Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2015-07-21

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

All times shown according to UTC.

Time Nick Message
00:10 dims_ joined #fuel-dev
00:14 angdraug joined #fuel-dev
00:38 user01 joined #fuel-dev
00:39 rmallah @barthalion 403  Forbidden\nErr http://10.20.0.2:8080/mos-ubuntu/ mos6.1-updates/main nailgun-net-check all 6.1.0-7132.1\n  403  Forbidden\n'
00:39 rmallah 2015-07-20 13:17:04.274 730 TRACE fuel_agent.cmd.agent Stderr: 'E: Failed to fetch http://10.20.0.2:8080/mos-ubuntu/pool/main/a/astute/nailgun-mcagents_6.1.0-642.1_all.deb  403  Forbidden\n\nE: Failed to fetch http://10.20.0.2:8080/mos-ubuntu/pool/main/n/nailgun/nailgun-agent_6.1.0-7132.1_all.deb  403  Forbidden\n\nE: Failed to fetch http://10.20.0.2:8080/mos-ubuntu/pool/main/n/nailgun/nailgun-net-check_6.1.0-7132.1_all.deb  403  Forbidden\n\nE: Unable
00:39 rmallah to fetch some archives, maybe run apt-get update or try with --fix-missing?\n'
00:39 rmallah @barthalion i find that many of the folders under mos-ubuntu have non-executable permissions.
00:47 Bomfunk joined #fuel-dev
02:08 neophy joined #fuel-dev
02:24 rmallah @barthalion , ubuntu got installed , now openstack getting installed.
02:26 tzn joined #fuel-dev
02:38 mgolub joined #fuel-dev
03:04 rmallah @barthalion install of openstack on controller failed with following message "Deployment has failed. Method granular_deploy. Deployment failed on nodes 1.
03:04 rmallah Inspect Astute logs for the details"
03:09 dims_ joined #fuel-dev
03:15 rmallah latest errors are:
03:15 rmallah (/Stage[main]/Main/Openstack::Network::Create_router[router04]/Neutron_router_interface[router04:net04__subnet]/ensure) change from absent to present failed: did not get expected message on interface addition, got
03:28 dims_ joined #fuel-dev
04:06 dims_ joined #fuel-dev
04:13 dims_ joined #fuel-dev
04:27 dims_ joined #fuel-dev
04:50 mkwiek joined #fuel-dev
04:56 angdraug joined #fuel-dev
05:02 dims_ joined #fuel-dev
05:22 dims_ joined #fuel-dev
05:36 dims_ joined #fuel-dev
05:58 ashtokolov joined #fuel-dev
06:07 alex_bh joined #fuel-dev
06:21 dims_ joined #fuel-dev
06:22 salmon_ joined #fuel-dev
06:27 rmallah joined #fuel-dev
06:27 rmallah bartha ayt?
06:34 ub joined #fuel-dev
06:42 stamak joined #fuel-dev
06:43 ub joined #fuel-dev
06:44 amnk joined #fuel-dev
06:47 hyperbaba joined #fuel-dev
06:56 dims_ joined #fuel-dev
07:05 tzn joined #fuel-dev
07:06 neophy joined #fuel-dev
07:11 dancn joined #fuel-dev
07:16 ub joined #fuel-dev
07:37 dims_ joined #fuel-dev
08:12 ddepaoli joined #fuel-dev
08:20 rmallah @barthalion http://paste.openstack.org/show/394771/
08:20 monester joined #fuel-dev
08:22 barthalion rmallah: could you upload diagnostic snapshot somewhere?
08:22 rmallah yep glad to hear from
08:22 rmallah u
08:29 hyperbaba joined #fuel-dev
08:31 dims_ joined #fuel-dev
08:32 pboldin joined #fuel-dev
08:39 rmallah @barthalion any preferred place where i can dump the snap.
08:39 barthalion rmallah: anywhere, really; dropbox, gdrive, whatever you use
08:40 rmallah i liked http://paste.openstack.org/
08:40 barthalion you can't upload a tarball there
08:40 rmallah :-) i undestand.
08:46 dims_ joined #fuel-dev
08:50 HeOS joined #fuel-dev
08:52 romcheg joined #fuel-dev
08:53 pboldin joined #fuel-dev
08:57 rmallah_ joined #fuel-dev
08:58 rmallah_ @barthalion i am uploading to a site its gonna take some time.
08:58 rmallah_ the file is around 112MB is it expected to be that fat?
