Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2014-08-21

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

All times shown according to UTC.

Time Nick Message
00:20 rmoe joined #fuel-dev
01:35 christopheraedo How do I enable "experimental" mode? I wanted to try the zabbix deployment, but I think I need to turn on experimental in a conf file on fuel master node, right?
01:42 Kupo24z1 joined #fuel-dev
01:47 ilbot3 joined #fuel-dev
01:47 Topic for #fuel-dev is now #fuel-dev Fuel dev docs http://docs.mirantis.com/fuel-dev/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel-dev/ | gerrit traffic @ #fuel-tracker
02:04 christopheraedo joined #fuel-dev
02:54 EeeBotu New bug #1359494 in fuel: "Explain using unsupported_hardware option to install Fuel on hardware not supported by Centos distro" [Undecided, New] https://bugs.launchpad.net/bugs/1359494
03:50 stevemac joined #fuel-dev
04:25 prmtl joined #fuel-dev
04:46 dshulyak_ joined #fuel-dev
05:26 prmtl joined #fuel-dev
05:57 adanin joined #fuel-dev
06:07 salmon_ joined #fuel-dev
06:23 teran joined #fuel-dev
06:25 prmtl joined #fuel-dev
06:25 teran_ joined #fuel-dev
06:25 prmtl joined #fuel-dev
06:29 teran joined #fuel-dev
06:30 teran_ joined #fuel-dev
06:36 teran joined #fuel-dev
06:38 teran_ joined #fuel-dev
06:39 sambork joined #fuel-dev
06:40 teran__ joined #fuel-dev
06:40 tatyana joined #fuel-dev
06:48 tuvenen joined #fuel-dev
06:51 shauno_ joined #fuel-dev
07:05 shauno_ joined #fuel-dev
07:10 dancn joined #fuel-dev
07:15 pasquier-s joined #fuel-dev
07:23 skolekonov joined #fuel-dev
07:29 dshulyak mihgen: evgeniyl: IOError socket closed https://bugs.launchpad.net/fuel/+bug/1358972 , it is not about nailgun, it is about rabbitmq container
07:29 dshulyak user can restart docker container and necessery configuration will be applied
07:30 dshulyak or
07:30 dshulyak rabbitmqctl add_vhost '/'
07:30 dshulyak rabbitmqctl add_user naily BlSv1yAr
07:30 dshulyak rabbitmqctl set_permissions -p '/' naily '.*' '.*' '.*'
07:32 mihgen dshulyak: I understand it's not nailgun thing, but I didn't read careful enough bug info so can't even guess what is the best way to fix..
07:34 teran joined #fuel-dev
07:38 shauno_ joined #fuel-dev
07:50 shauno_ joined #fuel-dev
08:08 prmtl joined #fuel-dev
08:10 Longgeek joined #fuel-dev
08:12 Longgeek joined #fuel-dev
08:21 EeeBotu New bug #1359601 in fuel: "Howto enable experimental mode" [High, Confirmed] https://bugs.launchpad.net/bugs/1359601
08:24 dancn joined #fuel-dev
08:38 e0ne joined #fuel-dev
08:39 msemenov joined #fuel-dev
08:40 tzn joined #fuel-dev
08:48 tzn joined #fuel-dev
08:57 tzn joined #fuel-dev
08:59 IvanBerezovskiy joined #fuel-dev
09:07 HeOS joined #fuel-dev
09:25 bookwar joined #fuel-dev
09:28 e0ne joined #fuel-dev
09:41 tzn joined #fuel-dev
09:51 tzn joined #fuel-dev
09:52 tzn joined #fuel-dev
09:53 tzn joined #fuel-dev
10:06 teran joined #fuel-dev
10:07 dmitryme Hey library people, we need review in https://review.openstack.org/#/c/115266/ , this fixes a critical bug
10:08 EeeBotu New bug #1359642 in fuel: "document how to backup/restore Fuel Master Node" [Undecided, New] https://bugs.launchpad.net/bugs/1359642
10:08 EeeBotu New bug #1359640 in fuel: "Document Experimental features implementation" [Undecided, New] https://bugs.launchpad.net/bugs/1359640
10:17 mihgen holser: I suggest to keep https://bugs.launchpad.net/fuel/+bug/1356812 open, see my comment #2 there
