| lifeless | SpamapS: is https://trello.com/c/M4lQQmb8/40-images-move-persistent-state-locations-to-mnt-state-same-path-unconditionally-mnt-always-exists done ? | 00:40 |
|---|---|---|
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-image-elements: Replace /var/lib/mysql path in the upstart script https://review.openstack.org/67742 | 00:52 |
| mordred | wow. that's such a small url, how do you know what it points to? | 01:03 |
| StevenK | lifeless: remote-operations is a string, anything is true, nothing is false. Perhaps I should RTFM about booleans and how they get set | 01:05 |
| lifeless | mordred: you don't, you have to guess | 01:05 |
| lifeless | StevenK: its fine as is, but perhaps documenting that (non-empty is true) is a good idea :) | 01:05 |
| StevenK | lifeless: +## A non-empty string means true, the default is '', which means false. | 01:06 |
| lifeless | yeah | 01:06 |
| lifeless | StevenK: or is that already there and I failed at reading? | 01:07 |
| lifeless | cause, thats my super power | 01:07 |
| StevenK | It wasn't. | 01:07 |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Add a new remote ops JSON value to boot-seed-vm https://review.openstack.org/67366 | 01:07 |
| StevenK | lifeless: I'll be disappearing into a place where time has no meaning -- the waiting room at the doctors; need a new script for asthma meds | 01:10 |
| lifeless | StevenK: may the breath be with you | 01:10 |
| lifeless | StevenK: hey on mirrors - might want to document a quick mini-howto (or a link to one) for mirroring just a given arch-release | 01:11 |
| StevenK | lifeless: To use the Soyuz nomenclature, a specific DAS? | 01:23 |
| *** nosnos has joined #tripleo | 01:24 | |
| lifeless | yeah: saucy amd64 main+universe+updates+security | 01:34 |
| StevenK | lifeless: My mirror script for the ubuntu element doesn't replace security, but everything tends to hit updates anyway, so it's probably fine | 01:56 |
| lifeless | StevenK: close enough then :). So yeah, one of those would be nice. | 01:57 |
| lifeless | StevenK: a related but different thing would be an element to install one of those on a machine so we can easily deploy a mirror into an environment | 01:57 |
| mordred | StevenK, lifeless: I could imagine that people inside of a terrible corporate firewall might want one inside of their firewall | 01:58 |
| StevenK | lifeless: I could be convinced to write a Debian/Ubuntu mirror element | 01:58 |
| StevenK | mordred: Yes, why do we block every high port :-( | 01:59 |
| lifeless | StevenK: every port :P | 01:59 |
| lifeless | StevenK: not my job to convince you :) | 01:59 |
| StevenK | Not every port, ssh is open | 01:59 |
| lifeless | StevenK: oh, ok. Nearly every :) | 01:59 |
| StevenK | lifeless: It will take very little convincing. :-) | 01:59 |
| lifeless | there is a socks proxy around | 01:59 |
| lifeless | StevenK: I'll let you exercise your vestigial but growing decision making powers :) | 02:00 |
| StevenK | Just considering the wisdom of a element that will pull down 30+ GiB | 02:00 |
| lifeless | StevenK: it would pull it down once deployed though, right ? | 02:00 |
| lifeless | StevenK: not at build time... | 02:00 |
| StevenK | I'd hope so | 02:00 |
| lifeless | :) | 02:00 |
| *** spzala has joined #tripleo | 02:06 | |
| lifeless | erm, something just took the ci-overcloud controller off the network | 02:29 |
| lifeless | wtf | 02:29 |
| lifeless | ddd | 02:31 |
| lifeless | asdasdasd | 02:31 |
| lifeless | SpamapS: btw I think you're wrong about the occ cache being needed to fix booting | 02:56 |
| lifeless | SpamapS: we have a lower level problemw which if fixed will fix occ too | 02:56 |
| lifeless | [ 17.776872] EXT4-fs (sda3): re-mounted. Opts: (null) | 02:56 |
| lifeless | Cloud-init v. 0.7.3 running 'init-local' at Mon, 20 Jan 2014 02:55:36 +0000. Up | 02:56 |
| lifeless | 22.99 seconds. | 02:56 |
| lifeless | cloud-init-nonet[24.68]: waiting 10 seconds for network device | 02:56 |
| lifeless | cloud-init-nonet[34.68]: waiting 120 seconds for network device | 02:56 |
| *** AaronGr_Zzz is now known as AaronGr | 03:03 | |
| *** AaronGr is now known as AaronGr_Zzz | 03:11 | |
| *** vkozhukalov has joined #tripleo | 03:28 | |
| *** ramishra has joined #tripleo | 03:33 | |
| *** rbrady has quit IRC | 03:43 | |
| mordred | lifeless: you down with occ? | 03:43 |
| StevenK | lifeless: So TARGET_ROOT is not set in pre-install.d | 03:48 |
| StevenK | Hm | 04:02 |
| StevenK | Having qemu-nbd running with a mounted FS over it while the file underneath qemu-nbd changed was a BAD idea. | 04:03 |
| openstackgerrit | Steve Kowalik proposed a change to openstack/diskimage-builder: Allow use of mirrors when building Ubuntu images https://review.openstack.org/66984 | 04:07 |
| lifeless | StevenK: you might like https://bugs.launchpad.net/diskimage-builder/+bug/1224669 | 04:12 |
| StevenK | lifeless: Looks good, just not sure where to start pulling threads from | 04:15 |
| *** akuznetsov has joined #tripleo | 04:16 | |
| lifeless | StevenK: as Ng says, poke around at the way the ramdisk-image-create stuff works | 04:16 |
| *** coolsvap has joined #tripleo | 04:19 | |
| *** spzala has quit IRC | 04:22 | |
| SpamapS | lifeless: hm | 05:14 |
| SpamapS | lifeless: dhcp-all-interfaces not working? | 05:14 |
| SpamapS | lifeless: when I looked into it the problem was that it _was_ working, but we still had to move the ip from eth0 -> br-ctlplane .. which would not happen until we had identified eth0 as the bridge.. which would not happen until we fetch the heat metadata. | 05:15 |
| *** rbrady has joined #tripleo | 05:22 | |
| lifeless | SpamapS: everything is dhcp right? | 05:23 |
| lifeless | SpamapS: and br-ctlplane is a bridge, right? | 05:23 |
| lifeless | SpamapS: so it will dhcp successfully, and we dont need to move the ip | 05:23 |
| lifeless | SpamapS: but I bet cloud-init is kicking in before ovs starts, so we end up with the ip address on both devices and things horked | 05:24 |
| lifeless | StevenK: C is now esconced for pre-bed tv, so if you want more detailed pointers at the kernel/ramdisk export stuff I can do that no | 05:24 |
| lifeless | now | 05:25 |
| StevenK | lifeless: Excellent. | 05:25 |
| StevenK | I've just finished fixing some stupidity I had with dropping disk-image-get-kernel in the same branch. | 05:26 |
| StevenK | lifeless: Here, or some other plan? | 05:30 |
| lifeless | hmm ? | 05:30 |
| * lifeless is now confused | 05:30 | |
| StevenK | lifeless: More detailed pointers conversation | 05:31 |
| lifeless | ah | 05:32 |
| lifeless | so here yes | 05:32 |
| lifeless | firstly, note that ramdisk-image-create is a symlink | 05:32 |
| StevenK | Yeah, I tripped over that already | 05:33 |
| StevenK | "Have I read this script before? It seems very famil... Oh, it's a symlink" | 05:33 |
| lifeless | oh yeah, backwards compat - we'll need to keep the get-kernel tool for a while... | 05:33 |
| SpamapS | lifeless: the thing I see on my local vms is not quite that. cloud-init is largely a no-op if it can't reach ec2 metadata. | 05:33 |
| lifeless | SpamapS: it ifups super early | 05:34 |
| lifeless | SpamapS: *that* is sufficient when both eth0 and bt-ctlplane are configured to dhcp, and ovs isn't running yet. | 05:34 |
| *** noslzzp has quit IRC | 05:34 | |
| lifeless | SpamapS: I'll let you chew on this for a bit | 05:34 |
| lifeless | SpamapS: I may be wrong, of course. | 05:35 |
| SpamapS | are they both dhcp? | 05:35 |
| lifeless | SpamapS: check your /e/n/i | 05:35 |
| SpamapS | I can't check because -rw-r--r-- 1 root root 0 Jan 16 17:59 seed.qcow2 | 05:35 |
| lifeless | ahahahahahahahahahahaha | 05:35 |
| StevenK | His devtest is broken | 05:36 |
| SpamapS | has the fix for the rsync bug landed? | 05:36 |
| StevenK | SpamapS: You should approve my patch to fix that, then | 05:36 |
| lifeless | there's a bunch of stuff in incubator/tie/tht that needs to land | 05:37 |
| lifeless | asap | 05:37 |
| lifeless | StevenK: btw, one of the most important things for you to do is to start reviewing; our culture is -very- reviewer biaswed | 05:37 |
| *** michchap has quit IRC | 05:37 | |
| lifeless | SpamapS: anyhow, yes, its all dhcp | 05:37 |
| *** michchap has joined #tripleo | 05:37 | |
| lifeless | SpamapS: eth0 gets dhcp from virbr0; br-ctlplane is configured for dhcp by dhcp-all-interfaces | 05:37 |
| lifeless | SpamapS: which configures *rather too much* as dhcp, but thats a different discussion | 05:38 |
| lifeless | StevenK: so, I had to dig around to find where the ramdisk stuff is done | 05:38 |
| lifeless | StevenK: git log -p FTW | 05:38 |
| lifeless | elements/ramdisk/cleanup.d/99-extract-ramdisk-files | 05:38 |
| lifeless | StevenK: cleanup.d runs while we have the thing in a regular file tree, outside the chroot | 05:38 |
