Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2014-07-10

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

All times shown according to UTC.

Time Nick Message
00:04 aglarendil yep. it is rabbitmq partitioning
00:09 aglarendil let's upload 307th ISO to CI
00:10 aglarendil we have Centos HA broken and >2 controllers Ubuntu broken, but it will work with our CI tests
00:31 IlyaE joined #fuel-dev
01:00 xarses joined #fuel-dev
01:38 eshumakher joined #fuel-dev
02:32 IlyaE joined #fuel-dev
03:04 IlyaE joined #fuel-dev
03:22 IlyaE joined #fuel-dev
04:45 IlyaE joined #fuel-dev
05:10 dshulyak_ joined #fuel-dev
05:37 AndreyDanin joined #fuel-dev
05:54 dshulyak_ joined #fuel-dev
06:01 aglarendil guys +1 for CentOS, please https://review.openstack.org/#/c/105912/
06:31 sambork joined #fuel-dev
06:47 IlyaE joined #fuel-dev
06:50 ddmitriev joined #fuel-dev
06:58 pasquier-s joined #fuel-dev
06:59 gilmeir joined #fuel-dev
07:22 Longgeek joined #fuel-dev
07:23 al_ex10 joined #fuel-dev
07:25 Longgeek joined #fuel-dev
08:15 pasquier-s joined #fuel-dev
08:18 guillaume__1 joined #fuel-dev
08:18 prmtl hi guys, we have some changes ready to merge for access-control-master-node bp
08:18 prmtl could someone take a look on them (review and merge)?
08:18 prmtl https://review.openstack.org/103897
08:18 prmtl https://review.openstack.org/105213
08:18 prmtl https://review.openstack.org/103673
08:18 prmtl https://review.openstack.org/104131
08:18 prmtl https://review.openstack.org/103554
08:18 prmtl https://review.openstack.org/105767
08:18 prmtl https://review.openstack.org/102795
08:18 prmtl https://review.openstack.org/105113
08:18 prmtl also it would be great to rerun tests for this changeset (and eventually merge):
08:18 prmtl https://review.openstack.org/104903
08:24 mattymo prmtl, to rerun tests, just modify commit message or hit rebase change
08:24 mattymo we don't have a simple gerrit trigger that would rerun Fuel CI
08:25 prmtl k, got it
08:29 mattymo prmtl, we have a little instability in fuel master, but maybe later today we can merge
08:29 mattymo I want to get some confidence from asledzinskiy first though
08:32 prmtl mattymo: ok, thanks for the info
08:37 e0ne joined #fuel-dev
08:44 mihgen prmtl: regarding https://review.openstack.org/#/c/104903/, unit tests failing, did you take a look?
08:44 asyriy joined #fuel-dev
08:45 IvanBerezovskiy joined #fuel-dev
08:52 brain461 joined #fuel-dev
09:05 mrasskazov joined #fuel-dev
09:05 aglarendil joined #fuel-dev
09:07 nurla joined #fuel-dev
09:12 prmtl mihgen: I've wrote to vkramskikh about them, I'll ping him on irc
09:15 accela-dev joined #fuel-dev
09:25 e0ne joined #fuel-dev
09:57 igajsin do we have any prognosis of when ubuntu-master will work?
09:58 bookwar igajsin: ubuntu master on fuel ci currently fixed, it passes empty test
09:59 bookwar igajsin: we have several tests passed on iso 307 by now https://fuel-jenkins.mirantis.com/job/master_fuellib_review_systest_ubuntu/
10:02 apodrepniy_ joined #fuel-dev
10:03 igajsin thanx
10:04 pasquier-s_ joined #fuel-dev
10:05 teran joined #fuel-dev
10:14 zynzel joined #fuel-dev
10:14 moshele joined #fuel-dev
10:16 scroiset joined #fuel-dev
10:21 brain461 joined #fuel-dev
10:24 Longgeek joined #fuel-dev
10:31 teran joined #fuel-dev
10:34 teran_ joined #fuel-dev
10:36 vkramskikh prmtl: i've uploaded https://review.openstack.org/#/c/104903/ rebased on https://review.openstack.org/#/c/102795/ but tests still fail
10:36 vkramskikh it seems we need to modify test runner to make it work against nailgun with enabled auth
10:45 AndreyDanin joined #fuel-dev
10:50 ddmitriev joined #fuel-dev
10:51 teran joined #fuel-dev
11:09 warpc_ joined #fuel-dev
11:11 teran joined #fuel-dev
11:12 teran_ joined #fuel-dev
11:13 teran__ joined #fuel-dev
11:14 teran___ joined #fuel-dev
11:19 ikalnitsky joined #fuel-dev
11:20 e0ne joined #fuel-dev
11:51 prmtl vkramskikh: ok, I'm looking on it
11:55 teran joined #fuel-dev
11:58 guillaume__1 joined #fuel-dev
11:58 pasquier-s joined #fuel-dev
12:19 ikalnitsky_ joined #fuel-dev
12:19 ikalnitsky_ left #fuel-dev
12:43 tnurlygayanov hi there!
