Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2014-05-06

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

All times shown according to UTC.

Time Nick Message
00:18 IlyaE joined #fuel-dev
00:20 e0ne joined #fuel-dev
00:33 rmoe joined #fuel-dev
00:51 artem_panchenko joined #fuel-dev
01:09 EeeBotu joined #fuel-dev
01:13 christopheraedo joined #fuel-dev
01:20 e0ne joined #fuel-dev
01:23 xarses joined #fuel-dev
01:33 EeeBotu New bug #1316377 in fuel: "Enabling vCenter integration disables Ceph for Glance and RadosGW" [High, New] https://bugs.launchpad.net/bugs/1316377
01:33 EeeBotu New bug #1316376 in fuel: "keystone client doesn't work with default openrc" [Low, New] https://bugs.launchpad.net/bugs/1316376
02:29 saibarspeis joined #fuel-dev
02:56 Ch00k joined #fuel-dev
03:08 igormarnat_ joined #fuel-dev
03:16 warpc joined #fuel-dev
03:20 e0ne joined #fuel-dev
03:29 IlyaE joined #fuel-dev
03:43 igormarnat_ joined #fuel-dev
04:20 e0ne joined #fuel-dev
04:23 Ch00k joined #fuel-dev
04:25 IlyaE joined #fuel-dev
05:21 dshulyak_ joined #fuel-dev
05:25 IlyaE joined #fuel-dev
05:50 anotchenko joined #fuel-dev
05:52 _daks_ joined #fuel-dev
06:11 sambork joined #fuel-dev
06:12 e0ne joined #fuel-dev
06:24 mirrorbox 254             'disk': 'disk/id/b00b135'})
06:24 mirrorbox ok
06:31 anotchenko joined #fuel-dev
06:34 Ch00k joined #fuel-dev
06:36 salmon_ joined #fuel-dev
06:37 Ch00k joined #fuel-dev
07:07 bogdando joined #fuel-dev
07:14 rzhnichkov joined #fuel-dev
07:25 EeeBotu New bug #1316446 in fuel: "UserWarning after image creation" [Medium, Confirmed] https://bugs.launchpad.net/bugs/1316446
07:28 IlyaE joined #fuel-dev
07:28 artem_panchenko1 joined #fuel-dev
07:49 IlyaE joined #fuel-dev
07:57 anotchenko joined #fuel-dev
08:00 akasatkin joined #fuel-dev
08:15 Ch00k joined #fuel-dev
08:25 EeeBotu New bug #1316453 in fuel: "Murano contains outadated version of core packages" [High, Confirmed] https://bugs.launchpad.net/bugs/1316453
08:29 anotchenko joined #fuel-dev
08:45 Tatyanka_Leontov joined #fuel-dev
08:46 Ch00k joined #fuel-dev
08:48 Ch00k joined #fuel-dev
08:51 rzhnichkov_ joined #fuel-dev
08:52 e0ne joined #fuel-dev
09:02 morale joined #fuel-dev
09:02 morale_ joined #fuel-dev
09:03 enikanorov joined #fuel-dev
09:10 teran joined #fuel-dev
09:15 dteselkin_ https://review.openstack.org/#/c/92319/
09:27 igormarnat_ joined #fuel-dev
09:32 rzhnichkov__ joined #fuel-dev
09:57 akasatkin joined #fuel-dev
09:59 tnurlygayanov joined #fuel-dev
10:20 Ch00k joined #fuel-dev
10:26 e0ne_ joined #fuel-dev
10:29 EeeBotu New bug #1316524 in fuel: "Ceph-deploy config pull issue" [Medium, Confirmed] https://bugs.launchpad.net/bugs/1316524
10:30 mihgen dburmistrov: hi
10:30 mihgen dburmistrov: I'd like to know the status of mirrors build
10:31 mihgen as tnurlygayanov and other #murano folks need it to include updated murano packages
10:31 salmon_ hi all, why neutron is not default option in wizzard? I thought nova-network will be deprecated
10:31 dburmistrov last mirror build was success at 06.05.2014 0:44:14
10:31 mihgen dburmistrov: teran I see #124 build is green there. Did we fix all the issues?
