Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2014-04-25

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

All times shown according to UTC.

Time Nick Message
00:05 tzn joined #fuel-dev
00:08 IlyaE joined #fuel-dev
00:16 e0ne joined #fuel-dev
00:52 rmoe joined #fuel-dev
01:16 e0ne joined #fuel-dev
02:02 xarses joined #fuel-dev
02:07 tzn joined #fuel-dev
02:16 e0ne joined #fuel-dev
02:29 BillTheKat joined #fuel-dev
03:16 e0ne joined #fuel-dev
03:58 IlyaE joined #fuel-dev
04:09 tzn joined #fuel-dev
04:16 e0ne joined #fuel-dev
04:16 BillTheKat joined #fuel-dev
04:34 IlyaE joined #fuel-dev
05:10 tzn joined #fuel-dev
05:16 e0ne joined #fuel-dev
05:26 e0ne joined #fuel-dev
05:53 IlyaE joined #fuel-dev
06:00 enikanorov_ joined #fuel-dev
06:09 tzn joined #fuel-dev
06:14 vkozhukalov joined #fuel-dev
06:29 e0ne joined #fuel-dev
06:34 sambork joined #fuel-dev
06:40 sambork joined #fuel-dev
06:43 rvyalov joined #fuel-dev
06:49 tatyana_ joined #fuel-dev
07:01 salmon_ joined #fuel-dev
07:04 artem_panchenko joined #fuel-dev
07:13 tzn joined #fuel-dev
07:19 warpc joined #fuel-dev
07:27 mrasskazov joined #fuel-dev
07:38 e0ne joined #fuel-dev
07:45 e0ne_ joined #fuel-dev
08:21 mrasskazov joined #fuel-dev
08:26 e0ne joined #fuel-dev
08:26 e0ne hi
08:26 e0ne does anybody know about such error during iso build?
08:26 e0ne Global symbol "$forceyes" requires explicit package name at /usr/sbin/multistrap line 989.
08:26 e0ne Execution of /usr/sbin/multistrap aborted due to compilation errors.
08:27 holser_ It’s known Ubuntu 14.04 issue
08:28 holser_ get multistrap package from 12.04
08:29 mihgen holser_: I bet we need to add a section to docs about it ..
08:30 mihgen e0ne: please respond to comments here: https://review.openstack.org/#/c/87767/5/nailgun/nailgun/test/unit/test_network_check.py
08:30 holser_ You are right
08:30 mihgen e0ne: I'd like to merge it soon..
08:31 e0ne mihgen, looking on it
08:31 e0ne holser_, what version of multistrap should i use?
08:32 holser_ fuel)sgolovatiuk@mc0n2-msk:~/workspace/fuel-main$ dpkg -l | grep multistrap
08:32 holser_ ii  multistrap                          2.1.6ubuntu3                        all
08:33 e0ne_ joined #fuel-dev
08:33 holser_ e0ne: there is another resolution
08:34 holser_ e0ne: Open ./mirror/ubuntu/createchroot.mk +23 and add || exit 0 to the end
08:34 holser_ e0ne: sudo multistrap -a amd64  -f $(LOCAL_MIRROR_UBUNTU_OS_BASEURL)/multistrap.conf -d $(LOCAL_MIRROR_UBUNTU_OS_BASEURL)/chroot || exit 0
08:36 e0ne_ this workaround doesn't works for me. i've got other errors
08:36 holser_ what errors?
08:45 EeeBotu New bug #1312600 in fuel: "virtualbox/launch.sh is broken" [Undecided, New] https://bugs.launchpad.net/bugs/1312600
08:45 EeeBotu New bug #1312585 in fuel: "Unavailable node panel has invalid cursor" [Low, Confirmed] https://bugs.launchpad.net/bugs/1312585
08:51 dshulyak dilyin: hi, take a look please https://review.openstack.org/#/c/89895/
09:04 vkozhukalov joined #fuel-dev
09:05 mattymo I wish we used a normal tool like mock for chroot
09:07 dshulyak mattymo: can we deploy mongo nodes, except for primary, in parallel with computes, cinder lvm, ceph, etc?
09:09 rvyalov joined #fuel-dev
09:11 aglarendil dshulyak: Nope, otherwise your ceilometer nodes will not work as they do not have configured database
09:13 dshulyak aglarendil: even if one mongo node will be deployed? which is primary in nailgun context
09:14 aglarendil dshulyak: I am not sure that we should do this anyway. you can ask max mazur for details, but I think, there is no benefit to do this just before the code freeze
09:14 mattymo all mongo first, then controllers, then storage/computes
09:16 EeeBotu New bug #1312602 in fuel: "Primary node selection when nodes are added into deployed cluster" [High, Confirmed] https://bugs.launchpad.net/bugs/1312602
09:19 e0ne_ holser_: Global symbol "$forceyes" requires explicit package name at /usr/sbin/multistrap line 989.
