi has quit IRC | 06:33 | |
| *** matsuhashi has joined #tripleo | 06:35 | |
| *** eghobo has quit IRC | 06:44 | |
| *** e0ne has quit IRC | 06:47 | |
| *** e0ne has joined #tripleo | 06:48 | |
| *** e0ne has quit IRC | 06:51 | |
| *** rdopieralski has joined #tripleo | 06:56 | |
| StevenK | 2014-03-26 06:22:04.504 | Waiting for keystone to initialise... | 07:01 |
| StevenK | 2014-03-26 06:22:05.966 | An unexpected error prevented the server from fulfilling your request. (HTTP 500) | 07:01 |
| StevenK | Gar | 07:01 |
| *** vkozhukalov has joined #tripleo | 07:02 | |
| *** matsuhas_ has joined #tripleo | 07:03 | |
| *** matsuhas_ has quit IRC | 07:04 | |
| *** jcoufal has joined #tripleo | 07:04 | |
| *** matsuhas_ has joined #tripleo | 07:06 | |
| *** matsuhashi has quit IRC | 07:07 | |
| *** mrunge has joined #tripleo | 07:32 | |
| *** ramishra_ has quit IRC | 07:34 | |
| *** jprovazn has joined #tripleo | 07:39 | |
| *** thouveng has joined #tripleo | 07:49 | |
| *** matsuhas_ has quit IRC | 07:59 | |
| *** matsuhashi has joined #tripleo | 08:04 | |
| GheRivero | morning all | 08:12 |
| lsmola_ | SpamapS: hello, are you around? | 08:17 |
| *** pblaho has joined #tripleo | 08:18 | |
| *** panda has joined #tripleo | 08:18 | |
| lifeless | GheRivero: morning! | 08:19 |
| GheRivero | lifeless: morning | 08:20 |
| lifeless | GheRivero: Ng: https://review.openstack.org/#/c/82683/ passed now | 08:21 |
| GheRivero | cool | 08:23 |
| *** hashar has joined #tripleo | 08:23 | |
| *** Matt2 has joined #tripleo | 08:31 | |
| rpodolyaka | morning all | 08:34 |
| killer_prince | lifeless: i am trying to fix bug id 1231946 (Fedora images built on a UEFI host are configured to boot UEFI). but as i understand, uefi needs GPT formatted disk. so is it okay to assume that if user is creating a uefi compatible disk image using vm element, vm element should create a GPT fromatted disk and configure UEFI boot partition, and in all other case, use MBR formatted disk and install MBR compatible boot loaders.. | 08:35 |
| lifeless | killer_prince: anything deployed via nova-bm will be partitioned MBR today. | 08:35 |
| lifeless | killer_prince: we haven't added GPT support. | 08:36 |
| *** jistr has joined #tripleo | 08:37 | |
| killer_prince | lifeless: is there any work going on for GPT support in nova-bm/ironic.. i just wanted to get involved | 08:37 |
| lifeless | killer_prince: I think so, don't know the status | 08:37 |
| lifeless | killer_prince: it would be ironic | 08:37 |
| lifeless | killer_prince: so, vm images will always be MBR for now, because booting UEFI is a pita with kvm (doable, but sheese) | 08:38 |
| lifeless | killer_prince: and non-vm will always be MBR partitions because thats what nova-bm creates. | 08:38 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Remove an excess cp of disk images. https://review.openstack.org/82683 | 08:38 |
| lifeless | GheRivero: Ng: can I get two hell yeahs for https://review.openstack.org/#/c/82983/ ? | 08:39 |
| lifeless | StevenK: https://review.openstack.org/#/c/78467/ | 08:39 |
| killer_prince | lifeless: k.. but i see some code related to efi in vm element.. is it going to be removed soon or should vm element always setup a mbr compatible boot loader.. | 08:39 |
| lifeless | killer_prince: you do ? | 08:40 |
| lifeless | killer_prince: if its from a1beca7b44a788dc627a210550ab2d96714f391c it seems to be disabling efi boot targets | 08:41 |
| lifeless | killer_prince: under some circumstances | 08:42 |
| *** ifarkas has joined #tripleo | 08:42 | |
| killer_prince | okay.. | 08:44 |
| killer_prince | so then how to go about bug id 1231946.. Fedora images built on a UEFI host are configured to boot UEFI | 08:44 |
| *** rpodolyaka1 has joined #tripleo | 08:45 | |
| *** rlandy has joined #tripleo | 08:48 | |
| *** rpodolyaka1 has quit IRC | 08:48 | |
| *** gcha has joined #tripleo | 08:48 | |
| rpodolyaka | a quick question on 'untriaged bugs bot'. Should we list 'Incomplete' bugs as untriaged? | 08:49 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Scrub ironic config in the seed config. https://review.openstack.org/82983 | 08:51 |
| openstackgerrit | Alexis Lee proposed a change to openstack/tripleo-incubator: Fall back to $0 if BASH_SOURCE is unset https://review.openstack.org/81539 | 08:53 |
| lifeless | rpodolyaka: incomplete_with_response yes | 08:56 |
| lifeless | rpodolyaka: incomplete_without_response, no | 08:57 |
| *** jcoufal has quit IRC | 08:59 | |
| *** jcoufal has joined #tripleo | 09:00 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Add Ironic support to incubator. https://review.openstack.org/81959 | 09:02 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Remove backwards compatability in setup-baremetal https://review.openstack.org/78467 | 09:02 |
| openstackgerrit | Matthew Macdonald-Wallace proposed a change to openstack/tripleo-image-elements: Install the check_mk Agent https://review.openstack.org/81485 | 09:04 |
| *** athomas has joined #tripleo | 09:10 | |
| *** martyntaylor has joined #tripleo | 09:10 | |
| *** derekh has joined #tripleo | 09:11 | |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/diskimage-builder: debian: install cloud-init on Wheezy https://review.openstack.org/82325 | 09:21 |
| rpodolyaka | lifeless: ack, thanks! | 09:22 |
| *** andreaf has joined #tripleo | 09:22 | |
| gonery | lifeless thanks for the review | 09:23 |
| *** gonery is now known as Goneri | 09:23 | |
| *** hashar has quit IRC | 09:28 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Adds "element-provides" support to element dependencies https://review.openstack.org/79421 | 09:30 |
| lifeless | StevenK: ^^^ was a rebase only | 09:30 |
| *** lucasagomes has joined #tripleo | 09:31 | |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/tripleo-incubator: doc: ReST syntax changes to improve readability https://review.openstack.org/82315 | 09:33 |
| *** howleyt_ has joined #tripleo | 09:34 | |
| *** hashar has joined #tripleo | 09:35 | |
| *** Rakesh5 has joined #tripleo | 09:36 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Remove unnecessary cinder metadata https://review.openstack.org/80600 | 09:38 |
| *** giulivo has joined #tripleo | 09:39 | |
| *** e0ne has joined #tripleo | 09:39 | |
| andrearosa | question about tripleo-geat-templates: can anyone tell me where the notcompute.yaml file is used? It's not used in the makefile and I can't see other places where it's used | 09:39 |
| andrearosa | s/tripleo-geat-templates/tripleo-heat-templates | 09:39 |
| *** derekh has quit IRC | 09:42 | |
| *** derekh has joined #tripleo | 09:42 | |
| *** pbelanyi has joined #tripleo | 09:44 | |
| *** rpodolyaka1 has joined #tripleo | 09:46 | |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-incubator: WIP - configure a node ephemeral disk https://review.openstack.org/82776 | 09:46 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add fedora-rdo-icehouse element https://review.openstack.org/80658 | 09:47 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add console kernel parameters to extlinux configuration https://review.openstack.org/82822 | 09:47 |
| *** bauzas has joined #tripleo | 09:51 | |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Add keepalive option to rabbitmq config https://review.openstack.org/81920 | 09:51 |
| *** rpodolyaka1 has quit IRC | 09:51 | |
| lifeless | GheRivero: hey - in https://bugs.launchpad.net/tripleo/+bug/1295732 | 09:54 |
| uvirtbot | Launchpad bug 1295732 in tripleo "tripleo devtest overcloud startup takes too long " [High,Triaged] | 09:54 |
| lifeless | GheRivero: virtio makes it fly for me :) | 09:54 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Fall back to $0 if BASH_SOURCE is unset https://review.openstack.org/81539 | 09:56 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: qpidd user should own sasldb file https://review.openstack.org/81908 | 09:57 |
| GheRivero | lifeless: will post data with virtio and no virtio to the bug later this morning just to compare | 09:57 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Add hostname of Test env to json https://review.openstack.org/80603 | 09:57 |
| GheRivero | apparently my builds where dying because the user image wasn't built with apt-sources and failed to fetch the mirrors | 09:59 |
| GheRivero | but it's solved now | 09:59 |
| derekh | lifeless: did you mean to approve https://review.openstack.org/#/c/79074/ , I didn't pass CI and I believe will cause us failures in all other CI runs | 09:59 |
| derekh | *It didn't | 10:00 |
| lifeless | GheRivero: cool | 10:00 |
| howleyt_ | GheRivero, lifeless: I believe that smulcahy tried with virtio and saw no noticeable difference | 10:00 |
| lifeless | howleyt_: I see 23m drop to 16m | 10:01 |
| lifeless | erm 16m50s, so 17m | 10:01 |
| GheRivero | but i don't like the solution. Passing elements via NODE_DIST will make to add those elements also to the build_ramdisk image | 10:01 |
| *** jp_at_hp has joined #tripleo | 10:01 | |
| smulcahy | ah, someone called? | 10:02 |
| smulcahy | :) | 10:02 |
| smulcahy | I need a communications channel multiplexer or something | 10:02 |
| smulcahy | or more eyes | 10:02 |
| lifeless | GheRivero: sure, I see you're working on a nicer thing | 10:02 |
| GheRivero | lifeless: https://review.openstack.org/#/c/82739/2 | 10:02 |
| lifeless | GheRivero: but - the ramdisk image needs the same apt-sources etc too :) | 10:02 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Inject admin username and tenant name in config https://review.openstack.org/81813 | 10:03 |
| lifeless | derekh: Apparently I was blind. | 10:03 |
| lifeless | derekh: merged at 5pm. Now 11pm. Have we had passes since then ? | 10:03 |
