| *** cody-somerville has quit IRC | 00:01 | |
| *** ramishra has joined #tripleo | 00:10 | |
| *** sdake_ has quit IRC | 00:11 | |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Move nova.conf script to nova-api directory https://review.openstack.org/128794 | 00:28 |
|---|---|---|
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Fix log settings for nova.conf https://review.openstack.org/128795 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Make nova.conf copies in each nova-controller container https://review.openstack.org/128796 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Remove unnecessary nova-controller which has no associated binary https://review.openstack.org/128797 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Have Dockerfile add nova.conf and have start.sh run nova.conf https://review.openstack.org/128798 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Make nova-controller/*/start.sh and nova-controller/*/nova.conf +x https://review.openstack.org/128799 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Import from fedora-rdo-base rather then fedora-rdo-nova-controller https://review.openstack.org/128800 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Use name nova-config.sh instead of nova.conf https://review.openstack.org/128801 | 00:28 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Nova has an undeclared dependency on python-cinderclient https://review.openstack.org/128802 | 00:29 |
| *** rainya has quit IRC | 00:29 | |
| *** otoolee has quit IRC | 00:31 | |
| *** otoolee has joined #tripleo | 00:42 | |
| *** cody-somerville has joined #tripleo | 00:43 | |
| *** cody-somerville has quit IRC | 00:44 | |
| *** ramishra has quit IRC | 00:51 | |
| *** ramishra has joined #tripleo | 00:52 | |
| *** ramishra has quit IRC | 01:02 | |
| *** cwolferh_ has quit IRC | 01:07 | |
| *** ramishra has joined #tripleo | 01:32 | |
| *** ramishra has quit IRC | 01:42 | |
| *** ramishra has joined #tripleo | 02:02 | |
| *** pradeep has joined #tripleo | 02:07 | |
| *** nosnos has joined #tripleo | 02:20 | |
| *** pradeep1 has joined #tripleo | 02:23 | |
| *** pradeep has quit IRC | 02:23 | |
| SpamapS | hm | 02:50 |
| SpamapS | no_proxy=127.0.0.1 break=after-error PYPI_MIRROR_URL=http://127.0.0.1/mirrors/pypi/ SEED_DIB_EXTRA_ARGS="pypi" DIB_NO_PYPI_PIP=1 scripts/devtest_seed.sh | 02:51 |
| SpamapS | I'd expect this to use my local mirror | 02:51 |
| SpamapS | which does have pip... but I get | 02:51 |
| SpamapS | + python /tmp/get-pip.py | 02:51 |
| SpamapS | Downloading/unpacking pip http://127.0.0.1/mirrors/pypi/pip/ uses an insecure transport scheme (http). Consider using https if 127.0.0.1 has it available Could not find any downloads that satisfy the requirement pip | 02:51 |
| SpamapS | Cleaning up... | 02:51 |
| StevenK | export PYPI_MIRROR_URL=http://silver/pypi/simple | 02:53 |
| StevenK | export DIB_COMMON_ELEMENTS="pypi common-venv stackuser" | 02:53 |
| StevenK | Is what I use | 02:53 |
| StevenK | SpamapS: Ah ha, are you missing a simple ? | 02:53 |
| SpamapS | doohh | 02:54 |
| *** cwolferh_ has joined #tripleo | 02:54 | |
| StevenK | SpamapS: That tripped me up too :-) | 02:55 |
| *** ramishra_ has joined #tripleo | 02:58 | |
| *** jerryz has joined #tripleo | 02:59 | |
| *** untriaged-bot has joined #tripleo | 03:00 | |
| untriaged-bot | Untriaged bugs so far: | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1381408 | 03:00 |
| *** untriaged-bot has quit IRC | 03:00 | |
| uvirtbot | Launchpad bug 1381408 in diskimage-builder ""visudo" always got error when building overcloud image" [Undecided,New] | 03:00 |
| *** ramishra has quit IRC | 03:00 | |
| *** spzala has quit IRC | 03:07 | |
| *** rhallisey has quit IRC | 03:09 | |
| *** ramishra_ has quit IRC | 03:10 | |
| *** noslzzp has quit IRC | 03:11 | |
| SpamapS | StevenK: ok, I had the wrong thing in my mirror's Alias. :-P | 03:14 |
| SpamapS | StevenK: and I was missing simple. :-P | 03:14 |
| *** pradeep has joined #tripleo | 03:21 | |
| *** pradeep1 has quit IRC | 03:22 | |
| *** pradeep1 has joined #tripleo | 03:24 | |
| *** pradeep has quit IRC | 03:24 | |
| *** nosnos has quit IRC | 03:26 | |
| *** nosnos has joined #tripleo | 03:27 | |
| *** pradeep1 has quit IRC | 03:29 | |
| *** nosnos has quit IRC | 03:31 | |
| *** ramishra has joined #tripleo | 03:32 | |
| tchaypo | So hp's abuse team have just /away | 03:38 |
| tchaypo | well done jaames | 03:38 |
| tchaypo | licenced to operate a car, cannot operate irc | 03:38 |
| tchaypo | so as i was saying, hp's abuse team have notified us that the undercloud control node is resolving DNS for randoms on the internet, and this makes them sad | 03:39 |
| * greghaynes wonders if rh is also doing this | 03:41 | |
| *** pradeep has joined #tripleo | 03:44 | |
| StevenK | tchaypo: Speaking of, I need VPN to talk to the bastions, don't I? | 03:45 |
| greghaynes | nope | 03:45 |
| tchaypo | StevenK: nope | 03:47 |
| *** cwolferh_ has quit IRC | 03:47 | |
| StevenK | Hmm | 03:47 |
| tchaypo | greghaynes: II'm *guessing* that RH has iptables rules that block the inbound requests | 03:47 |
| greghaynes | yea, hopefully | 03:47 |
| StevenK | Have you tried? | 03:47 |
| StevenK | So, uh, how do I log into these things? | 03:48 |
| tchaypo | which is what I've dropped in place for now, as it's a lot faster than tweaking dnsmasq to listen on all the other interfaces | 03:48 |
| *** yamahata has quit IRC | 03:48 | |
| greghaynes | ssh stack@bastion.hp2.tripleo.org | 03:48 |
| tchaypo | StevenK: but only after someone deploys your keys on the box | 03:48 |
| tchaypo | I've got a to-do to fix the deployment script on hp2 | 03:49 |
| tchaypo | so it's not likely to have dropped your key | 03:49 |
| StevenK | Right, so it is stack. I get permission denied, so I'll just wait :-) | 03:49 |
| tchaypo | no, you'll need to point me at your key so i can add it by hand, since I haven't fixed the script yet | 03:49 |
| * tchaypo fetches from gerrit | 03:50 | |
| StevenK | tchaypo: https://launchpad.net/~stevenk/+sshkeys | 03:50 |
| SpamapS | ssh-import-id stevenk | 03:50 |
| SpamapS | Oh I forgot to follow-up and do the steps to get your keys on | 03:51 |
| StevenK | Here I was being naive and thinking since my branch has landed, everything would magically work :-P | 03:51 |
| tchaypo | try now | 03:52 |
| greghaynes | StevenK: You think were the types to automate that sort of thing? | 03:52 |
| StevenK | Haha | 03:52 |
| tchaypo | if someone else can give you access to the gSheet that'd be handy, I have to run to the supermarket | 03:52 |
| StevenK | tchaypo: Works for hp2, not hp1 | 03:52 |
| * tchaypo is now nursing a wisdom teeth extractee | 03:52 | |
| tchaypo | i *think* hp1 is fixed and it should be applied there soon. If it's not I can handle it when i come back. | 03:53 |
| tchaypo | if you're looking for something to do | 03:53 |
| tchaypo | the next thing I was planning to do was run your new register-nodes over /home/shared/nodes.json into the undercloud, then deploy-ci-overcloud again | 03:54 |
| tchaypo | give me time to run to the shops and back and i can walk through that with you, if you don't mind | 03:55 |
| *** rbrady has quit IRC | 04:08 | |
| StevenK | greghaynes: You said in https://review.openstack.org/126456 LGTM, but then you -1'd? | 04:13 |
| greghaynes | StevenK: yes, id like to deprecate the original behavior | 04:14 |
| greghaynes | so print a warning or somesuch | 04:14 |
| greghaynes | because then we can remove it | 04:15 |
| StevenK | greghaynes: Sure. In a different branch, since that change is about PKI setup only | 04:15 |
| StevenK | devtest_overcloud still calls -s on init-keystone | 04:15 |
| greghaynes | actually, I should comment on the commit msg too. Currently we init the pki using another method so there isnt an incoming change that effects this | 04:16 |
| greghaynes | it currently just checks if the certs already exist so its essentially a no-op | 04:16 |
| jerryz | tchaypo: if i have a bunch of real baremetals, can i still make use of devtest script to deploy them? and how? change testenv.json? | 04:17 |
| StevenK | greghaynes: --public for init-keystone sets http:// | 04:17 |
| StevenK | So we need a Better Way | 04:18 |
| greghaynes | StevenK: Yes, im referring to the --pki-setup bits | 04:19 |
| StevenK | greghaynes: Oh. I'm only touching --ssl/--public in that patch because I add another mutually exclusive group | 04:20 |
| StevenK | greghaynes: And the PKI setup init-keystone does just calls keystone-manage pki_setup, is that clever enough to check for existing certs? | 04:21 |
| greghaynes | StevenK: http://logs.openstack.org/56/126456/1/check-tripleo/check-tripleo-ironic-undercloud-precise-nonha/5d9fa00/console.html.gz#_2014-10-07_03_02_46_376 | 04:21 |
| greghaynes | StevenK: Yep | 04:21 |
| *** nosnos has joined #tripleo | 04:23 | |
| StevenK | jerryz: Use --nodes for devtest.sh | 04:23 |
| SpamapS | I'm just doing that right now actually. :) | 04:24 |
| SpamapS | and just discovered the mac's I have are all wrong | 04:24 |
| * SpamapS ponders trying to use ilo pm driver | 04:24 | |
| *** ramishra has quit IRC | 04:24 | |
| greghaynes | StevenK: basically, the --pki-setup bits id like to remove because they really dont make sense there, but we have to wait a cycle after deprecation so deprecating ASAP seems ideal :) | 04:25 |
| jerryz | StevenK: thanks | 04:29 |
| StevenK | greghaynes: Huh. So we don't need the command line options | 04:30 |
| StevenK | greghaynes: Or we do, if we flip the default to False | 04:31 |
| greghaynes | Yea, we could just remove it alltogether if we didnt care about backwards compat | 04:31 |
| StevenK | Except we decided we do care about it | 04:32 |
| greghaynes | yep, so deprecate! | 04:32 |
| *** rushiagr_away is now known as rushiagr | 04:37 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/os-cloud-config: Teach init-keystone about PKI setup https://review.openstack.org/126456 | 04:38 |
| StevenK | greghaynes: Like ^? | 04:38 |
| StevenK | greghaynes: And then we can set --no-pki-setup for devtest once that lands and is released | 04:38 |
| greghaynes | yaya, deprecate all the things! | 04:38 |