12:44 tnurlygayanov I need help with Nailgun API
12:44 tnurlygayanov my script performs PUT request on url = 'http://{0}/api/clusters/{1}/network_configuration/neutron/'
12:45 tnurlygayanov and I get strange error
12:45 tnurlygayanov requests.exceptions.ConnectionError: HTTPConnectionPool(host='%7B0%7D', port=80): Max retries exceeded with url: /api/clusters/%7B1%7D/network_configuration/neutron/ (Caused by <class 'socket.gaierror'>: [Errno -2] Name or service not known)
12:46 tnurlygayanov so, I just want to update information about IP addresses configuration in my new OpenStack cluster
12:50 al_ex11 joined #fuel-dev
12:56 sambork joined #fuel-dev
13:01 aviramb joined #fuel-dev
13:11 tatyana joined #fuel-dev
13:12 e0ne joined #fuel-dev
13:43 pasquier-s joined #fuel-dev
13:47 msemenov_ joined #fuel-dev
13:47 guillaume__1 joined #fuel-dev
13:53 guillaume__1 joined #fuel-dev
14:02 jobewan joined #fuel-dev
14:21 teran joined #fuel-dev
14:29 al_ex11 joined #fuel-dev
14:31 Chlorum joined #fuel-dev
14:33 dpyzhov tnurlygayanov: %7B1%7D? Are you sure this is what you want?
14:35 Longgeek joined #fuel-dev
14:37 aviramb_ joined #fuel-dev
14:38 akuznetsova joined #fuel-dev
14:39 IgorYozhikov joined #fuel-dev
14:43 katyafervent joined #fuel-dev
14:43 meow-nofer_ joined #fuel-dev
14:44 dteselkin joined #fuel-dev
14:47 tsufiev joined #fuel-dev
14:47 igormarnat joined #fuel-dev
14:49 mattymo joined #fuel-dev
14:51 mihgen mattymo: what's the status of testing of access control?
14:51 mihgen who is doing that?
14:51 angdraug joined #fuel-dev
14:55 IlyaE joined #fuel-dev
14:56 mihgen sambork: dshulyak why https://review.openstack.org/#/c/103897/ is -1?
14:56 mihgen are we ready to merge the rest of the patches?
14:56 dshulyak mihgen: it should be mirror state of nailgun
14:56 mihgen actually from https://review.openstack.org/#/q/status:open+branch:master+topic:bp/access-control-master-node,n,z I see three patches with -1
14:57 asledzinskiy mihgen: I'm doing that. Found only minor issues. Everything works fine. But the only part I can't test is how OSTF works because my deploys failed with (/Stage[main]/Galera/Package[MySQL-server]/ensure) change from absent to 5.5.28_wsrep_23.7 failed: Could not update: Failed to update to version 5.5.28_wsrep_23.7, got version 5.5.28_wsrep_23.7-12.mira1 instead . I'm searching if this issue is reported, if not - report by myself
14:57 dshulyak mihgen: if auth is turned on in nailgun then it should be on for ostf
14:57 mihgen dshulyak: what do you mean by mirror state by nailgun?
14:57 mihgen aglarendil: take a look at the issue with MySQL ^^^
14:57 mihgen I believe it was fixed in master
14:57 mihgen asledzinskiy: is it about centos HA ?