| smulcahy | I need to get my workstation back on the network and run a bunch more speed tests with a few configs, to make sure I'm seeing consistent numbers | 10:03 |
| derekh | lifeless: don't think so http://goodsquishy.com/downloads/tripleo-jobs.html | 10:04 |
| smulcahy | but I need to wait for IT to fix my network first so I can't say when that will happen | 10:04 |
| lifeless | derekh: nuts; can has revert? | 10:04 |
| derekh | lifeless: we can revert, move image outside of tmpfs or increase DIB_MIN_TMPFS | 10:04 |
| lifeless | derekh: revert is fast | 10:05 |
| derekh | lifeless: ok, will do | 10:05 |
| lifeless | derekh: the rest we have to iterate on; so lets revert, and we can reinstate with the other stuff at the same time | 10:05 |
| lifeless | derekh: btw - moving image outside tmpfs, did you see someone suggesting using a new tmpfs? I like that idea | 10:05 |
| derekh | lifeless: Ya, I saw that, works for me | 10:06 |
| howleyt_ | lifeless: maybe I'm misrepresenting smulcahy and it was just that he still found it slow with virtio enabled. | 10:06 |
| openstackgerrit | Derek Higgins proposed a change to openstack/diskimage-builder: Revert "Only create a tmpfs big enough for DIB_MIN_TMPFS" https://review.openstack.org/83021 | 10:07 |
| *** matsuhashi has quit IRC | 10:09 | |
| howleyt_ | On a different note, in Mustache, can I differentiate empty vs. false values? | 10:09 |
| derekh | lifeless Ng ; revert to fix CI, should we wait on CI to prove that this was the problem? | 10:10 |
| lifeless | derekh: the failure is the one in the failed run on what I wrongly landed. No. | 10:10 |
| lifeless | howleyt_: empty is falsey. So no. | 10:11 |
| lifeless | at least, I don't know how to ;) | 10:11 |
| derekh | lifeless: yup it is | 10:11 |
| lifeless | derekh: I mean - no need to wait | 10:12 |
| howleyt_ | lifeless: ya, that's what I thought. | 10:12 |
| derekh | lifeless: ok, agreed, lets land it | 10:12 |
| *** yamahata has quit IRC | 10:15 | |
| howleyt_ | lifeless: any chance you could take a quick look at https://review.openstack.org/#/c/81920/ so that I know have the right idea on use of symlinks, etc. | 10:21 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Teach testenv about numbers and bridges https://review.openstack.org/80874 | 10:21 |
| dtantsur | slagle, ping re instack-undercloud | 10:30 |
| dtantsur | slagle, I'm interesting now in running instack-undercloud with Ironic. Am I right assuming by default it's going to use nova baremetal (judging by answers file) | 10:31 |
| *** akrivoka has joined #tripleo | 10:33 | |
| *** matsuhashi has joined #tripleo | 10:39 | |
| dtantsur | ... but the first instack question is what are VIRTUAL_POWER_HOST? "instack" vm IP? My laptop IP? | 10:40 |
| dtantsur | ifarkas, maybe you know ^^^ | 10:40 |
| lifeless | howleyt_: broadly yes; have reviewed the install.d file only, and given some detailed feedback | 10:42 |
| lifeless | dtantsur: what are you trying to do ? | 10:43 |
| dtantsur | lifeless, I'm trying to use instack for installing undercloud on VMs to develop Ironic | 10:43 |
| dtantsur | however, it's the first time I actually see instack, thus my questions :) | 10:44 |
| lifeless | ah | 10:44 |
| lifeless | I've never used instack either | 10:44 |
| lifeless | however trunk tripleo doesn't support ironic yet | 10:44 |
| lifeless | so it may be premature to use instack | 10:44 |
| lifeless | have you tried basic devtest ? | 10:45 |
| *** hashar has quit IRC | 10:45 | |
| dtantsur | lifeless, actually not | 10:46 |
| dtantsur | it seemed to me instack could be easier (which I'm no longer sure is true) | 10:46 |
| howleyt_ | lifeless: thank you kindly. | 10:47 |
| *** rpodolyaka1 has joined #tripleo | 10:47 | |
| *** hashar has joined #tripleo | 10:48 | |
| dtantsur | lifeless, instack-undercloud is based on tripleo mostly, right? So for me it does not matter whether I use devtest or instack, I still have to setup Ironic manually? | 10:50 |
| lifeless | dtantsur: not at all | 10:50 |
| *** howleyt_ has quit IRC | 10:50 | |
| lifeless | dtantsur: see the instructions at the top of https://bugs.launchpad.net/ironic/+bug/1297063 | 10:50 |
| uvirtbot | Launchpad bug 1297063 in ironic "failed to deploy overcloud node" [Undecided,New] | 10:50 |
| lifeless | that will deploy a seed vm with ironic in it, use it to deploy an undercloud with ironic in it, use that to deploy an overcloud etc | 10:51 |
| dtantsur | lifeless, cool, thanks | 10:51 |
| dtantsur | object has no attribute '__name__' <-- the same with devstack, oh | 10:52 |
| *** rpodolyaka1 has quit IRC | 10:52 | |
| lifeless | dtantsur: iz bug :) | 10:53 |
| *** hashar has quit IRC | 10:54 | |
| *** hashar has joined #tripleo | 10:54 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Revert "Only create a tmpfs big enough for DIB_MIN_TMPFS" https://review.openstack.org/83021 | 11:05 |
| openstackgerrit | Stuart McLaren proposed a change to openstack/tripleo-image-elements: Glance: Allow configuring default_log_levels https://review.openstack.org/76557 | 11:07 |
| *** noslzzp has joined #tripleo | 11:13 | |
| *** fandi has quit IRC | 11:21 | |
| openstackgerrit | Cian O'Driscoll proposed a change to openstack/tripleo-heat-templates: Update templates to default to preserve ephemeral https://review.openstack.org/83037 | 11:22 |
| *** e0ne has quit IRC | 11:23 | |
| *** e0ne has joined #tripleo | 11:23 | |
| *** e0ne has quit IRC | 11:29 | |
| *** yamahata has joined #tripleo | 11:40 | |
| *** hashar has quit IRC | 11:43 | |
| *** rpodolyaka1 has joined #tripleo | 11:47 | |
| *** jcoufal has quit IRC | 11:52 | |
| *** rpodolyaka1 has quit IRC | 11:52 | |
| *** IvanBerezovskiy has joined #tripleo | 11:54 | |
| *** jcoufal has joined #tripleo | 11:55 | |
| *** bogdando has quit IRC | 11:58 | |
| *** athomas has quit IRC | 11:59 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tuskar: Adding swift role back https://review.openstack.org/82499 | 11:59 |
| *** e0ne has joined #tripleo | 12:00 | |
| *** lblanchard has joined #tripleo | 12:00 | |
| *** e0ne has quit IRC | 12:04 | |
| *** e0ne has joined #tripleo | 12:05 | |
| *** Slower has quit IRC | 12:05 | |
| *** athomas has joined #tripleo | 12:05 | |
| derekh | jogo: I'm trying to plot the run times for successful tripleo runs, took a look at graphite and the graph you linked on monday, the timers don't seem to be returning values I'm expecting | 12:05 |
| derekh | jogo: any pointers ? http://graphite.openstack.org/render/?width=586&height=308&from=-36hours&target=movingAverage(stats.timers.zuul.pipeline.check-tripleo.resident_time.mean,%20%2712hours%27) | 12:05 |
| *** yamahata has quit IRC | 12:06 | |
| *** fandi has joined #tripleo | 12:09 | |
| *** e0ne has quit IRC | 12:10 | |
| *** matsuhashi has quit IRC | 12:10 | |
| *** bogdando has joined #tripleo | 12:10 | |
| *** matsuhashi has joined #tripleo | 12:11 | |
| *** matsuhas_ has joined #tripleo | 12:13 | |
| *** matsuhashi has quit IRC | 12:13 | |
| *** e0ne has joined #tripleo | 12:14 | |
| *** hewbrocca has joined #tripleo | 12:14 | |
| chmouel | i am still pretty confuse how the os-*-config tools works with heat, is there any image somewhere that describes how that stuff works | 12:18 |
| chmouel | i understand from the description of each projects in that they are applying configuration from a json/yaml file | 12:18 |
| chmouel | but where is that configuration come from? | 12:18 |
| hewbrocca | chmouel: configuration comes from cloud-init | 12:21 |
| chmouel | hewbrocca: or cfn right? | 12:21 |
| hewbrocca | When Heat launches each instance it passes config to nova | 12:21 |
| hewbrocca | which makes it available to the instance | 12:21 |
| hewbrocca | os-* tools can then retrieve it and apply it | 12:21 |
| chmouel | ok but that's only on provisioning, is it on stack-update as well? | 12:21 |
| hewbrocca | yes | 12:22 |
| hewbrocca | os-refresh-config polls, IIUC | 12:22 |
| hewbrocca | when it finds new config it pulls it down and apply it | 12:22 |
| chmouel | cool thanks, so what kind of config this is supposed to do? | 12:23 |
| *** e0ne_ has joined #tripleo | 12:23 | |
| chmouel | it can be anything or just launch other configuration management | 12:23 |
| chmouel | i.e: puppet/ansible etc.. | 12:23 |
| *** e0ne has quit IRC | 12:23 | |
| hewbrocca | for tripleo specifically, the os-* tools run scripts from each element | 12:24 |
| hewbrocca | but | 12:24 |
| hewbrocca | if you're not using tripleo, you can do whatever you want | 12:24 |
| hewbrocca | there's more new config stuff in Heat now | 12:25 |
| hewbrocca | cloud config | 12:25 |
| *** e0ne has joined #tripleo | 12:25 | |
| chmouel | yeah saw the software-config thing, and that's where my question were coming from, if that's tighten to the os-*-config tools | 12:26 |
| hewbrocca | it's not | 12:26 |
| hewbrocca | the os-* tools are tripleo specific | 12:26 |
| chmouel | (i indeed meant OS::Config::CloudConfig) | 12:27 |
| chmouel | ok | 12:27 |
| hewbrocca | so, yeah, definitely worth checking with Heat guys for the latest | 12:28 |
| chmouel | would you be able to use CloudConfig directly without the os-*-config? | 12:28 |
| hewbrocca | I *think* so | 12:28 |
| hewbrocca | but talk to the heat guys | 12:28 |
| *** e0ne_ has quit IRC | 12:28 | |
| *** jcoufal has quit IRC | 12:30 | |
| chmouel | cool, okay thanks a lot, i guess i have to test it to see how stuff fits together in real | 12:30 |
| *** jcoufal has joined #tripleo | 12:30 | |
| hewbrocca | no problem | 12:31 |
| hewbrocca | Steve Baker or Steve Hardy will be delighted to guide you through | 12:31 |
| hewbrocca | they're excited to get folks using this stuff I think | 12:31 |
| *** geerdest has joined #tripleo | 12:36 | |
| *** geerdest has joined #tripleo | 12:36 | |