| SpamapS | StevenK: https://review.openstack.org/#/c/67366/6 .. I don't see anything being added to json. ??? | 05:40 |
| StevenK | lifeless: Hm, did I miss that, and we should write out remote-operations? | 05:43 |
| lifeless | StevenK: this is the remote codepath only | 05:43 |
| lifeless | StevenK: it hooks inwith the test broker in tripleo-ci | 05:44 |
| lifeless | StevenK: it is probably possible to use the presence of env number to decide whether to do remote ops | 05:44 |
| StevenK | lifeless: After you said last week it was impossible. :-P | 05:45 |
| lifeless | StevenK: I didn't know about the env num variable last week | 05:45 |
| StevenK | Oh, I went from command line argument to JSON | 05:46 |
| *** akuznetsov has quit IRC | 05:46 | |
| StevenK | lifeless: Flipping back to the other conversation, I can see 99-extract-ramdisk-files depends on elements/ramdisk/post-install.d/99-build-ramdisk | 05:51 |
| *** nosnos has quit IRC | 05:52 | |
| *** nosnos_ has joined #tripleo | 05:52 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Change communication with remote host https://review.openstack.org/67689 | 05:53 |
| lifeless | StevenK: sure, but the principle doesn't | 05:54 |
| lifeless | StevenK: just copy the get-kernel-ramdisk logic into an element we can use when when we're not doing vm builds | 05:55 |
| lifeless | e.g. non-vm or akiari or some such name | 05:55 |
| *** rdopieralski has joined #tripleo | 05:58 | |
| StevenK | lifeless: Where did you get akiari from? | 06:00 |
| *** ramishra has quit IRC | 06:01 | |
| StevenK | SpamapS: So, for the simple case of "Let's run devtest, you don't want it or need it. For tripleo-ci, we'll build a JSON file there that has it. | 06:06 |
| lifeless | StevenK: oh, because nova baremetal takes ami + aki + ari | 06:08 |
| lifeless | machine image- one partition, kernel, ramdisk | 06:08 |
| *** e0ne has joined #tripleo | 06:12 | |
| *** akuznetsov has joined #tripleo | 06:13 | |
| *** akuznetsov has quit IRC | 06:13 | |
| SpamapS | StevenK: ok I think I understand. Approved. | 06:13 |
| *** akuznetsov has joined #tripleo | 06:13 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Add a new remote ops JSON value to boot-seed-vm https://review.openstack.org/67366 | 06:14 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Use BASH_SOURCE[0] to find scripts directory https://review.openstack.org/67687 | 06:14 |
| *** e0ne has quit IRC | 06:15 | |
| *** ramishra has joined #tripleo | 06:18 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Change communication with remote host https://review.openstack.org/67689 | 06:20 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Upgrade setuptools to 1.10 or more in a newly created virtualenv https://review.openstack.org/67688 | 06:20 |
| *** tzumainn has quit IRC | 06:26 | |
| *** marios has quit IRC | 06:28 | |
| *** marios has joined #tripleo | 06:29 | |
| *** vkozhukalov has quit IRC | 06:44 | |
| *** nosnos_ has quit IRC | 06:52 | |
| *** nosnos has joined #tripleo | 06:53 | |
| *** pblaho has joined #tripleo | 06:57 | |
| *** rlandy has joined #tripleo | 07:03 | |
| coolsvap | lifeless: can I create a dev environment for tripleo with devstack? | 07:08 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Deployment design form https://review.openstack.org/67155 | 07:28 |
| lifeless | coolsvap: not really; devstack installs a cloud. tripleo installs clouds... | 07:28 |
| lifeless | coolsvap: you can certainly use devstack when hacking on API services like tuskar ironic nova etc, which tripleo heavily depends on | 07:28 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Deployment design form https://review.openstack.org/67155 | 07:29 |
| *** jcoufal has joined #tripleo | 07:29 | |
| *** mrunge has joined #tripleo | 07:30 | |
| coolsvap | lifeless: okay, how to work on tripleo development, i kindof tried disk-image-builder by downloading from source | 07:31 |
| coolsvap | will it be same for other projects? | 07:31 |
| lifeless | coolsvap: start with tripleo-incubator | 07:32 |
| lifeless | coolsvap: its the entrypoint and user docs combined at the moment | 07:32 |
| coolsvap | lifeless: thanks cloning it | 07:33 |
| *** d0ugal has joined #tripleo | 07:37 | |
| *** d0ugal has joined #tripleo | 07:37 | |
| *** ifarkas has joined #tripleo | 07:43 | |
| *** jprovazn has joined #tripleo | 07:47 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Deployment design form https://review.openstack.org/67155 | 07:58 |
| rlandy | jprovazn: ping ... morning | 08:01 |
| *** e0ne has joined #tripleo | 08:01 | |
| *** jistr has joined #tripleo | 08:06 | |
| coolsvap | lifeless: I can start with http://docs.openstack.org/developer/tripleo-incubator/devtest.html, right? | 08:06 |
| lifeless | yeah | 08:08 |
| jprovazn | rlandy: morning | 08:10 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Install environment and templates for heat-engine https://review.openstack.org/67309 | 08:10 |
| *** bauzas has joined #tripleo | 08:12 | |
| *** ramishra has quit IRC | 08:14 | |
| rlandy | jprovazn: is this still valid https://bugs.launchpad.net/tripleo/+bug/1259546? Considering tripleo-heat-templates is included here: https://github.com/openstack/tripleo-image-elements/blob/master/elements/tuskar/source-repository-tuskar (and elements/tuskar/install.d/100-tuskar-api) - is there more required to add the dependency? | 08:15 |
| *** jtomasek has joined #tripleo | 08:16 | |
| *** ramishra has joined #tripleo | 08:18 | |
| jprovazn | rlandy: it should be OK, though it would be better check it with ifarkas | 08:20 |
| ifarkas | rlandy, jprovazn: it seems it's already fixed. tht is cloned and set to the proper path according to as tuskar.conf | 08:23 |
| ifarkas | rlandy, jprovazn: https://github.com/openstack/tripleo-image-elements/blob/master/elements/tuskar/install.d/100-tuskar-api#L9 | 08:24 |
| GheRivero | morning TripleO! | 08:24 |
| ifarkas | rlandy, jprovazn: best way to confirm is to build an image with tuskar | 08:25 |
| ifarkas | morning GheRivero! | 08:25 |
| rlandy | ifarkas: thanks ... can try that and confirm | 08:28 |
| *** vkozhukalov has joined #tripleo | 08:28 | |
| ifarkas | rlandy, cool, thanks | 08:28 |
| *** vkozhukalov has quit IRC | 08:34 | |
| *** kpavel has joined #tripleo | 08:40 | |
| *** kpavel_ has joined #tripleo | 08:40 | |
| *** kpavel has quit IRC | 08:44 | |
| *** kpavel_ is now known as kpavel | 08:44 | |
| *** jcoufal has quit IRC | 08:49 | |
| rpodolyaka | morning all | 08:50 |
| *** vkozhukalov has joined #tripleo | 08:51 | |
| lifeless | rpodolyaka: o/ | 08:53 |
| *** rha has quit IRC | 08:56 | |
| rpodolyaka | wow, zull is a mess today :( | 08:58 |
| rpodolyaka | *zuul | 08:59 |
| *** rha has joined #tripleo | 08:59 | |
| Ng | morning | 09:00 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Deployment design form https://review.openstack.org/67155 | 09:00 |
| rpodolyaka | o/ | 09:01 |
| Ng | lifeless: I admire your faith in lxc's containment (re openstack-dev) ;) | 09:01 |
| lifeless | Ng: indeed! | 09:04 |
| lifeless | rpodolyaka: think you can poke at https://review.openstack.org/#/c/59312/ ? dan has a -1 on it | 09:04 |
| rpodolyaka | lifeless: I believe I've already left comments to patch set 27 (the one dan left comments to). Will ping him today then | 09:05 |
| rpodolyaka | lifeless: not strong on sqlite though and can refactor that part | 09:06 |
| lifeless | rpodolyaka: ah, so yeah i don't think it will get other reviewers while it has the -1 | 09:06 |
| rpodolyaka | lifeless: agree | 09:07 |
| rpodolyaka | lifeless: ok, so I will prepare a fix for the sqlite comment and ping dan when he is online | 09:07 |
| *** jcoufal has joined #tripleo | 09:12 | |
| *** derekh has joined #tripleo | 09:15 | |
| *** markmc has joined #tripleo | 09:18 | |
| *** jistr has quit IRC | 09:31 | |
| *** jistr has joined #tripleo | 09:32 | |
| coolsvap | lifeless: how many nic's i need to run the devtest.sh script? | 09:32 |
| *** athomas has joined #tripleo | 09:34 | |
| lifeless | coolsvap: its all virt, just need lots of memory | 09:36 |
| *** lsmola_ has joined #tripleo | 09:38 | |
| coolsvap | lifeless: I am getting error "Network is already in use by interface eth0", so I thought if there is any requirement for multiple NICs | 09:38 |
| lifeless | coolsvap: some more context might be useful :)_ | 09:38 |
| lifeless | coolsvap: also its very late for me, you might want to chat with whoever is around (e.g. GheRivero / Ng / ...) | 09:38 |
| * Ng waves | 09:39 | |
| coolsvap | lifeless: sure, np! | 09:39 |
| coolsvap | Ng: hey hi | 09:39 |
| coolsvap | Ng: I am trying the devtest.sh script on a VM, got the error "Network is already in use by interface eth0", any ideas? | 09:40 |
| lifeless | oh hohoho keystone you naughty thing you | 09:40 |
| lifeless | keystone tenant-list | 09:40 |