09:20 holser_ I see
09:20 e0ne_ i'll try to build iso on our jenkins instead of desktop
09:20 holser_ just remove $forceyes from /usr/sbin/multistrap
09:20 holser_ it’s never specified
09:20 e0ne_ looks like update to ubuntu 14.04 was my mistake:(
09:21 holser_ install multistrap from 12.04
09:21 holser_ it will fix all your problems
09:22 holser_ wget http://mirrors.kernel.org/ubuntu/pool/universe/m/multistrap/multistrap_2.1.6ubuntu3_all.deb
09:22 holser_ dpkg -i multistrap_2.1.6ubuntu3_all.deb
09:24 MorAle joined #fuel-dev
09:26 mattymo holser_, alex_didenko apalkina anaboikina srv11 has a hung docker process :)
09:26 mattymo and bookwar :)
09:27 holser_ mattymo: our CI server should meet docker requierments
09:27 alex_didenko http://docs.docker.io/installation/ubuntulinux/#ubuntu-precise
09:27 mattymo oh it's on 3.5 kernel, not 3.8
09:27 vk joined #fuel-dev
09:27 alex_didenko yes
09:27 mattymo if we can get it upgraded to 3.8, these mount issues should be better
09:28 mattymo I definitely get this less in 3.10 than on 2.6
09:28 aglarendil hmm.
09:28 mattymo but there's still no aufs in fedora, but it is in ubuntu
09:28 aglarendil this was a stupid idea to update it ti 12.10
09:28 vkramskikh joined #fuel-dev
09:28 aglarendil 12.10 does not have officially backported kernels
09:29 alex_didenko I got zero hung/mount problems with docker on ubuntu-13.10 and 3.11.0-12-generic
09:29 vkramskikh joined #fuel-dev
09:29 bookwar mattymo: i'd rather not update srv11 right now, can we build the iso on one of the microclouds servers?
09:30 bookwar mattymo: which version do we need exactly?
09:30 mattymo bookwar, can you help get jenkins set up on my microcloud node?
09:30 alex_didenko ubuntu-13.10 or 14.04 should be fine
09:31 bookwar mattymo: yes, that's what i am thinking
09:31 mattymo 3.11.0-15-generic #25~precise1-Ubuntu this kernel is awesome
09:31 aglarendil this one is for 12.04
09:31 aglarendil and we have 12.10 on srv11
09:31 nurla on mc already installed U14.04
09:32 bookwar nurla: two days ago i've set up 12.04 node for mattymo, so we will use this one, just add it to jenkins
09:32 aglarendil we can try to install them from here: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.10.37-saucy/
09:32 aglarendil but these packages are guaranteed to work in 13.10
09:37 mattymo bookwar, can you put 172.18.8.7 on jenkins or point me to a doc for it?
09:38 bookwar mattymo: i am working on this right now, give me 5 minutes )
09:38 mattymo ok I will count
09:38 alex_didenko :)
09:39 mattymo alex_didenko, I will clone fuel_iso_with_gerrit_commits task so I can add our server to it for just docker build
09:39 mattymo I don't want it being chosen for normal isos
09:40 alex_didenko ok
09:41 BillTheKat joined #fuel-dev
09:47 EeeBotu New bug #1312620 in fuel: "Make shotgun to work inside of container" [High, Confirmed] https://bugs.launchpad.net/bugs/1312620
09:47 EeeBotu New bug #1312614 in fuel: "neutron nova notify service tenant id is missing on the primary controller" [Critical, Triaged] https://bugs.launchpad.net/bugs/1312614
09:49 bookwar mattymo: node enabled, you can restrict the jenkins job to run on label 'docker' then it will run on that particular server
09:53 sambork joined #fuel-dev
10:06 akupko joined #fuel-dev
10:07 vkramskikh is statement "today we'll have stable/5.0 branch so we can merge all pending requests" correct?
10:08 ykotko joined #fuel-dev
10:10 tnurlygayanov Hi there! :)
10:10 tnurlygayanov do we know the actual date for the release 5.0?
10:10 tnurlygayanov because I have different information, like May 9 and May 12...