| *** yamahata has joined #tripleo | 04:40 | |
| *** dshulyak_ has joined #tripleo | 04:48 | |
| * StevenK fixes the commit message. | 04:49 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/os-cloud-config: Deprecate PKI setup in init-keystone https://review.openstack.org/126456 | 04:56 |
| tchaypo | jerryz: yeah, as other people said, nodes.json | 05:03 |
| tchaypo | spamaps: where are your thingies? | 05:04 |
| tchaypo | I'm doing the same thing in hp2, except that I've just fixed 70% of my mac addresses | 05:04 |
| tchaypo | jerryz: if you can't figure out the format of nodes.json i have a sample i can give you | 05:04 |
| tchaypo | jerryz: but I'm pretty sure the docs have a sample | 05:04 |
| tchaypo | StevenK: greghaynes: lifeless: it seems t me that the issue with dnsmasq being available on the public interface is a special case of https://bugs.launchpad.net/tripleo/+bug/1188067 | 05:07 |
| uvirtbot | Launchpad bug 1188067 in tripleo "* listening services available on all addresses" [High,Triaged] | 05:07 |
| tchaypo | so I'm thinking of glomming onto that bug and raising the priority | 05:07 |
| jerryz | tchaypo: similar to testenv.json? | 05:07 |
| jerryz | tchaypo: please also send me the sample just for reference. | 05:08 |
| tchaypo | making a note of the work we've already done to allow the tripleo services to be available on a subset of interfaces, and then tackling dnsmasq. I'm not sure if we can handle dnsmasq the same way though - I don't think we can put it behind haproxy.. | 05:08 |
| *** otoolee has quit IRC | 05:09 | |
| tchaypo | jerryz: I'd really really like to just point you at the bit in the docs where there's a handy sample | 05:12 |
| tchaypo | jerryz: and indeed, devtest_testenv.sh says that "The schema for nodes can be found in the devtest_setup documentation." | 05:12 |
| tchaypo | jerryz: as far as I can tell that's a lie. | 05:12 |
| StevenK | It isn't | 05:12 |
| StevenK | A list of node metadata. Each node has "memory" in K, "cpu" in | 05:13 |
| StevenK | ## threads, "arch" (one of i386/amd64/etc), "disk" in GB, mac, a list of | 05:13 |
| StevenK | ## MAC addresses for the node and "pm_type", "pm_user", "pm_addr", and | 05:13 |
| StevenK | ## "pm_password" fields. | 05:13 |
| tchaypo | that's not a schema. | 05:13 |
| tchaypo | nor is it a useful sample. | 05:13 |
| tchaypo | it's a set of instructions for creating a sample | 05:13 |
| StevenK | http://docs.openstack.org/developer/os-cloud-config/usage.html#registering-nodes-with-a-baremetal-service | 05:14 |
| tchaypo | jerryz: there you go, stevenk found a sample :) | 05:16 |
| StevenK | Sure, because I *wrote* it :-) | 05:16 |
| jerryz | StevenK, tchaypo: the format is indeed the same as testenv.json, but a question: if it is a bare metal, what does a disk size and number of cpu mean? | 05:19 |
| jerryz | the disk size going to be formatted for OS partition? or just a number when registering the node | 05:20 |
| *** ramishra has joined #tripleo | 05:21 | |
| tchaypo | the former - if you set it to something bigger than the actual disk you'll get an error when it's trying to write the image | 05:23 |
| tchaypo | before a patch that stevenk is working in, that means you need to set all nodes to the size of the smallest node; after his patch it has the ability to generate custom-sized images for each size of disk you have | 05:24 |
| StevenK | tchaypo: Er? | 05:24 |
| tchaypo | StevenK: don't tell me i dreamed that patch that does baremetal flavors? | 05:25 |
| tchaypo | i even thought it might have landed already | 05:25 |
| StevenK | tchaypo: It's up for os-cloud-config | 05:26 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Import nova-ctr-base into every image to get kolla-common.sh https://review.openstack.org/128831 | 05:28 |
| *** k4n0 has joined #tripleo | 05:29 | |
| *** otoolee has joined #tripleo | 05:33 | |
| jerryz | tchaypo: gotcha. so right now, i can just use a common flavor for all the nodes despite of the physical spec difference as long as the disk size is sufficient? the nodes and flavor registered will use the numbers in my node.json? | 05:35 |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Add pointers to a sample nodes.json https://review.openstack.org/128832 | 05:37 |
| *** pensu has joined #tripleo | 05:38 | |
| tchaypo | jerryz: hang on, let me check something | 05:38 |
| StevenK | The flavor will use the memory, disk and cpu from the first node | 05:38 |
| tchaypo | for a flaver with an NGb root image, and MMMGb specifed in nodes.json, NGb will be partitioned for / and MMM-N for /mnt | 05:39 |
| jerryz | StevenK: what if the seed and under cloud are libvirt vms and the rest are real bare metals for the overcloud | 05:39 |
| tchaypo | if the value in nodes.json is smaller than the actual disk on the node you have some unused space; if it's bigger you have an error building | 05:40 |
| *** pcrews has quit IRC | 05:40 | |
| StevenK | jerryz: The setup-baremetal script takes a number of nodes -- it's the first one passed to setup-baremetal | 05:40 |
| StevenK | If it's split like that, you may need to do things by hand | 05:41 |
| StevenK | You might be able to get away with having the first node (or the first three for HA) be VMs, and then have the remainder as actual physical machines | 05:42 |
| jerryz | StevenK: i remember the nodes for setup-baremetal are the ones left after seed /under are built. | 05:42 |
| StevenK | jerryz: Actually, setup-baremetal runs during devtest_seed and devtest_undercloud to register nodes with a baremetal service and create the flavor | 05:43 |
| jerryz | StevenK: i mean the setup-baremetal in devtest_overcloud | 05:44 |
| StevenK | devtest_overcloud does not run setup-baremetal | 05:44 |
| *** bvivek has joined #tripleo | 05:51 | |
| tchaypo | StevenK: so. register-nodes.. | 05:57 |
| tchaypo | how to i figure out what to provide as for the SERVICE_HOST ? | 05:58 |
| *** rushiagr is now known as rushiagr_away | 06:00 | |
| tchaypo | oh, it's optional | 06:03 |
| *** mrunge has joined #tripleo | 06:08 | |
| * tchaypo discovers he can't use the new register-nodes goodness until he rebases | 06:11 | |
| * tchaypo sees the rebase | 06:11 | |
| * tchaypo cries | 06:11 | |
| StevenK | tchaypo: It's optional for ironic | 06:11 |
| StevenK | I'll be doing the same for -u in init-keystone once no-pki-setup lands | 06:11 |
| *** dshulyak_ has quit IRC | 06:16 | |
| tchaypo | http://paste.openstack.org/show/121398/ | 06:26 |
| tchaypo | oh, i see who's to blame for this mess. | 06:27 |
| tchaypo | the only thing my change has added there is NeutronPublicInterfaceTag | 06:31 |
| tchaypo | and https://review.openstack.org/#/c/103757/36/scripts/devtest_overcloud.sh seems to say taht that should be in the heat template | 06:31 |
| tchaypo | so presumably if I understood how to put it in heat I could avoid this merge.. | 06:32 |
| *** dsneddon has quit IRC | 06:36 | |
| *** dsneddon has joined #tripleo | 06:37 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add svc-map to cinder-api https://review.openstack.org/104198 | 06:44 |
| *** rbrady has joined #tripleo | 06:45 | |
| lifeless | tchaypo: ECONTEXT | 06:52 |
| *** ramishra has quit IRC | 06:56 | |
| *** dsneddon has quit IRC | 06:57 | |
| *** ramishra has joined #tripleo | 06:57 | |
| *** dsneddon has joined #tripleo | 06:58 | |
| *** jtomasek has joined #tripleo | 07:03 | |
| *** ifarkas has joined #tripleo | 07:11 | |
| *** jcoufal has joined #tripleo | 07:16 | |
| *** ramishra has quit IRC | 07:16 | |
| *** ramishra has joined #tripleo | 07:17 | |
| *** dsneddon has quit IRC | 07:18 | |
| *** dsneddon has joined #tripleo | 07:18 | |
| *** ramishra has quit IRC | 07:22 | |
| jerryz | anybody? | 07:28 |
| jerryz | lifeless: around? | 07:29 |
| lifeless | jerryz: somewhat, how can I help? | 07:36 |
| *** jprovazn has joined #tripleo | 07:36 | |
| jerryz | lifeless: where is neutron worker element located? is there a separate role for that? | 07:37 |
| jerryz | lifeless: or just inside controller? | 07:37 |
| *** ramishra has joined #tripleo | 07:38 | |
| *** dtantsur|afk is now known as dtantsur | 07:46 | |
| *** ramishra has quit IRC | 07:53 | |
| *** gfidente has joined #tripleo | 07:55 | |
| *** jp_at_hp has joined #tripleo | 07:57 | |
| *** jml has joined #tripleo | 08:01 | |
| *** bvivek has quit IRC | 08:01 | |
| *** jistr has joined #tripleo | 08:09 | |
| *** otoolee has quit IRC | 08:14 | |
| *** yuanying_ has quit IRC | 08:17 | |
| *** yuanying has joined #tripleo | 08:17 | |
| *** otoolee has joined #tripleo | 08:21 | |
| *** yuanying has quit IRC | 08:21 | |
| *** akrivoka has joined #tripleo | 08:22 | |
| *** lucasagomes has joined #tripleo | 08:27 | |
| *** yamahata has quit IRC | 08:36 | |
| *** bvivek has joined #tripleo | 08:37 | |
| *** kbyrne has quit IRC | 08:42 | |
| *** ramishra has joined #tripleo | 08:43 | |
| *** bashok has joined #tripleo | 08:46 | |
| *** killer_prince is now known as lazy_prince | 08:53 | |
| *** yuanying has joined #tripleo | 08:54 | |
| *** jang2 has joined #tripleo | 08:59 | |
| *** jang has quit IRC | 09:00 | |
| *** jang1 has quit IRC | 09:00 | |
| *** untriaged-bot has joined #tripleo | 09:00 | |
| untriaged-bot | Untriaged bugs so far: | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1381408 | 09:00 |
| uvirtbot | Launchpad bug 1381408 in diskimage-builder ""visudo" always got error when building overcloud image" [Undecided,New] | 09:00 |
| *** untriaged-bot has quit IRC | 09:00 | |
| *** ramishra has quit IRC | 09:01 | |
| *** lazy_prince has quit IRC | 09:05 | |
| lifeless | jerryz: worker element? | 09:06 |
| lifeless | jerryz: do you mean the neutron agent? | 09:06 |
| jerryz | lifeless: yes | 09:09 |
| tchaypo | lifeless: about dnsmasq? | 09:11 |
| *** pensu has quit IRC | 09:11 | |
| jerryz | lifeless: tchaypo: also yes | 09:14 |
| jerryz | lifeless, tchaypo: i mean where the external network is | 09:16 |
| *** lazy_prince has joined #tripleo | 09:17 | |
| jerryz | the baremetals that need internet access | 09:18 |
| openstackgerrit | Peter Belanyi proposed a change to openstack/tripleo-heat-templates: Add converted version of block and object storage https://review.openstack.org/123713 | 09:25 |