14:58 dshulyak mihgen: patches with fake keystone is not essential
14:58 dshulyak vkramskikh: right ?
14:58 asledzinskiy mihgen: yes, it's centos HA
14:58 teran_ joined #fuel-dev
14:58 holser asledzinskiy, use newer ISO
14:59 holser it’s already fixed
15:00 asledzinskiy holser: I'm using custom ISO with access-master-node patches so to use this fix I have to rebuild it
15:00 holser asledzinskiy: that’s correct
15:00 vkramskikh dshulyak: correct
15:01 holser most likely you built ISO when we merged MySQL 5.6 and then reverted it back due to issues
15:02 asledzinskiy holser: yeah, may be, I'd better deploy simple env to check OSTF than re-deploy ISO
15:10 salmon_ joined #fuel-dev
15:17 mihgen you can use simple mode to test authx for Fuel
15:21 tnurlygayanov a_teem ping
15:21 a_teem tnurlygayanov: pong
15:21 tnurlygayanov a_teem I have question about  Nailgun API. Do we have an example how to configure networks with this API?
15:22 tnurlygayanov I try to send PUT request to api/clusters/{1}/network_configuration/neutron/   and get 411 error
15:24 tnurlygayanov a_teem for 4.1.x it works, but for 5.x it doesn't work
15:27 xarses joined #fuel-dev
15:30 ikalnitsky joined #fuel-dev
15:30 tnurlygayanov http://paste.openstack.org/show/85966/   <<< a_teem, my code and error text
15:32 ikalnitsky mattymo: angdraug: https://review.openstack.org/#/c/105750/ https://review.openstack.org/#/c/105739/ could we merge these? i've tested it on iso, works fine.
15:33 a_teem tnurlygayanov: ok, give me some time to check your code
15:33 tnurlygayanov a_teem Thank you!
15:34 tnurlygayanov a_teem it reproduced even we don't change the JSON
15:35 tnurlygayanov it should reproduce with the following code: http://paste.openstack.org/show/85967/
15:40 ikalnitsky joined #fuel-dev
15:46 gilmeir_ joined #fuel-dev
15:52 tatyana joined #fuel-dev
15:53 xarses bookwar: please abort ubuntu for https://review.openstack.org/#/c/103280
15:54 mattymo she's not at her desk
15:54 mattymo if I see her, I'll let her know
15:57 AndreyDanin joined #fuel-dev
15:57 xarses bookwar: looks like I want you to kill https://fuel-jenkins.mirantis.com/job/master_fuellib_review_systest_ubuntu/ 1647
15:58 xarses I'm posting a new rev, and thats still in the queue
15:58 bookwar done
15:59 mihgen Folks, don't forget about the meeting in #openstack-meeting-alt
16:00 salmon_ joined #fuel-dev
16:03 dshulyak_ joined #fuel-dev
16:04 IvanBerezovskiy left #fuel-dev
16:06 xarses angdraug: aglarendil https://review.openstack.org/#/c/103947/ https://review.openstack.org/#/c/106053/
16:07 aglarendil I am testing xenolog's ML2 right now
16:07 aglarendil just to be sure it works
16:08 mattymo aglarendil, can you merge https://review.openstack.org/#/c/105739/ and https://review.openstack.org/#/c/105750/
16:08 mattymo it's got 2 +1s
16:09 aglarendil okay guys
16:09 aglarendil we are going to merge ML2 support by xenolog and build new ISO
16:10 aglarendil then we will merge access control and other fixes
16:10 aglarendil I wanna get intermediate FF stable ISO
16:10 mattymo ok
16:10 aglarendil for Mlnx and Vmware guys at least
16:10 mattymo just add it to your queue
16:16 angdraug aglarendil: if you want a stable iso, why are you not including fixes in it?
16:16 angdraug btw guys, meeting in #openstack-meeting-lat
16:16 angdraug s/lat/alt/
16:16 aglarendil all the critical fixes are in the ISO right now
16:16 aglarendil the next merge is access control and misc fixes
16:16 aglarendil which I do not want to break this particular ISO build
16:22 xarses bookwar: it looks like https://review.openstack.org/#/c/103280/20 doesn't have ubuntu scheduled, can you check?
16:23 angdraug xarses: meeting!