| lifeless | Unable to establish connection to http://10.10.16.175:35357/v2.0/tenants | 09:40 |
| lifeless | derekh: the six error | 09:41 |
| lifeless | derekh: what was the fix? Specifically for a *built image* with that defect | 09:41 |
| lifeless | derekh: I don't want to break the ci-overcloud, but it was built buggy | 09:42 |
| derekh | lifeless: for a alreaty built image I created a new venv without --system-site-packages | 09:42 |
| derekh | lifeless: for running clients from command line that was | 09:43 |
| lifeless | yeah | 09:43 |
| Ng | coolsvap: could you put the output of the devtest.sh run into a pastebin somewhere? | 09:44 |
| lifeless | coolsvap: note that running devtest *within* a VM is going to end in tears | 09:45 |
| Ng | slow, slow tears | 09:45 |
| lifeless | coolsvap: as the VM's it creates will be QEMU and -slow- | 09:45 |
| derekh | Good idea or bad idea ? http://paste.openstack.org/show/61545/ | 09:47 |
| coolsvap | i just wanted to get first feel of it, and i was not sure about the option | 09:48 |
| coolsvap | http://paste.openstack.org/show/61546/ | 09:48 |
| coolsvap | option "--trash-my-machine" | 09:48 |
| lifeless | coolsvap: yah, its intended to get you to stop and pause and read the docs so you can see what it does to setup the test environment | 09:49 |
| lifeless | derekh: seems likely to cause a proliferation of untested types | 09:49 |
| lifeless | derekh: and they're easy to get wrong | 09:49 |
| lifeless | derekh: where wrong == fail open | 09:49 |
| lifeless | derekh: I've pushed an update of your incubator stack | 09:50 |
| derekh | lifeless: yup, thats why I though it might be a bad idea, the reason I thought of it was to avoid having to use --raw while we add a new type | 09:50 |
| lifeless | derekh: adding a new type should be a couple hours end to end though :) | 09:50 |
| lifeless | write, grab two reviewers, fix, land, release. | 09:50 |
| derekh | lifeless: ok, fair enough | 09:51 |
| lifeless | hmm | 09:54 |
| lifeless | perhaps the admin endpoint needs to be internet accessible ? | 09:54 |
| *** derekh is now known as derekh_afk | 09:59 | |
| coolsvap | Ng: can I run devtest.sh on my desktop? what does "--trash-my-machine" indicate? | 10:01 |
| Ng | coolsvap: that we're going to install software, create VMs and change some networking. It shouldn't actually trash your machine, but we want to be very clear that this isn't 100% self-contained :) | 10:01 |
| lifeless | have a look at install-dependencies and devtest_setup, devtest_testenvironment | 10:02 |
| coolsvap | Ng: hmm hmm | 10:02 |
| lifeless | its' much less intruisive than devstack | 10:02 |
| *** rwsu has joined #tripleo | 10:03 | |
| *** vkozhukalov has quit IRC | 10:04 | |
| lifeless | derekh_afk: btw, I have nodepool building usable nodes now | 10:14 |
| lifeless | derekh_afk: and most of the patches landed | 10:14 |
| lifeless | https://review.openstack.org/#/c/67762/ needs to land for that to be the case with infra's nodepool | 10:14 |
| *** vkozhukalov has joined #tripleo | 10:16 | |
| *** max_lobur_afk is now known as max_lobur | 10:18 | |
| lifeless | derekh_afk: my test one is killing them too quickly though, probably I need to tweak the custom ssh key patch | 10:18 |
| lifeless | derekh_afk: and I've put a big braindump into https://trello.com/c/dZQl2lSJ/72-more-ci-overcloud-automation-prep-for-migrating-to-cd-overcloud | 10:19 |
| lifeless | derekh_afk: I need to make it sane and put stuff for the etherpad in the etherpad | 10:19 |
| *** e0ne_ has joined #tripleo | 10:22 | |
| *** e0ne has quit IRC | 10:26 | |
| *** ifarkas has quit IRC | 10:29 | |
| *** martyntaylor has joined #tripleo | 10:31 | |
| *** ifarkas has joined #tripleo | 10:32 | |
| *** akrivoka has joined #tripleo | 10:50 | |
| *** e0ne has joined #tripleo | 10:51 | |
| *** e0ne_ has quit IRC | 10:54 | |
| openstackgerrit | Dirk Mueller proposed a change to openstack/tripleo-heat-templates: Add license file https://review.openstack.org/67806 | 10:58 |
| *** e0ne_ has joined #tripleo | 11:07 | |
| *** e0ne has quit IRC | 11:07 | |
| *** martyntaylor has quit IRC | 11:07 | |
| *** martyntaylor1 has joined #tripleo | 11:08 | |
| *** lucasagomes has joined #tripleo | 11:09 | |
| *** jistr has quit IRC | 11:11 | |
| *** jistr has joined #tripleo | 11:14 | |
| *** martyntaylor has joined #tripleo | 11:15 | |
| *** martyntaylor1 has quit IRC | 11:15 | |
| *** e0ne has joined #tripleo | 11:17 | |
| *** e0ne_ has quit IRC | 11:17 | |
| coolsvap | Ng: got it working now, devtest_seed in progress | 11:19 |
| coolsvap | Ng: a machine with 8 core and 32g ram will be sufficient to test it out? | 11:20 |
| rpodolyaka | coolsvap: more than enough :) | 11:20 |
| coolsvap | rpodolyaka: cool, thx! | 11:21 |
| rpodolyaka | coolsvap: I used to run devtest on a machine with 8GB | 11:21 |
| rpodolyaka | coolsvap: but 16GB is a much nicer experience | 11:21 |
| * rpodolyaka blames chromium | 11:21 | |
| coolsvap | I have 8gb on my laptop as well, will give it a try on my laptop as well | 11:22 |
| Ng | coolsvap: 8 might be pushing it a bit :) | 11:26 |
| Ng | but the octocore 32GB machine should do a very nice job of it | 11:26 |
| Ng | my local runs for development are on a 16GB workstation that does nothing else. takes about an hour to do a full run with zero caching | 11:27 |
| coolsvap | Ng, rpodolyaka: hmm, do we have approx download that happens with zero caching? | 11:27 |
| Ng | download size? not offhand | 11:28 |
| coolsvap | Ng: actually i am on a limited download data plan, so wanted some approximations, I will check with first run | 11:31 |
| Ng | one ubuntu/fedora cloud image plus a bunch of packages, git repos and python modules | 11:31 |
| Ng | coolsvap: you might want to set up a local squid cache so you don't need to redownload as much stuff on subsequent runs | 11:31 |
| *** boris-42 has quit IRC | 11:31 | |
| coolsvap | hmm, I will do that, was reading about the same | 11:32 |
| Ng | the ubuntu/fedora cloud image will be cached anyway, until a new one is released upstream | 11:33 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Allow use of mirrors when building Ubuntu images https://review.openstack.org/66984 | 11:35 |
| *** derekh_afk is now known as derekh | 11:38 | |
| derekh | lifeless: sounds good, I'm going to do the whole thing again manually today to see what else might be missing | 11:38 |
| derekh | lifeless: your patch to add gear in prepare_tripleo looks familiar | 11:39 |
| derekh | lifeless: https://review.openstack.org/#/c/67685/1/modules/openstack_project/files/nodepool/scripts/prepare_tripleo.sh | 11:39 |
| *** boris-42 has joined #tripleo | 11:41 | |
| openstackgerrit | Dirk Mueller proposed a change to openstack/diskimage-builder: /usr/bin move: use /usr/bin/env https://review.openstack.org/67817 | 11:42 |
| *** pblaho has quit IRC | 11:47 | |
| *** rpodolyaka has quit IRC | 11:50 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-incubator: Change communication with remote host https://review.openstack.org/67689 | 11:51 |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-incubator: Upgrade setuptools to >=1 in a newly created virtualenv https://review.openstack.org/67688 | 11:51 |
| *** coolsvap has quit IRC | 12:07 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/os-apply-config: Add new username type https://review.openstack.org/67820 | 12:20 |
| *** CaptTofu has joined #tripleo | 12:25 | |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/tripleo-image-elements: Add Ceilometer elements https://review.openstack.org/64663 | 12:29 |
| *** rbrady has left #tripleo | 12:32 | |
| *** rbrady has joined #tripleo | 12:33 | |
| rbrady | Anyone know why this approved change isn't merging? Is there something else that needs to happen with gerrit/jenkins? https://review.openstack.org/#/c/67163/ | 12:40 |
| *** e0ne has quit IRC | 12:41 | |
| *** markmc has quit IRC | 12:41 | |
| *** e0ne has joined #tripleo | 12:41 | |
| *** markmc has joined #tripleo | 12:44 | |
| *** CaptTofu has quit IRC | 12:46 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Deployment design form https://review.openstack.org/67155 | 12:46 |
| *** pblaho has joined #tripleo | 12:50 | |
| *** e0ne_ has joined #tripleo | 12:53 | |
| openstackgerrit | A change was merged to openstack/tuskar: Sync with global requirements https://review.openstack.org/67061 | 12:55 |
| *** e0ne has quit IRC | 12:55 | |
| rlandy | lifeless: hi ... I'm running through the updated devtest. scripts/write-tripleorc does not store $TE_DATAFILE. Should it? | 12:57 |
| slagle | i don't think he wants it to. i had submitted a change for that and it got Nack'd | 12:57 |
| slagle | however if manually running devtest is explicitally referencing TE_DATAFILE we may need to revisit that | 12:58 |
