Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2016-08-04

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

All times shown according to UTC.

Time Nick Message
00:52 code-R joined #fuel
00:54 code-R_ joined #fuel
00:59 xarses joined #fuel
01:23 fatdragon joined #fuel
01:30 fatdragon joined #fuel
01:48 ilbot3 joined #fuel
01:48 Topic for #fuel is now Fuel 8.0 (Liberty) https://www.fuel-infra.org/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
02:00 fritchie joined #fuel
02:02 dmellado joined #fuel
04:00 fritchie joined #fuel
04:39 Alcest joined #fuel
04:59 javeriak joined #fuel
05:42 reddydodda joined #fuel
05:43 reddydodda Hi Everyone, Can some one guide me how to setup provider network using fuel9.0 ??
05:43 reddydodda or how i can add second external network using fuel9.0 ? can some one help me with some document or links.
05:52 maestropandy joined #fuel
05:57 kelepirci hello reddydodda. Adding second external network is not possible with fuel. but you can easily add second network from horizon or cli
05:58 kelepirci but adding second network is different depends on what type of network are you going to add.
06:00 fritchie joined #fuel
06:00 kelepirci you can use this question to solve your problem: https://ask.openstack.org/en/question/87867/after-fuel-7-deploy-how-to-create-multi-external-network/
06:01 javeriak joined #fuel
06:08 reddydodda joined #fuel
06:08 reddydodda kelepirci : Thank you for your value reply.
06:09 reddydodda kelepirci : i am using vlan
06:10 reddydodda and also my external network has vlan tagging for second external subnet. ex: Primary Network : 192.168.1.0/24 ( untagged on eth3 )
06:11 reddydodda second public network is 192.168.2.0/24 ( tagged 200 on eth3 )
06:11 code-R joined #fuel
06:16 Miouge joined #fuel
06:24 kelepirci that might be problem. I have not created tagged and untagged publice network before.
07:05 javeriak joined #fuel
07:05 code-R joined #fuel
07:13 vkulanov joined #fuel
07:17 code-R_ joined #fuel
07:25 HeOS joined #fuel
07:43 DavidRama joined #fuel
08:00 fritchie joined #fuel
08:08 DavidRama1 joined #fuel
08:15 noshankus joined #fuel
08:20 yassine joined #fuel
08:30 javeriak joined #fuel
08:31 fatdragon joined #fuel
09:01 fatdragon joined #fuel
09:09 javeriak joined #fuel
09:09 nokes joined #fuel
09:14 ekosareva joined #fuel
09:23 xek_ joined #fuel
09:40 Billias joined #fuel
09:41 fatdragon joined #fuel
09:49 permalac joined #fuel
10:00 fritchie joined #fuel
10:20 fatdragon joined #fuel
10:42 kelepirci joined #fuel
10:54 Miouge joined #fuel
11:21 ekosareva joined #fuel
11:41 xek__ joined #fuel
12:01 fritchie joined #fuel
12:21 fatdragon joined #fuel
12:31 vkulanov romcheg, can you please review/merge backport https://review.openstack.org/#/c/349984/ and here I've addressed your comments https://review.openstack.org/#/c/346454/
12:32 romcheg vkulanov: as soon as I'm back from the clinic
12:34 akscram Guys, take a look on one line backport to stable/mitaka for fuel-library: https://review.openstack.org/#/c/351150/
12:36 akscram sbog, mmalchuk, mwhahaha, bogdando ^^
12:51 Billias joined #fuel
12:57 javeriak joined #fuel
13:17 Liuqing joined #fuel
13:22 fatdragon joined #fuel
13:52 javeriak joined #fuel
13:52 v1k0d3n joined #fuel
14:00 fritchie joined #fuel
14:01 maestropandy left #fuel
14:09 code-R joined #fuel
14:11 code-R_ joined #fuel
14:18 Billias joined #fuel
14:20 johnavp19891 joined #fuel
14:26 Billias left #fuel
14:34 javeriak joined #fuel
14:38 xarses joined #fuel
14:48 Liuqing joined #fuel
14:55 code-R joined #fuel
14:56 code-R_ joined #fuel
15:04 thiagolib hello someone has set up the cinder-volume to run on OpenStack mirantis pacemaker that is installed through the fuel
15:05 Liuqing joined #fuel
15:24 fatdragon joined #fuel
15:30 krypto joined #fuel
15:33 fatdragon joined #fuel
15:43 noshankus joined #fuel
15:45 byrdog55 joined #fuel
15:45 code-R joined #fuel
15:48 byrdog55 Hello All, I’m troubleshooting a plugin that works in Fuel 8.0 but does not in 9.0 and I noticed that I have a required_for:
15:48 byrdog55 - openstack-network-server-config statement but I no loger see that Task ID anywhere.  It used to be in ./modules/osnailyfacter/modular/openstack-network/tasks.yaml
15:53 byrdog55 is there a yaml file that is created that lists the tasks, id and priorty that is generated by Nailgun?