08:59 aderyugin joined #fuel-dev
08:59 barthalion yeah, not surprising
09:00 rmallah_ bartha i had to set 755 to many folders inside the nginx in order to install the ubuntu 14.04 on nodes during provisioning . I feel it is a bug with 6.1 stable , pls comment ...
09:01 barthalion no idea – never needed it
09:02 ashtokolov joined #fuel-dev
09:02 rmallah_ i did post to this room lemme see the chat logs.
09:02 barthalion saw it this morning, but nothing comes to my mind
09:03 barthalion 755 doesn't sound right anyway, you don't need executable bit on file to download it
09:04 rmallah_ hmm not in files i said folders
09:04 rmallah_ http://irclog.perlgeek.de/fuel-dev/2015-07-21#i_10928001
09:04 barthalion ah
09:04 rmallah_ the folders were 644
09:04 rmallah_ i did
09:05 rmallah_ find <dir> -type d -exec chmod 755 {} \;
09:05 barthalion 644 should be fine too, actually
09:05 barthalion no idea, sorry
09:06 rmallah_ no 644 on dirs is not fine
09:06 BobBall joined #fuel-dev
09:06 rmallah_ it wont allow the contents of dir to be read , historically reading on dirs requires execution rights in unix/linux :-)
09:07 barthalion true that
09:07 rmallah_ and hence nginx responded with 403 (forbidden)
09:08 huazhihao joined #fuel-dev
09:08 BobBall Hi All - we're working with Andrey to set up a plugin that creates a new release so we can make substantial changes to the wizard.  However after creating this new release we can't deploy it as Fuel says "Deployment tasks not found" - how are deployment tasks created and why don't they exist for a custom release?
09:15 e0ne joined #fuel-dev
09:17 rmallah_ barthalion where / how should i be reporting the issue of folder permissions that atleast forced a manual intervention in my case
09:25 hyperbaba joined #fuel-dev
09:26 dims_ joined #fuel-dev
09:27 monester joined #fuel-dev
09:28 rmallah_ @barthalion the snap is here:  http://www.mediafire.com/download/2ch91dae3dgpu6r/fuel-snapshot-2015-07-21_08-24-42.tar.xz
09:28 tzn joined #fuel-dev
09:32 barthalion rmallah_: great, I'll try to take a look today
09:32 barthalion about bug reports… https://bugs.launchpad.net/fuel/
09:33 rmallah_ thanks barthalion
09:36 dims_ joined #fuel-dev
09:36 rmallah_ do u feeel i should file a bug about the folder issue or is it known?
09:43 monester joined #fuel-dev
09:54 romcheg joined #fuel-dev
09:58 tzn joined #fuel-dev
10:04 rmallah_ @barth i created a bug report https://bugs.launchpad.net/fuel/+bug/1476593
10:06 stamak joined #fuel-dev
10:16 dims_ joined #fuel-dev
10:58 monester joined #fuel-dev
11:10 diurchenko joined #fuel-dev
11:16 pboldin joined #fuel-dev
11:21 HeOS joined #fuel-dev
11:23 amnk joined #fuel-dev
12:13 _tzn joined #fuel-dev
12:23 Miouge joined #fuel-dev
12:23 ashtokolov joined #fuel-dev
12:30 amnk joined #fuel-dev
12:38 rmallah joined #fuel-dev
12:51 rmallah joined #fuel-dev
12:56 alex_bh joined #fuel-dev
13:23 aviramb joined #fuel-dev
13:46 ashtokolov joined #fuel-dev
13:46 swann Hi Fuelers, I've a question about hiera: what is the purpose of the /etc/hiera/nodes.yaml ? since the same data is present in astute.yaml
13:49 mwhahaha if you look at the hiera load order, it's an override
13:49 mwhahaha it overrides what is in astute and globals based on the hierarchy
13:50 mwhahaha go checkout /etc/hiera.yaml
13:50 rbrooker joined #fuel-dev
13:51 claflico joined #fuel-dev
13:53 stamak joined #fuel-dev
13:53 dims_ joined #fuel-dev
14:03 asteer joined #fuel-dev
14:04 swann mwhahaha: yes I saw the hierarchy, actualy that's my issue. when nodes properties are changed (IPs) before deployment, nodes.yaml is not updated but astute.yaml is ok
14:04 swann in the context of plugins this failed ('hiera nodes' return old values)
14:05 swann mwhahaha: see my comment on https://review.openstack.org/#/c/198839/2
14:05 mwhahaha i'm not that familiar with the nodes file
14:05 mwhahaha i wonder what generates it
14:06 mwhahaha so astute can't be higher than global because global uses astute to generate those
14:06 mwhahaha you could move nodes down, but astute must be last
14:06 swann nodes.yaml seems to be generated at post_deployment tasks, so OpenStack deployment is not impacted but plugins deployement are in bad shape
14:06 mwhahaha eh no one likes plugins anyway ;)
14:07 * mwhahaha keeps an eye out for Sheena
14:08 mwhahaha so what plugins needs that node list?
