Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2016-02-26

| Channels | #fuel-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:19 zigo xarses: Is Dima B. around? He's missing our 1:1 again.
02:00 rmoe joined #fuel-dev
02:36 xarses joined #fuel-dev
02:48 ilbot3 joined #fuel-dev
02:48 Topic for #fuel-dev is now Fuel Development http://wiki.openstack.org/wiki/Fuel | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel-dev/ | gerrit traffic @ #fuel-tracker
03:10 dims joined #fuel-dev
03:17 intr1nsic joined #fuel-dev
06:17 javeriak joined #fuel-dev
06:18 javeriak joined #fuel-dev
06:21 javeriak_ joined #fuel-dev
06:57 fzhadaev joined #fuel-dev
07:01 _nadya_ joined #fuel-dev
07:49 salmon_ joined #fuel-dev
07:52 ddepaoli joined #fuel-dev
08:16 zubchick joined #fuel-dev
08:20 dshulyak joined #fuel-dev
08:25 xarses Please review python-fuelclient patches https://review.openstack.org/#/c/280966/ https://review.openstack.org/#/c/280949
08:25 xarses Please review specs https://review.openstack.org/#/c/284294/ and https://review.openstack.org/#/c/281557/
08:31 HeOS joined #fuel-dev
08:31 amnk joined #fuel-dev
08:31 amnk joined #fuel-dev
08:41 sea`` joined #fuel-dev
08:41 sea`` left #fuel-dev
08:42 ekosareva joined #fuel-dev
08:54 amnk joined #fuel-dev
08:56 dnikishov joined #fuel-dev
09:00 Liuqing joined #fuel-dev
09:30 rvyalov_ joined #fuel-dev
09:32 xarses_ joined #fuel-dev
09:37 lezbar__ joined #fuel-dev
10:12 javeriak joined #fuel-dev
10:22 amnk joined #fuel-dev
10:23 amnk joined #fuel-dev
10:40 amnk joined #fuel-dev
10:55 agordeev zigo: hey. recently you asked about apt-key. Probably that could help https://review.openstack.org/#/c/275021
10:55 dpuchkin joined #fuel-dev
10:57 zubchick joined #fuel-dev
11:08 zubchick joined #fuel-dev
11:19 ekosareva joined #fuel-dev
11:40 zubchick joined #fuel-dev
12:39 amnk joined #fuel-dev
12:40 dshulyak joined #fuel-dev
12:51 ominakov joined #fuel-dev
12:54 ominakov_ joined #fuel-dev
13:28 zubchick joined #fuel-dev
13:54 dims joined #fuel-dev
14:19 ominakov Hi guys! I have comment from ikalnitsky in my patch for python-fuelclient (https://review.openstack.org/#/c/241990/) about Keystone endpoints. But I'm not sure what we should change in endpoints? Now we have http://{ip}:8000/api for nailgun, http://{ip}:8000/ostf for ostf, and auth url for Keystone we set in python-fuelclient https://github.com/openstack/python-fuelclient/blob/master/fuelclient/client.py#L41-L44
14:29 romcheg ominakov: looks in general it should work
14:37 ikalnitsky ominakov: i didn't get your question. change what? change where? in keystone we should register public nailgun/ostf endpoints, not internal one.
14:42 xarses_ joined #fuel-dev
14:48 ominakov ikalnitsky, which url we should use for public nailgun/ostf?
14:51 ikalnitsky ominakov: http://{ip}:8000/api for nailgun, http://{ip}:8000/ostf
14:51 ikalnitsky it's a correct one
14:52 ikalnitsky but we may have other ones registered in keystone
15:02 HeOS_ joined #fuel-dev
15:07 Liuqing joined #fuel-dev
15:08 ominakov ikalnitsky, looks like we have only this one https://github.com/openstack/fuel-library/blob/60d621b10c40a9c1c90a071e9d10da77ca5d1559/deployment/puppet/fuel/manifests/auth.pp#L62-L66 and https://github.com/openstack/fuel-library/blob/60d621b10c40a9c1c90a071e9d10da77ca5d1559/deployment/puppet/fuel/manifests/ostf/auth.pp#L62-L66
15:30 zigo agordeev: Thanks for pointing it out!
15:31 zigo agordeev: The issue with that patch, IMO, is that it's not enforcing the check of the key, therefore, there could be a man in the middle attack. We should 1/ download the key 2/ check it's fingerprint against something known, and ONLY THEN do 3/.
15:32 amnk joined #fuel-dev
15:32 zigo However, that's a pretty nice patch already.
15:34 zigo agordeev: How would I add my key (available as pubkey.gpg on my debian repo) when doing the provision then?
15:35 zigo agordeev: I have a use case where I could incorporate the patch right away and test it.
15:36 zigo agordeev: I see there's a --gpg-public-key new parameter, but then how do I get this passed to fa_build_image? Can I add such option in the openstack.yaml ?
15:38 zigo Oh, settings.yaml ?!?
15:38 zigo To me, it should go in openstack.yaml, because we may use a key for Ubuntu and one for Debian, for example...
15:40 xarses_ joined #fuel-dev
15:51 agordeev zigo: let me explain. there's a tool which will build a bootstrap image for you. So, you just pass this new parameter to it. https://github.com/openstack/fuel-specs/blob/master/specs/8.0/dynamically-build-bootstrap.rst
16:03 agordeev zigo: i missed the fact that proposed patch is about bootstrap images. Images for provisioning and the provisioning flow might not be ready for that at all. At least i didn't find any patches to nailgun which are related to gpg keys. it's the right time to file a bug about that.