| *** pensu has joined #tripleo | 09:26 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Add a big hammer https://review.openstack.org/98336 | 09:28 |
| *** funzo has quit IRC | 09:28 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Force empty $TMPDIR inside the chroot https://review.openstack.org/128696 | 09:28 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Support debug logging on seed https://review.openstack.org/128704 | 09:37 |
| *** derekh has joined #tripleo | 09:39 | |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Revert "Set tripleo ci control variables based on job name" https://review.openstack.org/123788 | 09:39 |
| *** pblaho has joined #tripleo | 09:43 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Remove first-boot.d support https://review.openstack.org/126836 | 09:48 |
| GheRivero | morning all | 09:57 |
| *** killer_prince has joined #tripleo | 10:03 | |
| openstackgerrit | Tom Cammann proposed a change to openstack/tripleo-image-elements: Disable SSLv3 in Stunnel https://review.openstack.org/128868 | 10:05 |
| *** jerryz has quit IRC | 10:13 | |
| *** pelix has joined #tripleo | 10:16 | |
| openstackgerrit | Loganathan Parthipan proposed a change to openstack/tripleo-image-elements: sets novnc endpoint uri to https if tls is enabled https://review.openstack.org/128871 | 10:16 |
| *** openstackgerrit has quit IRC | 10:19 | |
| *** openstackgerrit has joined #tripleo | 10:20 | |
| *** akrivoka has quit IRC | 10:21 | |
| *** lazy_prince has quit IRC | 10:24 | |
| *** killer_prince is now known as lazy_prince | 10:24 | |
| *** ramishra has joined #tripleo | 10:28 | |
| tchaypo | jerryz: I'm not sure how that works. I think maybe it's as simple as them getting a route to the seed by default, withthe ability to add in a different gateway IP if needed | 10:30 |
| tchaypo | but I don't actually know anything about it so I could be wildly wrong | 10:30 |
| *** ramishra has quit IRC | 10:32 | |
| *** yamahata has joined #tripleo | 10:36 | |
| *** jp_at_hp has quit IRC | 10:38 | |
| openstackgerrit | Nicholas Randon proposed a change to openstack/tripleo-image-elements: Fix RabbitMQ element clustering start and stop https://review.openstack.org/115524 | 10:48 |
| *** pensu has quit IRC | 10:57 | |
| *** ramishra has joined #tripleo | 10:57 | |
| *** akrivoka has joined #tripleo | 10:59 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Avoid overwritting of hooks https://review.openstack.org/126869 | 11:06 |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tuskar-ui: Make safe int cast https://review.openstack.org/128883 | 11:06 |
| *** otoolee has quit IRC | 11:28 | |
| *** otoolee has joined #tripleo | 11:34 | |
| *** rhallisey has joined #tripleo | 11:35 | |
| *** ramishra has quit IRC | 11:35 | |
| *** julim has quit IRC | 11:46 | |
| openstackgerrit | therese-mchale proposed a change to openstack/tripleo-image-elements: Use Percona Packages for Ubuntu and Debian https://review.openstack.org/90134 | 11:48 |
| *** bvivek has quit IRC | 12:00 | |
| *** radez_g0n3 is now known as radez | 12:04 | |
| *** ramishra has joined #tripleo | 12:16 | |
| *** jehb has joined #tripleo | 12:19 | |
| openstackgerrit | Ana Krivokapic proposed a change to openstack/tuskar-ui: Style the Deployment Confirmation dialog https://review.openstack.org/128903 | 12:19 |
| *** funzo has joined #tripleo | 12:23 | |
| *** morazi has quit IRC | 12:25 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Disable all interfaces on eni systems https://review.openstack.org/128213 | 12:27 |
| *** ramishra has quit IRC | 12:28 | |
| *** dprince has joined #tripleo | 12:32 | |
| *** jdob has joined #tripleo | 12:32 | |
| openstackgerrit | A change was merged to openstack/os-collect-config: Modify gitignore to ignore cover https://review.openstack.org/120994 | 12:48 |
| *** pensu has joined #tripleo | 12:53 | |
| *** ramishra has joined #tripleo | 12:54 | |
| *** pradk has quit IRC | 12:55 | |
| *** shadower39 has joined #tripleo | 12:57 | |
| *** noslzzp has joined #tripleo | 12:59 | |
| *** shadower39 has quit IRC | 13:01 | |
| *** shadower78 has joined #tripleo | 13:02 | |
| *** shadower has quit IRC | 13:03 | |
| *** shadower78 is now known as shadower | 13:03 | |
| *** jang2 has quit IRC | 13:05 | |
| *** tzumainn has joined #tripleo | 13:06 | |
| *** pblaho has quit IRC | 13:07 | |
| *** tsedovic has joined #tripleo | 13:08 | |
| *** tsedovic has quit IRC | 13:09 | |
| *** jang has joined #tripleo | 13:10 | |
| *** jang1 has joined #tripleo | 13:10 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: ensure that port is added to bridge after a reboot https://review.openstack.org/123534 | 13:10 |
| openstackgerrit | A change was merged to openstack/os-cloud-config: Switch to oslo.i18n https://review.openstack.org/128017 | 13:11 |
| *** morazi has joined #tripleo | 13:15 | |
| *** ramishra has quit IRC | 13:20 | |
| *** ramishra has joined #tripleo | 13:21 | |
| *** julim has joined #tripleo | 13:21 | |
| *** pensu has quit IRC | 13:25 | |
| *** bashok has quit IRC | 13:28 | |
| openstackgerrit | Erik Colnick proposed a change to openstack/tripleo-specs: TripleO DVR enablement https://review.openstack.org/113983 | 13:34 |
| SpamapS | tchaypo: my thingies are buried deep inside the corporate HP network unfortunately. :-P | 13:37 |
| *** jprovazn has quit IRC | 13:37 | |
| *** rbrady_ has joined #tripleo | 13:46 | |
| *** rbrady has quit IRC | 13:47 | |
| *** rbrady_ is now known as rbrady | 13:48 | |
| sdake | morning | 13:49 |
| openstackgerrit | Erik Colnick proposed a change to openstack/tripleo-image-elements: Create 'neutron-router' image element https://review.openstack.org/127541 | 13:49 |
| *** cody-somerville has joined #tripleo | 13:50 | |
| *** cody-somerville has joined #tripleo | 13:50 | |
| openstackgerrit | Erik Colnick proposed a change to openstack/tripleo-image-elements: Remove redundancies between neutron-router and neutron-network-node https://review.openstack.org/127618 | 13:56 |
| *** jp_at_hp has joined #tripleo | 13:57 | |
| openstackgerrit | Erik Colnick proposed a change to openstack/tripleo-image-elements: Add Neutron DVR configuration variables https://review.openstack.org/127988 | 13:58 |
| *** pradk has joined #tripleo | 13:59 | |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-image-elements: Fix race between apache2 starting and o-r-c configuring apache2 https://review.openstack.org/112039 | 14:00 |
| *** ramishra has quit IRC | 14:00 | |
| *** noslzzp has quit IRC | 14:00 | |
| sdake | larsks rhaillisey can you ahe a look at the nova-conductor changes | 14:03 |
| sdake | nova-conductor is almost working with these | 14:03 |
| larsks | sdake: sure, will look in just a few. | 14:03 |
| *** TheJulia has joined #tripleo | 14:05 | |
| *** ramishra has joined #tripleo | 14:09 | |
| openstackgerrit | Giulio Fidente proposed a change to openstack/tripleo-image-elements: Ensure neutron rootwrap.d symlink is not nested https://review.openstack.org/128935 | 14:11 |
| *** jprovazn has joined #tripleo | 14:12 | |
| *** spzala has joined #tripleo | 14:14 | |
| sdake | nova-conductor/nova-controller above | 14:17 |
| *** pradeep has quit IRC | 14:17 | |
| *** spzala has quit IRC | 14:18 | |
| *** spzala has joined #tripleo | 14:18 | |
| *** weshay has quit IRC | 14:20 | |
| *** athomas has quit IRC | 14:23 | |
| rhallisey | sdake, were you able to see nova.conf when you ran the api? | 14:23 |
| *** morazi has quit IRC | 14:23 | |
| greghaynes | SpamapS: TheJulia Heres the pattern that im thinking for bootstrap_host group https://review.openstack.org/#/c/128938/ | 14:23 |
| *** jang1 has quit IRC | 14:24 | |
| *** jang has quit IRC | 14:24 | |
| *** jang has joined #tripleo | 14:25 | |
| SpamapS | greghaynes: yes that is fantastic | 14:26 |
| SpamapS | greghaynes: so what you're saying is.. add another layer of abstraction.. ;) | 14:26 |
| greghaynes | heh | 14:26 |
| greghaynes | basically | 14:26 |
| greghaynes | or at least use the already existing one | 14:27 |
| *** marun has joined #tripleo | 14:27 | |
| *** jang has quit IRC | 14:28 | |
| *** athomas has joined #tripleo | 14:30 | |
| *** jang has joined #tripleo | 14:30 | |
| *** morazi has joined #tripleo | 14:31 | |
| *** jang has quit IRC | 14:32 | |
| *** jang has joined #tripleo | 14:33 | |
| *** jang has quit IRC | 14:33 | |
| *** jang has joined #tripleo | 14:34 | |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-image-elements: Fix race between apache2 starting and o-r-c configuring apache2 https://review.openstack.org/112039 | 14:34 |
| *** jang has quit IRC | 14:36 | |
| *** jang has joined #tripleo | 14:37 | |
| *** jang has quit IRC | 14:37 | |
| *** weshay has joined #tripleo | 14:38 | |
| *** jang has joined #tripleo | 14:38 | |
| *** noslzzp has joined #tripleo | 14:38 | |
| larsks | sdake: question about some of your changes in e.g. https://review.openstack.org/#/c/128800/: for other things (like glance), we are creating a glance-base image that installs the common stuff required by all the other glance images. We were doing the same for nova; any reason not to? | 14:40 |
| *** jang has quit IRC | 14:40 | |
| larsks | It'll work either way, of course. Just curious. | 14:40 |
| *** k4n0 has quit IRC | 14:40 | |
| *** jang has joined #tripleo | 14:41 | |
| *** jang has quit IRC | 14:41 | |
| rhallisey | sdake, so for the sake of getting it working we can do it the way you have it | 14:42 |
| larsks | Yup. | 14:42 |
| rhallisey | sdake, but after I might go back and change it to make it similar to the others | 14:42 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-image-elements: Fix race between apache2 starting and o-r-c configuring apache2 https://review.openstack.org/112039 | 14:44 |
| *** jang has joined #tripleo | 14:44 | |
| *** cody-somerville has quit IRC | 14:44 | |
| *** akrivoka has quit IRC | 14:47 | |
| *** openstackgerrit has quit IRC | 14:48 | |
| *** openstackgerrit has joined #tripleo | 14:49 | |
| *** dmellado has joined #tripleo | 14:50 | |
| *** akrivoka has joined #tripleo | 14:50 | |