| slagle | or fix devtest to not reference it | 12:58 |
| rlandy | I am manually running devtest - because it changed from my last run and I wanted to see what the differences were | 12:59 |
| *** shadower has quit IRC | 12:59 | |
| rlandy | I have not yet gotten to a point where TE_DATAFILE is referenced but I can see it's not stored | 12:59 |
| *** shadower has joined #tripleo | 12:59 | |
| rlandy | let's see what I hit finishing devtest setup | 13:00 |
| *** markmc has quit IRC | 13:07 | |
| rlandy | slagle: the last command on this page: http://docs.openstack.org/developer/tripleo-incubator/devtest_testenv.html gives "-bash: $JSONFILE: ambiguous redirect". I don't see $JSONFILE defined anywhere. | 13:09 |
| rlandy | http://docs.openstack.org/developer/tripleo-incubator/devtest_setup.html says: "Devtest uses a JSON file to describe the test environment that OpenStack will run within. The JSON file path is given by $TE_DATAFILE" | 13:10 |
| rlandy | maybe it's not critical? | 13:10 |
| slagle | rlandy: so, there's a fix in review for that: https://review.openstack.org/#/c/67535/1 | 13:11 |
| slagle | but apparently there's some confusion around it | 13:11 |
| rlandy | ah ok ... looking | 13:11 |
| slagle | maybe add any more details you have to that review | 13:11 |
| slagle | since not everyone is happy with the proposed solution :) | 13:11 |
| *** rpodolyaka has joined #tripleo | 13:15 | |
| *** markmc has joined #tripleo | 13:15 | |
| rlandy | sure - $JSONFILE never seems to be defined - while $TE_DATAFILE is defined and then not stored | 13:16 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add os-svc-restart https://review.openstack.org/67827 | 13:17 |
| openstackgerrit | James Slagle proposed a change to openstack/diskimage-builder: Add map-services https://review.openstack.org/67828 | 13:18 |
| openstackgerrit | James Slagle proposed a change to openstack/diskimage-builder: Copy hook directories with -r into the chroot https://review.openstack.org/67829 | 13:18 |
| *** max_lobur is now known as max_lobur_afk | 13:26 | |
| lucasagomes | lifeless, ping. About bug #1261644, after a success deploy we still can check the available ISCSI nodes via "iscsiadm -m node" , should we logout and delete from that session after successfully dd'ing the image to the iscsi disk as well? | 13:29 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Deployment design form https://review.openstack.org/67155 | 13:32 |
| *** martyntaylor has quit IRC | 13:33 | |
| *** jdob has joined #tripleo | 13:33 | |
| *** lblanchard has joined #tripleo | 13:34 | |
| *** martyntaylor has joined #tripleo | 13:36 | |
| *** ramishra has quit IRC | 13:40 | |
| *** Ng_ has joined #tripleo | 13:43 | |
| rbrady | slagle: do you know if there is some sort of manual step involved or something left on the change that needs to be done to get it to merge? https://review.openstack.org/#/c/67163/ | 13:46 |
| *** Ng_ has quit IRC | 13:47 | |
| *** Ng_ has joined #tripleo | 13:49 | |
| *** Ng_ has quit IRC | 13:51 | |
| *** Ng_ has joined #tripleo | 13:51 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Adds iptables element https://review.openstack.org/67163 | 13:54 |
| slagle | rbrady: hmm, not honestly sure | 13:54 |
| slagle | i approved it again | 13:54 |
| slagle | there it goes. i guess it was a bogon | 13:54 |
| rbrady | slagle: thanks | 14:00 |
| *** morazi has joined #tripleo | 14:01 | |
| rbrady | slagle: these changes have been hanging on that other change: https://review.openstack.org/#/c/66569/ , https://review.openstack.org/#/c/66571/ , https://review.openstack.org/#/c/67458/ , https://review.openstack.org/#/c/67460/ | 14:02 |
| slagle | rbrady: gerrit is saying those need to be rebased | 14:05 |
| *** Ng_ has quit IRC | 14:08 | |
| *** ramishra has joined #tripleo | 14:12 | |
| *** Ng_ has joined #tripleo | 14:12 | |
| *** Ng has quit IRC | 14:13 | |
| *** Ng_ is now known as Ng | 14:13 | |
| *** ramishra has quit IRC | 14:15 | |
| *** ramishra has joined #tripleo | 14:15 | |
| *** ramishra has quit IRC | 14:16 | |
| *** tzumainn has joined #tripleo | 14:21 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates cinder iptables rules https://review.openstack.org/67458 | 14:21 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates iptables rule for MySQL element https://review.openstack.org/66569 | 14:24 |
| *** matty_dubs|gone is now known as matty_dubs | 14:25 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates boot-stack iptables file https://review.openstack.org/66571 | 14:27 |
| *** dprince has joined #tripleo | 14:27 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates gearman iptables rule https://review.openstack.org/67460 | 14:27 |
| *** jrist has quit IRC | 14:28 | |
| *** pblaho1 has joined #tripleo | 14:31 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-incubator: Add $SEED_IP back into $no_proxy https://review.openstack.org/67842 | 14:31 |
| *** morazi_ has joined #tripleo | 14:32 | |
| *** pblaho has quit IRC | 14:33 | |
| *** max_lobur_afk is now known as max_lobur | 14:33 | |
| *** morazi has quit IRC | 14:34 | |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Add unstyled overcloud resource category page https://review.openstack.org/67632 | 14:39 |
| *** coolsvap has joined #tripleo | 14:41 | |
| openstackgerrit | Marios Andreou proposed a change to openstack/tuskar: Using merge.py from tuskar to generate overcloud.yaml https://review.openstack.org/52045 | 14:49 |
| openstackgerrit | Marios Andreou proposed a change to openstack/tuskar: Using merge.py from tuskar to generate overcloud.yaml https://review.openstack.org/52045 | 14:50 |
| *** pblaho1 has quit IRC | 14:52 | |
| *** morazi_ has quit IRC | 14:52 | |
| *** pblaho has joined #tripleo | 14:55 | |
| *** jrist has joined #tripleo | 15:02 | |
| openstackgerrit | Jay Dobies proposed a change to openstack/tuskar: Cleanup for Icehouse requirements https://review.openstack.org/67848 | 15:04 |
| *** morazi_ has joined #tripleo | 15:06 | |
| *** nosnos has quit IRC | 15:06 | |
| *** mrunge has quit IRC | 15:09 | |
| *** bcrochet has quit IRC | 15:11 | |
| *** bcrochet has joined #tripleo | 15:13 | |
| *** morazi_ is now known as morazi | 15:13 | |
| marios | jdob: ping | 15:14 |
| jdob | marios: ahoy | 15:14 |
| marios | hey - just looking at https://review.openstack.org/67848 now.will put comments there. are you planning on writing new controllers? has that bee dicussed at all? i understand tzumain has some mock api calls | 15:16 |
| jdob | marios: yes, that's literally my next task. i went to do it and realized all that stuff was there and figured I'd do it in two review steps | 15:16 |
| jdob | going to add them for the domain model entries, so resource categories and overcloud config | 15:16 |
| marios | jdob: yeah, imo they should go together | 15:16 |
| openstackgerrit | Dirk Mueller proposed a change to openstack/diskimage-builder: fedora/RHEL: use env from /usr/bin https://review.openstack.org/67817 | 15:17 |
| marios | jdob: i mean, if we rip out all controllers now, tuskar becomes completely useless, i mean even for debugging/dev purposes | 15:17 |
| jdob | but no one's using it right now anyway, the UI isn't calling anything in there | 15:17 |
| marios | jdob: cos there are no controllers at all. perhaps just write the stubs and return strings for now | 15:17 |
| jdob | the stubs make sense | 15:18 |
| marios | jdob: right agree, but for example to move forward with https://review.openstack.org/#/c/52045/7 i need at least a controller otherwise i cant test/work with the code | 15:18 |
| jdob | i can totally get the stubs in quickly | 15:19 |
| jdob | marios: i'll get started on the stubs this morning and get them in as soon as I can | 15:20 |
| marios | jdob: thanks, again, imo, i just put my name to it on the review too | 15:20 |
| *** martyntaylor has quit IRC | 15:23 | |
| jdob | marios: i dont understand the need to block this one though. the old calls are completely incorrect, so you'd rather me leave in an invalid API rather than delete it before writing the right one? | 15:23 |
| jdob | that seems more misleading than anything if someone were to look at the API and assume those calls were going to stick around | 15:23 |
| *** martyntaylor has joined #tripleo | 15:23 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates gearman iptables rule https://review.openstack.org/67460 | 15:25 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates boot-stack iptables file https://review.openstack.org/66571 | 15:26 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates iptables rule for MySQL element https://review.openstack.org/66569 | 15:26 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates cinder iptables rules https://review.openstack.org/67458 | 15:26 |