16:04 zigo agordeev: My problem is that when I click on the deploy button, nailgun starts fa_build_image, and it fails because I'm using my own brewed Debian repo.
16:04 zigo agordeev: So, your patch wouldn't address that?
16:05 agordeev zigo: unfortunately not, it wouldn't
16:05 zigo :/
16:05 zigo agordeev: Any idea how I could do it? (a quick temporary hack so I can go further is fine...)
16:14 zubchick joined #fuel-dev
16:17 agordeev zigo: why did it require gpg check? i think that at least few components need to be changed in order to allow you to pass gpg key to provisioning images. it sounds like a feature request.
16:18 zigo agordeev: It's required, because "apt-get -y update" just crashes the whole process for me.
16:18 zigo It's supposed to be ignoring GPG errors, but in fact, it's not.
16:21 amnk joined #fuel-dev
16:21 agordeev could you share the log of fa-build-image? i suspect that it has nothing to deal with GPG errors (as those are probably only harmless warnings). it must be entirely another reason behind that.
16:31 IvanBerezovskiy left #fuel-dev
16:43 zigo agordeev: The problem is that now that I have fixed another issue, I get nothing happening, surprisingly.
16:44 zigo I get the deploying progress bar, and nothing more... :(
16:44 zigo Let me check for it again though.
16:45 zigo oh, I got this interesting one too:
16:46 zigo 2016-02-26 06:27:04.058 9513 TRACE fuel_agent.cmd.agent Stderr: "W: Failed to fetch http://192.168.4.115:9999/security/dists/jessie/updates/InRelease  Unable to find expected entry 'universe/binary-amd64/Packages' in Release file (Wrong sources.list entry or malformed file)\n\nE: Some index files failed to download. They have been ignored, or old ones used instead.\n"
16:46 zigo Not sure how to fix it...
16:46 zigo OF COURSE I don't have "universe", this is not Ubuntu ... :P
16:56 angdraug joined #fuel-dev
17:00 amnk joined #fuel-dev
17:01 zigo agordeev: http://sid.gplhost.com/fuel/fuel-agent-env-1.log.txt
17:01 zigo BTW, it'd be nice if fuel was not logging $OS_PASSWORD which I have in my env vars ... :/
17:03 zigo Also, is there a place where I can see the image, even if it failed?
17:05 agordeev zigo: if image building gets failed, then all leftovers (if any) will be removed as well.
17:05 zigo agordeev: Is there a way to get this feature disabled?
17:05 zigo Because it's hard to debug then...
17:07 agordeev zigo, have you sorted out the issue with wrong repo settings?
17:07 zigo agordeev: Yeah, I've rewritten openstack.yaml and use Debian stuff instead.
17:08 zigo ie: scrapped - pk: 1 and - pk: 2, then wrote my stuff there.
17:08 agordeev zigo, afaik, repo settings must be exposed somewhere at settings tab of your cluster in fuel UI
17:09 zigo Yes, but I don't really want to edit that each time I'm doing a test.
17:09 dims joined #fuel-dev
17:29 ilbot3 joined #fuel-dev
17:29 Topic for #fuel-dev is now Fuel Development http://wiki.openstack.org/wiki/Fuel | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel-dev/ | gerrit traffic @ #fuel-tracker
17:30 agordeev what lines does output of mco ping from the master node contain?
17:30 agordeev $ mco ping
17:32 zigo # mco ping
17:32 zigo 1 time=35.10 ms
17:32 zigo 2 time=37.54 ms
17:32 zigo fuelmit time=39.95 ms
17:33 zigo 1 and 2 are my slave nodes.
17:33 zigo fuelmit is the Debian master node.
17:33 zigo It's just the hostname of it.
17:33 zigo I don't really understand where 1 and 2 are from ...
17:35 zigo agordeev: Astute uses "bunny" no?
17:36 agordeev what is bunny?
17:37 zigo agordeev: It looks like a new lib replacing AMQP.
17:37 zigo ruby-amqp -> ruby-bunny
17:38 agordeev zigo, once slave node gets registered in nailgun, then nailgun will give the unique id for it, after that nailgun-agent reconfigure mcollective service on it. so, this is the way how it gets its id
17:39 zigo Gosh, the new astute got a bunch of new depends to work on... :/
17:39 * zigo packages that to Debian.
18:35 dshulyak joined #fuel-dev
18:42 krobzaur joined #fuel-dev
18:46 rvyalov_ joined #fuel-dev
19:09 krobzaur joined #fuel-dev
19:16 amnk joined #fuel-dev
19:17 xarses_ joined #fuel-dev
19:47 _nadya_ joined #fuel-dev
20:15 javeriak joined #fuel-dev
20:16 javeriak joined #fuel-dev
20:21 javeriak_ joined #fuel-dev
20:48 dims joined #fuel-dev
21:31 zubchick joined #fuel-dev
23:29 angdraug joined #fuel-dev
23:34 dims joined #fuel-dev
23:55 angdraug https://docs.fuel-infra.org/fuel-dev/develop/system_tests/tree.html
23:55 angdraug hm...

| Channels | #fuel-dev index | Today | | Search | Google Search | Plain-Text | summary