| larsks | sdake: actually, I think we need to rework those changes. | 14:53 |
| larsks | I think we need to keep the nova-controller base image (we can rename it if you like)...because in https://review.openstack.org/#/c/128796/, we ended up with three different shell scripts all doing the same thing. | 14:54 |
| larsks | So for every change we need to make, we need to make it in three places. | 14:54 |
| rhallisey | larsks, https://review.openstack.org/#/c/128796/ we can just do the config in the controller | 14:57 |
| larsks | rhallisey: I think that is what I was just saying, or are you saying something else? | 14:57 |
| larsks | I am not sure I followed that comment. | 14:57 |
| rhallisey | larsks, yes I agree but that's what it was before | 14:58 |
| rhallisey | I was wondering why it was changed | 14:58 |
| rhallisey | or didn't work | 14:58 |
| larsks | rhallisey: I am still not following you! Are you agreeing that we should go back to having a base image from which the other images inherit? | 14:58 |
| rhallisey | yes | 14:58 |
| larsks | Yay! :) | 14:58 |
| rhallisey | :) | 14:58 |
| larsks | The reason it was changed is that the start script was trying to start the nova-controller image, which was an error. The correct fix would have been to fix the start script. | 14:59 |
| larsks | I am working on that, and trying to integrate changes from this series of patches. | 14:59 |
| rhallisey | larsks, ok, that answers my question | 14:59 |
| *** ramishra has quit IRC | 15:00 | |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1381408 | 15:00 |
| uvirtbot | Launchpad bug 1381408 in diskimage-builder ""visudo" always got error when building overcloud image" [Undecided,New] | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| larsks | we do need to remove start.sh from the nova-controller image (that will just get tromped on by start scripts in other images). | 15:00 |
| larsks | we can handle common setup as in glance (where the base image installs a configure-glance.sh script that is called by start.sh in other images) | 15:00 |
| *** jprovazn is now known as jprovazn_afk | 15:00 | |
| *** lazy_prince is now known as killer_prince | 15:01 | |
| *** sdake_ has joined #tripleo | 15:03 | |
| *** sdake_ has quit IRC | 15:03 | |
| *** sdake_ has joined #tripleo | 15:03 | |
| openstackgerrit | Nicholas Randon proposed a change to openstack/tripleo-image-elements: Fix RabbitMQ element clustering start and stop https://review.openstack.org/115524 | 15:04 |
| *** noslzzp has quit IRC | 15:07 | |
| *** noslzzp has joined #tripleo | 15:09 | |
| *** ramishra has joined #tripleo | 15:10 | |
| *** pcrews has joined #tripleo | 15:12 | |
| *** ifarkas has quit IRC | 15:14 | |
| sdake | larsks I ended up making that change later | 15:14 |
| sdake | but happy to rework the change set if it matters | 15:14 |
| *** daneyon has quit IRC | 15:14 | |
| sdake | personally I dont trhink it does, as we don't have any real ci :( | 15:14 |
| *** daneyon has joined #tripleo | 15:15 | |
| *** noslzzp has quit IRC | 15:15 | |
| *** ifarkas has joined #tripleo | 15:16 | |
| larsks | This isn't a question of CI. | 15:16 |
| larsks | This is a maintenance issue. Someone reports a bug, you fix it in nova-api/start.sh or something, but not in nova-conductor/start.sh, and then someone reports the same bug in that image... | 15:16 |
| *** noslzzp has joined #tripleo | 15:17 | |
| greghaynes | TheJulia: SpamapS For some reason I end up with a "All hosts have already failed" but I dont see any failing tasks and the ply recap says failed=0 for every host.... | 15:18 |
| greghaynes | youall seen this before? | 15:19 |
| *** pensu has joined #tripleo | 15:19 | |
| *** pradeep has joined #tripleo | 15:21 | |
| *** ramishra has quit IRC | 15:22 | |
| openstackgerrit | Erik Colnick proposed a change to openstack/tripleo-image-elements: Add logic to ensure vlan configuration is persisted https://review.openstack.org/125118 | 15:27 |
| *** killer_prince has quit IRC | 15:28 | |
| *** bvivek has joined #tripleo | 15:30 | |
| *** avozza has joined #tripleo | 15:33 | |
| *** pradeep has quit IRC | 15:36 | |
| *** killer_prince has joined #tripleo | 15:37 | |
| *** killer_prince is now known as lazy_prince | 15:37 | |
| lxsli | greghaynes: at your convenience please would you see my response on https://review.openstack.org/#/c/113546 ? Cheers! | 15:38 |
| *** jistr has quit IRC | 15:40 | |
| *** bvivek has quit IRC | 15:42 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Refactor Heat dependencies for nova compute https://review.openstack.org/122384 | 15:42 |
| *** bvivek has joined #tripleo | 15:43 | |
| greghaynes | lxsli: will do | 15:43 |
| sdake | larsks look later in the patch stream | 15:45 |
| sdake | if we had ci, this would probably bust | 15:45 |
| sdake | but later in the patch stream I do exactly as you mention | 15:46 |
| sdake | I'd rather not have to rewrite all the patches but can if necessary... | 15:46 |
| larsks | sdake: I see https://review.openstack.org/#/c/128831/ | 15:46 |
| larsks | ...which seems to introduce even more duplication. errr, triplication. | 15:46 |
| larsks | I am in the midst of reworking these patches. | 15:46 |
| sdake | it copies the same pattern of glance | 15:46 |
| sdake | exact same pattern | 15:47 |
| sdake | nova-ctr-base is like glance-base | 15:47 |
| sdake | notice the D next t o nova-config.sh in each of the subdirs | 15:47 |
| SpamapS | greghaynes: no that should not happen | 15:48 |
| SpamapS | greghaynes: you should have at least one failed | 15:48 |
| SpamapS | greghaynes: it is a misnomer "all" .. it really means "the max fail percentage has been exceeded" | 15:49 |
| larsks | Yup. But it's not clear why you discarded the nova-controller-base only to re-introduce the same thing. | 15:49 |
| larsks | It's kind of a conusing patch series. | 15:49 |
| larsks | *confusing | 15:49 |
| sdake | ok I'll rework the patch set then | 15:50 |
| larsks | Like my typing. | 15:50 |
| greghaynes | SpamapS: hrm, well it looks like the command at the top of galera_status is skipping the hosts because the removed=, which is fine, but then it stops running the playbook there and reports fail | 15:50 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: changes to base nova configuration https://review.openstack.org/128966 | 15:52 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: fix nova-controller pod https://review.openstack.org/128967 | 15:52 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: have start.sh source in common configuration script https://review.openstack.org/128968 | 15:52 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: Nova has an undeclared dependency on python-cinderclient https://review.openstack.org/128969 | 15:52 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: introduce conductor to common configuration https://review.openstack.org/128970 | 15:52 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: introduce scheduler to common configuration https://review.openstack.org/128971 | 15:52 |
| larsks | sdake: see what you think of that series ^^^^^ | 15:52 |
| greghaynes | SpamapS: Heres an end portion of my log http://paste.ubuntu.com/8574580/ | 15:52 |
| sdake | larsks you need MY_IP | 15:53 |
| sdake | hence you need kolla-common in the start scripts | 15:53 |
| larsks | Take a closer look. | 15:53 |
| larsks | They are all getting kolla-common. | 15:53 |
| larsks | Via config-nova-controller. | 15:53 |
| larsks | (.sh) | 15:53 |
| sdake | i see | 15:53 |
| sdake | diffeerente then glance | 15:53 |
| sdake | I think you need to have separate ops because you want to check between common import and nova-config | 15:54 |
| sdake | check for variables that is | 15:54 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Set tripleo ci control variables based on job type https://review.openstack.org/128972 | 15:54 |
| larsks | There are start-specific checks where appropriate (e.g., we only check DB_ROOT_PASSWORD in the nova-conductor start script) | 15:55 |
| larsks | (via check_for_db) | 15:55 |
| sdake | I think we want nova-conductor creating the db/keystone users | 15:55 |
| sdake | rather then nova-api | 15:55 |
| larsks | Okay. Why for? | 15:56 |
| sdake | api could horizontally scale | 15:56 |
| sdake | conductor not so | 15:56 |
| larsks | It does create the database and database users. But why keystone? | 15:56 |
| sdake | it is something we can fix later I guess | 15:56 |
| SpamapS | greghaynes: I'd share that in #ansible .. it looks weird to me. | 15:56 |
| TheJulia | greghaynes: would seem like it thinks it can't proceed but no single host failed... possibly a bad when conditional | 15:56 |
| larsks | I think the fix for that and elsewhere is going to be gus' suggestion re: moving one-time init completely outside of the service images. | 15:57 |
| *** rhallisey has quit IRC | 15:57 | |
| greghaynes | hrm | 15:57 |
| larsks | E.g., keystone and glance have the same problem right now. | 15:57 |
| TheJulia | greghaynes: that or it doesn't know about the host you've referenced on the hosts line | 15:57 |
| greghaynes | TheJulia: Yea, my thinking is its some wierd interaction between serial: 1 and when | 15:57 |
| SpamapS | sdake: have you considered having an ephemeral container that is just brought up to bootstrap things, and then removed. | 15:57 |
| greghaynes | well, it skipped one host which seems correct | 15:57 |
| SpamapS | larsks: ahh I see you have considered that :) | 15:58 |
| TheJulia | greghaynes: I'd put name tags in between each step to give you more visibility | 15:58 |
| larsks | SpamapS: Yeah, exactly. | 15:58 |
| sdake | SpamapS yes we have - enounicorns :) | 15:58 |
| larsks | We can use kubernetes restart policies to create run-once containers. | 15:58 |
| SpamapS | Separate orchestration of end goal from workflow :) | 15:59 |
| larsks | SpamapS: That is totally the direction in which I would like to move. | 15:59 |
| sdake | yup but keep in mind wehave 2 days to finish getting nova working | 15:59 |
| greghaynes | TheJulia: ah, good call | 15:59 |
| SpamapS | you have a bring-up workflow which is [put things in a pre-bootstrapped state][bootstrap][put things in ready state] | 15:59 |
