Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2015-08-31

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

All times shown according to UTC.

Time Nick Message
00:06 DevStok joined #fuel
00:06 DevStok hi
00:06 DevStok i'm getting
00:06 DevStok 10.20.0.14
00:06 DevStok ERROR neutron.agent.linux.async_process [-] An error occurred while communicating with async process [['ovsdb-client', 'monitor', 'Interface', 'name,ofport', '--format=json']].
00:25 tzn joined #fuel
00:48 nurla joined #fuel
01:34 xarses joined #fuel
01:40 gongysh joined #fuel
02:27 tzn joined #fuel
02:52 hakimo_ joined #fuel
03:27 Samos123 joined #fuel
04:28 tzn joined #fuel
04:41 stamak joined #fuel
04:44 vsedelnik joined #fuel
04:45 vsedelnik joined #fuel
04:46 vsedelnik joined #fuel
04:50 sergmelikyan joined #fuel
04:59 nurla joined #fuel
06:01 championofcyrodi joined #fuel
06:13 xarses_ joined #fuel
06:27 sergmelikyan joined #fuel
06:28 tzn joined #fuel
06:43 mkwiek joined #fuel
06:49 xarses_ joined #fuel
06:50 xarses_ joined #fuel
06:51 xarses_ joined #fuel
06:53 xarses joined #fuel
06:53 xarses joined #fuel
07:20 youellet__ joined #fuel
07:23 JoeStack joined #fuel
07:24 bazilza joined #fuel
07:39 ktychkova joined #fuel
07:41 jkirnosova_ joined #fuel
07:45 samuelBartel joined #fuel
08:36 alex_didenko left #fuel
08:41 saibarspeis joined #fuel
08:50 stamak joined #fuel
08:58 e0ne joined #fuel
09:09 e0ne joined #fuel
09:26 JoeStack joined #fuel
09:53 tzn joined #fuel
10:46 teran joined #fuel
10:52 vsedelnik joined #fuel
10:58 jaypipes joined #fuel
11:01 vsedelnik joined #fuel
11:30 vsedelnik joined #fuel
11:35 jaypipes joined #fuel
11:47 hyperbaba joined #fuel
11:48 kda joined #fuel
11:58 sergmelikyan joined #fuel
11:59 e0ne joined #fuel
12:11 kda hello chat, I have a question concerning Fuel. So far the deployment is x86/amd64 only. My goal is to use fuel for deployment on aarch64 (arm64). For that I got an x86 Fuel master node and I'm wondering what's the best way to go about deploying the functional nodes? So far I've tried manually installing nailgun-agent+mcollective on the arm64 node and got it detected (after some patches to Ohai) but I'm afraid I am configuring things improperly because no deployment
12:11 kda action is possible.
12:13 kda The Openstack itself works already on arm64, I've been able to deploy both Juno and Kilo using Devstack and even integrate them with OpenDaylight.
12:34 dklepikov joined #fuel
12:42 kda Different question then, is it possible to initiate fuel deployment skipping the OS installation? What would I need to put on the target node to prepare it for openstack deployment step?
12:47 DevStok joined #fuel
13:09 tzn joined #fuel
13:17 claflico joined #fuel
13:29 e0ne joined #fuel
13:36 DevStok hi cannot understand why, after modification to the network I cant associate a floatingIP to a VM(old) but I can associate a floatingIP to a NEW VM
13:36 DevStok I've set multiple network on icehouse
13:50 DevStok ERROR neutron.agent.linux.async_process [-] An error occurred while communicating with async process [['ovsdb-client', 'monitor', 'Interface', 'name,ofport', '--format=json']].
13:50 DevStok in neutron-plugin-openvswitch-agent.log
14:06 meow-nofer_ joined #fuel
14:06 asilenkov joined #fuel
14:43 sbog kda: regarding to initiate fuel deployment skipping os installation - i'm afraid it will be really tricky, cause yaml files created for every node and you cannot apply puppet manifests w/o data for them
14:45 sbog goukos: /usr/lib/python2.7/dist-packages/nova/scheduler/ on controller node
14:47 kda sbog: So that would mean that I am left with an option of creating an alternative bootstrap and base image for my platform (arm64). I presume the best place to start would be the fuel-main repo on github?
14:48 sbog shane_: for 6.1 there is slightly another process for deploy. First, files for image for target OS should be downloaded (in case of Ubuntu), then image will builded on master node, then it will transferred to target node, writing on it and then node will be rebooted. It get some time (much time in case of Ubuntu and slow connection to upstream repos)
14:52 sbog kda: probably, yes. Or, for example, you can try next trick: deploy configuration you need on VMs and then get data sources for puppet (/etc/astute.yaml should be enough, I think). Then you can place this data onto your arm nodes with preinstalled OS and run puppet manifests one by one, step by step. Right sequence for puppet apply for every node you can get from https://github.com/stackforge/fuel-library/blob/master/utils/task_graph/task_graph.py
14:54 sbog jidar: hey. I'm afraid that answer is 'no'. But try to ask more narrow question - dev's can try to explain how it works.
14:56 kda sbog: thanks, I will try the manual approach
14:57 rmoe joined #fuel
14:59 sbog kda: np. If something will go wrong - ask here, devs will be glad to help.
15:22 blahRus joined #fuel
15:34 fedexo joined #fuel
15:35 jidar sbog: yea, me and another guy tracked down some of the VRRP stuff and how it relates to namespaces but it's a tricky (and interesting) setup for sure
15:53 pbrooko joined #fuel
16:01 sergmelikyan joined #fuel
16:22 jaypipes joined #fuel
16:22 skylerberg joined #fuel
16:26 sergmelikyan joined #fuel
16:47 xarses joined #fuel
17:10 tzn joined #fuel
17:17 sergmelikyan joined #fuel
17:26 e0ne joined #fuel
17:34 tzn joined #fuel
17:38 e0ne joined #fuel
17:50 tzn joined #fuel
18:20 sergmelikyan joined #fuel
18:58 tzn joined #fuel
19:43 stamak joined #fuel
20:17 sergmelikyan joined #fuel
20:19 kaliya joined #fuel
20:20 HeOS joined #fuel
20:27 osryan joined #fuel
20:40 xarses joined #fuel
20:56 teran joined #fuel
21:59 sbfox joined #fuel
22:04 sergmelikyan joined #fuel
22:08 sergmelikyan joined #fuel
22:18 sergmelikyan joined #fuel
22:21 sergmelikyan joined #fuel
22:52 rmoe joined #fuel
22:53 jobewan joined #fuel
23:01 tzn joined #fuel
23:50 tzn joined #fuel

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