10:14 bookwar there is this link https://wiki.openstack.org/wiki/Fuel/5.0_Release_Schedule but i'd ask mihgen if it is up to date
10:22 tnurlygayanov https://bugs.launchpad.net/fuel/+bug/1312638
10:22 tnurlygayanov bookwar: thank you!
10:23 aglarendil tnurlygayanov:  check your firewall
10:23 mihgen tnurlygayanov: yep, schedule is all right
10:23 tnurlygayanov root@comp:~/fuel-main/virtualbox# iptables -L
10:24 tnurlygayanov Chain INPUT (policy ACCEPT)
10:24 tnurlygayanov target     prot opt source               destination
10:24 tnurlygayanov ACCEPT     udp  --  anywhere             anywhere             udp dpt:domain
10:24 tnurlygayanov ACCEPT     tcp  --  anywhere             anywhere             tcp dpt:domain
10:24 tnurlygayanov ACCEPT     udp  --  anywhere             anywhere             udp dpt:bootps
10:24 tnurlygayanov ACCEPT     tcp  --  anywhere             anywhere             tcp dpt:bootps
10:24 tnurlygayanov ACCEPT     udp  --  anywhere             anywhere             udp dpt:domain
10:24 tnurlygayanov ACCEPT     tcp  --  anywhere             anywhere             tcp dpt:domain
10:24 tnurlygayanov ACCEPT     udp  --  anywhere             anywhere             udp dpt:bootps
10:24 tnurlygayanov ACCEPT     tcp  --  anywhere             anywhere             tcp dpt:bootps
10:24 mihgen tnurlygayanov: please use paste.openstack.org
10:24 tnurlygayanov Chain FORWARD (policy ACCEPT)
10:24 tnurlygayanov target     prot opt source               destination
10:24 tnurlygayanov ACCEPT     all  --  anywhere             anywhere
10:24 tnurlygayanov ACCEPT     all  --  anywhere             anywhere
10:24 tnurlygayanov ACCEPT     all  --  anywhere             anywhere
10:24 tnurlygayanov sorry.
10:24 mihgen tnurlygayanov: and don't do this  ever again here please
10:24 tnurlygayanov http://paste.openstack.org/show/77068/
10:25 mihgen tnurlygayanov: about schedule - if we find criticals after hard code freeze, we will have to respin a cycle of testing, create a new release candidate and test it
10:25 mihgen so it can move release date
10:25 tnurlygayanov mihgen: ok, thank you!
10:26 mihgen tnurlygayanov: np. more info is here: http://www.mail-archive.com/fuel-dev@lists.launchpad.net/msg00882.html
10:27 EeeBotu New bug #1312638 in fuel: "Fuel slave nodes can not reach the DHCP server" [High, Confirmed] https://bugs.launchpad.net/bugs/1312638
10:27 EeeBotu New bug #1312637 in fuel: "mongo setup should be before ceilometer setup" [Undecided, New] https://bugs.launchpad.net/bugs/1312637
10:27 EeeBotu New bug #1312631 in fuel: "cinder doesn't send notifications to ceilometer" [Undecided, In Progress] https://bugs.launchpad.net/bugs/1312631
10:27 EeeBotu New bug #1312627 in fuel: "HA scalability doesn't work" [High, New] https://bugs.launchpad.net/bugs/1312627
10:27 tnurlygayanov aglarendil: looks like I have no any bad firewall rules
10:28 dteselkin_ https://review.openstack.org/#/c/90313/
10:29 vkramskikh is statement "today we'll have stable/5.0 branch so we can merge all pending requests" correct?