14:08 swann mwhahaha: I like plugins ;) and several external contributors like plugins as well
14:09 swann mwhahaha: I guess several plugins need to use hiera('nodes')
14:09 mwhahaha so for the plugin, why can't you add your own task to generate nodes early?
14:09 swann lma collector use nodes list for example
14:10 swann mwhahaha: because it's not the responsability of plugins to manage hiera
14:10 swann s/manage/configure/
14:10 mwhahaha uh sometimes it is
14:11 mwhahaha we're doing it to split out some roles and we're managing that data
14:11 mwhahaha So either we move the nodes.yaml generation in the deployment earlier, or you can do your own in the plugin or you find another way to get that information
14:12 swann mwhahaha: or astute update in the same time nodes.yaml correctly like it does for astutes.yaml
14:12 mwhahaha no astute is run pre-deployment and generated once and never updated again
14:12 mwhahaha hence the hierarchy
14:13 swann but I like the first solution: move down 'nodes' in the hierarchy
14:13 swann astute doesn't run post-deployment tasks ?
14:14 mwhahaha no
14:14 swann who does?
14:14 mwhahaha astute.yaml is generated first
14:14 mwhahaha and that's what gets shipped to start the deployment
14:14 mwhahaha this seems like a good topic for openstack-dev so we can get some history behind the hiera stuff and the ordering
14:14 mwhahaha my understanding is only what i've figured out so far
14:15 mwhahaha but my understanding is that astute.yaml is generated as part of the deployment, shipped to the nodes to kick off the deployment
14:15 mwhahaha that's why it has the tasks an stuff
14:15 mwhahaha globals is generated as part of a pre-deployment task
14:15 mwhahaha and nodes later, so if you kill a controller or move things around the node.yaml gets updated
14:16 mwhahaha so then tasks later will get the updated environment node information
14:17 swann ok, I understand the purpose of nodes.yaml now, thanks.
14:18 swann But in the particular case where the env is updated before the deployment, nodes.yaml is wrong
14:18 mwhahaha so i'm not sure what exactly you're running into where you're geting the wrong nodes information
14:20 swann - not me but a user of our LMA plugins - runs into this issue because he updates all IPs of nodes before deployment
14:22 swann I now that it's not a usual way to deploy a cluster but Fuel allow it (updating IPs of nodes) before deployment
14:22 mwhahaha that sounds like a terrible idea
14:24 swann mwhahaha: yes I guess .. but who knows their constraints/requirements !?
14:24 mwhahaha i'm sure it's (not) documented somewhere
14:24 mwhahaha ;)
14:26 swann mwhahaha: indeed I didn't find any doc.. :(
14:26 swann so to summarize:
14:27 swann 1/ we cannot move down 'nodes' after 'astute' to fix the issue because it will break every things after an update of the env
14:27 mwhahaha if the plugin is making ip address updates, i would argue that it should be running a task to update nodes.yaml
14:28 swann it's not the plugin who updates ip addresses ..
14:28 mwhahaha or triggering something to do that as it's fundamentally altering the environment
14:28 mwhahaha who is then?
14:28 swann the end-user
14:28 mwhahaha how
14:29 swann because he needs it I guess
14:29 mwhahaha but how is he making that change
14:29 mwhahaha brb
14:29 swann fuel --env 1 deployment default
14:29 pasquier-s mwhahaha, swann, definitely need to discuss that on the ml so samuelBartel will be able to provide all the details :-)
14:29 swann .. updating ip addresses ..