16:00 fritchie joined #fuel
16:01 romcheg vkulanov: ping
16:01 romcheg vkulanov: could you please join #openstack-meeting-alt ?
16:05 yottatsa joined #fuel
16:30 Miouge joined #fuel
16:36 kelepirci joined #fuel
16:37 fritchie joined #fuel
16:41 Billias joined #fuel
16:43 gomarivera joined #fuel
17:01 xarses dpyzhov: ikutukov kozhukalov do you want to continue discussing?
17:01 agordeev folks, i have something to add
17:01 dpyzhov xarses: so our plan is to deliver 9.2 as a portable release that can be used on newton master node, right?
17:01 dpyzhov or maybe 9.3
17:01 xarses dpyzhov: that's my vision
17:02 ikutukov why not
17:02 dpyzhov that's another pro for limiting stable/mitaka scope )
17:03 dpyzhov xarses: but what about creation of list of bugs for backport in order to see the scope of work for 9.1?
17:03 kozhukalov i am afraid this approach to consume non library stuff from master does contradict to out intent to deliver very stable and short releases 9.x
17:03 kozhukalov because this approach requires a lot of testing
17:04 xarses we'd be releasing the non-library components more frequently
17:04 ikutukov dpyzhov, well, i think one minor release we will spend on stabilizing of  decoupled release format and its interaction with Nailgun. Recall the tasks.yaml as an example of early format.
17:04 xarses why, we release frequently and test well
17:04 kozhukalov xarses: for community it is ok, but not for stable 9.x
17:04 fritchie joined #fuel
17:04 xarses from the not library release standpoint, its the most risk, not the rest of fuel
17:04 kozhukalov it usually takes 2 months to stabilize and test code
17:05 xarses because we don't release often enough to force ourselves to not have those problems
17:06 dpyzhov there are bugs outside library. and we need stable version of nailgun, astute and other stuff. we can't take fresh master and call it 'stable because we have stable puppet manifests'
17:06 kozhukalov no, that's because we have 3000 test cases which require a lot of time to run and then  analize
17:07 kozhukalov every test cycle takes usually at least couple days
17:08 ikutukov automation of 90% scenarios is a great investment. The last 10% - is a loss, that is cheaper to cover manually otherwise we will need to re-create all system logic as tests)
17:08 kozhukalov yes, there are bugs in python, it is not only library
17:11 xarses It is about automation, if we want to be more stable and predictable then we need to get more rigorous with the code that goes into master. If we have a long release cycle we can be lax about automation then we have to make otherways to control what is stable and how to keep stable that way
17:11 kozhukalov fyi, we are only couple months of divergence in master from mitaka and swarm is now only 55% before merger of xenial patches
17:12 kozhukalov it was 97% when we released 9.0
17:13 kozhukalov xarses: yes, you are right about automation but reality is that we don't have enough resources to improve this soon
17:15 xarses ok, so this isn't solving anything
17:20 agordeev folks, will we have time for open discussions?