10:32 bookwar vkramskikh: no
10:32 vkramskikh ;_;
10:33 bookwar vkramskikh: afaik we will have stable/5.0 branch on hard code freeze
10:33 mattymo alex_didenko, holser_ updated bootstrap script to use "dockerctl" not the undefined variable $dockerctl
10:47 EeeBotu New bug #1312652 in fuel: "No such file or directory - /etc/murano/murano.conf" [Critical, New] https://bugs.launchpad.net/bugs/1312652
10:47 EeeBotu New bug #1312645 in fuel: "outdated websockify package at least in Ubuntu" [Critical, Confirmed] https://bugs.launchpad.net/bugs/1312645
10:49 tzn joined #fuel-dev
11:00 e0ne joined #fuel-dev
11:00 tzn joined #fuel-dev
11:07 teran joined #fuel-dev
11:28 EeeBotu New bug #1312671 in fuel: "Bare-metal: CentOS provisioning hangs on DELL servers" [Critical, New] https://bugs.launchpad.net/bugs/1312671
11:28 EeeBotu New bug #1312666 in fuel: "OS controller cannot be deployed (ends with error)" [High, New] https://bugs.launchpad.net/bugs/1312666
11:28 EeeBotu New bug #1312663 in fuel: "tmp directories should be dynamic in build scripts" [Undecided, New] https://bugs.launchpad.net/bugs/1312663
11:28 EeeBotu New bug #1312662 in fuel: "IP Range controls have different vertical spacing" [Low, Confirmed] https://bugs.launchpad.net/bugs/1312662
11:28 EeeBotu New bug #1312657 in fuel: "/etc/cron.daily/logrotate:  error: 20-fuel.conf:3 duplicate log entry for /var/log/nailgun-agent.log" [Undecided, New] https://bugs.launchpad.net/bugs/1312657
11:28 EeeBotu New bug #1312654 in fuel: "build iso on Ubuntu 14.04 (multistrap issue)" [Medium, New] https://bugs.launchpad.net/bugs/1312654
11:28 EeeBotu New bug #1312653 in fuel: "HA-mode for vCenter doesn't work properly" [Undecided, New] https://bugs.launchpad.net/bugs/1312653
11:30 MorAle joined #fuel-dev
11:32 tnurlygayanov I checked this patch set https://review.openstack.org/#/c/90313/ with bug fix and it works on my environment
11:46 mattymo dshulyak, sorry 1 more fix
11:46 mattymo "by fuelmenu nad puppet on deploy" you misspelled and
11:53 mattymo alex_didenko, holser_ I added -L 2 to lrzip so it can decompress faster
11:53 mattymo deployment is really slow if you use standard compression level
11:54 alex_didenko ok, btw lrz archive made with "make docker" is 700+ Mb, while you had 200+
11:54 alex_didenko lrzip info shows unpacked size 5G
11:55 holser_ mattymo: ok
12:00 mattymo yes because you're not using -U option
12:00 mattymo you must tar into a big file, then run a separate lrzip not from stdin
12:01 tzn joined #fuel-dev
12:01 mattymo actually the iso I made is 230mb
12:09 roman_vyalov joined #fuel-dev
12:25 mihgen alex_didenko: do you know if we will have separate LVM for /var/log/remote  on master node?
12:25 mihgen alex_didenko: there is bug https://bugs.launchpad.net/fuel/+bug/1283821
12:25 mihgen which I want to raise up to High… don't want master node to be polluted by logs
12:26 alex_didenko mihgen: I'm not aware of any changes regarding moving logs to a separate LVM
12:29 steale joined #fuel-dev
12:29 b-zone joined #fuel-dev
12:29 EeeBotu New bug #1312692 in fuel: "[UI] Show mcollective logs in Logs for provisioned/deployment" [Wishlist, New] https://bugs.launchpad.net/bugs/1312692
12:29 EeeBotu New bug #1312686 in fuel: "Can't upload image with ceph deploy" [Medium, New] https://bugs.launchpad.net/bugs/1312686
12:29 morale_ joined #fuel-dev
12:30 MorAle left #fuel-dev
12:30 vkramskikh guys, we have created channel #fuel-ui, if you have any questions feel free to join
12:34 e0ne joined #fuel-dev
12:36 warpc joined #fuel-dev
12:37 vkramskikh murano guys, why do you add 10+ random reviewers for every CR?
12:40 * mihgen ohhh
12:40 mihgen who broke ISO ??
12:40 mihgen I see smoke test failed
12:41 mihgen nurla: did you see it? any ideas? ^^
12:42 alex_didenko mattymo: ok, thx for the info, I'll check/update makefiles
12:43 mattymo alex_didenko, starting supervisord before dockerctl build all causes containers to get built and then start in the wrong order
12:43 nurla mihgen:  we have many failures on ostf tests
12:44 alex_didenko mattymo: ok, I'll look at it too
12:45 jobewan joined #fuel-dev
12:45 mihgen nurla: we need a critical bug about it, and get someone looking into it…
12:46 mihgen nurla: any ideas what could lead to this breakage ?
12:46 mihgen particular team, so we can assign bug to someone immediately?
12:46 mattymo alex_didenko, I'm just going to puppetize all the steps between mkdir -p /var/www/nailgun/dump and dockerctl build all
12:47 mattymo copying /etc/astute.yaml during puppet run is probably a bad idea since it may influence operation during run
12:47 alex_didenko mattymo: ok, I was thinking about the same thing :)
12:47 mattymo lrunzip and docker load..... I'm not sure if we should do it in puppet or not
12:49 mihgen monester: https://bugs.launchpad.net/fuel/+bug/1312238 should we add devops tag to this issue?