| *** dprince has joined #tripleo | 12:41 | |
| *** lucasagomes is now known as lucas-hungry | 12:44 | |
| openstackgerrit | Trent Geerdes proposed a change to openstack/diskimage-builder: Adding pypi-mirror dependencies for redhat https://review.openstack.org/82513 | 12:44 |
| *** matty_dubs|gone is now known as matty_dubs | 12:46 | |
| *** rpodolyaka1 has joined #tripleo | 12:48 | |
| *** hashar has joined #tripleo | 12:49 | |
| *** rpodolyaka1 has quit IRC | 12:52 | |
| *** weshay has joined #tripleo | 13:05 | |
| *** e0ne_ has joined #tripleo | 13:07 | |
| *** matsuhas_ has quit IRC | 13:07 | |
| *** CaptTofu has quit IRC | 13:07 | |
| *** mrunge has quit IRC | 13:08 | |
| *** CaptTofu has joined #tripleo | 13:08 | |
| *** CaptTofu has quit IRC | 13:09 | |
| *** CaptTofu has joined #tripleo | 13:09 | |
| *** e0ne has quit IRC | 13:10 | |
| *** CaptTofu has quit IRC | 13:14 | |
| *** CaptTofu_ has joined #tripleo | 13:14 | |
| *** julim has joined #tripleo | 13:16 | |
| *** yamahata has joined #tripleo | 13:18 | |
| SpamapS | hewbrocca: you got a few details wrong in relaying to chmouel how os-*-config works | 13:27 |
| SpamapS | chmouel: ^^ | 13:27 |
| hewbrocca | SpamapS: LOL I'm confident that is the case | 13:28 |
| SpamapS | os-collect-config gathers information from multiple sources | 13:28 |
| hewbrocca | I hope I didn't completely butcher them... | 13:28 |
| SpamapS | ec2 metadata, /var/lib/heat-cfntools/cfn-init-data (which is written at the request of Heat by cloud-init), and the CFN API of Heat. | 13:28 |
| SpamapS | I think its time I make a picture actually :) | 13:29 |
| chmouel | SpamapS: heh thanks, yeah that would be very nice for newcommer as this is the confusing part of tripleO :) | 13:29 |
| SpamapS | hewbrocca: os-refresh-config is really really dumb, and just runs like run-parts through 4 dirs, /opt/stack/os-config-refresh/{pre-configure.d,configure.d,post-configure.d,migration.d} | 13:29 |
| chmouel | SpamapS: is the multiple sources very important in the context of tripleO ? | 13:29 |
| hewbrocca | SpamapS: picture worth a thousand words, yes... | 13:30 |
| SpamapS | chmouel: yes, we use them all | 13:30 |
| SpamapS | whenever any of them change (as determined by comparing to the previously known version), the command is run with an environment variable pointing to the new metadata. | 13:30 |
| SpamapS | the command == os-refresh-config in our scenario | 13:31 |
| chmouel | SpamapS: ah ok i see, that's because there is no a unifomize way at the moment to communicate with instance among openstack service ? | 13:31 |
| SpamapS | but could very well be puppet, chef, or "your magical configuration machine of choice" | 13:31 |
| SpamapS | chmouel: there is, but it is through ec2 metadata which has.. some issues. ;) | 13:32 |
| chmouel | hehe yeah i mean openstack services in an native openstack way ;) | 13:32 |
| *** yassine_ has joined #tripleo | 13:32 | |
| *** noslzzp has quit IRC | 13:34 | |
| *** nosnos has quit IRC | 13:35 | |
| hewbrocca | SpamapS: OK, so I had the basic idea right, I just messed up the delivery mechanism | 13:37 |
| hewbrocca | I didn't realize there were 4 of them... | 13:37 |
| * hewbrocca boggles | 13:37 | |
| *** Rakesh5 has quit IRC | 13:39 | |
| Ng | morning | 13:44 |
| *** killer_prince has quit IRC | 13:46 | |
| *** lazy_prince is now known as killer_prince | 13:46 | |
| *** rpodolyaka1 has joined #tripleo | 13:49 | |
| *** lucas-hungry is now known as lucasagomes | 13:50 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Install python-keystoneclient in the swiftclient venv https://review.openstack.org/75932 | 13:52 |
| *** rpodolyaka1 has quit IRC | 13:53 | |
| *** noslzzp has joined #tripleo | 13:56 | |
| openstackgerrit | Thom Leggett proposed a change to openstack/diskimage-builder: Uses policy-rc.d to prevent dpkg starting daemons https://review.openstack.org/82074 | 13:56 |
| Goneri | chmouel, hewbrocca, SpamapS: Do I get the idea right? http://people.debian.org/~goneri/stuff/os-refresh-config-flow.pdf | 14:02 |
| ifarkas | slagle, ping | 14:04 |
| lsmola_ | SpamapS: hello | 14:07 |
| *** rdopieralski has quit IRC | 14:07 | |
| lsmola_ | SpamapS: so seems to me like there is not an easy way around https://bugs.launchpad.net/tripleo/+bug/1291905 | 14:07 |
| uvirtbot | Launchpad bug 1291905 in tripleo "new resources are not available for signalling during an update -> heat stack update timeout" [High,In progress] | 14:07 |
| lsmola_ | SpamapS: as the os-collect-config relies also on non-existent resource | 14:08 |
| lsmola_ | SpamapS: i could hack it around by explicitly sending the signal defined in the template, like shardy advised me http://paste.fedoraproject.org/88017/60519139/ | 14:10 |
| lsmola_ | SpamapS: but then the completion condition would say: os-apply-config was finished | 14:10 |
| lsmola_ | SpamapS: which is probably what we want | 14:10 |
| lsmola_ | SpamapS: what do you htink? | 14:10 |
| *** lblanchard has quit IRC | 14:12 | |
| *** lblanchard has joined #tripleo | 14:13 | |
| *** CaptTofu_ has quit IRC | 14:15 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tuskar: Adding swift role back https://review.openstack.org/82499 | 14:15 |
| *** CaptTofu has joined #tripleo | 14:16 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Adding pypi-mirror dependencies for redhat https://review.openstack.org/82513 | 14:17 |
| *** CaptTofu has quit IRC | 14:20 | |
| lsmola_ | ifarkas: jistr guys can you look at https://review.openstack.org/#/c/81200/ | 14:33 |
| jistr | lsmola_: does it pick nodes with less RAM but only those with *at least as much RAM as the flavor specifies*? | 14:35 |
| lsmola_ | jistr: yes | 14:36 |
| lsmola_ | jistr: the filter is unchanged | 14:36 |
| *** e0ne has joined #tripleo | 14:36 | |
| lsmola_ | jistr: this just sets Weight, so basically what to pick first | 14:36 |
| jistr | right. cool, thx. | 14:37 |
| lsmola_ | jistr: but only from the set of already filleted machines :-) | 14:37 |
| ifarkas | lsmola_, do you have a bug against nova for this^? | 14:37 |
| lsmola_ | ifarkas: yes, the one it partially fix | 14:38 |
| lsmola_ | ifarkas: the wanted solution is to create exact_match filter | 14:38 |
| lsmola_ | ifarkas: and put that to nova.conf | 14:38 |
| ifarkas | lsmola_, cool, thx | 14:39 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mysql-dev and mariadb-dev elements https://review.openstack.org/79886 | 14:39 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mariadb element https://review.openstack.org/79884 | 14:39 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mysql-common element https://review.openstack.org/79885 | 14:39 |
| *** e0ne_ has quit IRC | 14:39 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tuskar: Adding swift role back https://review.openstack.org/82499 | 14:43 |
| lsmola_ | ifarkas: jistr and please also this, tuskar is currently broken https://review.openstack.org/#/c/82499/ | 14:44 |
| derekh | Is it just me or is devtest fedora now completely broken ? on the undercloud I got eth0 with e1000 and ens4 with virtio , non of these if em1 https://github.com/openstack/tripleo-incubator/blob/master/scripts/devtest_undercloud.sh#L39 | 14:45 |
| jistr | lsmola_: i'll try at some point, but currently i'm being pulled in different hi-prio directions :( | 14:45 |
| lsmola_ | derekh: it was broken yesterday AFAIK, akrivoka tried it | 14:45 |
| lsmola_ | derekh: I will try today | 14:45 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Setting negative RamWeigher https://review.openstack.org/81200 | 14:45 |
| *** mestery has joined #tripleo | 14:46 | |
| akrivoka | yes, I tried multiple times, each time it timed out while waiting for the undercloud heat stack | 14:47 |
| *** mclaren has joined #tripleo | 14:48 | |
| lsmola_ | akrivoka: any log errors? | 14:48 |
| *** rpodolyaka1 has joined #tripleo | 14:49 | |
| derekh | dprince: was https://review.openstack.org/#/c/72003/4 intended to set net devices to em* ? | 14:49 |
| *** CaptTofu has joined #tripleo | 14:50 | |
| derekh | lsmola_: thanks | 14:51 |
| dprince | derekh: yes. But there was a fix to that because I messed up. | 14:51 |
| dprince | derekh: https://review.openstack.org/#/c/82093/ | 14:52 |
| akrivoka | lsmola_: unfortunately I wiped my env, but I will try again today | 14:52 |
| *** hashar has quit IRC | 14:52 | |
| dprince | derekh: both of those should be landed now though | 14:52 |
| derekh | dprince: I seem to have that patch aswell | 14:52 |
| lsmola_ | akrivoka: ok, I will try | 14:53 |
| dprince | derekh: this too? https://review.openstack.org/#/c/82173/ | 14:53 |
| *** rpodolyaka1 has quit IRC | 14:53 | |
| derekh | Mar 26 12:53:32 localhost systemd-udevd[219]: renamed network interface eth0 to ens4 | 14:53 |
| lsmola_ | dprince: cool | 14:53 |
| derekh | dprince: no I don't have that last one, will try it now | 14:54 |
| dprince | derekh: there is a race without that | 14:54 |
| derekh | looks like I was betting on the slow horse | 14:54 |
| dprince | derekh: now you've got me thinking I forgot something else? Man, we need Fedora in the CI... | 14:56 |
| derekh | dprince: agreed we do, the commits to add it are all up there :-( | 14:56 |
| dprince | derekh: That udev rules rename was a bad link. I thought it was something else. You won't hit any issues with that. | 14:58 |
| dprince | derekh: So disregard my link to 82173 | 14:58 |
| derekh | dprince: okie dokie | 14:58 |
| dprince | derekh: what issues are you seeing? | 14:58 |
| dprince | derekh: presumably you just aren't seeing em1 | 14:59 |
| derekh | dprince: my undercloud came up with ens4 instead of em1 | 14:59 |
| dprince | derekh: oh... | 14:59 |
| dprince | derekh: so biosdevname is working then. | 14:59 |