| sdake | so lets not complicate things in the short term for our milestone 1 deadline | 15:59 |
| larsks | sdake: and that is why I am not suggesting we make those changes right now :) | 15:59 |
| SpamapS | that [bootstrap] state is the reason os-cloud-config exists. ;) | 16:00 |
| larsks | SpamapS: well, this seems like a completely separate issue. Regardless of how we bootstrap things, we need to move that process outside of the service containers. | 16:00 |
| *** jistr has joined #tripleo | 16:00 | |
| SpamapS | 'kube-magic-thing run os-cloud-config' | 16:00 |
| SpamapS | larsks: yeah ignore my how .. the where and when are more interesting. :) | 16:01 |
| SpamapS | though I'm still pretty sure we shouldn't have two ways to bootstrap OpenStack clouds if we can avoid it. | 16:01 |
| sdake | larsks I've reviewed the patch set with what I think would make it proper - if yuou can rework quickly we can merge | 16:03 |
| *** rameshg87-afk has joined #tripleo | 16:03 | |
| larsks | I'm on it. I would like to test before we merge. | 16:03 |
| rameshg87-afk | dprince, SpamapS, hi | 16:03 |
| sdake | larsks have to keep in mind I spent all day debugging that stuff yesterday so that is why my patch set looks disorganized | 16:03 |
| larsks | That is what rebase is for :) | 16:03 |
| rameshg87-afk | dprince, SpamapS, request you to have a look at the review https://review.openstack.org/123642 . | 16:04 |
| sdake | yes, but ~9pm time for bed rather then rebase ;) | 16:04 |
| larsks | You are obviously better at time management that I am. | 16:04 |
| sdake | also you want to fix the busted crudini as a dependency somewhere in that chain | 16:04 |
| * larsks REBASES ALL THE THINGS | 16:04 | |
| sdake | because set -e will cause the script to error immediately durnig setup | 16:04 |
| larsks | That is the idea, yes. | 16:05 |
| sdake | what I mean is there is a broken crudini operation | 16:05 |
| sdake | which will not return 0 | 16:05 |
| larsks | If you have noted it in comments, I will fix it. | 16:05 |
| sdake | I'm not sure where it s hould go sec | 16:05 |
| larsks | There were some typos in the existing configs (a --del with a value, and some operations that were referring to /etc/nova/nova instead of /etc/nova/nova.conf). I fixed in these existing patches. | 16:06 |
| greghaynes | SpamapS: TheJulia So in wierder news, if I do --limit to one host it fails in the step after the last one it prints out | 16:07 |
| greghaynes | makes me think its actually running the next step and just not printing it | 16:07 |
| sdake | larsks - ok then that should be fine | 16:07 |
| *** pcrews has quit IRC | 16:08 | |
| larsks | sdake: re: 31 character limit on docker image names, where is that? I can successfully build "kollaglue/fedora-rdo-nova-controller", which has 36 characters. | 16:08 |
| sdake | try nova-controller-base | 16:08 |
| larsks | Or is that a limitation in the docker hub? | 16:08 |
| sdake | docker hub i guess | 16:08 |
| sdake | it spits out an error on push | 16:08 |
| larsks | That works also. | 16:09 |
| larsks | Ah, okay. | 16:09 |
| *** pcrews has joined #tripleo | 16:09 | |
| sdake | i'm not sure what the limit encompasses | 16:09 |
| sdake | I just know nova-controller-base no worky | 16:09 |
| *** rainya has joined #tripleo | 16:09 | |
| larsks | Yup, I see the error from the hub now. Seems like a silly limit, but what can ya do. | 16:09 |
| *** nosnos has quit IRC | 16:12 | |
| *** nosnos has joined #tripleo | 16:12 | |
| larsks | Hmmm, kollaglue/fedora-rdo-nova-ctr-base is 34 characters, but seems to push anyway. I guess kollaglue/fedora-rdo-n-c-base is safer. | 16:13 |
| *** rhallisey has joined #tripleo | 16:14 | |
| *** openstackstatus has joined #tripleo | 16:14 | |
| *** ChanServ sets mode: +v openstackstatus | 16:14 | |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: renamed nova-controller -> n-c-base https://review.openstack.org/128985 | 16:15 |
| *** nosnos has quit IRC | 16:17 | |
| *** rameshg87-afk is now known as rameshg87 | 16:17 | |
| *** noslzzp has quit IRC | 16:19 | |
| *** dtantsur is now known as dtantsur|afk | 16:19 | |
| rameshg87 | derekh: can you please have a look at this review https://review.openstack.org/#/c/123642/ . this is required for an ironic driver landed in juno. | 16:19 |
| *** akrivoka has quit IRC | 16:19 | |
| -openstackstatus- NOTICE: An error in a configuration change to mitigate the poodle vulnerability caused a brief outage of git.openstack.org from 16:06-16:12. The problem has been corrected and git.openstack.org is working again. | 16:20 | |
| *** ramishra has joined #tripleo | 16:22 | |
| *** jcoufal has quit IRC | 16:26 | |
| *** pradeep has joined #tripleo | 16:28 | |
| *** lsmola has quit IRC | 16:29 | |
| *** lazy_prince has quit IRC | 16:30 | |
| *** dmellado has quit IRC | 16:31 | |
| sdake | larsks I went aheaed and abanonded my change set, so if you can power out a new set, tha twouldrock so I can get some dev done today :) | 16:33 |
| *** dshulyak_ has joined #tripleo | 16:33 | |
| larsks | sdake: I responded to several of your comments. I think things are in good shape. | 16:33 |
| larsks | Let me know. I am building images for testing right now. | 16:34 |
| rhallisey | larsks, oh you're the culprit :P. I'm waiting for you to finish | 16:34 |
| larsks | rhallisey: Huh? I'm building into my larsks/ namespace. | 16:34 |
| *** ramishra has quit IRC | 16:34 | |
| rhallisey | Repository kollaglue/fedora-rdo-base already being pulled by another client. Waiting | 16:35 |
| rhallisey | how is this.. I'm building from my namespace.. | 16:35 |
| openstackgerrit | Jeff Peeler proposed a change to stackforge/kolla: WIP: Current heat service work https://review.openstack.org/128994 | 16:36 |
| *** rameshg87 has quit IRC | 16:36 | |
| larsks | rhallisey: that error is referring to operations happening on your system. | 16:36 |
| larsks | s/error/"informative message" | 16:37 |
| larsks | That means your local docker daemon is engaged in a pull operation. | 16:37 |
| rhallisey | larsks, I just rebuilt these minions.. weird.. | 16:37 |
| * rhallisey builds again | 16:38 | |
| greghaynes | SpamapS: TheJulia So, it seems like setting max_fail_percentage: 0 is a no-nop | 16:39 |
| *** jp_at_hp has quit IRC | 16:40 | |
| sdake | larsks I went back through, check the -1's out | 16:41 |
| larsks | checking | 16:41 |
| sdake | pretty minor stuff | 16:41 |
| sdake | daneyon your changes rock :) - I hada fit nitpicks about spaces vs tabs | 16:41 |
| sdake | I prefer tabs, but everyone else loves spaces for whatever reason | 16:41 |
| jpeeler | people have their editors still setup for python | 16:42 |
| larsks | sdake: the comment on https://review.openstack.org/#/c/128968/ really belongs to https://review.openstack.org/#/c/128966/, since this is the patch that introduces the crudini operations. | 16:42 |
| larsks | sdake: I love tabs for shell scripts, especially because some operations (indented "here" documents) can't be used with spaces. | 16:43 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Ensure neutron rootwrap.d symlink is not nested https://review.openstack.org/128935 | 16:43 |
| daneyon | sdake: thanks. now i'm working on the difficult part... neutron agents and plugins. u cool with focusing on ml2/ovs/gre to align with the os docs or mls/ovs/vxlan or something else? | 16:44 |
| *** lsmola has joined #tripleo | 16:44 | |
| larsks | daneyon: I think vxlan is the preferred tunneling solution these days, no? | 16:46 |
| larsks | Or am I imagining? | 16:46 |
| greghaynes | SpamapS: TheJulia https://review.openstack.org/#/c/128997/ | 16:47 |
| *** lucasagomes has quit IRC | 16:47 | |
| * larsks needs lunch. back in a bit. | 16:48 | |
| sdake | daneyon I can barely get neutron working typically so I'd go with your judgement :) | 16:49 |
| *** derekh has quit IRC | 16:50 | |
| sdake | larsks I thought you might just introduce a patch between the two | 16:50 |
| sdake | rather then ham it iinto one or the other | 16:51 |
| *** yamahata has quit IRC | 16:52 | |
| *** dshulyak_ has quit IRC | 16:53 | |
| *** pradeep1 has joined #tripleo | 16:53 | |
| *** pradeep has quit IRC | 16:56 | |
| *** bvivek has quit IRC | 16:57 | |
| daneyon | sdake: I agree vxlan is preferred. I will shoot for mls/ovs/vxlan and use gre as a backup plan. ok? | 16:59 |
| *** avozza is now known as zz_avozza | 17:03 | |
| *** killer_prince has joined #tripleo | 17:04 | |
| *** killer_prince is now known as lazy_prince | 17:05 | |
| SpamapS | greghaynes: nice work. :) | 17:06 |
| *** yamahata has joined #tripleo | 17:08 | |
| *** rainya has quit IRC | 17:08 | |
| *** rainya has joined #tripleo | 17:09 | |
| sdake | daneyon speaking french but wfm ;-) | 17:09 |
| *** pradeep has joined #tripleo | 17:09 | |
| *** pradeep1 has quit IRC | 17:09 | |
| *** noslzzp has joined #tripleo | 17:09 | |
| *** rainya has quit IRC | 17:14 | |
| sdake | larsks I notice with mion count of 1 the heat template works great but minion count of 2 thiungs don't seem to work | 17:14 |
| larsks | I run with 2 minions all the time, no problems. | 17:15 |
| sdake | yup | 17:15 |
| sdake | maybe you can investigate my setup after we release | 17:15 |
| larsks | Is the overlay network up? | 17:15 |
| *** rainya has joined #tripleo | 17:15 | |
| sdake | so I can get with 2 minions | 17:15 |
| sdake | 1 minion is working fine | 17:15 |
| larsks | You should be able to ping the address of kbr0 on one minion from the other minion. | 17:15 |
| sdake | I am not sure, I stopped trying with 2 minions a couple days ago | 17:15 |
| larsks | Okay. Well, when you have the chance, give that a shot. | 17:15 |
| sdake | I think we would be wise to recommend 1 minion in the docs | 17:15 |
| sdake | after release I should be able to get into it | 17:15 |
| *** ifarkas has quit IRC | 17:18 | |
| sdake | docs/release info | 17:18 |
| *** dshulyak_ has joined #tripleo | 17:19 | |
| larsks | sdake: I haven't seen a problem with multiple minions elsewhere. I don't want to discourage testing with multiple minions unless we have a real problem, because I think it helps flush out configuration issues that might "work by accident" on a single minion. | 17:21 |