10:18 mihgen bogdando: dilyin: alex_didenko: holser please help in reviewing as per dmitryme's request above ^^^
10:18 holser mihgen, I marked that bug as confirmed
10:18 holser thanks for tips
10:20 mihgen holser: I actually read carefully. This is rather a task than a bug, and description doesn't intend to solve initial issue which is galera failure
10:21 mihgen it's about of stopping deployment, not fixing galera, and we need it just to get faster feedback on failure in CI
10:21 * holser nods
10:21 mihgen holser: I think we should move it to 6.0 and address a real bug
10:21 mihgen in 5.1 - which I believe you described in fuel-dev ML
10:24 aepifanov joined #fuel-dev
10:26 aepifanov joined #fuel-dev
10:27 eshumkher joined #fuel-dev
10:33 eshumkher Folks, please review the commit: https://review.openstack.org/#/c/114935/
10:34 e0ne joined #fuel-dev
10:35 Longgeek joined #fuel-dev
10:41 evgeniyl mattymo: http://jasonwilder.com/blog/2014/08/19/squashing-docker-images/ maybe it will help us to reduce the size of images
10:43 mattymo evgeniyl, , docker-squash looks cool
10:43 mattymo it would speed up the decompression a lot
10:43 mattymo and running docker load
10:44 mattymo the only question I have is does it preserve the CMD param from the original image
10:45 mattymo now I know why I haven't heard of it :) it was started only 2 weeks ago
10:55 teran joined #fuel-dev
11:06 Longgeek joined #fuel-dev
11:06 prmtl joined #fuel-dev
11:09 Longgeek joined #fuel-dev
11:20 e0ne joined #fuel-dev
11:25 teran joined #fuel-dev
11:26 teran_ joined #fuel-dev
11:28 tzn joined #fuel-dev
11:37 evgeniyl mattymo: hm, it looks like I have a problem with upgrade from 5.0.1, do you restart the containers after docker upgrade in the puppet manifests?
11:45 prmtl joined #fuel-dev
11:45 sambork joined #fuel-dev
11:50 mattymo evgeniyl, containers are started first from docker build all. if docker package gets upgraded, puppet runs dockerctl start all
11:50 mattymo do you have an env or fuel_upgrade.log?
11:51 tzn joined #fuel-dev
11:51 eshumkher joined #fuel-dev
11:51 asledzinskiy joined #fuel-dev
11:53 evgeniyl mattymo: does `dockerctl start all` restart container if it was running?
11:55 mattymo no
11:55 mattymo in 5.1 manifests: if it was never built, it builds it. if it was stopped, it starts it and runs pre/post hooks. if it is running, it just warns that it's already running and skips all hooks
11:57 shauno__ joined #fuel-dev
11:59 dshulyak_ joined #fuel-dev
12:03 adanin_ joined #fuel-dev
12:03 teran joined #fuel-dev
12:04 dancn joined #fuel-dev
12:04 shauno_ joined #fuel-dev
12:04 teran_ joined #fuel-dev
12:08 sambork joined #fuel-dev
12:08 tzn joined #fuel-dev
12:09 dancn joined #fuel-dev
12:10 dshulyak_ joined #fuel-dev
12:11 prmtl joined #fuel-dev
12:12 Longgeek joined #fuel-dev
12:12 EeeBotu New bug #1359705 in fuel: "[update] oslo.messaging is not updated by patching" [Critical, New] https://bugs.launchpad.net/bugs/1359705
12:12 EeeBotu New bug #1359696 in fuel: "There is wrong hint in tab 'Storage Backends ' in wizard." [Undecided, New] https://bugs.launchpad.net/bugs/1359696
12:14 Longgeek joined #fuel-dev
12:14 dancn joined #fuel-dev
12:14 aepifanov joined #fuel-dev
12:18 pasquier-s joined #fuel-dev
12:39 aepifanov joined #fuel-dev
12:42 testmaniac joined #fuel-dev
12:47 testmaniac is the iso build from last night going to be the fuel 5.1 final?