10:32 mihgen dburmistrov: ok, thanks.
10:32 mihgen tnurlygayanov: Timur, I think you would need to ping dburmistrov or someone else from osci team to trigger the build when you are ready
10:33 mihgen dburmistrov: is it the plan how we update packages in mirrors? or what's the workflow?
10:34 dburmistrov mirror building starts by hand on demand due to huge amount of needed time
10:38 e0ne joined #fuel-dev
10:44 anotchenko joined #fuel-dev
10:45 rzhnichkov joined #fuel-dev
10:46 EeeBotu New bug #1316534 in fuel: "Murano telnet deployment doesn't finish" [High, Confirmed] https://bugs.launchpad.net/bugs/1316534
10:47 tnurlygayanov mihgen, ok
10:47 tnurlygayanov Dmitry Teselkin will update Murano packages and after that we need to mirror repositories with new packages.
10:54 teran mihgen: all problems are because of server fault. If it reproduces once again - we just reinstall OS there.
10:55 teran mihgen: currently all should work fine, but it fails thrice for last 16 hours
11:23 tnurlygayanov need review: https://review.openstack.org/#/c/92319/
11:23 tnurlygayanov Thank you :)
11:28 dshulyak joined #fuel-dev
11:29 EeeBotu New bug #1316554 in fuel: "Nodes aren't bootstrapped after stopping deployment on provisioning" [High, New] https://bugs.launchpad.net/bugs/1316554
11:29 EeeBotu New bug #1316537 in fuel: "Controller goes offline on deploying phase If deletion of cluser take place previously" [High, New] https://bugs.launchpad.net/bugs/1316537
11:42 igormarnat_ joined #fuel-dev
11:47 rzhnichkov_ joined #fuel-dev
11:48 IvanBerezovskiy1 joined #fuel-dev
11:49 AndreyDanin_ joined #fuel-dev
11:49 salmon_ joined #fuel-dev
11:50 xarses joined #fuel-dev
11:51 xdeller_ joined #fuel-dev
12:12 morale joined #fuel-dev
12:20 xarses joined #fuel-dev
12:20 Ch00k joined #fuel-dev
12:21 mihgen dburmistrov: Dima, can you please verify https://review.openstack.org/#/c/92319/ ? We need OSCI confirmation that packages are prepared, etc.
12:23 mihgen dburmistrov: I can merge or dpyzhov after your approval, or we need someone else - brain461, mrasskazov, SergeyOtpuschennikov
12:24 dburmistrov 1 min
12:25 dburmistrov +1
12:25 teran joined #fuel-dev
12:25 mihgen dburmistrov: ok, thanks, then I merge..
12:26 mihgen dburmistrov: will you retrigger mirror build or it's not needed now?
12:26 _daks_ joined #fuel-dev
12:26 dburmistrov it needs. i can retrigger
12:27 dburmistrov >Wiping out workspace first.
12:28 dburmistrov mirroring started
12:30 artem_panchenko joined #fuel-dev
12:30 rzhnichkov__ joined #fuel-dev
12:32 EeeBotu New bug #1316581 in fuel: "Shotgun cannot log into the host system" [High, Confirmed] https://bugs.launchpad.net/bugs/1316581
12:35 warpc joined #fuel-dev
12:37 salmon_1 joined #fuel-dev
12:38 dburmistrov folks, do not merge changes for code which should be packages while mirroring job is runnung. it breaks mirroring stage
12:39 dburmistrov *should be packaged
12:39 teran_ joined #fuel-dev
12:42 sambork1 joined #fuel-dev
12:44 mihgen bogdando: are you working on https://bugs.launchpad.net/bugs/1316204 ?
12:44 mihgen dburmistrov: > do not merge changes for code which should be packages while mirroring job is runnung -- it's hard to know
12:44 mihgen we need something automated to prevent it then
12:45 mihgen ideally we would rather create a workaround for mirroring job instead
12:45 mihgen why it breaks mirroring stage?