| dprince | derekh: man, I hate VMs for baremetal testing | 15:00 |
| *** julim_ has joined #tripleo | 15:00 | |
| dprince | derekh: this is why the Fedora cloud image uses eth0 | 15:00 |
| *** julim has quit IRC | 15:01 | |
| derekh | dprince: I also tried with virtio and the device was eth0 (still not em1) | 15:01 |
| *** fungi has joined #tripleo | 15:03 | |
| *** yamahata has quit IRC | 15:03 | |
| dprince | derekh: well that might be something different | 15:03 |
| *** fungi has left #tripleo | 15:03 | |
| dprince | derekh: so you've got this https://review.openstack.org/#/c/81596/ | 15:03 |
| dprince | derekh: basically, just wanting to know if everything is up to date | 15:03 |
| dprince | derekh: there is nothing unlanded that you should need (that I know of) | 15:04 |
| dprince | derekh: The ens4 interface name means it is working. And switching the virtio type would have been my suggestion. | 15:05 |
| *** mestery has quit IRC | 15:05 | |
| *** weshay has quit IRC | 15:05 | |
| *** yamahata has joined #tripleo | 15:05 | |
| *** hashar has joined #tripleo | 15:06 | |
| *** gcha has quit IRC | 15:07 | |
| dprince | derekh: The complication here is we now require the 'baremetal' element in devtest for other reasons (extracting kernel and ramdisk). So I can't just remove it... but I could split out the stable interfaces stuff into another element... which is actually what I initially did but during review it was decided living under the 'baremetal' element is best. | 15:08 |
| dprince | derekh: Just sayin, if we need to step back from this for VM testing we can. | 15:08 |
| *** yamahata has quit IRC | 15:09 | |
| *** yamahata has joined #tripleo | 15:10 | |
| *** UtahDave has joined #tripleo | 15:13 | |
| derekh | dprince: sorry was on phone, ya I'm uptodate I have a diskimage-builder from this morning, so if we split it out were would need a different element set for real baremetal and virt? | 15:15 |
| SpamapS | are vms unhappy w/ biosdevname changes? | 15:15 |
| dprince | derekh: yes, and is why I sort of disliked adding this to baremetal... cause it is used for other required things that aren't really baremetally anyway | 15:16 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Alphabetize all element-deps https://review.openstack.org/81635 | 15:17 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Add dib-lint script https://review.openstack.org/81634 | 15:17 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: WIP: Make sure all scripts are set -e https://review.openstack.org/81637 | 15:17 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Enforce alphabetical ordering of element-deps https://review.openstack.org/81636 | 15:17 |
| derekh | SpamapS: devtest fedora net device name is now eth0 (e1000) or ens4 (virtio), devtest is looking for em1 | 15:17 |
| dprince | derekh: Like I said, with virtio the ens4 means it is working. We could just default devtest to use that. | 15:18 |
| bnemec | Grr, pushed the wrong branch. | 15:18 |
| derekh | SpamapS: while your there you may want to resubmit the tmpfs change it was merged this morning and then reverted (as it brought us back to a 4G tmpfs and CI started failing) | 15:18 |
| *** lsmola_ has quit IRC | 15:19 | |
| *** markmc has joined #tripleo | 15:19 | |
| derekh | dprince: but virtio isn't the default in devtest we need the default to work, lifeless doesn't want to make that switch until we have tested are full matrix of support distros with it | 15:20 |
| derekh | dprince: which I could do | 15:20 |
| derekh | given time | 15:20 |
| *** UtahDave has left #tripleo | 15:20 | |
| *** weshay has joined #tripleo | 15:21 | |
| derekh | so I'll need to check virtio works with F20, ubuntu precise and raring on both F20 and Ubuntu Hosts, 6 runs I think | 15:23 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Alphabetize all element-deps https://review.openstack.org/81635 | 15:23 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Add dib-lint script https://review.openstack.org/81634 | 15:23 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: WIP: Make sure all scripts are set -e https://review.openstack.org/81637 | 15:23 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Enforce alphabetical ordering of element-deps https://review.openstack.org/81636 | 15:23 |
| bnemec | I really hosed up that series of commits. :-/ | 15:24 |
| bnemec | Should be fixed now though. | 15:24 |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-incubator: Add clarification on required permissions https://review.openstack.org/83102 | 15:24 |
| dprince | derekh: I think the quickest thing to unbreak it but keep the feature is to split this out into a new element. | 15:25 |
| dprince | derekh: I say lets go with that which should be much less disruptive to everyone I think. And then we can revert my change to devtest that landed yesterday too | 15:26 |
| derekh | dprince: ok, in the meantime I'm gonna test virtio and see what happens | 15:26 |
| dprince | derekh: cool. Sorry for the disruptions man | 15:26 |
| derekh | dprince: no prob, needed a poke to finally get the virt io work done anyways | 15:27 |
| slagle | dprince: if you're going to be poking around in there, i'm seeing that all the dhcp-interface@.services for the generated interfaces are coming up as failed | 15:27 |
| dprince | slagle: oh? | 15:28 |
| slagle | dprince: the ExecStartPre=/usr/local/sbin/dhcp-all-interfaces.sh %I, does not exit non-zero | 15:28 |
| slagle | for the generated interfaces | 15:28 |
| slagle | so then the ifup command still runs, and fails | 15:29 |
| slagle | doesn't seem to cause any issue, fwiw | 15:30 |
| slagle | but does not look very nice in "systemctl --failed" | 15:30 |
| dprince | slagle: which node? | 15:30 |
| slagle | dprince: any of them | 15:31 |
| slagle | i see it on control, compute, blockstorage, and swiftstorage | 15:31 |
| * dprince wants to log off IRC now and go back to bed | 15:32 | |
| *** lsmola_ has joined #tripleo | 15:34 | |
| *** vkozhukalov has quit IRC | 15:34 | |
| dprince | slagle: so systemd is very aggressive about starting things as soon as it sees an interface. | 15:35 |
| dprince | slagle: udev rule fires off, tags teh systemd device, which causes systemd to startup that interface (or try to) | 15:35 |
| dprince | slagle: I wonder if dhcp-all-interfaces.sh is timing out for beeing too slow | 15:36 |
| *** ifarkas has quit IRC | 15:36 | |
| *** IvanBerezovskiy has left #tripleo | 15:37 | |
| dprince | slagle: I'll mess w/ it as best I can. The new way of doing things (using a udev rule to instantiate the systemd service) is much cleaner IMO. But we may need to tweak the service or dhcp-all-interfaces.sh script a bit... | 15:37 |
| *** petertoft has joined #tripleo | 15:39 | |
| dprince | derekh: have you tried getting a substring w/ the jq version in Fedora? | 15:39 |
| * dprince needs to slice and dice the json | 15:40 | |
| *** ifarkas has joined #tripleo | 15:40 | |
| derekh | dprince: nope, I haven't tried much jq vodoo much at all | 15:40 |
| dprince | derekh: who is our resident expert then? | 15:40 |
| derekh | dprince: lifeless and StevenK seem to have read the manual | 15:41 |
| dprince | StevenK: around? | 15:41 |
| derekh | dprince: probably not, I think its night in his TZ | 15:41 |
| dprince | derekh: okay, well I read the man page on Fedora and we don't have a couple functions that upstream jq has... makes me sad but I'll figure something out | 15:45 |
| derekh | :-( | 15:45 |
| *** spzala has joined #tripleo | 15:46 | |
| SpamapS | dprince: I've done a bit of jq | 15:46 |
| SpamapS | dprince: whats up? | 15:46 |
| dprince | SpamapS: http://paste.openstack.org/show/74366/ | 15:48 |
| *** ifarkas has quit IRC | 15:48 | |
| dprince | SpamapS: So I reformatted the jq in devtest_seed.sh so I can inject the baremetal-network JSON into it (for example if I don't want to use 192.0.2.0/24) | 15:49 |
| dprince | SpamapS: All is well really, except on line 14 | 15:49 |
| dprince | SpamapS: in Fedora we have JQ 1.3 and that last [-3:3] is meant to slice off the cidr netmask | 15:50 |
| *** rpodolyaka1 has joined #tripleo | 15:50 | |
| dprince | SpamapS: What I get is jq: error: Cannot index string with object | 15:50 |
| dprince | SpamapS: I've tinkered around a bit, it looks like upstream has a toarray function and has also added some more capabilities to slice/dice maybe. We just don't have those functions yet. | 15:51 |
| dprince | SpamapS: So the question is am I missing something? Or... should I just bang in an individual cidr/netmask parameter and be done w/ it. | 15:52 |
| *** rpodolyaka1 has quit IRC | 15:52 | |
| SpamapS | dprince: hm | 15:53 |
| SpamapS | dprince: It is a bit lazy to just slice it with [-2:3] | 15:53 |
| SpamapS | since /8 will not work | 15:53 |
| SpamapS | err [-3:3] | 15:53 |
| dprince | SpamapS: good point, gotta start somewhere though | 15:54 |
| SpamapS | so I think adding it as a field is a good way to go unless there is a split function where we can split on / | 15:54 |
| dprince | SpamapS: heh. And I thought the lazy approach was to add the extra field. | 15:54 |
| dprince | SpamapS: FWIW the split function is what I really wanted (everything after the /) | 15:55 |
| dprince | SpamapS: but thanks for the confidence man | 15:56 |
| SpamapS | dprince: well you could also defer to shell for that | 15:56 |
| SpamapS | jq is just for parsing the json IMO | 15:56 |
| SpamapS | doing more than that with it is fun, but kind of evil | 15:56 |
| SpamapS | "do one thing" | 15:56 |
| dprince | SpamapS: right, would significantly dirty the jq statement though as we'd need to switch from '' to "" and I'd have to escape like mad | 15:57 |
| tteggel | ci poorly? lots of hudson.remoting errors. | 15:57 |