| larsks | I guess I am happy to make the default number_of_minions=1, without otherwise discouraging people from trying more. | 17:23 |
| larsks | Does that work for you? | 17:23 |
| sdake | wfm | 17:25 |
| *** cody-somerville has joined #tripleo | 17:25 | |
| sdake | that sounds great | 17:25 |
| sdake | thatway we dont have to mention in our directions about that point | 17:25 |
| sdake | larsks I think generally all the devs except you are havingtrouble with multiple minions | 17:27 |
| sdake | so maybe its environmental? | 17:27 |
| larsks | I don't know. That's the first I've heard about problems. | 17:27 |
| larsks | No reports == no chance to diagnose. | 17:27 |
| sdake | I have been complainign for a week ;-) | 17:27 |
| *** yamahata has quit IRC | 17:27 | |
| sdake | but its all good | 17:27 |
| sdake | I'm satisified in single minion setup the code we have minus nova is in good working order | 17:27 |
| larsks | I'm satisfied that it's in good working order on a multi-minion system. | 17:28 |
| sdake | once the nova changes land, we can fixed it | 17:28 |
| sdake | cool | 17:28 |
| larsks | But again, if someone can actually show me a problem, I would love to take a look. | 17:28 |
| sdake | i'll show you tuesday/wed of next week | 17:28 |
| sdake | maybe you can debug | 17:28 |
| larsks | Awesome. | 17:28 |
| sdake | I'm not quite clear on how lanmanager fits into the equation | 17:28 |
| sdake | or how to debug all the things | 17:28 |
| sdake | i've managed to learn to debug k8s and that is about it | 17:29 |
| sdake | and docker | 17:29 |
| larsks | linkamanager sets up the overlay network that permits any pod to be accessible from any minion. That's pretty much all it does. | 17:29 |
| sdake | but neutron+linkmanager nope | 17:29 |
| sdake | yup I got that | 17:29 |
| sdake | not sure if I'm hitting a problem with that or neutron or something else | 17:29 |
| sdake | but I don't want to distrub my env for t hat atm | 17:29 |
| larsks | https://github.com/larsks/heat-kubernetes/commit/40f943cc4a1aac98ae3e54f1153c6b984ed10458 | 17:30 |
| sdake | cool thanks! | 17:30 |
| sdake | I think goal of announce email is just to get people to try openstack is the simplest possible way | 17:31 |
| sdake | (on k8s+docker that is) | 17:31 |
| larsks | Remember that the k8s folks provide instructions for setting up a vagrant-based test environment. I'm only using Heat because I already had an appropriate openstack environment up and running. | 17:32 |
| sdake | yup heat rocks for this | 17:32 |
| sdake | and it fits in with the target audience | 17:32 |
| larsks | Well, my point was that for folks who do not already have openstack, it might be easier to follow the k8s vagrant instructions. | 17:32 |
| sdake | oh - right I guess we need a faq for that then :) | 17:32 |
| larsks | Because then they don't need to learn about nova and neutron and heat just to learn about k8s and the kolla images. | 17:32 |
| sdake | or a link to theier setup docs | 17:33 |
| sdake | I find launching a heat template with a git pull and heat create easier then following the directions from k8s upstream | 17:33 |
| larsks | Even so, you've had some neutron issues to overcome. The templates are great for people who already have openstack up and running, but not necessarily the best starting point for other folks. | 17:35 |
| sdake | right | 17:35 |
| larsks | I haven't looked closely at the upstream docs though, so I don't know how hard/easy is that process. | 17:35 |
| sdake | agree / agree :) | 17:35 |
| sdake | so where are you wiht fixing the -1s | 17:35 |
| sdake | so we can get this merged and start fixing nova-controller to actually work | 17:36 |
| sdake | because right now, it doesn't even after my change set (which ends i nthe same state as yours) | 17:36 |
| *** bkero has quit IRC | 17:36 | |
| larsks | There are only two -1s left. | 17:36 |
| larsks | On https://review.openstack.org/#/c/128968/, I don't know which, if any, crudini op is failing. | 17:37 |
| larsks | But if there is a failure I should be able to replicate it real soon now. | 17:37 |
| sdake | something to do with a log operation I think | 17:37 |
| larsks | On https://review.openstack.org/#/c/128985/, I like "nova-ctr-base" also, but the resulting image name is 34 characters. | 17:37 |
| sdake | and the logs change /etc/nova/nova vs /etc/nova/nova.conf | 17:38 |
| larsks | sdake: I think that is already fixed. | 17:38 |
| sdake | nova-ctr-base works I've pushed it | 17:38 |
| larsks | I realize that it works, but it violates the 31 character limit, which means it might stop working without warning. | 17:38 |
| sdake | I don't know what the 31 character limit is | 17:38 |
| larsks | If docker hub says "limit your image names to 31 characters", I think we should do that. | 17:38 |
| sdake | perhaps its after the "user"( kolla) | 17:38 |
| larsks | Hmmm, maybe. | 17:38 |
| larsks | Let me experiment. | 17:38 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Don't re-install cloud-init for centos7 images https://review.openstack.org/127411 | 17:39 |
| sdake | daneyon if you want perms to push images, sign up on docker.io hub and let me know your id | 17:40 |
| *** zz_avozza is now known as avozza | 17:40 | |
| larsks | The error suggests that it is the entire thing: "2014/10/16 13:40:13 Error: Status 400 trying to push repository larsks/this-is-a-long-name-so-stop-whi: "this-is-a-long-name-so-stop-whi is too long, needs to be < 31 characters" | 17:40 |
| larsks | (and len("this-is-a-long-name-so-stop-whi") == 31) | 17:41 |
| daneyon | sdake: i have an acct: danehans | 17:41 |
| sdake | thats 38 chars larsks | 17:41 |
| sdake | 38 -5 = 32 not 31 | 17:41 |
| openstackgerrit | A change was merged to stackforge/kolla: changes to base nova configuration https://review.openstack.org/128966 | 17:41 |
| openstackgerrit | A change was merged to stackforge/kolla: fix nova-controller pod https://review.openstack.org/128967 | 17:41 |
| sdake | ya maybe it needs to be 30 chars | 17:41 |
| larsks | Right. You were suggesting "perhaps its after the "user"". | 17:41 |
| larsks | I was testing that theory. | 17:41 |
| sdake | maybe it includes the / ;) | 17:42 |
| larsks | And it appears to be not the case. | 17:42 |
| sdake | try 1 less character | 17:42 |
| larsks | That seems to work. I almost suspect that their length check is off. But given that, I'll go change https://review.openstack.org/#/c/128985/ | 17:42 |
| *** ccrouch has quit IRC | 17:43 | |
| *** ccrouch has joined #tripleo | 17:43 | |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: renamed nova-controller -> nova-ctr-base https://review.openstack.org/128985 | 17:44 |
| larsks | sdake: that is the last -1; the one on https://review.openstack.org/#/c/128968/ it's not clear to me that there's actually anything there to fix. | 17:45 |
| *** athomas has quit IRC | 17:46 | |
| sdake | daneyon your all set | 17:46 |
| sdake | larsks ok I'll +2 an we can get it merged and work from there | 17:47 |
| sdake | ya off by 1 ftw :) | 17:47 |
| daneyon | sdake: great. would you like me to push my neutron-base and neutron-server images to kollaglue and update my patch accordingly? | 17:47 |
| sdake | they probably count the / | 17:48 |
| sdake | daneyon ya go ahead and push em | 17:48 |
| openstackgerrit | A change was merged to stackforge/kolla: have start.sh source in common configuration script https://review.openstack.org/128968 | 17:48 |
| sdake | sudo ./build --release --push | 17:48 |
| openstackgerrit | A change was merged to stackforge/kolla: Nova has an undeclared dependency on python-cinderclient https://review.openstack.org/128969 | 17:48 |
| daneyon | sdake: will do | 17:48 |
| sdake | daneyon not sure we have enough time to sort out neutron integration before release | 17:48 |
| sdake | can SUID processes run in containers? | 17:49 |
| sdake | I am getting a permission error on something that should run using rootwrap | 17:49 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: renamed nova-controller -> nova-ctr-base https://review.openstack.org/128985 | 17:49 |
| openstackgerrit | A change was merged to stackforge/kolla: introduce conductor to common configuration https://review.openstack.org/128970 | 17:50 |
| openstackgerrit | A change was merged to stackforge/kolla: introduce scheduler to common configuration https://review.openstack.org/128971 | 17:50 |
| sdake | cool all merged | 17:51 |
| sdake | I'll do a build for kolla now and push it | 17:51 |
| larsks | sdake: https://review.openstack.org/#/c/128985/ doesn't seem to have merged yet. | 17:53 |
| sdake | maybe it needs a rebase? | 17:53 |
| *** wuhg has quit IRC | 17:53 | |
| larsks | Nope, gerrit would flag that. | 17:53 |
| sdake | it takes a moment for gerrit to flag - v isn't +2 yet | 17:54 |
| larsks | Try adding another +2+1 and see what happens, maybe? | 17:54 |
| sdake | the gate is still running I suspect | 17:54 |
| larsks | Oh, right, waiting on jenkins. | 17:54 |
| larsks | Yup. | 17:54 |
| sdake | although the gate should be immediate | 17:54 |
| sdake | since we hae very little gating | 17:54 |
| sdake | you code reviewed your own change :) | 17:55 |
| sdake | naughty | 17:55 |
| sdake | jenkins takes forever to sort out rebases | 17:56 |
| sdake | i'll bbi 15 | 17:56 |
| *** cody-somerville has quit IRC | 17:57 | |
| openstackgerrit | A change was merged to stackforge/kolla: renamed nova-controller -> nova-ctr-base https://review.openstack.org/128985 | 17:57 |
| daneyon | sdake: should i use any specific tag for the neutron images or just leave it as CID? | 18:04 |
| *** marun has quit IRC | 18:05 | |
| larsks | daneyon: Right now you are welcome to push as :latest, although at some point (I hope) that will only be available via an automated workflow. | 18:07 |
| *** marun has joined #tripleo | 18:07 | |
| *** noslzzp has quit IRC | 18:07 | |
| daneyon | larsks: thx. | 18:07 |
| *** akrivoka has joined #tripleo | 18:07 | |
| *** pradeep1 has joined #tripleo | 18:08 | |
| *** pradeep has quit IRC | 18:08 | |
| *** dshulyak_ has quit IRC | 18:08 | |