12:47 Ch00k joined #fuel-dev
12:47 dburmistrov new megre - new package in obs while mirroring script expect old package
12:48 dburmistrov i think it can be done with our osci jenkins
12:48 dburmistrov srv11 looks broken. again http://jenkins-product.srt.mirantis.net:8080/job/fuel_5_0_mirror/127/console
12:48 teran joined #fuel-dev
12:49 mihgen teran: ^^^
12:51 EeeBotu New bug #1316583 in fuel: "Destroy filesystem of provisioned node if call stop provision when node was reboot with installed os" [High, New] https://bugs.launchpad.net/bugs/1316583
12:55 bogdando mihgen, yes I'm reproducing it now, and was apparently reproduced https://bugs.launchpad.net/fuel/+bug/1316093
13:00 salmon_ joined #fuel-dev
13:02 teran_ joined #fuel-dev
13:12 teran_ dburmistrov: I just restarted 128th, looks like it's building
13:13 teran joined #fuel-dev
13:13 dburmistrov teran: thanx!
13:20 morale mihgen: I've fixed that string issue with plurals - https://review.openstack.org/#/c/92142/2
13:22 mihgen morale: ok, thanks. dpyzhov - can you also take a look
13:28 dpyzhov morale, mihgen do we have a bug for this?
13:28 mihgen we don't
13:29 mihgen I just commented this issue here, and Sasha fixed it
13:29 dpyzhov merged
13:30 anotchenko left #fuel-dev
13:30 brain461 teran: srv11 got bad again http://jenkins-product.srt.mirantis.net:8080/job/fuel_iso_with_gerrit_commits/567/console
13:30 anotchenko joined #fuel-dev
13:30 mihgen evgeniyl: can you take a look pls at https://review.openstack.org/#/c/92315/2 and merge if it's ok?
13:30 dpyzhov mihgen: please review https://review.openstack.org/#/c/90521/
13:30 dpyzhov this bug really annoys me
13:31 mihgen teran: can we move fuel-iso-with-gerrit-blabla to another server?
13:32 teran brain461: it's not about srv11, it's about some bullshit previously done on that server. Like unmounted filesystems inside chroot. Wait a minute...
13:33 brain461 teran: same happened on the srv08-srt server. we might have faulty jenkins job?
13:34 mihgen brain461: teran folks, new ISO with docker, as far as I know, can not be deployed on srv11
13:34 mihgen due to issues with versions of ubuntu packages there
13:34 mihgen so we used supermicro another server for fuel-5.0 iso job
13:35 brain461 mihgen: I'm attempting to build 4.1 iso
13:35 brain461 might be artifacts left from unsuccessful 5.0 builds
13:35 EeeBotu New bug #1316604 in fuel: "Murano system tests fail with assert on muranoconductor count" [High, Confirmed] https://bugs.launchpad.net/bugs/1316604
13:37 teran brain461: is it ok for your 4.1 iso cephrepo,http://osci-obs.vm.mirantis.net:82/centos-fuel-5.0-stable-15138/centos/ ?
13:39 brain461 package versions are the same as in 4.1, should be no conflicts because of that
13:43 teran brain461: http://jenkins-product.srt.mirantis.net:8080/job/fuel_iso_with_gerrit_commits/576/console -- should work
13:44 Ch00k joined #fuel-dev
13:45 Ch00k joined #fuel-dev
13:46 bogdando https://bugs.launchpad.net/fuel/+bug/1316093 looks realy strange. I've uploaded new logs from recently reproduced lab
13:47 bogdando I can see tcpdump flows from rsyslog client and server, but remote logs for puppet are absent in /var/log/remote
13:47 bogdando this behavior looks new since docker was introduced
13:48 artem_panchenko joined #fuel-dev
14:05 mihgen bogdando: interesting that I didn't see this issue, at least on centos
14:06 bogdando I wasn't able to reproduce it for the first try
14:06 bogdando but this one was done at centos
14:06 bogdando 1316204 is reproducing also
14:10 bogdando s/reproduced