14:30 swann fuel --env 1 deployment upload
14:30 swann fuel deploy-changes
14:31 julien_ZTE joined #fuel-dev
14:31 pasquier-s iiuc they're doing this way since Fuel 5.x because the IP addressing done by Fuel was too inflexible for their requirements
14:31 mwhahaha so that deploy-changes should trigger a yaml update somehow
14:32 swann pasquier-s: just discussing on the right channel before
14:32 mwhahaha that's probably the more prudent way to do this
14:32 swann mwhahaha: yes it triggers only the update of astute.yaml
14:32 mwhahaha something like this is totally one of those *there be dragons* things
14:32 pasquier-s swann, sure I wasn't trying to stop the discussion ;-)
14:34 swann pasquier-s: whew :D
14:36 mwhahaha anyway yea we shouldn't be moving the hiera ordering as that's going to break a whole lot of stuff
14:37 mwhahaha we need to properly identify what is occuring and see if we need to trigger additional updates
14:40 swann mwhahaha: yes, so I'll open a new bug with the use case and the description to reproduce, and continue the discussion elsewhere
14:40 swann thanks mwhahaha
14:40 mwhahaha sounds good
14:49 Miouge joined #fuel-dev
14:50 mattymo joined #fuel-dev
14:51 mattymo joined #fuel-dev
14:52 mattymo_ joined #fuel-dev
14:53 mattymo joined #fuel-dev
14:53 Miouge joined #fuel-dev
14:54 mattymo_ joined #fuel-dev
14:55 skolekonov joined #fuel-dev
14:55 mattymo joined #fuel-dev
14:57 mattymo_ joined #fuel-dev
14:57 mattymo joined #fuel-dev
14:57 skolekonov left #fuel-dev
14:58 skolekonov joined #fuel-dev
14:58 mattymo_ joined #fuel-dev
14:59 mattymo joined #fuel-dev
15:01 mattymo_ joined #fuel-dev
15:01 angdraug joined #fuel-dev
15:01 mattymo joined #fuel-dev
15:03 mattymo_ joined #fuel-dev
15:03 ashtokolov joined #fuel-dev
15:03 mattymo joined #fuel-dev
15:04 ashtokol_ joined #fuel-dev
15:05 mattymo_ joined #fuel-dev
15:05 ashtokolov joined #fuel-dev
15:05 mattymo joined #fuel-dev
15:07 mattymo_ joined #fuel-dev
15:07 xarses joined #fuel-dev
15:07 mattymo joined #fuel-dev
15:09 mattymo_ joined #fuel-dev
15:09 mattymo joined #fuel-dev
15:11 mattymo joined #fuel-dev
15:13 mattymo joined #fuel-dev
15:15 mattymo joined #fuel-dev
15:21 mattymo joined #fuel-dev
15:23 xenolog joined #fuel-dev
15:31 mattymo joined #fuel-dev
15:33 mattymo joined #fuel-dev
15:50 julien_ZTE joined #fuel-dev
15:51 julien_ZTE joined #fuel-dev
16:28 amnk joined #fuel-dev
16:51 julien_ZTE joined #fuel-dev
17:07 mattymo1 joined #fuel-dev
17:17 tzn joined #fuel-dev
17:37 monester joined #fuel-dev
17:39 romcheg joined #fuel-dev
17:48 bitblt joined #fuel-dev
17:50 HeOS joined #fuel-dev
17:51 bitblt joined #fuel-dev
17:57 HeOS joined #fuel-dev
17:58 bitblt joined #fuel-dev
18:02 jaypipes joined #fuel-dev
18:04 amnk joined #fuel-dev
18:12 rvyalov joined #fuel-dev
18:12 Samos123 joined #fuel-dev
18:13 izinovik joined #fuel-dev
18:14 akislitsky joined #fuel-dev
18:15 vkramskikh joined #fuel-dev
18:15 amnk joined #fuel-dev
18:16 meow-nofer joined #fuel-dev
18:16 xenolog joined #fuel-dev
18:16 monester joined #fuel-dev
18:17 mihgen joined #fuel-dev
18:17 skylerberg joined #fuel-dev
18:18 evg joined #fuel-dev
18:19 sbog joined #fuel-dev
18:20 MorAle joined #fuel-dev
18:21 MiroslavAnashkin joined #fuel-dev
18:22 apalkina joined #fuel-dev
18:32 ashtokolov joined #fuel-dev
18:35 ashtokol_ joined #fuel-dev
18:42 ashtokolov joined #fuel-dev
18:44 amnk joined #fuel-dev
18:47 xarses need reviewers for https://review.openstack.org/#/c/203270/
18:48 rmoe joined #fuel-dev
18:48 xarses also https://review.openstack.org/#/c/203847/
18:52 julien_ZTE joined #fuel-dev
19:04 pboldin1 joined #fuel-dev
19:04 e0ne joined #fuel-dev
19:15 stamak joined #fuel-dev
19:30 ashtokolov joined #fuel-dev
19:35 pboldin joined #fuel-dev
19:47 amnk joined #fuel-dev
20:10 mmosesohn_ joined #fuel-dev
20:15 amnk_ joined #fuel-dev
20:17 amnk joined #fuel-dev
20:19 amnk_ joined #fuel-dev
20:21 salmon_ joined #fuel-dev
20:21 amnk joined #fuel-dev
20:24 amnk_ joined #fuel-dev
20:27 amnk joined #fuel-dev
20:27 e0ne joined #fuel-dev
20:28 amnk_ joined #fuel-dev
20:31 skylerberg joined #fuel-dev
20:32 amnk joined #fuel-dev
20:33 skylerberg I am trying to set up a fuel master to test out a plugin I am developing, but I am having no luck. Is there a guide to install a fuel master? I have tried using the fuel ISO, and my VM hangs on boot. I also tried fuel-dev-tools vagrant, and salt-call state.highstate --local has 4 failures.