| *** akrivoka has quit IRC | 18:14 | |
| daneyon | I'm doing a "docker push kollaglue/fedora-rdo-neutron-base" and I still receive "Access Denied: Not allowed to create Repo at given location" | 18:15 |
| sdake | larsks +1 | 18:19 |
| sdake | daneyon are you using sudo ./build --push --release? | 18:19 |
| larsks | danehans was not a member of kollaglue. | 18:20 |
| larsks | daneyon: I just fixed that. Try now. | 18:20 |
| sdake | wierd I added it a minute ago | 18:20 |
| sdake | maybe buggy docker.io | 18:20 |
| larsks | There is a "save" button that I almost missed. | 18:20 |
| sdake | it put your mug shot up and everything | 18:20 |
| sdake | oh maybe that was the problem | 18:20 |
| larsks | But you can see it here now: https://hub.docker.com/account/organizations/kollaglue/ | 18:21 |
| sdake | apologies daneyon | 18:21 |
| sdake | rebuilding nova-controller now | 18:22 |
| sdake | so we can have a joint debug session if possible :) | 18:22 |
| *** pelix has quit IRC | 18:25 | |
| *** noslzzp has joined #tripleo | 18:28 | |
| *** noslzzp has quit IRC | 18:28 | |
| daneyon | pushing to kollaglue works now. thx! | 18:31 |
| *** ramishra has joined #tripleo | 18:35 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Remove first-boot.d support https://review.openstack.org/126836 | 18:37 |
| *** spzala has quit IRC | 18:37 | |
| *** jdob has quit IRC | 18:38 | |
| *** jdob has joined #tripleo | 18:38 | |
| *** ramishra has quit IRC | 18:40 | |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Make shell scripts executible https://review.openstack.org/129017 | 18:40 |
| sdake | current images are busted larsks :( | 18:40 |
| larsks | Need details sdake :( | 18:41 |
| sdake | see last review | 18:41 |
| sdake | that should get em booting | 18:41 |
| larsks | Looks good. | 18:42 |
| sdake | I don't know why docker doesn't just fix that | 18:42 |
| sdake | seems easy enough | 18:42 |
| sdake | or atleast spit out a warning | 18:43 |
| larsks | Hard to spit out a warning. You may have "CMD /does/not/exist", and expect that to be provided by images based on yours. | 18:43 |
| sdake | I see | 18:43 |
| *** saurabhs has joined #tripleo | 18:43 | |
| sdake | well that makes sense then | 18:43 |
| sdake | out of pain meds | 18:44 |
| sdake | fkn dentists and their tight ass pain med practicies | 18:44 |
| sdake | 3 crowns = suck | 18:44 |
| sdake | I havet o drive over there if I want more - not sure if I'm motivated enough | 18:45 |
| sdake | guess we will see how lunch goes | 18:45 |
| larsks | Good luck. | 18:45 |
| openstackgerrit | A change was merged to stackforge/kolla: Make shell scripts executible https://review.openstack.org/129017 | 18:45 |
| sdake | does crux get installed in fedora-rdo-base? | 18:46 |
| larsks | Yes. | 18:47 |
| larsks | https://github.com/stackforge/kolla/blob/master/docker/fedora-rdo-base/Dockerfile#L14 | 18:47 |
| sdake | who sorted out that dependency lsit? | 18:47 |
| *** jdob has quit IRC | 18:49 | |
| *** jdob has joined #tripleo | 18:49 | |
| *** derekh has joined #tripleo | 18:50 | |
| larsks | https://github.com/stackforge/kolla/blame/master/docker/fedora-rdo-base/Dockerfile#L14 | 18:50 |
| larsks | "sorted out" in what sense? | 18:51 |
| *** bnemec has quit IRC | 18:52 | |
| *** bnemec has joined #tripleo | 18:53 | |
| *** ramishra has joined #tripleo | 18:53 | |
| *** pradeep has joined #tripleo | 18:55 | |
| *** pradeep1 has quit IRC | 18:55 | |
| *** ramishra has quit IRC | 18:58 | |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Wrong shell script specified in start.sh https://review.openstack.org/129019 | 18:58 |
| *** rainya has quit IRC | 18:58 | |
| *** jistr has quit IRC | 18:59 | |
| sdake | i sapmmed the logs and docker seems to keep the whole thing | 19:00 |
| sdake | and doesn't crash on my end | 19:00 |
| sdake | might oom at some point tho | 19:00 |
| larsks | I guess you could spam the logs and watch disk usage on /var/lib/docker. Possibly it's keeping it on disk and not in memory. | 19:00 |
| sdake | yup possible | 19:01 |
| *** spzala has joined #tripleo | 19:01 | |
| *** boris-42 has quit IRC | 19:02 | |
| *** rainya has joined #tripleo | 19:04 | |
| *** boris-42 has joined #tripleo | 19:04 | |
| sdake | /opt/kolla/config-nova-controller.sh: line 14: NOVA_DB_PASSWORD: command not found | 19:04 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: correct missing backslash in config-nova-controller.sh https://review.openstack.org/129022 | 19:05 |
| *** pradeep1 has joined #tripleo | 19:06 | |
| sdake | that patch is in merge conflict whichever that means :) | 19:06 |
| larsks | means it needs a rebase on master. just a sec. | 19:07 |
| sdake | oh right, thought you were working on master | 19:08 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: correct missing backslash in config-nova-controller.sh https://review.openstack.org/129022 | 19:08 |
| larsks | I thought so too :) Apparently I needed a | 19:09 |
| larsks | 'git pull' | 19:09 |
| *** otoolee has quit IRC | 19:09 | |
| *** pradeep has quit IRC | 19:09 | |
| *** cody-somerville has joined #tripleo | 19:09 | |
| sdake | need SSD for my minidell | 19:12 |
| sdake | these builds take forever | 19:12 |
| sdake | i have 4 raid 0 ssd on my main workstation | 19:12 |
| *** jprovazn_afk has quit IRC | 19:12 | |
| openstackgerrit | A change was merged to openstack/os-cloud-config: Add support for creating baremetal flavors https://review.openstack.org/115872 | 19:19 |
| *** otoolee has joined #tripleo | 19:22 | |
| *** mrunge has quit IRC | 19:23 | |
| openstackgerrit | Daneyon Hansen proposed a change to stackforge/kolla: Adds Neutron Server Support https://review.openstack.org/128558 | 19:29 |
| *** dprince has quit IRC | 19:32 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Migrate ceilometer elements to svc-map https://review.openstack.org/128689 | 19:39 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add ironic to enable-packages-install https://review.openstack.org/127390 | 19:39 |
| jpeeler | larsks: so are we using spaces or tabs? | 19:41 |
| larsks | sdake wants to use "bashate", which seems to like spaces. | 19:42 |
| larsks | I prefer tabs, but not enough to whinge it. So spaces, I guess. | 19:42 |
| jpeeler | i'm going to approve this change, but it has tabs | 19:43 |
| larsks | Go ahead. At some point sdake will fix the gate and we will have a massive fix-all-the-bashate-warnings day. | 19:43 |
| openstackgerrit | A change was merged to stackforge/kolla: correct missing backslash in config-nova-controller.sh https://review.openstack.org/129022 | 19:44 |
| jpeeler | larsks: do you know if i delete a pod and docker is still downloading the image, if it can be resumed upon relaunching the pod? | 19:45 |
| larsks | jpeeler: I am having trouble parsing that question. Here is my best answer: the "docker pull" operation is done by the daemon, and will generally continue even if you cancel the client that initiated the request. | 19:46 |
| *** cody-somerville has quit IRC | 19:46 | |
| larsks | So a "docker pull" will probably continue even if you stop the pod that triggered it. | 19:46 |
| larsks | Does that help? | 19:47 |
| sdake | I like tabs as well | 19:47 |
| sdake | openstack is anti tab for some reason :( | 19:47 |
| jpeeler | yes, i figured that was the case | 19:49 |
| openstackgerrit | Daneyon Hansen proposed a change to stackforge/kolla: Adds Neutron Server Support https://review.openstack.org/128558 | 19:50 |
| *** ramishra has joined #tripleo | 19:54 | |
| *** derekh has quit IRC | 19:54 | |
| *** cody-somerville has joined #tripleo | 19:58 | |
| *** cody-somerville has joined #tripleo | 19:58 | |
| *** ramishra has quit IRC | 19:59 | |
| *** gfidente has quit IRC | 20:00 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Force empty $TMPDIR inside the chroot https://review.openstack.org/128696 | 20:04 |
| *** noslzzp has joined #tripleo | 20:05 | |
| SpamapS | sdake: https://etherpad.openstack.org/p/kilo-tripleo-summit-topics wants Kolla topics. :) | 20:06 |
| *** daneyon has quit IRC | 20:11 | |
| *** pensu has quit IRC | 20:12 | |
| *** daneyon has joined #tripleo | 20:12 | |
| *** pradeep has joined #tripleo | 20:14 | |
| *** pradeep1 has quit IRC | 20:14 | |
| *** jehb has quit IRC | 20:23 | |
| greghaynes | SpamapS: I notice there is no name by the ansible for updates session... | 20:24 |
| greghaynes | hint hint hint | 20:24 |
| SpamapS | greghaynes: Do we need names? | 20:25 |
| SpamapS | greghaynes: it sprang forth from the mountain like a rainbow of truth | 20:25 |
| greghaynes | *shrug* all the otheres have names. Last time there had to be a person who actually 'proposed' each one | 20:25 |
| greghaynes | Ah, the session of immaculate conception | 20:25 |
| * SpamapS credits the flying spaghetti monster | 20:26 | |
| slagle | that guy is busy | 20:34 |
| greghaynes | SpamapS: re: https://review.openstack.org/#/c/113358/ - wha? | 20:35 |
| greghaynes | does code exist to do that? | 20:36 |
| *** noslzzp has quit IRC | 20:37 | |
| *** noslzzp has joined #tripleo | 20:40 | |
| openstackgerrit | Steve Baker proposed a change to openstack/os-collect-config: Collector which does a simple HTTP GET https://review.openstack.org/128479 | 20:40 |
| sdake | SpamapS I won't be at kilo summit, but larsks will be | 20:41 |
| *** ChuckC has quit IRC | 20:42 | |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: fix typo in crudini command https://review.openstack.org/129042 | 20:43 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: add /check.sh script to nova-api image https://review.openstack.org/129043 | 20:44 |
| *** cody-somerville has quit IRC | 20:46 | |
| *** noslzzp has quit IRC | 20:46 | |
| sdake | larsks can you ack this https://review.openstack.org/#/c/129019/ | 20:46 |
| openstackgerrit | A change was merged to stackforge/kolla: Wrong shell script specified in start.sh https://review.openstack.org/129019 | 20:47 |
| *** noslzzp has joined #tripleo | 20:52 | |
| openstackgerrit | A change was merged to stackforge/kolla: fix typo in crudini command https://review.openstack.org/129042 | 20:53 |