14:17 jobewan joined #fuel-dev
14:32 teran joined #fuel-dev
14:33 teran_ joined #fuel-dev
14:33 tsduncan__ joined #fuel-dev
14:35 EeeBotu joined #fuel-dev
14:38 teran joined #fuel-dev
14:56 aglarendil bogdando: did you reset your environment
14:57 aglarendil bogdando: there is a bug related to rsyslogd not reloaded after environment is deleted or reset
14:57 bogdando agrendil, recreated some now ones
14:57 bogdando new*
14:57 aglarendil bogdando: in this case it continues to write to old files
14:58 bogdando aglarendil, can you link it? I will set 131609 as a dupe then
14:58 aglarendil https://bugs.launchpad.net/fuel/+bug/1315382
14:58 aglarendil bogdando: ^^
14:59 IlyaE joined #fuel-dev
15:01 bogdando nice catch, thanks
15:03 rzhnichkov_ joined #fuel-dev
15:24 morale dpyzhov: mihgen: this bug should be fixed in 5.0, please, review https://review.openstack.org/#/c/92389/ slight condition change from != to !==
15:27 IvanBerezovskiy1 left #fuel-dev
15:30 fandi joined #fuel-dev
15:35 warpc joined #fuel-dev
15:37 crshman joined #fuel-dev
15:37 EeeBotu New bug #1316640 in fuel: "[system tests] test_cobbler_alive test fails with timeout on checking cobbler API" [High, New] https://bugs.launchpad.net/bugs/1316640
15:43 dpyzhov morale: I can barely understand commit message
15:46 morale ok, will add commas and line breaks
15:48 morale uff, already merged :(
15:51 crshman is master the branch that'll be packaged up and deployed at the end of the week?
15:54 teran joined #fuel-dev
16:00 e0ne joined #fuel-dev
16:04 rmoe joined #fuel-dev
16:14 dshulyak_ joined #fuel-dev
16:31 e0ne joined #fuel-dev
16:32 e0ne joined #fuel-dev
16:42 e0ne joined #fuel-dev
16:45 e0ne_ joined #fuel-dev
16:52 xarses joined #fuel-dev
16:56 angdraug joined #fuel-dev
17:42 e0ne joined #fuel-dev
18:02 IlyaE joined #fuel-dev
18:10 e0ne joined #fuel-dev
18:16 tatyana joined #fuel-dev
18:58 EeeBotu New bug #1316741 in fuel: "Wrong loging of failed ostf tests in system test" [High, In Progress] https://bugs.launchpad.net/bugs/1316741
18:58 EeeBotu New bug #1316727 in fuel: "[OSTF][system_test] Sahara test failed with error in ostf that net04_ext pool not found" [High, Confirmed] https://bugs.launchpad.net/bugs/1316727
18:58 EeeBotu New bug #1316725 in fuel: "Pecan erros after adapter was started" [High, New] https://bugs.launchpad.net/bugs/1316725
18:58 EeeBotu New bug #1316720 in fuel: "We should increase time to live for messages and queues for max value in mcollective" [Undecided, New] https://bugs.launchpad.net/bugs/1316720
19:03 Ch00k joined #fuel-dev
19:46 IlyaE joined #fuel-dev
20:07 EeeBotu New bug #1316771 in fuel: "System test deploy_simple_mongo failed: instances failed to become active" [Critical, Confirmed] https://bugs.launchpad.net/bugs/1316771
20:07 EeeBotu New bug #1316761 in fuel: "Nodes failed to reboot for HA env" [High, New] https://bugs.launchpad.net/bugs/1316761
20:38 e0ne joined #fuel-dev
21:21 ilbot3 joined #fuel-dev
21:21 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
21:42 xarses angdraug: rmoe https://bugs.launchpad.net/fuel/+bug/1316812
21:48 aglarendil xarses: I have never faced this issue
21:48 aglarendil xarses: it looks pretty strange
21:49 aglarendil xarses: we have system tests running all the required stuff without this problem
21:50 xarses it only occurs if your host supports kvm nested, but kernel module kvm wont allow kvm_intel | kvm_amd to load
21:50 xarses in guest
21:50 xarses the kvm submodules won't load if kvm didn't create /dev/kvm
21:51 aglarendil you mean, if kvm_intel or kvm_amd is blacklisted?