| marios | jdob: heh - am not trying to block anything. i think this is great movement (overall concept of moving functionality out to the right places to simplify the tuskar deploy process) | 15:27 |
| *** rdopieralski has quit IRC | 15:27 | |
| jdob | er, sorry, by "block" i meant the -1, wasnt trying to make it sound like a bigger deal than it was | 15:27 |
| marios | jdob: yeah, just my opinion on the review. also re the heat client - don't know why my comment isn't showing there, let me try re add it | 15:28 |
| jdob | was gonna ask about that one too, I thought I was doing something dumb and couldn't find it :) | 15:29 |
| marios | jdob: no idea, perhaps i added it right at the end or something... sec | 15:29 |
| *** maksimkins has joined #tripleo | 15:31 | |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Add unstyled overcloud resource category page https://review.openstack.org/67632 | 15:31 |
| marios | jdob: put a comment instead mate cos not showing on heat/client.py | 15:33 |
| *** lucasagomes is now known as lucas-hungry | 15:33 | |
| *** vkozhukalov has quit IRC | 15:34 | |
| *** tzumainn has joined #tripleo | 15:38 | |
| *** kpavel has quit IRC | 15:38 | |
| *** tzumainn has joined #tripleo | 15:38 | |
| marios | tzumainn: hey | 15:39 |
| marios | :) | 15:39 |
| tzumainn | marios, heya : ) | 15:39 |
| marios | tzumainn: so, about that 'list of mock calls' , is that in Tuskar UI master already? | 15:39 |
| tzumainn | marios, yep: https://github.com/openstack/tuskar-ui/blob/master/tuskar_ui/api.py | 15:39 |
| tzumainn | marios, they're just best guesses, it can be changed if you think they should be changed | 15:40 |
| marios | tzumainn: excellent, thanks! | 15:40 |
| marios | jdob: ^^^ controllers :) | 15:40 |
| jdob | marios: ya, that's what I planned on going from | 15:40 |
| jdob | i worked with tzumainn to get those written in the first place :) | 15:41 |
| marios | jdob: perfect, let me know what i can do if you can identify a clean break point in something. i could take one of the resource types for example | 15:41 |
| jdob | ok, i'll let you know if I need a hand, but there really isn't too much to do for basic REST stuff | 15:42 |
| marios | jdob: re, the heat client and tuskar vs the UI invoking heat, was that on a mailing list thread (do you remember?) | 15:42 |
| marios | jdob: sure thanks | 15:42 |
| jdob | ya, back in december there was a big thread about... "cli architecture" in the subject I think | 15:42 |
| marios | jdob: in any case re heat, if that is the current consensus then np, i will change in next revision of my review | 15:43 |
| jdob | where do you stand on the heat template generation? is it pulling data from the domain models yet? | 15:44 |
| marios | jdob: we need to see it in practice and i absolutely agree we should revisit post I, assuming this lets us do what we need for I | 15:44 |
| marios | jdob: you mean on my review? | 15:44 |
| jdob | on the feature in general | 15:44 |
| marios | jdob: well, i'm not sure. given the simplicity of the proposed tuskar functionality ('generate overcloud.yaml')... perhaps all we need is the data from the UI on that particular call... | 15:45 |
| jdob | not all of it is coming from the UI, the res categories are saved in tuskar itself | 15:46 |
| *** saneagle has joined #tripleo | 15:46 | |
| jdob | so you'll grab those from the DB | 15:46 |
| marios | jdob: well, thats kinda my question | 15:46 |
| marios | jdob: what are we storing. i saw the model for resource category didn't have a 'count' | 15:47 |
| jdob | thats because its just the definition of the category. the counts are assigned on a per overcloud basis, so its a separate table | 15:47 |
| marios | jdob: so does that mean that user will say 'category foo' and a count (foo having been earlier defined and stored in tuskar db) | 15:47 |
| jdob | exactly | 15:47 |
| jdob | for icehouse, the categories are defined by the installer and we wont have add/edit in the UI | 15:48 |
| marios | jdob: ok... so right now my review is using a fictional data structure (had to make an assumption so i could start) | 15:48 |
| marios | jdob: i will be revisiting that and pulling out the relevant info | 15:48 |
| marios | jdob: for example, the image_id for a given category | 15:48 |
| jdob | cool | 15:48 |
| marios | jdob: thanks for your time mate | 15:48 |
| jdob | np, feel free to ping me with any other questions once you dig into there | 15:49 |
| marios | ack | 15:49 |
| jdob | the review you have now, is that a workinprogress? | 15:49 |
| marios | jdob: well yes, but i removed the wip to attract eyes and reviews ;) | 15:49 |
| jdob | hahaha, gotcha. i'll take a look later on today :) | 15:49 |
| marios | jdob: so the official answer is | 15:49 |
| marios | <pokerface> no </pokerface> | 15:50 |
| jdob | hahaha | 15:50 |
| *** saneagle has quit IRC | 15:51 | |
| *** maksimkins has quit IRC | 15:51 | |
| *** marun has joined #tripleo | 15:57 | |
| *** jprovazn has quit IRC | 16:08 | |
| rpodolyaka | dprince: ping | 16:08 |
| *** ifarkas has quit IRC | 16:08 | |
| dprince | rpodolyaka: hi | 16:09 |
| rpodolyaka | dprince: hey! may I ask you to revisit https://review.openstack.org/#/c/59312/ ? I left comments to patch set 27 (the one you left comments to) | 16:09 |
| rpodolyaka | dprince: if you insist on sqlite one, I'll introduce a helper function | 16:10 |
| dprince | rpodolyaka: The sqlite issue wasn't why I added a -1. The main issue I have with it is the fact that you change task states in the driver when I think the compute/manager should already take care of that for you with a rebuild. | 16:12 |
| rpodolyaka | dprince: so usually yes, but not in the case of rebuild | 16:13 |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-incubator: Replace rsync with the copyseed command https://review.openstack.org/67867 | 16:13 |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-incubator: Replace usage of qemu+ssh with ssh <command> https://review.openstack.org/67868 | 16:13 |
| rpodolyaka | dprince: as we override the default implementation of rebuild | 16:13 |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-incubator: Read env-num from TE_DATAFILE https://review.openstack.org/67689 | 16:13 |
| *** jcoufal has quit IRC | 16:15 | |
| *** spzala has joined #tripleo | 16:16 | |
| dprince | rpodolyaka: I'm not sure we are talking about the same things. When performing a rebuild task state should change here: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L2245 | 16:17 |
| dprince | rpodolyaka: The driver actually gets invoked here: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L2280 | 16:17 |
| dprince | rpodolyaka: Have you actually tried it without that code to verify it is required? | 16:18 |
| *** marun has quit IRC | 16:20 | |
| rpodolyaka | dprince: I haven't :) so it should work indeed, but instance will never be in REBUILD_SPAWNING state, it will go to directly to ACTIVE from REBUILDING | 16:21 |
| *** lucas-hungry is now known as lucasagomes | 16:21 | |
| rpodolyaka | dprince: so it's a question of 'state granularity' from what I see | 16:22 |
| *** marun has joined #tripleo | 16:22 | |
| rpodolyaka | dprince: we can't put instance to REBUILD_SPAWNING in manager, because it depends on the concrete driver, what actions are performed before spawning | 16:22 |
| rpodolyaka | dprince: this is why we put this state update to rebuild implementation | 16:23 |
| *** e0ne has joined #tripleo | 16:23 | |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Add unstyled overcloud resource category page https://review.openstack.org/67632 | 16:25 |
| *** e0ne has quit IRC | 16:25 | |
| *** e0ne has joined #tripleo | 16:26 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Add more detail to the json file https://review.openstack.org/67686 | 16:26 |
| *** e0ne_ has quit IRC | 16:27 | |
| *** e0ne has quit IRC | 16:30 | |
| *** AaronGr_Zzz is now known as AaronGr | 16:37 | |
| *** UtahDave has joined #tripleo | 16:38 | |
| SpamapS | hm.. I wonder if there is another virtualenv problem | 16:40 |
| SpamapS | http://paste.openstack.org/show/61576 | 16:41 |
| SpamapS | looks like heatclient is picking up the system python-six | 16:41 |
| SpamapS | which is too old | 16:41 |
| SpamapS | ['', '/opt/stack/python-heatclient', '/usr/lib/python2.7/dist-packages', '/opt/stack/venvs/python-heatclient/lib/python2.7', '/opt/stack/venvs/python-heatclient/lib/python2.7/plat-i386-linux-gnu', '/opt/stack/venvs/python-heatclient/lib/python2.7/lib-tk', '/opt/stack/venvs/python-heatclient/lib/python2.7/lib-old', '/opt/stack/venvs/python-heatclient/lib/python2.7/lib-dynload', '/usr/lib/python2.7', '/usr/lib/python2.7/plat-i386-linux-gnu', '/usr/lib/ | 16:42 |
| dprince | SpamapS: how old are your clients. Mine is working locally from this weekend. | 16:42 |