16:27 xarses angdraug: yep
16:28 IlyaE joined #fuel-dev
16:29 bookwar xarses: it is in the queue, jenkins posts comments on job start only
16:30 xarses bookwar: ok, thanks
16:30 brain461 joined #fuel-dev
16:33 evgeniyl mattymo: https://review.openstack.org/#/c/106097/
16:36 mattymo evgeniyl, ouch. it's sad we need this
16:36 mattymo but it's totally safe to run this
16:37 evgeniyl mattymo: I was trying to find other way to do it, but didn't find
16:37 mattymo and it's better to run here than trying to cram this into some weird space in puppet and expect it to work for every case
16:38 mattymo +1
16:38 evgeniyl mattymo: also when I was working in previous company, we had to do the same stuff for deb package manager, and it was upgrade feature too
16:39 evgeniyl mattymo: thanks
16:40 mattymo I've fixed every kind of rpm and yum problem that can exist before. it's easy to just shut down/suspend computer in the middle of a package install. I had to write a really annoying program just to block users from breaking rpm
16:44 wrale joined #fuel-dev
16:48 e0ne joined #fuel-dev
17:04 aglarendil guys, we still have issues with >2 controllers and rabbitmq
17:04 aglarendil the cluster is still partitioned
17:04 aglarendil we will deal with it tomorrow
17:04 angdraug so no stable ISO today?
17:05 aglarendil all you need is to do following commands during deployment:
17:05 aglarendil rabbitmqctl stop_app
17:05 aglarendil rabbitmqctl force_reset
17:05 aglarendil rabbitmqctl join_cluster rabbit@<master_node>
17:05 aglarendil rabbitmqctl start_app
17:05 aglarendil on all the secondary nodes
17:06 angdraug is there a bug?
17:06 aglarendil https://bugs.launchpad.net/fuel/+bug/1339080
17:06 aglarendil xenolog is dealing with it right now
17:08 aglarendil anyway, 1-controller HA deployment is fine now.
17:09 aglarendil so you can test simple fixes
17:09 aglarendil and xenolog's ML2 was merged
17:11 aglarendil I will merge access control tomorrow
17:11 aglarendil also you can test it on your own
17:12 aglarendil and merge it
17:12 aglarendil if it works :-)
17:15 tatyana joined #fuel-dev
17:17 dshulyak_ joined #fuel-dev
17:30 prmtl joined #fuel-dev
17:31 tatyana joined #fuel-dev
17:50 teran joined #fuel-dev
18:11 IlyaE joined #fuel-dev
18:32 prmtl joined #fuel-dev
18:32 xarses angdraug: aglarendil: https://review.openstack.org/#/c/106053/
18:39 xarses bookwar: I filed https://bugs.launchpad.net/fuel/+bug/1340336 about the puppet-lint test stopping on the first error
18:40 AndreyDanin joined #fuel-dev
18:43 angdraug https://review.openstack.org/#/c/106053/ approved
19:03 AndreyDanin joined #fuel-dev
19:20 e0ne joined #fuel-dev
19:26 sanek joined #fuel-dev
19:28 e0ne joined #fuel-dev
19:46 angdraug joined #fuel-dev
19:50 salmon_ joined #fuel-dev
19:54 xarses bookwar: aglarendil mihgen poke
19:56 xarses I think we need to change the ubuntu_ha pull request test to use ceph instead of swift so we don't keep getting close or exceeding timeout because controller role keeps re-running
19:58 angdraug aglarendil: any news about https://bugs.launchpad.net/fuel/+bug/1339080 ?
20:08 xarses Also, I think deployment takes longer just because of crm, the only reason I can think for this is that the 1 vcpu is too little
20:25 sanek joined #fuel-dev
20:34 aglarendil guys
20:34 aglarendil https://review.openstack.org/106164 this is critical
20:34 angdraug ?
20:34 aglarendil no, angdraug, I do not have any news on rabbitmq bug
20:35 angdraug :(
20:35 angdraug https://review.openstack.org/login/c/106164/1/deployment/puppet/osnailyfacter/manifests/cluster_ha.pp
20:35 angdraug can you make it consistent with cluster_simple?