| openstackgerrit | A change was merged to stackforge/kolla: add /check.sh script to nova-api image https://review.openstack.org/129043 | 20:53 |
| *** rainya has quit IRC | 20:54 | |
| *** ramishra has joined #tripleo | 20:55 | |
| *** ramishra has quit IRC | 20:59 | |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1381408 | 21:00 |
| uvirtbot | Launchpad bug 1381408 in diskimage-builder ""visudo" always got error when building overcloud image" [Undecided,New] | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| *** jdob has quit IRC | 21:01 | |
| *** rainya has joined #tripleo | 21:03 | |
| openstackgerrit | greghaynes proposed a change to openstack/tripleo-image-elements: Add SSL PKI properties for keystone https://review.openstack.org/129049 | 21:05 |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-incubator: Check for firewalld before starting seed https://review.openstack.org/129051 | 21:09 |
| openstackgerrit | greghaynes proposed a change to openstack/tripleo-heat-templates: Add SSL PKI properties for keystone https://review.openstack.org/129052 | 21:11 |
| *** ChuckC has joined #tripleo | 21:13 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Consolidate lsb_release source-repositories https://review.openstack.org/128434 | 21:14 |
| *** noslzzp has quit IRC | 21:16 | |
| sdake | nova-api working !! | 21:17 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Run nova-api as privileged so it can muck with iptables https://review.openstack.org/129054 | 21:19 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: fix sql typo in nova-conductor start script https://review.openstack.org/129056 | 21:22 |
| *** andreaf has joined #tripleo | 21:23 | |
| *** rainya has quit IRC | 21:24 | |
| *** cody-somerville has joined #tripleo | 21:25 | |
| *** cody-somerville has joined #tripleo | 21:25 | |
| *** rainya has joined #tripleo | 21:25 | |
| tchaypo | StevenK: it came to me in a dream | 21:25 |
| tchaypo | that your register-nodes work is in os-cloud-config | 21:26 |
| tchaypo | I'm hoping my dream is accurate | 21:26 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: configure nova database connection https://review.openstack.org/129059 | 21:26 |
| openstackgerrit | A change was merged to stackforge/kolla: fix sql typo in nova-conductor start script https://review.openstack.org/129056 | 21:27 |
| openstackgerrit | greghaynes proposed a change to openstack/os-cloud-config: Make keystone_pki less keystone specific https://review.openstack.org/129062 | 21:28 |
| greghaynes | StevenK: ^ Your favorite patch! | 21:28 |
| *** rainya has quit IRC | 21:31 | |
| *** rainya has joined #tripleo | 21:31 | |
| *** cody-somerville has quit IRC | 21:35 | |
| openstackgerrit | Clint 'SpamapS' Byrum proposed a change to openstack/tripleo-image-elements: Ensure password is set correctly on existing users https://review.openstack.org/92207 | 21:39 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed a change to openstack/tripleo-image-elements: Add an element to restore ssh keys from /mnt/state https://review.openstack.org/114360 | 21:42 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Migrates from map-services to svc-map https://review.openstack.org/129065 | 21:45 |
| *** julim has quit IRC | 21:47 | |
| *** spzala has quit IRC | 21:47 | |
| *** spzala has joined #tripleo | 21:48 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Migrate from map-services to svc-map https://review.openstack.org/129066 | 21:49 |
| *** ramishra has joined #tripleo | 21:55 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Allow for multiple image outputs from raw source https://review.openstack.org/125475 | 21:56 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Migrates corosync element to svc-map https://review.openstack.org/129072 | 21:56 |
| *** cody-somerville has joined #tripleo | 22:00 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Migrate geard to svc-map https://review.openstack.org/129073 | 22:00 |
| *** cody-somerville has quit IRC | 22:00 | |
| *** cody-somerville has joined #tripleo | 22:00 | |
| *** ramishra has quit IRC | 22:00 | |
| sdake | nova kollaglue/fedora-rdo-nova-conductor,kollaglue/fedora-rdo-nova-scheduler,kollaglue/fedora-rdo-nova-api 10.0.0.4/ name=nova-controller Running | 22:00 |
| sdake | larsks ^^ | 22:00 |
| sdake | RUNNING ftw | 22:00 |
| sdake | now for compute | 22:01 |
| *** rhallisey has quit IRC | 22:02 | |
| *** radez is now known as radez_g0n3 | 22:04 | |
| sdake | jpeeler radez_g0ne can you g uys take a quick look at https://review.openstack.org/#/c/129054/ | 22:04 |
| sdake | or if larsks is around as well | 22:04 |
| *** rainya has quit IRC | 22:05 | |
| *** pradeep1 has joined #tripleo | 22:05 | |
| *** rainya has joined #tripleo | 22:06 | |
| *** pradeep has quit IRC | 22:07 | |
| *** rwsu has quit IRC | 22:07 | |
| openstackgerrit | Clint 'SpamapS' Byrum proposed a change to openstack/tripleo-heat-templates: Create RabbitMQ users on controller https://review.openstack.org/92204 | 22:08 |
| larsks | sdake: not actually running, because it needs at least https://review.openstack.org/#/c/129059/ | 22:11 |
| larsks | I will take a look at that review... | 22:12 |
| *** funzo has quit IRC | 22:15 | |
| *** funzo has joined #tripleo | 22:15 | |
| *** pradk has quit IRC | 22:18 | |
| *** sdake has quit IRC | 22:21 | |
| *** larstobi has quit IRC | 22:23 | |
| *** pcrews has quit IRC | 22:24 | |
| *** larstobi has joined #tripleo | 22:29 | |
| *** pcrews has joined #tripleo | 22:30 | |
| *** rhallisey has joined #tripleo | 22:32 | |
| openstackgerrit | Ian Wienand proposed a change to openstack/diskimage-builder: Standarise tracing for scripts https://review.openstack.org/119023 | 22:33 |
| SpamapS | http://logs.openstack.org/60/114360/9/check-tripleo/check-tripleo-ironic-overcloud-f20-nonha/d5638f0/console.html#_2014-10-16_22_20_56_974 | 22:37 |
| SpamapS | has anybody seen this fail? | 22:37 |
| *** sdake has joined #tripleo | 22:37 | |
| *** sdake has quit IRC | 22:37 | |
| *** sdake has joined #tripleo | 22:37 | |
| greghaynes | SpamapS: hrm, its annoying we dont have haproxy logs | 22:40 |
| sdake | larsks oh yes, I had that locally, I guess I missed your review | 22:42 |
| sdake | can you set max_retries to something like 5, so the container exits if the db isn't available | 22:43 |
| sdake | otherwise the container will spin forever and the db may never come up | 22:43 |
| sdake | which would make diagnostics a bit difficult I think | 22:43 |
| sdake | \o/ yay for controller | 22:48 |
| sdake | now to beat compute into submission | 22:48 |
| SpamapS | greghaynes: very | 22:49 |
| greghaynes | SpamapS: that does really look like a haproxy doesnt have a backend error | 22:49 |
| greghaynes | SpamapS: since theres not other errors in any of the logs | 22:49 |
| SpamapS | I like our little mini-signal implementation with nc | 22:52 |
| SpamapS | had forgotten somebody got that working. :-P | 22:52 |
| greghaynes | yea, thats impressive. My big gripe is that it doesnt respect ctrl+c | 22:52 |
| *** ramishra has joined #tripleo | 22:56 | |
| greghaynes | SpamapS: So there is a really interesting issue ive run into with our update playbook. We have a lot of sets of two tasks that both register a variable foo, one if helion defined, one if !defined | 22:58 |
| SpamapS | greghaynes: also notice in that error series, os-collect-config log ends 7 minutes earlier, with an error... that is weird | 22:58 |
| greghaynes | turns out whatever runs last overwites the value of the variable | 22:58 |
| SpamapS | greghaynes: yeah a lot of that was because we used to have two distinct sets of services | 22:59 |
| SpamapS | well we do still actually | 22:59 |
| greghaynes | yep, so for registering variables that doesnt work so much | 22:59 |
| SpamapS | I think the end goal is to actually not have a 'helion' mode, but a helion branch that has roles that override this stuff. | 22:59 |
| greghaynes | yea, If only include had a when: property... | 23:00 |
| SpamapS | it does sort of | 23:00 |
| SpamapS | use roles :) | 23:00 |
| SpamapS | I'm learning about them now and I have a restructuring plan | 23:00 |
| SpamapS | but I want to have CI first | 23:00 |
| greghaynes | haha, ok, I should also read up on that | 23:00 |
| *** ramishra has quit IRC | 23:01 | |
| SpamapS | So the idea is to have the role be 'tripleo' and then helion can override that with 'helion' and provide a helion role that diverges. | 23:01 |
| SpamapS | I think | 23:01 |
| SpamapS | anyway... brain is pretty much cooked today | 23:02 |
| SpamapS | and still no CI.. I think I need to give up trying to deploy a cloud on these blades and just setup a bunch of identical jenkins slaves. :-P | 23:02 |
| SpamapS | too hard when it's not my network | 23:02 |
| greghaynes | yea, sounds like what they call environments in salt | 23:03 |
| SpamapS | it's more contained than that | 23:04 |
| SpamapS | greghaynes: actually let me back up. The service list thing is a little wonky IMO. I think we might want to define that ordering in the elements. | 23:05 |
| SpamapS | greghaynes: what we want to use roles for is to encapsulate the steps each class of servers need to go through. That way Helion with it's special groupings can just write a new workflow that fits in vsa and whatever else where it needs to be, but all of the "magic" is happening in roles underneath. | 23:07 |
| SpamapS | greghaynes: so the idea is, let them patch the high level logical workflow, instead of having all this low level stuff in update_cloud.yml | 23:08 |
| *** yamahata has joined #tripleo | 23:10 | |
| *** ChuckC___ has joined #tripleo | 23:13 | |
| *** marun has quit IRC | 23:26 | |
| *** ChuckC___ has quit IRC | 23:34 | |
| *** morazi has quit IRC | 23:35 | |
| *** marun has joined #tripleo | 23:37 | |
| *** cody-somerville has quit IRC | 23:39 | |
| *** saurabhs has quit IRC | 23:42 | |
| *** juice has quit IRC | 23:46 | |
| *** yamahata has quit IRC | 23:53 | |
| *** juice has joined #tripleo | 23:53 | |
| *** ramishra has joined #tripleo | 23:57 | |
| tchaypo | reboot undercloud control node, come up without /mnt/state, yay | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!