| SpamapS | dprince: I just built a new seed | 16:42 |
| SpamapS | dprince: and heat cli does not work on it | 16:42 |
| SpamapS | sys.path has /usr/lib/python2.7/dis-patckages in front of /opt/stack/venvs/python-heatclient/lib/python2.7/site-packages | 16:43 |
| dprince | SpamapS: okay. Me too, and on Fedora mine works. | 16:43 |
| SpamapS | dprince: Perhaps Fedora has six >= 1.4.1 | 16:44 |
| SpamapS | Ubuntu 13.10 is at 1.3.0 | 16:45 |
| dprince | SpamapS: ah, snap yeah I just started using F20 and it does have 1.4.1 by default. | 16:46 |
| dprince | SpamapS: F19 has 1.3 though... at least it did for me last week | 16:46 |
| *** jistr has quit IRC | 16:48 | |
| SpamapS | well this seems like virtualenv's problem | 16:48 |
| SpamapS | afk for a bit.. but this may need to be a firedrill as everybody is likely broken on Ubuntu. | 16:48 |
| derekh | SpamapS: https://review.openstack.org/#/c/67178/ | 16:51 |
| *** GheRiver1 has joined #tripleo | 16:53 | |
| *** GheRiver1 has quit IRC | 16:53 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-image-elements: Remove --system-site-packages from clients virtualenv https://review.openstack.org/67178 | 16:53 |
| dprince | rpodolyaka: I'm good w/ it for now. | 16:54 |
| dprince | rpodolyaka: ideally we'd make that state update call in manager (like we do everywhere else)... but since this is the only driver that actually implements rebuild ATM it is okay I think. | 16:54 |
| rpodolyaka | dprince: agree. thanks! | 16:55 |
| *** pblaho has quit IRC | 16:57 | |
| *** AaronGr is now known as AaronGr_Zzz | 16:58 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-image-elements: Remove --system-site-packages from clients virtualenv https://review.openstack.org/67178 | 17:02 |
| *** pblaho has joined #tripleo | 17:04 | |
| *** pblaho has quit IRC | 17:04 | |
| rpodolyaka | hmm, interesting, anyone else seeing this http://paste.openstack.org/show/61581/ when deploying undercloud? | 17:07 |
| *** ifarkas has joined #tripleo | 17:08 | |
| *** vkozhukalov has joined #tripleo | 17:08 | |
| *** bauzas has quit IRC | 17:13 | |
| SpamapS | derekh: https://review.openstack.org/#/c/67178/ .. -1'd.. commit message was a bit of a mess.. but the rest looks good. | 17:17 |
| derekh | SpamapS: fixing | 17:17 |
| SpamapS | rpodolyaka: yes that is, I believe, also the 67178 issue | 17:17 |
| SpamapS | rpodolyaka: basically nothing is working because of virtualenv fail | 17:17 |
| rpodolyaka | SpamapS: hmm, if so, it's very subtle | 17:18 |
| dprince | SpamapS: I'm sorry I didn't point that out earlier. I had grabbed Derek's fix as well... but again once I upped to F20 it wasn't needed. Total memory fail.. | 17:18 |
| SpamapS | rpodolyaka: keystoneclient is raising exceptions I think | 17:18 |
| SpamapS | dprince: np .. easy fix for now | 17:19 |
| SpamapS | note that virtualenv 1.11.1 is due "soon" so then we should be able to use 1.11 and this won't be an issue | 17:19 |
| *** ramishra has joined #tripleo | 17:19 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/tripleo-image-elements: Remove --system-site-packages from clients virtualenv https://review.openstack.org/67178 | 17:21 |
| derekh | SpamapS: done, thanks | 17:21 |
| SpamapS | derekh: thanks, +A'd | 17:24 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Remove --system-site-packages from clients virtualenv https://review.openstack.org/67178 | 17:26 |
| *** AaronGr has joined #tripleo | 17:29 | |
| *** AaronGr has quit IRC | 17:30 | |
| *** AaronGr_Zzz is now known as AaronGr | 17:30 | |
| *** martyntaylor has quit IRC | 17:38 | |
| *** matty_dubs is now known as matty_dubs|lunch | 17:38 | |
| *** marun has quit IRC | 17:40 | |
| *** marun has joined #tripleo | 17:41 | |
| *** matty_dubs|lunch has quit IRC | 17:43 | |
| bnemec_ | So regarding https://review.openstack.org/#/c/67535/ | 17:52 |
| bnemec_ | Is there a reason that in this one case we're passing TE_DATAFILE as a parameter instead of just using the environment variable? | 17:53 |
| *** bnemec_ is now known as bnemec | 17:54 | |
| derekh | Nice | 17:56 |
| derekh | + ssh root@192.0.2.1 'cat /opt/stack/boot-stack/virtual-power-key.pub' | 17:56 |
| derekh | Run completed. | 17:56 |
| *** ramishra has quit IRC | 17:59 | |
| *** boris-42 has quit IRC | 18:00 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Add $SEED_IP back into $no_proxy https://review.openstack.org/67842 | 18:00 |
| derekh | lifeless: ^ just got first CI run to complete, looks like we're getting v. close, gotta run but will rebuild the test env host later as I had to manually tweak a few things since it was last built | 18:01 |
| SpamapS | yeah my seed seems to be working now | 18:03 |
| *** derekh has quit IRC | 18:03 | |
| *** boris-42 has joined #tripleo | 18:04 | |
| *** CaptTofu has joined #tripleo | 18:06 | |
| *** matty_dubs|lunch has joined #tripleo | 18:07 | |
| *** e0ne has joined #tripleo | 18:10 | |
| * Ng dinners, bbl | 18:10 | |
| *** athomas has quit IRC | 18:10 | |
| SpamapS | whoa | 18:13 |
| SpamapS | kernel oops or panic on resizing root fs | 18:13 |
| openstackgerrit | Ralf Haferkamp proposed a change to openstack/diskimage-builder: Fix mysql package mappings for opensuse https://review.openstack.org/67899 | 18:13 |
| *** rlandy is now known as rlandy|bbl | 18:16 | |
| *** vkozhukalov has quit IRC | 18:19 | |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/tripleo-image-elements: Add Ceilometer elements https://review.openstack.org/64663 | 18:20 |
| *** matty_dubs|lunch is now known as matty_dubs | 18:26 | |
| *** marun has quit IRC | 18:32 | |
| *** marun has joined #tripleo | 18:33 | |
| *** marun has quit IRC | 18:40 | |
| lifeless | morning | 18:41 |
| *** marun has joined #tripleo | 18:41 | |
| phschwartz | lifeless: morning | 18:45 |
| rbrady | bnemec: did you resolve your cinder error? | 18:47 |
| *** newell has joined #tripleo | 18:47 | |
| rbrady | hey newell | 18:47 |
| bnemec | rbrady: I haven't actually gotten very far on that. Keep getting distracted with other stuff. | 18:48 |
| newell | howdy rbrady | 18:50 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add Ceilometer elements https://review.openstack.org/64663 | 18:50 |
| rbrady | bnemec: did it have anything to do with iniparse? | 18:52 |
| bnemec | rbrady: I don't think so. I seem to recall it was an issue with the .ok file for the firewall rules. | 18:52 |
| *** CaptTofu has quit IRC | 18:52 | |
| SpamapS | hm | 18:52 |
| SpamapS | I sourced devtest_variables.sh but I still got errors when trying to upload my newly built undercloud image to the seed's glance.. | 18:53 |
| rbrady | bnemec: the latest cinder clement changes that landed this morning eliminate the usage of the .ok file, so you should be good | 18:53 |
| bnemec | rbrady: Cool, I was thinking that would be the case. Thanks. | 18:53 |
| rbrady | bnemec: np | 18:53 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: fedora/RHEL: use env from /usr/bin https://review.openstack.org/67817 | 18:54 |
| lifeless | phschwartz: hi! | 18:55 |
| lifeless | phschwartz: https://bugs.launchpad.net/ubuntu/+source/openvswitch/+bug/1270649 :) | 18:56 |
| phschwartz | lifeless: Any luck on the ovs+gre slowdown? | 18:56 |
| phschwartz | beat me by 1 second. lol | 18:56 |
| lifeless | phschwartz: so yeah, two fun bugs emerged | 18:56 |
| lifeless | that one that I linked | 18:56 |
| lifeless | and | 18:57 |
| lifeless | https://bugs.launchpad.net/neutron/+bug/1270646 which is still in progress | 18:57 |
| *** spzala has quit IRC | 18:57 | |
| lifeless | I can workaround the gre performance issue by clamping the mtu on the VMs to 1458 bytes | 18:57 |
| *** spzala_ has joined #tripleo | 18:57 | |
| *** spzala_ is now known as spzala | 18:58 | |
| lifeless | SpamapS: devtest_variables doesn't include '. seedrc' does it? Thats up to you to switch between the cloud you want to use? | 18:58 |
| lifeless | phschwartz: or by disabling GRO on the external interfaces | 18:58 |
| lifeless | phschwartz: I'm fairly certain that DF is not being honoured | 18:58 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Fix the curl command in the ironic-deploy element https://review.openstack.org/66677 | 18:58 |
| phschwartz | lifeless: interesting | 18:58 |
| lifeless | phschwartz: since ovs removed exporting df to the tunnel wrapper in 1.10.2 which I completely fail to understand | 18:59 |
| lifeless | the tunnel is effectively pretending to be infinitely wide | 18:59 |
| phschwartz | And that would defn cause an issue. | 19:00 |
| SpamapS | lifeless: hm, seedrc wants $SEED_IP | 19:03 |
| lifeless | SpamapS: _variables can only export that if its in the datafile (vs obtained by mac + arp introspection) | 19:04 |
| lifeless | perhaps we should use 192.0.2.1 there | 19:04 |
| lifeless | I dunno | 19:04 |
| *** CaptTofu has joined #tripleo | 19:05 | |