21:52 aglarendil actually, loading of this modules create /dev/kvm
21:52 aglarendil AFAIK
21:52 xarses no, it just won't load in guest because it thinks it's not supported
21:52 xarses root@node-1:/etc/init# modprobe kvm
21:52 xarses root@node-1:/etc/init# modprobe kvm_amd
21:52 xarses FATAL: Error inserting kvm_amd (/lib/modules/3.11.0-18-generic/kernel/arch/x86/kvm/kvm-amd.ko): Operation not supported
21:52 aglarendil but kvm_intel will load just fine and that's it
21:52 aglarendil and udev will create corresponding device
21:52 xarses interperting debian init script implies that kvm will create /dev/kvm
21:53 xarses it does similar loading of kvm_intel and kvm_amd, but exits 0 if /dev/kvm doesn't exist
21:53 xarses beforehand
21:54 xarses its not a problem in centos, because it doesn't try to do this module loading service
21:55 aglarendil actually, kvm submodule loading will always create /dev/kvm if kvm is supported
21:55 aglarendil it will fail only of it is not supported
21:56 aglarendil guys, we have a really awesome issue with cinder
21:56 aglarendil https://bugs.launchpad.net/fuel/+bug/1316204
21:56 aglarendil I would like you to investigate it
21:56 aglarendil because my mind is blowing off
21:57 aglarendil cinder-scheduler does not schedule the volumes
21:57 aglarendil it simply does not send the message to corresponding cinder-volume.<host_name> queue
21:57 aglarendil I have waste 4 hours on debugging this and had no success
21:57 angdraug aglarendil: rmoe is looking at the cinder volume bug
21:58 teran joined #fuel-dev
21:58 aglarendil xarses: let's return to KVM bug. am I right that you are saying that qemu-kvm fails only if nested virt is not supported ?
22:01 xarses qemu-kvm fails to install in ubuntu on our host yes, probably because it does no support it well. So we cant test ubuntu at the moment
22:03 aglarendil do you mean the host node or the slave node ?
22:03 aglarendil I did not face this issue - we have a plenty of environments running ubuntu both as a host and as slave nodes
22:03 aglarendil I have my workstation with ubuntu and nested virt - no problems at all
22:04 aglarendil interesting moment about https://bugs.launchpad.net/fuel/+bug/1316204
22:04 aglarendil I created a loop device on the controller, configured volume group and started cinder-volume service
22:04 aglarendil it works just fine
22:05 aglarendil it looks like networking issue
22:06 xarses why are we combining cinder and mongo role?
22:06 xarses do we expect this to work? they use the same lvm volume
22:07 xarses i raised this issue before in mongo reviews
22:08 aglarendil actually, this is not the problem
22:08 aglarendil you can deploy even without mongo role - there is comment from Mike
22:08 xarses ok, cinder create works OK with ceph backend
22:09 EeeBotu New bug #1316812 in fuel: "ubuntu qemu-kvm package attempts to start a service and fails" [High, Confirmed] https://bugs.launchpad.net/bugs/1316812
22:16 teran joined #fuel-dev
22:31 aglarendil I have nailed it
22:32 aglarendil the problem is we somehow magically missed control_exchange configuration of cinder
22:32 aglarendil it should be set to cinder, not openstack on cinder-volume node
22:33 rmoe that's odd, the controllers have that set correctly but not any cinder node
22:34 aglarendil they need it to be synchronised between the nodes
22:34 aglarendil controller config had a commit setting this to cinder for rpc notifications for ceilometer
22:35 aglarendil and cinder-volume node did not have it
22:36 xarses ok, so that should fix that, how do we want to address mongo using all of the cinder vg?
22:37 xarses ban combining mongo and cinder-lvm?
22:37 aglarendil mongo is not using cinder VG, AFAIK.