20:36 moshele joined #fuel-dev
20:36 angdraug LGTM otherwise
20:38 aglarendil done
20:38 angdraug +2, waiting on CI to approve
20:42 aglarendil also this one: https://review.openstack.org/106165
20:44 angdraug aglarendil: why comment it out?
20:44 angdraug we have git for that :)
20:44 aglarendil we have the only version of mysql-wsrep
20:44 aglarendil okay-okay)
20:44 xarses shouldn't the version number come from versions.yaml?
20:45 aglarendil it will come
20:45 aglarendil it will be caught by package provider
20:45 aglarendil second patch-set
20:46 xarses ok, ya ha deployments need two vcpu
20:46 xarses otherwise load is verry high, and deployment is slow
20:46 angdraug +2 on mysql version too
20:47 aglarendil it ends within an hour with 1 vcpu machines for me
20:47 aglarendil although I have ssd disks as a backend
20:48 aglarendil crm does not affect much
20:49 xarses i do to, but the CI job times out at 50 min
20:49 xarses about an hour is too long
20:50 xarses the average runtime of ubuntu ha is 42 min
20:50 xarses so there isn't much room if you overrun
20:50 aglarendil if CI job times out - there is a bug
20:50 aglarendil that leads to endless puppet retries
20:51 aglarendil guys, I would really appreciate if you merged these requests without CI
20:51 aglarendil as they are trivial
20:51 aglarendil or you will need to retrigger ISO built
20:51 aglarendil it will take place in 9 minutes
20:51 aglarendil it cannot get worse as smoke test is not passing
20:53 xarses angdraug: will merge 106164 now as it has already completed syntax check
20:54 xarses 106165 we should wait for syntax check
20:56 angdraug 165 merged
20:57 angdraug fuel_5_iso This project is currently disabled
20:57 angdraug aglarendil: what's up with ^
20:58 aglarendil why are you concerned about 5.0.1 ISO ?
20:58 aglarendil I guess, it was disabled by our QA team
20:58 aglarendil as they are going to freeze it
20:59 angdraug we've not reached code freeze criteria yet
20:59 angdraug we can't have RC at least 856764 is fixed
21:00 aglarendil what is the bug number you are talking about ?
21:02 angdraug also we need https://review.openstack.org/105996 merged and backported
21:02 angdraug https://bugs.launchpad.net/oslo.messaging/+bug/856764
21:02 dshulyak1 joined #fuel-dev
21:02 angdraug we discussed it in the meeting today
21:02 aglarendil oh!
21:02 angdraug http://eavesdrop.openstack.org/meetings/fuel/2014/fuel.2014-07-10-16.00.log.html
21:02 aglarendil angdraug: I do not really know why this build is disabled
21:02 aglarendil anyway, 856764 will be in packages
21:03 aglarendil exchanges declaration will be backported tomorrow, I hope
21:03 aglarendil what I am concerned now is master ISO with stupid bugs
21:03 aglarendil which somehow passed CI
21:04 aglarendil while they could not
21:04 angdraug yeah
21:04 aglarendil especially the one with vcenter_hash
21:04 aglarendil it passed CI for both HA and simple
21:05 aglarendil and I cannot imagine how it could happen
21:05 AndreyDanin joined #fuel-dev
21:05 aglarendil my best guess is that there was some modification in deployment serializer in nailgun
21:05 aglarendil and it was not caught by CI as it only checks manifests changes
21:06 aglarendil ok. I'm gonna get some good 8 hours of sleep - this was a really long day
21:06 aglarendil also, try to find why rabbitmq is not clustering properly
21:07 aglarendil OCF script is pretty simple
21:07 aglarendil it leverages notify mechanism of pacemaker
21:07 angdraug speaking of rabbitmq cluster,
21:07 angdraug xarses found out yesterday that vip across multiple controllers stopped working
21:07 angdraug was that fixed since then?
21:07 aglarendil I could not reproduce this issue
21:07 aglarendil the only problem I had was rabbitmq partitioning
21:08 aglarendil after manual rebuilding of cluster I was able to run OSTF tests successfully
21:08 aglarendil so, OCF script for rabbitmq just starts beam
21:08 aglarendil and then uses notify mechanism to join cluster.