| SpamapS | lifeless: this has become very unclear to me. I'm reading the new devtest and it doesn't really explain how to not start from point 0. :-/ | 19:06 |
| lifeless | SpamapS: it may be buggy | 19:06 |
| lifeless | SpamapS: you should of course be able to source the tripleorc that gets written as well | 19:07 |
| lifeless | we didn't (intentionally anyhow) break that | 19:07 |
| SpamapS | I don't have a tripleorc .. or I don't know where it is | 19:13 |
| SpamapS | lifeless: isn't that written at the very end? | 19:13 |
| *** akrivoka has quit IRC | 19:14 | |
| *** rpodolyaka1 has joined #tripleo | 19:15 | |
| lifeless | yes | 19:15 |
| lifeless | see write-tripleorc | 19:15 |
| SpamapS | ok I had a failure during devtest_undercloud | 19:15 |
| SpamapS | so didn't want to start over at devtest.sh ... | 19:15 |
| lifeless | ok so | 19:15 |
| lifeless | we didn't previously have any way to do that | 19:16 |
| *** markmc has quit IRC | 19:16 | |
| lifeless | because we were sourcing everything | 19:16 |
| SpamapS | seems to work now I had to export SEED_IP=looked.up.manually and then source seedrc | 19:16 |
| lifeless | now it should be possible but you've found the rough edges | 19:16 |
| lifeless | you know what to do :) | 19:16 |
| SpamapS | lifeless: right I thought that was why we split everything up so was thinking this was thought through :) | 19:16 |
| lifeless | SpamapS: we split it up for maintenance sanity as much as anything | 19:17 |
| lifeless | SpamapS: I wouldn't claim that it's 'done' | 19:17 |
| SpamapS | works fo rme | 19:17 |
| *** jtomasek has quit IRC | 19:17 | |
| SpamapS | also why aren't we defaulting to using pypi-openstack ? | 19:17 |
| lifeless | because thats a terrible default | 19:18 |
| lifeless | its a CI default not a production default | 19:18 |
| SpamapS | but its faster, and more stable | 19:18 |
| lifeless | for CI we'll want that | 19:18 |
| lifeless | SpamapS: for whom | 19:18 |
| SpamapS | me! | 19:18 |
| lifeless | SpamapS: sample size of 1 :) | 19:18 |
| SpamapS | well it gets better squid its for one | 19:19 |
| SpamapS | hits | 19:19 |
| bnemec | We have stackuser enabled by default in devtest... | 19:19 |
| bnemec | related to production versus test defaults. | 19:19 |
| lifeless | bnemec: devtest is in incubator | 19:19 |
| lifeless | bnemec: but sure, I see your point | 19:19 |
| lifeless | I want use to be pushing towards production defaults with opt-in CI or developer stuff | 19:20 |
| lifeless | not the other way around | 19:20 |
| lifeless | for production folk will probably run a pypi mirror | 19:20 |
| SpamapS | but why would I want pypi direct vs. "the thing that works with openstack" in production image building? | 19:20 |
| SpamapS | thats really confusing to me. | 19:20 |
| lifeless | SpamapS: because infra don't consider the pypi mirror to be a production service | 19:20 |
| lifeless | SpamapS: its a CI service, not 'anyone in the world should use this' | 19:21 |
| SpamapS | ah that I did not know. | 19:21 |
| lifeless | when I asked about rsyncing it that came up | 19:21 |
| lifeless | also that mirror is very restrictive in contents | 19:22 |
| lifeless | users that customise elements are almost certainly going to want things openstack itself doesn't carry | 19:22 |
| lifeless | - and won't until an openstack project includes them | 19:23 |
| lifeless | it would be like having a mirror of only the apt/yum packages we install and pointing at that by default | 19:23 |
| lifeless | for all builds | 19:23 |
| lifeless | also the pypi mirror is on HTTP and thus insecure - so not production ready in that way either | 19:23 |
| lifeless | I would argue against using it outside of CI for any of those reasons, let alone all of them | 19:24 |
| SpamapS | lifeless: this all makes sense. I think what may make sense is to have a good collection of tips for "going faster" which those points made so we know why its not the default. | 19:25 |
| lifeless | that would be good | 19:25 |
| SpamapS | cause its hard to get a local pypi mirror setup.. but its pretty easy to just use squid + a fast http-only mirror. :P | 19:26 |
| lifeless | I know :) | 19:26 |
| lifeless | I see our job as making those hard things simple and easty | 19:27 |
| lifeless | easy. | 19:27 |
| SpamapS | but the former is the better production option | 19:27 |
| lifeless | (I don't know what an easty is). | 19:27 |
| lifeless | I thought we had an element to setup a local pypi mirror:) | 19:27 |
| SpamapS | right now I'm just trying to get through devtest so I can test some heat stuff. :-P | 19:28 |
| SpamapS | and yeah, if thats our job, we have some work to do. :) | 19:28 |
| SpamapS | (note that today is a holiday... but I'm itching to get at this stuff :-P | 19:28 |
| lifeless | don't let me stop you :) | 19:28 |
| SpamapS | lifeless: has anybody suggested pushing pip / pypi toward signed files instead of https? | 19:30 |
| SpamapS | or at least, as an alternative to it | 19:30 |
| lifeless | SpamapS: they are doing that too via TUF | 19:30 |
| clarkb | SpamapS: yes, cheeseshop v2 will have it | 19:30 |
| SpamapS | ah good | 19:30 |
| clarkb | called warehouse | 19:30 |
| *** spzala has quit IRC | 19:35 | |
| *** edmund has joined #tripleo | 19:35 | |
| *** spzala has joined #tripleo | 19:36 | |
| * SpamapS gives up trying to resume after fail | 19:40 | |
| *** CaptTofu has quit IRC | 19:46 | |
| *** estsoft has joined #tripleo | 19:49 | |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Add unstyled overcloud resource category page https://review.openstack.org/67632 | 19:50 |
| *** marun has quit IRC | 19:55 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Disable uncompressed images for tripleo-cd. https://review.openstack.org/67336 | 19:57 |
| *** AaronGr is now known as Aarongr_afk | 19:59 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-incubator: Add documentation of JSONFILE variable https://review.openstack.org/67535 | 20:02 |
| *** marun has joined #tripleo | 20:06 | |
| *** ifarkas has quit IRC | 20:08 | |
| *** morazi has quit IRC | 20:16 | |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-incubator: Add devtest support for power management https://review.openstack.org/67926 | 20:17 |
| *** jcoufal has joined #tripleo | 20:18 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Preserve 'did we init openstack' state file. https://review.openstack.org/67339 | 20:25 |
| *** bauzas has joined #tripleo | 20:28 | |
| *** rlandy|bbl is now known as rlandy | 20:30 | |
| *** morazi has joined #tripleo | 20:32 | |
| *** derekh has joined #tripleo | 20:48 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Replace /var/lib/mysql path in the upstart script https://review.openstack.org/67742 | 20:50 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Fixes for use behind proxy https://review.openstack.org/67159 | 20:50 |
| *** funzo has quit IRC | 20:54 | |
| *** funzo has joined #tripleo | 20:54 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add haproxy element https://review.openstack.org/61374 | 20:57 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Move volume-resize to cinder-volume element https://review.openstack.org/67936 | 20:58 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Destroy the seed domain https://review.openstack.org/67937 | 20:58 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Fixup the handling of ssh keys https://review.openstack.org/67938 | 20:58 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Add tripleo-incubator/scripts to $PATH https://review.openstack.org/67939 | 20:58 |
| openstackgerrit | Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Added initial tuskar api tests https://review.openstack.org/67940 | 20:59 |
| *** CaptTofu has joined #tripleo | 21:05 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Moves iptables rule from cinder to cinder-api https://review.openstack.org/67942 | 21:06 |
| *** dprince has quit IRC | 21:10 | |