| dprince | SpamapS: Well, it appears jq already crossed that bridge... | 15:57 |
| dprince | SpamapS: we just don't have it yet | 15:57 |
| SpamapS | I don't think it is a good idea to lean on it for that. | 15:57 |
| SpamapS | I get _why_ it has it | 15:58 |
| *** matty_dubs is now known as matty_dubs|lunch | 15:58 | |
| SpamapS | but to me if jq expressions get more complicated than pulling out keys.. they're reaching too far. | 15:58 |
| *** lazy_prince has joined #tripleo | 15:58 | |
| dprince | SpamapS: fair enough, I've gotten quite a bit of pushback in TripleO when adding new params though... so go figure :) | 15:58 |
| *** killer_prince has quit IRC | 15:59 | |
| *** lazy_prince is now known as killer_prince | 15:59 | |
| dprince | SpamapS: funny how we switch sides on things sometimes :) | 15:59 |
| dprince | SpamapS: lifeless wasn't a big fan of splitting the network cidr off FWIW | 16:00 |
| dprince | SpamapS: Already tried it in the ensure-bridge thing... even though all the distro scripts use it sepearately | 16:00 |
| *** mestery has joined #tripleo | 16:01 | |
| *** eghobo has joined #tripleo | 16:04 | |
| *** eghobo has quit IRC | 16:04 | |
| *** pblaho has quit IRC | 16:04 | |
| *** eghobo has joined #tripleo | 16:05 | |
| *** eghobo has quit IRC | 16:05 | |
| *** eghobo has joined #tripleo | 16:06 | |
| *** lsmola_ has quit IRC | 16:06 | |
| SpamapS | dprince: we should probably stop smoking crack if we want consistency.. but.. whats the fun in that? | 16:06 |
| SpamapS | hewbrocca: https://wiki.openstack.org/wiki/OsCollectConfig | 16:07 |
| SpamapS | chmouel: https://wiki.openstack.org/wiki/OsCollectConfig | 16:07 |
| SpamapS | pictures! | 16:07 |
| chmouel | SpamapS: woo woo awesome! | 16:07 |
| chmouel | Goneri: ^ | 16:07 |
| Goneri | outch :) | 16:07 |
| Goneri | SpamapS: thanks | 16:08 |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/os-collect-config: Add diagram to os-collect-config README.rst https://review.openstack.org/83115 | 16:13 |
| SpamapS | anybody know a quick way to render rst -> HTML the same way the markdown command does? | 16:14 |
| slagle | i wish i did | 16:14 |
| slagle | i admit to using the github live edit as a "scratch space" | 16:14 |
| SpamapS | slagle: hah nice :) | 16:14 |
| *** weshay has quit IRC | 16:14 | |
| SpamapS | maybe I should just fork os-collect-config on github and push my changes into that | 16:15 |
| *** weshay has joined #tripleo | 16:15 | |
| slagle | well, it's got a Preview button | 16:15 |
| slagle | so you can edit, preview, w/o committing | 16:15 |
| *** killer_prince has quit IRC | 16:17 | |
| SpamapS | hm, github doesn't seem to like the .. image:: syntax | 16:18 |
| *** killer_prince has joined #tripleo | 16:18 | |
| clarkb | SpamapS: rst2html? | 16:20 |
| *** lsmola_ has joined #tripleo | 16:20 | |
| dprince | SpamapS: a simple jq --arg did the trick (/me has his variable and avoids adding a new parameter) | 16:20 |
| *** e0ne_ has joined #tripleo | 16:21 | |
| SpamapS | dprince: not sure I understand but I look forward to your patch.. ? | 16:22 |
| dprince | SpamapS: cool, I munged a variable in bash, and then can pass it into jq w/ --arg | 16:23 |
| *** e0ne has quit IRC | 16:24 | |
| *** boris-42 has quit IRC | 16:24 | |
| SpamapS | hrm, looks like github won't do any fancy tricks to let you use a non-absolute link to their raw image URL in the README | 16:25 |
| SpamapS | clarkb: thanks.. so simple ;) | 16:26 |
| openstackgerrit | Ryan Moore proposed a change to openstack/tripleo-image-elements: Allow setting snapshots image format and directory https://review.openstack.org/83122 | 16:27 |
| *** bauzas has quit IRC | 16:29 | |
| *** epim has joined #tripleo | 16:34 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-incubator: Write network configuration into seeds config.json https://review.openstack.org/83125 | 16:35 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-incubator: Don't hard code the baremetal seed IP in seedrc https://review.openstack.org/83126 | 16:35 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-incubator: Make the baremetal-network configurable https://review.openstack.org/82327 | 16:35 |
| *** rlandy has quit IRC | 16:35 | |
| *** boris-42 has joined #tripleo | 16:36 | |
| openstackgerrit | Peter Belanyi proposed a change to openstack/tuskar-ui: Fixed a few typos in runtest.sh https://review.openstack.org/83127 | 16:36 |
| jogo | derekh: I had trouble using graphite for that as well | 16:37 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Drop dnsmasq_range from the seeds config.json. https://review.openstack.org/83128 | 16:38 |
| jogo | derekh: I just looked at the jenkinses https://jenkins02.openstack.org/job/check-tripleo-undercloud-precise/buildTimeTrend | 16:38 |
| derekh | jogo: ya, I'm trying to get that consolidates all the jenkins servers together, guess I'll have to dig a bit more | 16:40 |
| derekh | jogo: do you know if there is any precedence for ci jobs sending their own statsd stats? | 16:40 |
| jogo | derekh: ask in #infra | 16:40 |
| derekh | will do, thanks | 16:41 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Drop dnsmasq_range from the seeds config.json. https://review.openstack.org/83128 | 16:41 |
| bnemec | SpamapS: http://www.tele3.cz/jbar/rest/rest.html ? | 16:41 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-heat-templates: Drop dnsmasq_range from the undercloud source. https://review.openstack.org/83130 | 16:42 |
| jogo | derekh: I was never sure how to get a good time measurement on these things myself | 16:43 |
| derekh | jogo: ok | 16:45 |
| *** pbelanyi has quit IRC | 16:47 | |
| *** hashar has quit IRC | 16:48 | |
| *** mestery has quit IRC | 16:49 | |
| smulcahy | I just logged https://bugs.launchpad.net/tripleo/+bug/1297948 - happy to discuss here, I think its a critical | 16:49 |
| uvirtbot | Launchpad bug 1297948 in tripleo "tripleo devtest fails sometimes when running /tmp/in_target.d/install.d/01-install-pip" [High,Triaged] | 16:49 |
| smulcahy | it makes using pypi mirrors and proxy for tripleo devtest totally unreliable | 16:49 |
| *** rpodolyaka1 has joined #tripleo | 16:50 | |
| *** vkozhukalov has joined #tripleo | 16:51 | |
| GheRivero | I think i fix it by export PIP_INDEX_URL=$PYPI_MIRROR_URL | 16:52 |
| GheRivero | but have to test in the server, once the current tripleo run finish | 16:52 |
| *** matty_dubs|lunch is now known as matty_dubs | 16:53 | |
| *** saurabhs has joined #tripleo | 16:54 | |
| SpamapS | bnemec: thanks, but I needed something that I can put in the same web root | 16:54 |
| GheRivero | smulcahy: ok. just test it ad that fix it. | 16:54 |
| *** rpodolyaka1 has quit IRC | 16:55 | |
| bnemec | SpamapS: Ah, gotcha. I had just used that a couple of times to verify doc output on changes. | 16:56 |
| smulcahy | gherhivero: I can give that a try | 16:58 |
| smulcahy | I'd like someone to explain what that fixes and why its neccesary though | 16:59 |
| *** nati_ueno has joined #tripleo | 17:00 | |
| *** rollerj__ has quit IRC | 17:02 | |
| *** mestery has joined #tripleo | 17:09 | |
| *** jistr has quit IRC | 17:10 | |
| *** e0ne_ has quit IRC | 17:15 | |
| smulcahy | I suspect the bug I just raised https://bugs.launchpad.net/tripleo/+bug/1297970 may be related to the use of pypi mirrors and/or wheels and maybe related to https://bugs.launchpad.net/tripleo/+bug/1297948 also | 17:23 |
| uvirtbot | Launchpad bug 1297970 in tripleo ""No local packages or download links found for six>=1.4.1" error when using a pypi mirror" [Undecided,New] | 17:23 |
| *** cwolferh has joined #tripleo | 17:24 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-incubator: Remove extra TMP variable from devtest_seed.sh https://review.openstack.org/83139 | 17:41 |
| GheRivero | smulcahy: I just remember the cause of your first problem | 17:42 |
| GheRivero | the images are built with the pypi element, but somehow, it cannot find the env variable to the mirror, so use the local cache (~/.cache/image-create/pypi) for it | 17:43 |
| GheRivero | in my case was a misspelled PYPI_MIRROR instead of PYPI_MIRROR_URL | 17:44 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add dib-lint script https://review.openstack.org/81634 | 17:44 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Alphabetize all element-deps https://review.openstack.org/81635 | 17:45 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Enforce alphabetical ordering of element-deps https://review.openstack.org/81636 | 17:45 |
| *** petertoft has quit IRC | 17:46 | |
| bnemec | \o/ | 17:47 |
| *** saurabhs has quit IRC | 17:50 | |
| *** morganfainberg_Z is now known as morganfainberg | 17:50 | |
| *** rpodolyaka1 has joined #tripleo | 17:51 | |
| *** epim has quit IRC | 17:54 | |
| *** rpodolyaka1 has quit IRC | 17:56 | |
| slagle | bnemec: i had my +A finger going there for a moment... | 17:58 |
| bnemec | slagle: Indeed, thanks :-) | 17:58 |
| *** epim has joined #tripleo | 18:00 | |
| *** saurabhs has joined #tripleo | 18:00 | |
| greghaynes | lifeless: in your comment on https://bugs.launchpad.net/tripleo/+bug/1295732 you have PYPI_MIRROR_URL as the non-wheel mirror and _1 as the wheel mirror, that seems backwards to me? | 18:01 |
| uvirtbot | Launchpad bug 1295732 in tripleo "tripleo devtest overcloud startup takes too long " [High,Triaged] | 18:01 |
| *** athomas has quit IRC | 18:02 | |
| *** lucasagomes is now known as lucas-afk | 18:02 | |
| *** derekh has quit IRC | 18:04 | |
| *** jcoufal has quit IRC | 18:07 | |
| * SpamapS lunches | 18:10 | |
| SpamapS | btw for those that missed it.. I made a pretty: https://wiki.openstack.org/wiki/OsCollectConfig | 18:11 |
| greghaynes | oo such pretty | 18:12 |