12:52 eshumkher_ joined #fuel-dev
13:07 jaypipes joined #fuel-dev
13:11 aepifanov joined #fuel-dev
13:14 EeeBotu New bug #1359728 in fuel: "[OSTF] Wrong discovering cd -tests" [Undecided, New] https://bugs.launchpad.net/bugs/1359728
13:14 EeeBotu New bug #1359725 in fuel: "Fuel upgrade from 5.0.1->5.1 on env deployed via cli finished with error" [High, New] https://bugs.launchpad.net/bugs/1359725
13:14 EeeBotu New bug #1332058 in fuel: "keystone behavior when one memcache backend is down" [High, Triaged] https://bugs.launchpad.net/bugs/1332058
13:15 EeeBotu New bug #1356921 in fuel: "[Update] Murano Dashboard update fails" [Critical, Fix Committed] https://bugs.launchpad.net/bugs/1356921
13:24 prmtl_ joined #fuel-dev
13:25 EeeBotu New bug #1359739 in fuel: "nova.servicegroup.drivers.db [-] model server went away" [Critical, New] https://bugs.launchpad.net/bugs/1359739
13:27 jobewan joined #fuel-dev
13:55 kaliya joined #fuel-dev
14:06 prmtl joined #fuel-dev
14:13 eshumkher joined #fuel-dev
14:13 ikalnitsky joined #fuel-dev
14:15 EeeBotu New bug #1359773 in fuel: "zabbix deployment error when default password is changed" [Undecided, New] https://bugs.launchpad.net/bugs/1359773
14:15 EeeBotu New bug #1359771 in fuel: "implement patching tests on CI" [Critical, Confirmed] https://bugs.launchpad.net/bugs/1359771
14:15 EeeBotu New bug #1359746 in fuel: "[Upgrade] After upgrade and re-deploy cluster Platform services functional tests failed with HTTPBadRequest: ERROR: Property error : Server: image Multiple physical resources were found with name (TestVM)." [High, New] https://bugs.launchpad.net/bugs/1359746
14:15 EeeBotu New bug #1359744 in fuel: "neutron.openstack.common.rpc.common: Failed to publish message to topic 'q-plugin': [Errno 32] Broken pipe" [Critical, New] https://bugs.launchpad.net/bugs/1359744
14:44 eshumkher joined #fuel-dev
14:55 iosstef joined #fuel-dev
15:10 tzn joined #fuel-dev
15:25 mihgen hey folks let's fill up agenda for iRC meeting: https://etherpad.openstack.org/p/fuel-weekly-meeting-agenda
15:31 agrebennikov joined #fuel-dev
15:31 agrebennikov evgeniyl, hi
15:32 agrebennikov evgeniyl, just wanted to know whether it is possible to update fuel node up to the latest 5.0.1 if it has been built from source (some pre 5.0.1)
15:35 mihgen dpyzhov: bogdando are you guys working on the same issue? https://bugs.launchpad.net/fuel/+bug/1359705
15:35 mihgen dilyin: sorry you, not dpyzhov ^^
15:36 evgeniyl agrebennikov: hi, what is the version in /etc/fuel/version.yaml? If the version is 5.0.1, then it's impossible to run upgrade from 5.0.1 to 5.0.1, we use this version as postfix for images and containers, so, it can break something, and there is validation prevent user to run upgrade.
15:37 dilyin oops, looks like we do
15:38 agrebennikov evgeniyl, yes, it shows 5.0.1 now
15:39 agrebennikov evgeniyl, so it means the only way to do it is to upgrade the repositories manually, then upgrade manifests, right?
15:43 prmtl joined #fuel-dev
15:45 prmtl joined #fuel-dev
15:47 evgeniyl agrebennikov: yes, the only way is to make upgrade manually, or build new upgrade tar ball which has version 5.0.1.1 with images with this version
15:49 agrebennikov evgeniyl, another question - am I right that once we updated Fuel node to the next version (say 5.1) it would be possible to create only new environments wth it?
15:49 agrebennikov evgeniyl, existing environments will still stay unchanged
15:51 evgeniyl agrebennikov: existing envs will still be unchanged if you will not run patching, and you can create env with new version of packages and with previous
15:52 evgeniyl agrebennikov: actually, we were thinking to disable ability to create new cluster with old version of openstack
15:54 jaypipes joined #fuel-dev
15:56 mihgen holser: bogdando do we still need changes in manifests for https://bugs.launchpad.net/keystone/+bug/1332058 ?
15:56 mihgen according to holser , we don't need to..