| *** max_lobur is now known as max_lobur_afk | 21:10 | |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to swift https://review.openstack.org/67943 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to openstack-clients https://review.openstack.org/67944 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to novnc https://review.openstack.org/67945 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to nova https://review.openstack.org/67946 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install nova-manage into /usr/local/bin https://review.openstack.org/67947 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install nova-baremetal-manage into /usr/local/bin https://review.openstack.org/67948 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to neutron https://review.openstack.org/67949 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to keystone https://review.openstack.org/67950 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install keystone-manage into /usr/local/bin https://review.openstack.org/67951 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to horizon https://review.openstack.org/67952 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to heat https://review.openstack.org/67953 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install heat-manage into /usr/local/bin https://review.openstack.org/67954 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add package install support to glance https://review.openstack.org/67955 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install glance-manage into /usr/local/bin https://review.openstack.org/67956 | 21:11 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Add os-svc-restart https://review.openstack.org/67827 | 21:11 |
| *** lucasagomes has quit IRC | 21:12 | |
| slagle | (BOOM) | 21:13 |
| Ng | wowser | 21:14 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Removes iptables from cinder element https://review.openstack.org/67957 | 21:15 |
| slagle | Ng: i contemplated doing one giant commit, but figured that'd get -1'd :) | 21:16 |
| Ng | heh :) | 21:18 |
| *** morazi has quit IRC | 21:22 | |
| mordred | wow | 21:24 |
| derekh | lifeless: you added seed-route-dev to the json file (or at least the reading of it) , in CI it needs to be eth1, had you anything in mind for how the value is retrieved? | 21:25 |
| derekh | lifeless: just before I make something up :-) | 21:25 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Convert -dev to -devel in fedora/map-packages https://review.openstack.org/67397 | 21:26 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Allow adding packages to debootstrap https://review.openstack.org/66430 | 21:26 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Fix mysql package mappings for opensuse https://review.openstack.org/67899 | 21:27 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add debian-upstart for experimenting with upstart https://review.openstack.org/66431 | 21:27 |
| *** CaptTofu has quit IRC | 21:28 | |
| *** rpodolyaka1 has quit IRC | 21:28 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Update to Fedora 20 https://review.openstack.org/67505 | 21:30 |
| *** morazi has joined #tripleo | 21:34 | |
| *** Shrews has quit IRC | 21:41 | |
| *** Shrews has joined #tripleo | 21:42 | |
| *** spzala has quit IRC | 21:43 | |
| * Ng ejects for the day | 21:46 | |
| derekh | lifeless: does it need to be in the json file http://paste.openstack.org/show/61601/ | 21:47 |
| openstackgerrit | Ralf Haferkamp proposed a change to openstack/diskimage-builder: Setup ccache symlinks on openSUSE https://review.openstack.org/67965 | 21:51 |
| *** tserong has quit IRC | 21:53 | |
| *** tserong has joined #tripleo | 21:53 | |
| *** tserong has quit IRC | 21:54 | |
| *** tserong has joined #tripleo | 21:54 | |
| *** estsoft has quit IRC | 21:56 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-incubator: Use TE_DATAFILE in devtest_testenv https://review.openstack.org/67535 | 21:59 |
| *** martyntaylor has joined #tripleo | 22:00 | |
| *** matty_dubs is now known as matty_dubs|gone | 22:01 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Add tripleo-incubator/scripts to $PATH https://review.openstack.org/67939 | 22:02 |
| *** lblanchard has quit IRC | 22:02 | |
| *** jdob has quit IRC | 22:10 | |
| *** marun has quit IRC | 22:14 | |
| openstackgerrit | Derek Higgins proposed a change to openstack/os-apply-config: Add new username type https://review.openstack.org/67820 | 22:24 |
| *** rlandy has quit IRC | 22:29 | |
| *** michchap has quit IRC | 22:30 | |
| *** michchap has joined #tripleo | 22:31 | |
| lifeless | derekh: hi | 22:33 |
| derekh | lifeless: hi ya | 22:33 |
| lifeless | uhm, seed dev route | 22:33 |
| lifeless | so I don't know where it has to live :) | 22:34 |
| lifeless | derekh: I'm not quite sure what your patch does | 22:35 |
| derekh | lifeless: that makes two of us :-) | 22:35 |
| lifeless | oh, I see | 22:35 |
| lifeless | so no, we use a non-routable range | 22:35 |
| derekh | lifeless: I'm grabbing the device from the arp table | 22:35 |
| lifeless | I -really- don't want that accidentally being routed somewhere real | 22:36 |
| lifeless | that would be an RFC MUST NOT violation | 22:36 |
| *** jcoufal has quit IRC | 22:36 | |
| derekh | so maybe we could just set it to $DEVICE in toci_devtest.sh ? and set DEVICE=eth1 in jjb ? | 22:37 |
| lifeless | derekh: well we have something thats working, no? | 22:38 |
| lifeless | what problem are you solving? | 22:38 |
| derekh | in CI we need the value to be eth1 , not the default | 22:38 |
| lifeless | right, so we set it in the testenv json ? | 22:39 |
| derekh | the json is created by the TE hosts and doesn't now the device name of the jenkins worker | 22:39 |
| derekh | *know | 22:39 |
| lifeless | oh | 22:39 |
| lifeless | so a coupling problem | 22:39 |
| lifeless | uhm | 22:39 |
| derekh | lifeless: so we be just hard codeing to eth1 and asumming its correct | 22:40 |
| lifeless | I suggest we set it in the TE host for now, because the jenkins worker setup is dedicated to us | 22:40 |
| derekh | lifeless: okiedokie willdo | 22:40 |
| lifeless | and its a trivial patch to do that, and we can always rearrange later | 22:40 |
| lifeless | vs the concerns with dynamic detection | 22:40 |
| derekh | ok | 22:41 |
| *** e0ne has quit IRC | 22:43 | |
| *** panda has joined #tripleo | 22:46 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Add more detail to the json file https://review.openstack.org/67686 | 22:46 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Destroy the seed domain https://review.openstack.org/67937 | 22:46 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Add tripleo-incubator/scripts to $PATH https://review.openstack.org/67939 | 22:46 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Fixup the handling of ssh keys https://review.openstack.org/67938 | 22:46 |
| *** panda___ has quit IRC | 22:49 | |
| *** panda_ has joined #tripleo | 22:55 | |
| lifeless | derekh: so we need to figure out how we run a proposed branch (given a ZUUL_REF) - of incubator etc etc | 22:55 |
| *** panda has quit IRC | 22:58 | |
| bnemec | Hmm: https://github.com/openstack-infra/tripleo-ci/blob/master/elements/testenv-worker/bin/ci_commands#L46 | 22:58 |
| bnemec | Isn't that going to hang waiting for input? | 22:58 |
| derekh | bnemec: yup, we give it input, one sed | 22:59 |
| derekh | bnemec: https://review.openstack.org/#/c/67867/1/scripts/boot-seed-vm | 22:59 |
| *** e0ne has joined #tripleo | 23:00 | |
| derekh | lifeless: yup, we'll need to set DIB_REPOLOCATION_* for all of the project and DIB_REPOREF_ for the commit once we got everything working it should be fairly easy | 23:01 |
| * derekh jinxed it | 23:02 | |
| bnemec | derekh: Huh, so that automatically picks up stdin. Cool. :-) | 23:02 |
| derekh | stdin ? is it not just a envvar? | 23:03 |
| derekh | lifeless: already sounds like there is more to it then I thought | 23:03 |
| *** panda has joined #tripleo | 23:04 | |
| derekh | lifeless: woops read the comment from bnemec as being from you | 23:05 |
| bnemec | heh | 23:05 |
| derekh | bnemec: yup thats how its happening | 23:05 |
| *** martyntaylor has quit IRC | 23:07 | |
| *** panda_ has quit IRC | 23:08 | |
| lifeless | derekh: so where are we at? ready to tag over? | 23:17 |
| *** e0ne has quit IRC | 23:24 | |
| derekh | lifeless: yup, gotta run, so as far as I know everything needed to get a full run today is now either in gerrit https://review.openstack.org/#/dashboard/1926 or in the notes https://etherpad.openstack.org/p/tripleo-test-cluster (under Some Notes on manually runni...) | 23:25 |
| derekh | I am just after rebuilding a new testenv with the branch waiting to be reviewed and starting a new TE heat stack | 23:26 |
| derekh | but I forgot to add this workaround, echo 10.10.16.175 ci-overcloud.tripleo.org >> /etc/hosts | 23:26 |
| *** panda_ has joined #tripleo | 23:27 | |
| derekh | lifeless: so when its up if you intend to use it you'll need to fix that DNS entry and rerun 60-setup-testenvs (after removing the completed file) | 23:27 |
| derekh | lifeless: that it I think its all up there wait for review, I probably mised something as I've been running manually but hopefully nothing major | 23:29 |
| derekh | gotta go | 23:30 |
| *** panda has quit IRC | 23:30 | |
| *** derekh has quit IRC | 23:30 | |
| *** e0ne has joined #tripleo | 23:55 | |
| *** e0ne has quit IRC | 23:56 | |
| *** e0ne has joined #tripleo | 23:57 | |
| *** e0ne has quit IRC | 23:58 | |
| *** e0ne has joined #tripleo | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!