20:34 mwhahaha https://wiki.openstack.org/wiki/Fuel/How_to_contribute#Fuel_development_environments
20:39 skylerberg mwhahaha: Okay, I will try the iso again. I was wondering if I was missing something.
20:39 mwhahaha which iso and how are you using it?
20:39 mwhahaha i've been successful using virtualbox on osx
20:39 mwhahaha other use the full fuel-devops stuff with kvm on ubuntu
20:41 skylerberg mwhahaha: I downloaded the 6.0 iso from torrent and I am using virtualbox on osx. Last time I installed it hanged on boot after installing. There might have been a problem with my virtual disk not being entirely clean though, so I am running again now.
20:41 mwhahaha you should try 6.1
20:41 mwhahaha unless you're targeting 6.0 for your plugin
20:42 skylerberg mwhahaha: I was just following a link I found. Would I be better off looking for a 6.1 iso, or should I build the iso from trunk?
20:43 mwhahaha depends on what you're targeting. 6.1 is current stable
20:44 mwhahaha trunk is currently being worked on so your milage may vary depending on the day
20:44 skylerberg mwhahaha: Thanks. It looks like it is working better this time. I will start looking for a 6.1 image and try developing against that.
20:46 mwhahaha there have been a bunch of improvements for 6.1 and 7.0
20:46 mwhahaha so depending on what you are trying to do 7.0 (master) may be better
20:53 julien_ZTE joined #fuel-dev
20:57 skylerberg Well, this plugin should be pretty straightforward, so I am hoping I wont need any newly added features.
21:00 HeOS joined #fuel-dev
21:14 e0ne joined #fuel-dev
21:15 ashtokolov joined #fuel-dev
21:16 ashtokolov joined #fuel-dev
21:16 e0ne joined #fuel-dev
21:25 julien_ZTE joined #fuel-dev
21:26 mwhahaha ok who is the wise guy who thought to make sort -u the way to sort the requirements-rpm.txt file for test-requirements-rpm? sort -u on OSX evidently != sort -u on Linux
21:28 obcecado what's the difference? the output order?
21:29 mwhahaha https://review.openstack.org/#/c/204266/2..3/requirements-rpm.txt
21:29 mwhahaha yea
21:30 mwhahaha case senstivity as well
21:35 obcecado the problem lies in the symbols
21:36 obcecado eg: @
21:36 mwhahaha 2 is OSX, 3 is linux
21:36 obcecado otherwise sort -fu should do it
21:36 mwhahaha yea
21:36 mwhahaha i think the symbols should be at the top anyway
21:37 mwhahaha -ufd
21:37 mwhahaha fixes it
21:37 mwhahaha i guess -d is the default for linux
21:40 obcecado seems osx sort is the same as openbsd's
21:40 obcecado :)
22:04 ashtokolov joined #fuel-dev
22:19 omolchanov joined #fuel-dev
22:38 barthalion joined #fuel-dev
22:41 _stowa joined #fuel-dev
22:48 mihgen xarses: holser: do you guys know anything about this https://bugs.launchpad.net/fuel/7.0.x/+bug/1456383 ?
22:49 mihgen it looks more like a feature to me, performance improvement
22:50 mihgen holser: what do we do about "ten Openstack components require 16k connection" in https://bugs.launchpad.net/mos/7.0.x/+bug/1475433? Sounds like we need to cross check this with oslo.db team & puppet-openstack teams
23:20 rmallah joined #fuel-dev
23:32 rmallah left #fuel-dev
23:33 xarses joined #fuel-dev
23:53 mihgen sgolovatiuk: ^^^
23:54 sgolovatiuk mihgen: it's a feature. it's not a bug
23:54 mihgen sgolovatiuk: which one?
23:55 sgolovatiuk https://bugs.launchpad.net/fuel/7.0.x/+bug/1456383
23:55 mihgen should we mark it with "feature" tag then.. ?
23:55 mihgen also, I'd say it's rather incomplete. In my opinion. I don't really see performance measurements / reasons why we need it

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