17:21 xarses agordeev: go ahead
17:21 agordeev folks. perhaps it's the right time to get rid of cloud-init out of provisioning flow at least for 10 as this change may be way too invasive for 9.x
17:21 agordeev cloud-init only added a lot of serious issues that we were unable to forecast in the early days of IBP. This may sound strange, but,frankly speaking, i can't find any pros of its use in case of Fuel.
17:21 agordeev i just only started etherpad about that, there's no spec/bp yet. https://etherpad.openstack.org/p/fuel-agent-without-cloud-init
17:21 agordeev i'll raise the same questions about the usage of cloud-init in dev mailing list to get the wider audience.
17:22 agordeev i mean, there's nothing wrong with the cloud-init itself. i'm not blaming it. it just very barely fits to Fuel use cases. We've picked up a wrong tool and using it in a wrong way.
17:23 agordeev it's getting harder and trickier to adopt the use of cloud-init for the sake of provisioning. that's all
17:26 yottatsa_ joined #fuel
17:31 javeriak joined #fuel
17:32 Miouge joined #fuel
17:36 xarses so what is ironic doing?
17:38 agordeev ironic driver for fuel-agent?
17:39 agordeev what ironic entity you're referring to? could you clarify?
17:40 xarses no, i mean how does Ironic / ironic-agent deal with configuring the network after the node has it's image written
17:45 agordeev i'm not SME for Ironic/IPA, but seems like it doesn't perform that. perhaps, it thoroughly relies on dhcp and neutron.
17:46 agordeev that's fine for cloud use case. but it's not suitable for fuel.
18:00 fritchie joined #fuel
18:02 xarses we only need it to run long enough for puppet to enfoce the configuration, we where just re-using dhcp
18:02 xarses previously
18:03 xarses so what are you proposing that we use instead of cloud-init?
18:04 agordeev i'm proposing puppet of course. since it gonna be sitting inside of bootstrap image too it's worth to re-use it for that purposes.
18:05 agordeev so then, node configuration could be thoroughly delegated to fuel-library. and will be managed from one place. it's better that what we have for now.
18:06 agordeev well, technically, it should be a separate task in the middle of provisioning flow.
18:22 javeriak joined #fuel
18:38 fatdragon joined #fuel
18:49 yottatsa joined #fuel
19:03 fritchie joined #fuel
19:16 javeriak joined #fuel
19:19 AJatDS joined #fuel
19:19 AJatDS hello
19:20 AJatDS hopefully qq for the channel, for some reason this morning the pacemaker cluster on my controllers was completely stopped
19:20 AJatDS I have most of it recovered except I can't get mysql to start on one of the controllers
19:21 AJatDS no idea why
19:21 AJatDS hopefully this is the right channel
19:24 DavidRama joined #fuel
19:31 DavidRama joined #fuel
19:31 preilly joined #fuel
19:39 fatdragon joined #fuel
20:00 fritchie joined #fuel
20:09 gomarivera joined #fuel
20:11 DavidRama joined #fuel
20:11 fatdragon joined #fuel
20:14 DavidRama joined #fuel
20:22 xarses you can try looking at the pacemaker log and see if it has something interesting to say
20:23 xarses otherwise you can use pcs_resouce to start the the serivce with the error log in copied to the terminal, odds are something is hiding there
20:25 HeOS joined #fuel
20:28 DavidRama left #fuel
20:57 gomarivera joined #fuel
21:02 Egyptian[Home] joined #fuel
21:47 GMAzrael joined #fuel
22:00 fritchie joined #fuel
22:31 DavidRama joined #fuel
22:31 DavidRama left #fuel
23:14 code-R joined #fuel
23:20 johnavp1989 joined #fuel

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