21:09 angdraug no promises about making a lot of progress with rabbitmq OCF, i've got a pile of ceph and 5.0.1 bugs to sort through, and xarses still has ml2-neutron work cut out for him
21:09 aglarendil so the main source of the bug is located somewhere in notify action
21:09 aglarendil ok. just in case you would want to switch the context
21:09 aglarendil I squashed down one bug with pacemaker sending nodenames with trailing whitespaces
21:09 aglarendil that's why bash comparisons were not working
21:10 angdraug what about reverting rabbitmq ocf until it works reliably?
21:10 aglarendil I hope, we will fix this bug tomorrow
21:10 aglarendil without any need to revert
21:10 aglarendil and there is nova module rebased on it
21:10 aglarendil it will be more difficult to revert it than to fix it
21:11 aglarendil xenolog will work on it today/tomorrow and squash it. And I will pay close attention to it
21:12 aglarendil actually, after the rebuild of cluster HA works seamlessly - it is just the matter of initial configuration, I guess
21:12 aglarendil something is broken in bash code of the script
21:12 aglarendil and I thinj i is something really simple
21:12 angdraug ok, I was just saying :)
21:12 aglarendil as always :)
21:12 angdraug if we track it down by tomorrow, definitely better to keep it
21:13 aglarendil yep. and we will do it. it is not the rocket science
21:13 aglarendil okay, good luck today, guys.
21:14 xarses if its not rocket science, why did your missle blow up?
21:14 xarses erm, cluster
21:15 xarses same difference =P
21:34 prmtl joined #fuel-dev
21:35 angdraug btw, fuel_master_iso didn't start yet
21:40 bookwar it is scheduled for 2:00 MSK
21:43 angdraug since you're still online, do you know why fuel_5_iso is disabled?
21:45 angdraug christop1eraedo: https://blueprints.launchpad.net/fuel/+spec/monitoring-system
21:55 angdraug xarses: https://review.openstack.org/106180
21:58 bookwar angdraug: unfortunately i don't, probably nurla knows
21:59 bookwar angdraug: there is rc1 iso, and i guess we need to run swarm test against it this night
22:05 angdraug we can't have RC without a fix for the oslo.messaging bug
22:11 sanek joined #fuel-dev
22:12 teran joined #fuel-dev
22:12 salmon_ joined #fuel-dev
22:15 bookwar angdraug: i can't tell you anything about it, sorry, as i don't have the info
22:33 bas_ joined #fuel-dev
22:33 dmitryme joined #fuel-dev
22:33 AndreyDanin joined #fuel-dev
22:34 SergeyLukjanov joined #fuel-dev
22:34 aignatov joined #fuel-dev
22:35 prmtl joined #fuel-dev
22:36 ruhe joined #fuel-dev
22:36 wrale joined #fuel-dev
22:36 mihgen joined #fuel-dev
22:36 MorAle joined #fuel-dev
22:36 higgins joined #fuel-dev
22:36 openstackstatus joined #fuel-dev
22:36 bookwar joined #fuel-dev
22:36 dburmistrov joined #fuel-dev
22:36 evg_ joined #fuel-dev
22:36 christop1eraedo joined #fuel-dev
22:36 monester_ joined #fuel-dev
22:36 akislitsky joined #fuel-dev
22:36 MiroslavAnashkin joined #fuel-dev
22:36 rvyalov joined #fuel-dev
22:36 meow-nofer joined #fuel-dev
22:36 sbog_ joined #fuel-dev
22:36 izinovik joined #fuel-dev
22:36 holser joined #fuel-dev
22:36 alex_didenko joined #fuel-dev
22:36 DinaBelova joined #fuel-dev
22:36 vkramskikh joined #fuel-dev
22:36 apalkina joined #fuel-dev
22:43 sbanka joined #fuel-dev
23:37 prmtl joined #fuel-dev
23:45 eshumakher joined #fuel-dev
23:48 intLabs joined #fuel-dev
23:50 intLabs Hi, I’ve been using fuel as part of a larger project that I am now looking to release as an ISO for easy installation by users, modified to install a custom dashboard and additonal services. Does anyone have any advice on the appropriate way to credit the fuel developers; my understanding is that all code within the fuel project is licenced under Apache 2.0 (as is my own code), but I would appreciate any input ◕‿◕

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