15:56 holser we do
15:57 holser actually, I am wrong
15:57 holser manifests are ready for pylibmc, keystone doesn't
15:57 holser that’s the problem
15:57 bogdando mihgen, yes, we definitely will, at least for backend, backend_arguments
15:57 holser it breaks deployment as 2 of 3 nodes are down
15:57 holser bogdando, we need to tweak, not to rewrite
15:58 tzn joined #fuel-dev
15:58 bogdando holser, yes, but keystone_config would have be issued, so manifests would have be updated
15:58 * holser nods
16:00 agrebennikov evgeniyl, emm... do you mean with 5.0.1 it will be already possible to do openstack patching? Or it's 5.1 feature?
16:00 mihgen let's go to #openstack-meeting-alt
16:01 evgeniyl agrebennikov: I'm talking about 5.1, in 5.0.1 we didn't have patching feature.
16:01 angdraug joined #fuel-dev
16:02 amnk joined #fuel-dev
16:02 agrebennikov evgeniyl, right, so I mean if Now we have an environment from 5.0.1 - it will be unchanged forever
16:03 evgeniyl agrebennikov: yep, but after 5.1 upgrade, you will be able to patch 5.0 and 5.0.1 envs
16:03 evgeniyl agrebennikov: to 5.0.2 version
16:04 agrebennikov evgeniyl, oh, that's cool)) and, it consequently mean I'll be able to update our custom-built iso to 5.0.2 as well (once it's being released)
16:05 agrebennikov evgeniyl, oh, sorry, just realized that I said stupid thing =)
16:06 agrebennikov evgeniyl, so 1. update my custom -> 5.1 when it comes; 2. update env to 5.0.2
16:06 agrebennikov evgeniyl, something like that?
16:08 agrebennikov angdraug, hi there! Could you tell me who is our best in glance?
16:09 evgeniyl agrebennikov: yeah, you are right, 5.1 release will contain openstack with version some-version-5.0.2 and you will be able to patch your 5.0 and 5.0.1 envs, also this release will include some-version-5.1 release which has new features, and you will be able to create _new_ env with this release.
16:09 agrebennikov evgeniyl, gotcha.
16:10 agrebennikov evgeniyl, what is the way you are going to patch openstack itself? distribute new packages or straight patching?
16:10 IvanBerezovskiy left #fuel-dev
16:13 evgeniyl agrebennikov: 5.1 will contain two repositories with new packages, during the patching procedure we use this new repositories and run deployment again, there are hacks in puppet which are trying to install latest packages if there are any
16:13 slava joined #fuel-dev
16:14 xarses joined #fuel-dev
16:15 agrebennikov evgeniyl, how are you going to work on any additional changes in config files in this case?
16:16 slava hi, guys! Is there any api call to Fuel to get openstack.yaml configuration file as json?
16:16 EeeBotu New bug #1359839 in fuel: "Fuel  does not support OpenStack patching per node or per node role" [Undecided, New] https://bugs.launchpad.net/bugs/1359839
16:16 EeeBotu New bug #1359834 in fuel: "[custom iso] Ceph volumes don't work on ubuntu/neutron-vlan" [High, Incomplete] https://bugs.launchpad.net/bugs/1359834
16:16 EeeBotu New bug #1359833 in fuel: "HA deployment fails with Neutron GRE" [Undecided, New] https://bugs.launchpad.net/bugs/1359833
16:16 EeeBotu New bug #1359818 in fuel: "[Upgrade] Fuel cli requests return 401 error after upgrade to 5.1" [High, Confirmed] https://bugs.launchpad.net/bugs/1359818
16:18 evgeniyl agrebennikov: we rsync new manifests on slave nodes, new == from stable/5.0, and for 5.1 envs from master == future stable/5.1
16:20 agrebennikov evgeniyl, emm... it's actually what happens all the time. My point is that customer may want to make some changes to config files after the deployment. In this case puppet will definitely revert them
16:21 mihgen agrebennikov: if you are talking about commercial support, then how can you support something being hacked?
16:22 evgeniyl agrebennikov: yes it will, and I'm not sure if want change this behaviour
16:23 agrebennikov mihgen, well... ok, let me ask in some other way - what if We did any customizations for the particular customer? I mean did you try to put such case into the management procedure?