| *** martyntaylor has left #tripleo | 18:15 | |
| lifeless | greghaynes: and yet, thats what I have | 18:19 |
| lifeless | greghaynes: I reported what I'm using, not what I should be using :0 | 18:19 |
| greghaynes | hah | 18:19 |
| greghaynes | I mentioned that I was surprised you have numbers to back up your settings | 18:19 |
| lifeless | SpamapS: so the reason I wanted the cidr is that its much harder to get skew in the config | 18:20 |
| lifeless | SpamapS: and we foot gunned ourselvs repeatedly without it | 18:20 |
| lifeless | dprince: ^ | 18:21 |
| dprince | lifeless: no foot gunning here. I'm just questioning a config that might should get split up because most of the places that uses it need individual peicies (not the whole) | 18:22 |
| dprince | lifeless: in any case I've got my jq issue sorted so we can skip it for now | 18:23 |
| dprince | lifeless: it was the ensure-bridge change the I really wanted to change this for anyways. But that still hasn't landed yet | 18:23 |
| lifeless | ack | 18:29 |
| lifeless | dprince: btw passing in a new network ++ | 18:29 |
| *** vkozhukalov has quit IRC | 18:29 | |
| dprince | lifeless: :) Hasn't made it guantlet yet but it worked for me locally... (perhaps I messed it up before submitting it) | 18:30 |
| *** e0ne has joined #tripleo | 18:30 | |
| *** spzala has quit IRC | 18:41 | |
| openstackgerrit | Jon-Paul Sullivan proposed a change to openstack/diskimage-builder: Write a dpkg manifest to list installed packages https://review.openstack.org/83171 | 18:46 |
| hewbrocca | BTW -- marios_ made a couple of tripleo-in-action videos yesterday | 18:49 |
| hewbrocca | https://www.youtube.com/watch?v=Ranfkx34dhg shows using the UI setting up the overcloud | 18:49 |
| lifeless | dprince: I want to land https://review.openstack.org/#/c/78467/ today - its blocking for Ironic | 18:50 |
| hewbrocca | and https://www.youtube.com/watch?v=s2DAngZ8__E | 18:50 |
| lifeless | dprince: what tests do you need done? | 18:50 |
| hewbrocca | shows using the the UI to launch it | 18:50 |
| dprince | lifeless: I just haven't switched over to actually use 'nodes' yet. | 18:51 |
| dprince | lifeless: should be easy to do, just wanted to kick the tires on it first | 18:51 |
| lifeless | dprince: can you do that today? | 18:51 |
| *** rpodolyaka1 has joined #tripleo | 18:52 | |
| lifeless | dprince: my goal is to get Ironic useful check results from toci asap so they stop breaking :) | 18:52 |
| dprince | lifeless: I'll try | 18:52 |
| lifeless | dprince: thanks | 18:53 |
| lifeless | slagle: SpamapS: https://review.openstack.org/#/c/82741/ has all the requested changes and passed CI | 18:54 |
| *** rpodolyaka1 has quit IRC | 18:56 | |
| *** rpodolyaka1 has joined #tripleo | 18:57 | |
| lifeless | GheRivero: ^ | 18:59 |
| *** epim has quit IRC | 19:01 | |
| greghaynes | I love when I leave a long review comment only to see that it was addressed in the next file in the review | 19:03 |
| lifeless | :) | 19:03 |
| greghaynes | I left a nice one about "why dont we make a userguide?" in the readme | 19:03 |
| lifeless | greghaynes: thats a great idea | 19:03 |
| *** spzala has joined #tripleo | 19:08 | |
| *** epim has joined #tripleo | 19:10 | |
| *** jp_at_hp has quit IRC | 19:20 | |
| *** andreaf has quit IRC | 19:34 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Make it possible to pass real hardware in. https://review.openstack.org/83188 | 19:34 |
| lifeless | dprince: ^ may be useful to you | 19:34 |
| dprince | lifeless: cool, I may come in handy sometime, I typically don't call the top level devtest.sh however | 19:36 |
| *** spzala has quit IRC | 19:38 | |
| *** rdopieralski has joined #tripleo | 19:39 | |
| *** rdopieralski has quit IRC | 19:43 | |
| lifeless | dprince: it also lets you pass --nodes to devtest_testenv.sh too | 19:44 |
| SpamapS | lifeless: /win 8 | 19:50 |
| jprovazn | greghaynes, Hi, are you around? | 19:50 |
| SpamapS | doh | 19:50 |
| SpamapS | lifeless: ACK | 19:50 |
| greghaynes | jprovazn: Hey | 19:50 |
| openstackgerrit | Coleman Corrigan proposed a change to openstack/tripleo-incubator: Add set-pip-vars script to simplify pip manifests https://review.openstack.org/83194 | 19:51 |
| jprovazn | greghaynes, https://review.openstack.org/#/c/79884/4/elements/mariadb/install.d/10-mariadb - you mean add rhel-based check because of percona-toolkit package is being installed or something else? | 19:51 |
| lifeless | GheRivero: Ng: hey so hp build perf | 19:52 |
| greghaynes | jprovazn: because of the mariadb-galera | 19:52 |
| lifeless | GheRivero: Ng: I note 600 seconds to build a ramdisk; thats 10x slower than for me: Can we fix that? | 19:52 |
| *** hashar has joined #tripleo | 19:52 | |
| greghaynes | jprovazn: I assume theres really no future in that line succeeding in !redhat packages, right? | 19:53 |
| Ng | lifeless: orly, where are you seeing 600s? | 19:53 |
| jprovazn | greghaynes, the mariadb-galera pkgs are available once mariadb.org repo is added - this is done in mariadb-dev pre-install script | 19:54 |
| greghaynes | ah, I was looking for something like that but forgot about the patch dependency | 19:55 |
| * greghaynes rechecks | 19:55 | |
| jprovazn | greghaynes, no worries - you remind me I should check package names for ubuntu | 19:56 |
| greghaynes | Yep, I was just doing that :) | 19:56 |
| jprovazn | :) great | 19:56 |
| greghaynes | jprovazn: looks like they are 'mariadb-galera-server galera' on ubuntu | 19:58 |
| jprovazn | yea | 19:58 |
| jprovazn | will add mapping | 19:59 |
| jprovazn | actually, IIRC yum should be case insensitive | 19:59 |
| greghaynes | haha | 20:00 |
| jprovazn | and it's not | 20:01 |
| lifeless | Ng: https://bugs.launchpad.net/tripleo/+bug/1295732 | 20:01 |
| uvirtbot | Launchpad bug 1295732 in tripleo "tripleo devtest overcloud startup takes too long " [High,Triaged] | 20:01 |
| lifeless | Ng: comment 7 | 20:02 |
| *** jang1 has joined #tripleo | 20:03 | |
| *** julim_ has quit IRC | 20:03 | |
| *** jang1 has quit IRC | 20:04 | |
| *** jang1 has joined #tripleo | 20:04 | |
| Ng | lifeless: attempting to reproduce | 20:05 |
| Ng | I wonder if this is the recent trend of overusing things like NODE_DIST, causing weird elements to leak into the ramdisk build, but I have no evidence of anything yet :) | 20:06 |
| *** e0ne has quit IRC | 20:07 | |
| *** bauzas has joined #tripleo | 20:10 | |
| lifeless | Ng: indeed thats possible | 20:10 |
| lifeless | Ng: OTOH 10 minutes sounds like either downloading a new base ubuntu image or some such cruft to me | 20:11 |
| Ng | possibly | 20:11 |
| Ng | a run on the lab hardware that has minimal elements ("ubuntu deploy") and no http proxy, but that didn't need to fetch a new base image, just finished after 4m17s | 20:12 |
| lifeless | GheRivero: when you say -u in NODE_DIST doesn't work... what happens? | 20:12 |
| lifeless | Ng: so thats 4 times slower still | 20:13 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/diskimage-builder: Adds package mapping for mariadb packages https://review.openstack.org/83199 | 20:13 |
| *** epim has quit IRC | 20:13 | |
| lifeless | Ng: is it using a local mirror ? | 20:13 |
| Ng | lifeless: no, I wanted to start from the very basics and add things one at a time to see where it gets better or worse | 20:14 |
| lifeless | ++ | 20:14 |
| Ng | lifeless: even going through the http proxy, fetching the cloud image is never getting even dangerously close to 500KB/s | 20:15 |
| Ng | I'd guess the average is about 465 | 20:15 |
| lifeless | Ng: well you know whats serving that :P | 20:16 |
| lifeless | Ng: (so you're in a better place than I to comment on whether thats expected) | 20:16 |
| Ng | lifeless: that is not the ubuntu.com servers :) | 20:17 |
| Ng | I'm in the same city as them and I can max out my line | 20:17 |
| Ng | average of 11.6MB | 20:17 |
| Ng | 18 seconds to grab the cloud image :) | 20:17 |
| *** yassine_ has quit IRC | 20:18 | |
| lifeless | Ng: random thought, mayeb the office egress to internet is via NY or something | 20:19 |
| lifeless | Ng: anyhow, why is it downloading the base image again? versimilitude? | 20:19 |
| Ng | lifeless: it wasn't doing so before, but you made the good point that it might be that, so I forced the issue :) | 20:21 |
| Ng | but the cloud image was updated about 7 hours ago, going by the mtime | 20:21 |
| *** rpodolyaka1 has quit IRC | 20:21 | |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mysql-dev and mariadb-dev elements https://review.openstack.org/79886 | 20:22 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mariadb element https://review.openstack.org/79884 | 20:22 |
| SpamapS | For me, cloud-images.ubuntu.com is always a _lot_ slower than archive.ubuntu.com | 20:22 |
| SpamapS | Perhaps we should mirror it into hp cloud's swift | 20:22 |
| Ng | lifeless: anyway, if this is a consistent internet speed that I'm seeing, we can rule this out, because the total time for downloading a cloud image to these lab machines appears to be 16 minutes | 20:22 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mysql-dev and mariadb-dev elements https://review.openstack.org/79886 | 20:22 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mariadb element https://review.openstack.org/79884 | 20:22 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Adds mysql-common element https://review.openstack.org/79885 | 20:22 |
| lifeless | dprince: so https://review.openstack.org/#/c/78467/ has now passed full CI; is there any chance you can migrate to --nodes now ? I hate to pressure you, but getting Ironic good testing - well, its been a long road. | 20:23 |
| SpamapS | Ng: just for giggles, try ssh -D to a box you know is fast, and then bounce through it using tsocks | 20:23 |