22:37 aglarendil it uses its own VG
22:37 aglarendil called mongo
22:37 aglarendil xarses: Andrew, could you please update KVM bug description
22:38 aglarendil xarses: with more detailed description
22:38 aglarendil xarses: because it is still difficult even for me, let alone OSCI guys
22:39 xarses I'll create a diff with the fix then
22:42 aglarendil xarses: I mean, I have not yet understood, why it is important, what is the environment setup
22:42 aglarendil xarses: and what is the difference between actual and expected behavious
22:43 rmoe if your server (like srv11) has nested KVm enabled ubuntu deployments will all fail
22:43 aglarendil xarses: may be it is because it is already 3 AM in Moscow and I am too dumb for this period of time :)
22:43 aglarendil they do not fail
22:43 aglarendil I have my own workstation with nested kvm and have ubuntu tests passing
22:43 aglarendil also
22:43 rmoe it fails for us if we have nested kvm enabled on our dev server
22:44 aglarendil http://jenkins-product.srt.mirantis.net:8080/job/master_fuelmain.system_test.ubuntu.bvt_2/41/
22:44 aglarendil look into this job
22:44 rmoe does that server have nested kvm enabled?
22:44 aglarendil it is ha nova-network on srv11
22:45 aglarendil http://paste.openstack.org/show/79248/
22:45 rmoe ok
22:45 rmoe maybe that's a red herring then
22:46 aglarendil may be it is some local bug of your ubuntu
22:46 aglarendil and this package you want OSCI to alter is not the one that host system is gonna use
22:46 rmoe yeah
22:47 IlyaE joined #fuel-dev
22:47 aglarendil https://review.openstack.org/92484
22:47 aglarendil +1, guys :)
22:48 xarses on our debian SID dev server, we have kvm and kvm_amd, for whatever reason, nested kvm wont load in guest. The service tries to load kvm_amd (because flag is in /proc/cpuinfo) , but it shouldn't because /bin/kvm didn't get created which means that kvm harware acceleration wont work
22:49 xarses is that understandable?
22:50 rmoe you mean /dev/kvm not /bin/kvm right?
22:51 xarses correct, /dev/kvm
22:53 aglarendil and if you do manual load of kvm_intel on the slave nodes, is /dev/kvm created ?
22:53 xarses no, loading kvm_intel is invalid on amd, and loading kvm_amd errors with message in description
22:56 aglarendil this means you have nested virt not working on the slave nodes
22:56 aglarendil and this is expected behavious, I guess
22:57 aglarendil s/behavious/behavious/
22:57 aglarendil s/behavious/behaviour/
23:02 aglarendil xarses: I still do not think this is a bug. that's why I ask you to provide more detailed description
23:02 aglarendil xarses: it looks more like your local environment issue
23:02 xarses how our server got this way is not the issue
23:02 xarses the problem is that it causes the service to fail
23:03 xarses when it should just ignore the failure to load kvm acceleration
23:08 aglarendil AFAIK, it causes service to fail only if you choose KVM virt type on the slave nodes
23:08 aglarendil and puppet installs qemu-kvm instead of qemu
23:09 xarses we selected qemu virt type
23:09 xarses puppet fails because
23:10 xarses Setting up qemu-kvm (1.2.0+noroms-0ubuntu7.12.10) ...
23:10 xarses start: Job failed to start
23:10 xarses invoke-rc.d: initscript qemu-kvm, action "start" failed.
23:10 xarses dpkg: error processing qemu-kvm (--configure):
23:10 xarses subprocess installed post-installation script returned error exit status 1
23:10 aglarendil oh, I see. yep, that's a problem
23:11 aglarendil all we need, then, is to do this check right before we try to chown /dev/kvm
23:12 xarses if we ! -e /dev/kvm we would be fine
23:12 xarses before doing these tests
23:12 aglarendil not instead of /usr/bin/kvm checking
23:13 xarses Debian upstream only checks /dev/kvm
23:13 aglarendil /dev/kvm is created after modprobe is done
23:13 aglarendil it may rely on newer udev which creates /dev/kvm instead of module
23:13 rmoe debian just checks for /dev/kvm, and debian is better than ubuntu so therefore checking /dev/kvm is the right solution :)
23:14 aglarendil in this case /dev/kvm will not exist until we do modprobe
23:14 aglarendil and script will always exit without modprobing of corresponding module
23:15 angdraug dilyin: why did you comment out everything in https://review.openstack.org/#/c/92167?
23:16 aglarendil angdraug: I think Dmitry is sleeping right now. You can express your surprise in code review :)
23:30 teran joined #fuel-dev
23:31 angdraug I don't know when you guys go to sleep anymore ;)
23:58 IlyaE joined #fuel-dev

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