16:25 mihgen agrebennikov: if so, then you would do another "release", like 5.0.3
16:26 mihgen or might be 5.0.1.1
16:30 aepifanov joined #fuel-dev
16:37 teran joined #fuel-dev
16:41 dshulyak_ joined #fuel-dev
16:55 jaypipes joined #fuel-dev
17:00 akasatkin Let's review advanced-networking spec: https://review.openstack.org/115340 !
17:06 apodrepniy joined #fuel-dev
17:15 salmon_ joined #fuel-dev
17:17 rmoe joined #fuel-dev
17:31 EeeBotu New bug #1359884 in fuel: "[update] Centos ha with ceph for volumes fails on ostf" [Critical, New] https://bugs.launchpad.net/bugs/1359884
17:59 e0ne joined #fuel-dev
18:44 aepifanov joined #fuel-dev
19:12 testmaniac What is the new target release date for the Fuel 5.1 iso?
19:14 testmaniac never mind... i see it as Sept 4th...
20:03 jobewan joined #fuel-dev
20:03 salmon_ joined #fuel-dev
20:04 prmtl joined #fuel-dev
20:42 EeeBotu New bug #1359955 in fuel: "installation error while installing 3 controller nodes" [Undecided, New] https://bugs.launchpad.net/bugs/1359955
20:46 xarses joined #fuel-dev
21:02 adanin joined #fuel-dev
21:48 AlexF joined #fuel-dev
21:56 EeeBotu New bug #1359989 in fuel: "Hypervisor summary shows incorrect total storage" [Undecided, New] https://bugs.launchpad.net/bugs/1359989
21:59 agrebennikov joined #fuel-dev
22:03 aepifanov joined #fuel-dev
22:32 AlexF joined #fuel-dev
22:33 xarses joined #fuel-dev
22:49 tnurlygayanov_ joined #fuel-dev
22:49 tnurlygayanov_ hi there
22:49 tnurlygayanov_ anybody online? :)
22:50 tnurlygayanov_ I have a question about options
22:50 tnurlygayanov_ Ceph RDB for volumes (Cinder)
22:50 tnurlygayanov_ Ceph RDB for images (Glance)
22:50 tnurlygayanov_ Ceph RDB for ephemeral volumes (Nova)
22:50 tnurlygayanov_ Ceph RadosGW for objects (Swift API)
22:50 angdraug tnurlygayanov: just ask your question and we'll ignore it :)
22:50 tnurlygayanov_ where we can find what Fuel will configure on the cloud if we will select each of this option?
22:51 tnurlygayanov_ angdraug hi :)    why?
22:52 angdraug tnurlygayanov: just kidding, we'll try to answer it
22:53 angdraug there's several places to look, mostly around fuel-library
22:53 angdraug are you looking for something specific?
22:53 tnurlygayanov_ angdraug now I can see that Ceph GW for Swift works in different modes when I select only Ceph RadosGW option and when I select all these options
22:54 tnurlygayanov_ and if I select all these components bulk api requests don't work
22:55 tnurlygayanov_ so, I want to check what we will change in configuration files for each parameter to find the root of this strange ehaviour
22:58 angdraug strange indeed
22:58 angdraug here's nova parts:
22:58 angdraug https://github.com/stackforge/fuel-library/blob/master/deployment/puppet/ceph/manifests/ephemeral.pp
22:58 angdraug https://github.com/stackforge/fuel-library/blob/master/deployment/puppet/ceph/manifests/nova_compute.pp
22:58 EeeBotu New bug #1360000 in fuel: "VM fails to schedule if ephemeral drive is going to be used" [Undecided, New] https://bugs.launchpad.net/bugs/1360000
22:59 angdraug cinder:
22:59 angdraug https://github.com/stackforge/fuel-library/blob/master/deployment/puppet/cinder/manifests/volume/rbd.pp
22:59 angdraug glance:
22:59 angdraug https://github.com/stackforge/fuel-library/blob/master/deployment/puppet/glance/manifests/backend/rbd.pp
23:04 tnurlygayanov_ angdraug Thank you!
23:23 AlexF joined #fuel-dev
23:36 teran joined #fuel-dev
23:41 angdraug xarses: https://bugs.launchpad.net/fuel/+bug/1359884

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