| lifeless | Ng: right, so 10m is < 16m. | 20:23 |
| Ng | lifeless: I'm not going to totally rule it out, because maybe the lab internet is faster at other times, but right now it's not an obvious contender | 20:24 |
| lifeless | Ng: shockingly scientific thinking there | 20:24 |
| dprince | lifeless: go for it, if it breaks me I'll fix it | 20:25 |
| lifeless | dprince: gentleman and a scholar@ | 20:25 |
| lifeless | bah, s/@/!/ | 20:25 |
| *** jprovazn is now known as jprovazn_afk | 20:25 | |
| Ng | lifeless: we shall be victorious through the judicious application of science :) | 20:25 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Remove backwards compatability in setup-baremetal https://review.openstack.org/78467 | 20:27 |
| *** mrunge has joined #tripleo | 20:27 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Update documentation. https://review.openstack.org/82741 | 20:30 |
| lifeless | EWAT | 20:31 |
| lifeless | 2014-03-26 05:29:04.980 | File "/opt/stack/venvs/keystone/local/lib/python2.7/site-packages/netaddr/ip/__init__.py", line 13, in <module> | 20:31 |
| lifeless | 2014-03-26 05:29:04.981 | from netaddr.strategy import ipv4 as _ipv4, ipv6 as _ipv6 | 20:31 |
| lifeless | 2014-03-26 05:29:04.981 | ValueError: bad marshal data (unknown type code) | 20:31 |
| lifeless | http://logs.openstack.org/51/79051/7/check-tripleo/check-tripleo-overcloud-precise/49d7695/console.html | 20:31 |
| lifeless | even weirder | 20:33 |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper File "/opt/stack/venvs/nova/local/lib/python2.7/site-packages/nova/cmd/baremetal_deploy_helper.py", line 167, in notify | 20:33 |
| *** testing__ has joined #tripleo | 20:33 | |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper s.connect((address, port)) | 20:33 |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper File "/usr/lib/python2.7/dist-packages/eventlet/greenio.py", line 194, in connect | 20:33 |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper socket_checkerr(fd) | 20:33 |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper File "/usr/lib/python2.7/dist-packages/eventlet/greenio.py", line 46, in socket_checkerr | 20:33 |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper raise socket.error(err, errno.errorcode[err]) | 20:33 |
| lifeless | 2014-03-26 15:27:25.292 4470 TRACE nova.virt.baremetal.deploy_helper error: [Errno 111] ECONNREFUSED | 20:34 |
| lifeless | we're starting to see lovely corner cases here :) | 20:34 |
| *** rpodolyaka1 has joined #tripleo | 20:38 | |
| *** julim has joined #tripleo | 20:45 | |
| Ng | interesting, curl was predicting badly, the actual download time was 12 minutes, which is by far the worst case I've been able to induce so far. Interestingly, the best case is 1m21s | 20:46 |
| Ng | that's with local ubuntu and pypi mirrors | 20:46 |
| lifeless | Ng: 1m21 to build it? | 20:46 |
| Ng | local to the lab network, I mean | 20:46 |
| lifeless | so 80s vs 60, thats tolerable | 20:47 |
| Ng | lifeless: just for the raw diskimage-builder command to build a deploy ramdisk | 20:47 |
| lifeless | Ng: could you use 'time testenv_ramdisk.sh' ? | 20:47 |
| lifeless | Ng: that should get the same thing the stuff in the bug measures | 20:47 |
| *** testing__ has quit IRC | 20:47 | |
| *** cwolferh has quit IRC | 20:48 | |
| *** cwolferh has joined #tripleo | 20:48 | |
| Ng | lifeless: I will do so, but there are still things like NODE_DIST/DIB_COMMON_ELEMENTS that could be affecting it | 20:48 |
| *** rpodolyaka1 has quit IRC | 20:50 | |
| lifeless | Ng: indeed | 20:51 |
| lifeless | q | 20:51 |
| lifeless | woo I68d46c4da8715df03c3a88393b55665dc57045a3 has landed | 20:51 |
| lifeless | now the tie patch and incubator and we have ironic ... wheeeee | 20:51 |
| Ng | lifeless: 1m41s | 20:52 |
| SpamapS | sweeeeeeeeeeet | 20:52 |
| lifeless | SpamapS: assuming it actualy, yknow, works | 20:53 |
| SpamapS | I ALWAYS assume that | 20:53 |
| SpamapS | American Optimism™ | 20:53 |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/tripleo-image-elements: Update the "arch" value in config.json to match $ARCH https://review.openstack.org/83211 | 20:55 |
| *** weshay has quit IRC | 20:56 | |
| Ng | lifeless: I have posted my numbers and a request for some settings info, to the bug | 20:56 |
| greghaynes | somehow I made a heat template that heat stack-create accepted, but causes the cfn api to return http://paste.ubuntu.com/7159007/ when called from os-collect-config | 20:56 |
| greghaynes | as a side note, only way I found that error message is via tcpdump which seemed less than ideal | 20:56 |
| *** e0ne has joined #tripleo | 20:57 | |
| greghaynes | I assume that means somewhere in heat it barfed on that resource? | 20:58 |
| *** boris-42 has quit IRC | 20:59 | |
| openstackgerrit | Thom Leggett proposed a change to openstack/diskimage-builder: Uses policy-rc.d to prevent dpkg starting daemons https://review.openstack.org/82074 | 20:59 |
| GheRivero | and back | 21:04 |
| GheRivero | lifeless: the problem is that NODE_DIST is used to call boot-seed-vm | 21:05 |
| GheRivero | boot-seed-vm -a $NODE_ARCH $NODE_DIST | 21:05 |
| *** epim has joined #tripleo | 21:05 | |
| GheRivero | but boot-seed-vm doesn't recognize the -u param | 21:05 |
| *** akrivoka has quit IRC | 21:06 | |
| greghaynes | ah, found the actual error... bad type to a fn::join. This does seem like a pretty crappy way to fail for that though | 21:06 |
| *** e0ne has quit IRC | 21:07 | |
| GheRivero | but it we move the user image to use cirros, there is no need for using NODE_DIST and the -u can be moved to DIB_COMMON_ELEMENTS | 21:08 |
| *** jp_at_hp has joined #tripleo | 21:10 | |
| *** boris-42 has joined #tripleo | 21:15 | |
| *** jp_at_hp has quit IRC | 21:16 | |
| *** jprovazn_afk has quit IRC | 21:16 | |
| *** marun has left #tripleo | 21:17 | |
| *** lifeless has quit IRC | 21:19 | |
| *** dprince has quit IRC | 21:20 | |
| *** e0ne has joined #tripleo | 21:22 | |
| *** CaptTofu has quit IRC | 21:24 | |
| *** CaptTofu has joined #tripleo | 21:25 | |
| *** CaptTofu has quit IRC | 21:26 | |
| *** CaptTofu_ has joined #tripleo | 21:26 | |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/diskimage-builder: Fix "(None)" seed hostname on Debian https://review.openstack.org/77301 | 21:27 |
| geerdest | noticing nova console-log not working. is there any open bug for that? not seeing one. | 21:31 |
| *** markmc has quit IRC | 21:32 | |
| SpamapS | geerdest: It works on our CI overcloud. | 21:35 |
| SpamapS | geerdest: you getting blank consoles? | 21:35 |
| *** markmc has joined #tripleo | 21:35 | |
| geerdest | just built all of this today | 21:35 |
| geerdest | ERROR (HTTPNotImplemented): Unable to get console log, functionality not implemented (HTTP 501) (Request-ID: req-5ff06cb5-8a55-46d8-99a6-137c30381651) | 21:35 |
| geerdest | thats on undercloud though | 21:38 |
| geerdest | looks like my latest build still doesn’t have a working overcloud… so much for today | 21:41 |
| SpamapS | oh | 21:42 |
| SpamapS | undercloud would not work no | 21:42 |
| SpamapS | undercloud is baremetal | 21:42 |
| SpamapS | or ironic | 21:42 |
| SpamapS | and thus needs a console driver for that | 21:42 |
| *** stevehuang has joined #tripleo | 21:43 | |
| geerdest | SpamapS: ah ok.. found an issue and yes.. it does work on my overcloud. thanks | 21:43 |
| SpamapS | There is some intention to have console-log work on baremetal servers.. it's just.. complicated :) | 21:44 |
| *** hashar has quit IRC | 21:45 | |
| *** matty_dubs is now known as matty_dubs|gone | 21:46 | |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-incubator: Add clarification on required permissions https://review.openstack.org/83102 | 21:46 |
| *** rpodolyaka1 has joined #tripleo | 21:50 | |
| *** rpodolyaka1 has quit IRC | 21:55 | |
| *** untriaged-bot has joined #tripleo | 22:00 | |
| untriaged-bot | Untriaged bugs so far: | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1296847 | 22:00 |
| uvirtbot | Launchpad bug 1296847 in tripleo "Patch needed to build undercloud on baremetal" [Undecided,New] | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1290488 | 22:00 |
| uvirtbot | Launchpad bug 1290488 in tripleo "Baremetal: Invalid credentials" [Undecided,Incomplete] | 22:00 |
| *** untriaged-bot has quit IRC | 22:00 | |
| bnemec | ^Nice! | 22:00 |
| Ng | cool | 22:03 |
| *** blamar has quit IRC | 22:03 | |
| *** lynxman has quit IRC | 22:04 | |
| greghaynes | \O/ | 22:04 |
| *** lynxman has joined #tripleo | 22:04 | |
| tchaypo | morning *grunts* | 22:07 |
| *** jang1 has quit IRC | 22:08 | |
| *** lynxman has quit IRC | 22:11 | |
| *** lynxman has joined #tripleo | 22:12 | |
| *** robertc has joined #tripleo | 22:13 | |
| robertc | my home net is offline for a few hours while we get the house spider-cleaned | 22:13 |
| *** robertc is now known as lifeless | 22:13 | |
| greghaynes | Meanwhile, in NZ... | 22:14 |
| devananda | lifeless: so https://bugs.launchpad.net/ironic/+bug/1295503 landed | 22:16 |
| uvirtbot | Launchpad bug 1295503 in ironic "nova-compute errors on startup when ironic user isn't registered in keystone" [High,Fix committed] | 22:16 |
| devananda | *the fix for | 22:16 |
| lifeless | devananda: yup | 22:16 |
| lifeless | just waiting on https://review.openstack.org/81627 to complete a check run | 22:17 |
| * devananda goes t ofind the other bug | 22:17 | |
| devananda | ah, thx | 22:17 |
| * Ng shivers. I do not even why we inhabit any part of the world that needs "spider-cleaning" to be a thing ;o | 22:17 | |
| bnemec | Wrong country, but I can't help thinking of http://www.sluniverse.com/php/vb/attachments/zomgwtfbbqgtfololcats/11529d1272801642-why-australia-sucks-health_bar_spider.jpg | 22:17 |
| lifeless | is the last remaining patch https://jenkins07.openstack.org/job/check-tripleo-overcloud-precise/298/ | 22:17 |
| lifeless | s/patch/job/ | 22:17 |
| lifeless | which it should, its running the qemu vm up atm | 22:18 |
| devananda | awesome | 22:18 |
| lifeless | slagle: SpamapS: for great irony: https://review.openstack.org/#/c/81627/ | 22:19 |
| lifeless | devananda: then we can recheck the incubator patch that enables ironic | 22:20 |
| lifeless | slagle: SpamapS: thats passed CI and is one of only two patches needed to get Ironic support for TripleO (and tripleo CI for Ironic) | 22:23 |
| *** lblanchard has quit IRC | 22:27 | |
| *** bauzas has quit IRC | 22:27 | |
| *** blamar has joined #tripleo | 22:42 | |
| *** cwolferh has quit IRC | 22:44 | |
| *** cwolferh has joined #tripleo | 22:49 | |
| *** e0ne has quit IRC | 22:49 | |
| *** rpodolyaka1 has joined #tripleo | 22:51 | |
| *** blamar has quit IRC | 22:53 | |
| *** lifeless has quit IRC | 22:54 | |
| *** rpodolyaka1 has quit IRC | 22:55 | |
| Lotus907efi | if I want to build overcloud images with precise do I need a pypi mirror built on a precise system? | 22:57 |
| Lotus907efi | i.e. if I have a pypi mirror that I build usinf run-mirror on a saucy system will that not work for a precise system due to wheel files? | 22:58 |
| Lotus907efi | or will the image get built using the tar.gz files to build stuff for the new precise image? | 22:58 |
| SpamapS | Lotus907efi: building _with_ precise, no. Building overcloud images that _are_ precise.. well.. IMO, don't do that. ;) | 22:58 |
| Lotus907efi | ok, but why? | 22:58 |
| SpamapS | Lotus907efi: libvirt.. ovs.. kvm... kernel... | 22:59 |
| SpamapS | the list goes on and on and on | 22:59 |
| Lotus907efi | hmm, I used vmbuilder earlier today and built a qcow2 image from a precise apt repo | 22:59 |
| SpamapS | Lotus907efi: it is entirely possible, but a bit futile given that we're designing a system that _embraces_ rapid constant change.. rather than "ship it and forget it" :) | 22:59 |
| Lotus907efi | yeah, but all of our current infrastructure is built around precise | 23:00 |
| StevenK | tchaypo: That's what namecheap.com is for. | 23:00 |
| Lotus907efi | and I have been "directed" to see if it will work | 23:00 |
| SpamapS | It will work. | 23:01 |
| SpamapS | Just.. | 23:01 |
| SpamapS | You'll spend quite a large percentage of your time backporting things. | 23:01 |
| SpamapS | The cloud archive should help. | 23:01 |
| StevenK | Except for libvirt, which infra keep tripping over? | 23:01 |
| SpamapS | But at that point, you're 99% already at saucy. :-P | 23:01 |
| SpamapS | so like.. why bother? :-P | 23:01 |
| Lotus907efi | so I deleted my current overcloud, exported DIB_RELEASE=precise and pointed DIB_APT_SOURCES at a sources.list file that points at our internal precise apt repo | 23:02 |
| Lotus907efi | and then ran devtest_overcloud.sh | 23:02 |
| Lotus907efi | just to see what would happen | 23:02 |
| *** mrunge has quit IRC | 23:02 | |
| SpamapS | Lotus907efi: did you feed that cloud after midnight.. because... didn't we tell you not to do that? ;) | 23:03 |
| Lotus907efi | what/where is cloud archive ? | 23:03 |
| Lotus907efi | hehehehe | 23:03 |
| Lotus907efi | very funny, I like that | 23:03 |
| SpamapS | Lotus907efi: http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0CGAQFjAB&url=http%3A%2F%2Fwww.ubuntu.com%2Fdownload%2Fcloud%2Fcloud-archive-instructions&ei=4VwzU6KNIJH9oATVhoGQDQ&usg=AFQjCNHHU1SzdtLcSXQDLExe2KmcT6JsYQ&sig2=etsNwAlk6VBe0n-tJpgZtw&bvm=bv.63808443,d.cGU | 23:04 |
| SpamapS | gah | 23:04 |
| SpamapS | damn you google | 23:04 |
| SpamapS | http://www.ubuntu.com/download/cloud/cloud-archive-instructions | 23:04 |
| Lotus907efi | tinyurl is your friend | 23:04 |
| SpamapS | Lotus907efi: that has all the stuff backported for havana on precise. | 23:04 |
| *** geerdest has quit IRC | 23:04 | |
| Lotus907efi | ah, ok | 23:04 |
| SpamapS | Lotus907efi: that wasn't a tinyurl problem, that was google all up in our bizness problem.. | 23:04 |
| Lotus907efi | yeah, google rules all, nine rings and all the rest | 23:05 |
| *** julim has quit IRC | 23:05 | |
| *** markmc has quit IRC | 23:06 | |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-incubator: Add instructions for setting up sudo https://review.openstack.org/83238 | 23:06 |
| *** lifeless has joined #tripleo | 23:06 | |
| SpamapS | Lotus907efi: so what happened? | 23:07 |
| *** noslzzp has quit IRC | 23:09 | |
| *** juice has quit IRC | 23:09 | |
| *** hewbrocca has quit IRC | 23:10 | |
| *** lifeless has quit IRC | 23:12 | |
| *** lifeless has joined #tripleo | 23:12 | |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-heat-templates: WIP: Switch overcloud to software-config https://review.openstack.org/81666 | 23:15 |
| Lotus907efi | well it quit after pulling quite a few packages | 23:18 |
| lifeless | dammit need one more core | 23:19 |
| Lotus907efi | and it quit after spitting out an error about a couple of packages could not be authenticated | 23:19 |
| lifeless | Ng: don't support you are still around? | 23:19 |
| lifeless | suppose | 23:19 |
| Lotus907efi | is Ng your covert core supplier? :) | 23:20 |
| lifeless | ng / slagle / SpamapS / GheRivero atm yeah | 23:20 |
| Lotus907efi | Psst! hey kid. wanna buy a core? | 23:20 |
| SpamapS | real cheap | 23:22 |
| SpamapS | high quality | 23:22 |
| SpamapS | priced to MOVE | 23:22 |
| *** nati_ueno has quit IRC | 23:23 | |
| *** nati_ueno has joined #tripleo | 23:23 | |
| Lotus907efi | this core was only driven on sundays by a little old lady from Pasadena | 23:24 |
| *** jtomasek has quit IRC | 23:25 | |
| *** juice has joined #tripleo | 23:26 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: debian: install cloud-init on Wheezy https://review.openstack.org/82325 | 23:28 |
| tchaypo | names | 23:34 |
| tchaypo | gah | 23:34 |
| tchaypo | i hate today. | 23:34 |
| lifeless | because its Monday? | 23:34 |
| tchaypo | no, thursday has always been the day i hate | 23:35 |
| tchaypo | anything that goes wrong on monday is caused by thursday making initial plans; tuesday is a small-scale test run for thursday; wednesday and friday get affected by ambient leakage from thursday, and the weekend is when thursday just breaks things for fun | 23:36 |
| tchaypo | in other news | 23:37 |
| *** yamahata has quit IRC | 23:37 | |
| tchaypo | https://review.openstack.org/#/c/82807/ | 23:37 |
| lifeless | tchaypo: there speaks a sysadmin | 23:38 |
| tchaypo | jps seems to be assuming the devtest_* scripts will be run from $TRIPLEO_ROOT/tripleo-incubator/scripts - that's where his script looks for tripleo-overcloud-passwords | 23:38 |
| tchaypo | JB has his own script which assumes they'll be in $TRIPLEO_ROOT/ | 23:39 |
| tchaypo | in my case they're in ~/src/tripleo-incubator | 23:39 |
| lifeless | tchaypo: what do you think of my suggestion? | 23:39 |
| *** saurabhs has quit IRC | 23:39 | |
| tchaypo | is there any standard location where we suggest people run the scripts from? | 23:39 |
| lifeless | that this is a nascent 'manage my cloud credentials' thing | 23:39 |
| lifeless | tchaypo: $TRIPLEO_ROOT for incubator stuff | 23:39 |
| lifeless | other things we assume path / installed | 23:39 |
| lifeless | except where we don't | 23:39 |
| lifeless | but we should | 23:40 |
| tchaypo | taking a look at http://docs.openstack.org/developer/tripleo-incubator/devtest.html for instance, it just says to do a git clone, then cd down to tripleo-incubator and run things from there | 23:40 |
| lifeless | yes | 23:41 |
| lifeless | the incubator isn't installable | 23:41 |
| tchaypo | which is why my stuff has ended up in ~/src/tripleo-incubator rather than $TRIPLEO_ROOT | 23:41 |
| lifeless | deliberately - thats a forcing function to divide 'experiment' and 'thing we know we want' | 23:41 |
| tchaypo | but I notice that all the environment rc files do get written to $TRIPLEO_ROOT/tripleo-incubator/seedrc etc - maybe we should be dumping the password files there as well? | 23:42 |
| lifeless | we should be putting the passwords into a heat json environment file which heat can use directly and we can query via jq | 23:44 |
| lifeless | probably in $TRIPLEO_ROOT | 23:44 |
| lifeless | IMO | 23:44 |
| greghaynes | and the private ssh key | 23:45 |
| greghaynes | for nova bm | 23:45 |
| greghaynes | I love when that thing scrolls by on me screen passed as a shell arg | 23:46 |
| tchaypo | I love it when I paste some logs to someone and it includes a private ssh key and i have a little panic | 23:48 |
| lifeless | greghaynes: the private key is already in JSON :) | 23:48 |
| lifeless | at least with Ironic (when someone gets time to change the patches) we can post that directly to Ironic and Heat won't know it at all | 23:49 |
| tchaypo | I love it when I ask my registrar why their UI tells me that I must urgently renew my domain, but that I'm not allowed to renew my domain, and their only response is to ask me if I've got access to the email address I just responded to their last query from | 23:49 |
| lifeless | 'yes I do' | 23:49 |
| StevenK | "Nope, I'm an autoresponder." | 23:49 |
| *** epim has quit IRC | 23:51 | |
| openstackgerrit | Steve Baker proposed a change to openstack/os-collect-config: When extracting deployments, only delete config https://review.openstack.org/83248 | 23:52 |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Use ip route replace in devtest_seed https://review.openstack.org/83249 | 23:54 |
| *** epim has joined #tripleo | 23:55 | |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Standardise location of environment password files. https://review.openstack.org/83250 | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