12:49 b-zone left #fuel-dev
12:49 mihgen monester: can we move it to 5.1 or it should be done in 5.0 ?
12:51 monester mihgen: we can move this bug to 5.1
12:52 mihgen aglarendil: dpyzhov: do you guys know what we merged over 3 last hours that could cause ISO breakage ?
12:55 ikalnitsky joined #fuel-dev
12:59 xdeller_ joined #fuel-dev
13:01 sambork left #fuel-dev
13:01 sambork joined #fuel-dev
13:02 sambork joined #fuel-dev
13:03 xdeller joined #fuel-dev
13:15 tnurlygayanov why we moved https://bugs.launchpad.net/fuel/+bug/1288135 to 5.1? we already have fix fir this problem and I verified this fix on my environment
13:16 mihgen tnurlygayanov: today is last day when we can address medium and low priority bugs
13:16 mihgen tnurlygayanov: so if you have a fix, please propose
13:17 mihgen tnurlygayanov: I'll return it back to 5.0
13:17 BillTheKat joined #fuel-dev
13:19 tnurlygayanov *sorry, it is another issue.
13:19 tnurlygayanov I mean this fix https://review.openstack.org/#/c/90313/
13:20 tnurlygayanov this is for another bug
13:20 mihgen tnurlygayanov: ok. I will move that one to 5.1 ?
13:26 tnurlygayanov mihgen:I suggest to merge fix for https://bugs.launchpad.net/fuel/+bug/1312600 in 5.0
13:26 tnurlygayanov we alsredy have fix for it
13:26 tnurlygayanov for https://bugs.launchpad.net/fuel/+bug/1288135 - ok, we can fix it in the next releases.
13:26 tnurlygayanov *already
13:26 mihgen MiroslavAnashkin: can you please test it? ^^
13:27 mihgen tnurlygayanov: we will get required +1th, and merge it
13:27 tnurlygayanov ok, thank you!
13:29 MiroslavAnashkin Oh, one more incorrect solution for VBox...
13:29 EeeBotu New bug #1312711 in fuel: "outdated packages in OSCI repository" [High, Triaged] https://bugs.launchpad.net/bugs/1312711
13:33 MiroslavAnashkin tnurlygayanov: From our docs: `iptables -t nat -A POSTROUTING -s 172.16.1.0/24 \! -d 172.16.1.0/24 -j MASQUERADE`
13:40 evgeniyl mattymo: found another problem, maybe it already solved in newest containers, we have symlink /etc/astute.yaml -> /etc/fuel/astute.yaml , I shared /etc/fuel with containers, and when I try to run a container, it fails because /etc/fuel/astute.yaml is symlink on nonexistent file.
13:40 mattymo newer container has it fixed)
13:41 mattymo but the latest has issues with supervisor starting before dockerctl builds all
13:41 evgeniyl mattymo: cool
13:47 tzn joined #fuel-dev
13:51 mihgen vkozhukalov: did we merge any code around disk changes?
13:51 mihgen smoke fails with this issue: http://paste.openstack.org/show/77089/
13:51 mihgen this is CRITICAL one which must be fixed ASAP
13:52 mihgen https://bugs.launchpad.net/fuel/+bug/1312755 this one actually
13:53 EeeBotu New bug #1312758 in fuel: "Master node bootstrapping yum issue" [Medium, New] https://bugs.launchpad.net/bugs/1312758
13:53 EeeBotu New bug #1312755 in fuel: "Instance failed to get Active state" [Critical, New] https://bugs.launchpad.net/bugs/1312755
13:53 EeeBotu New bug #1312751 in fuel: "OS controller fails to deploy: Deployment has failed. Error occurred while running method 'deploy'. Inspect Orchestrator logs for the details." [Undecided, New] https://bugs.launchpad.net/bugs/1312751
13:53 EeeBotu New bug #1312743 in fuel: "Murano conf issue" [Medium, New] https://bugs.launchpad.net/bugs/1312743
13:53 EeeBotu New bug #1312741 in fuel: "Focus is lost when editing a field on Network tab with visible network verification result" [Medium, Confirmed] https://bugs.launchpad.net/bugs/1312741
13:53 MiroslavAnashkin mihgen: Disk changes tests are in progress.
13:58 mihgen wait, we merged changes in scheduler
13:58 mihgen could it be an issue?
13:58 mihgen nurla: Tatyanka_Leontov aglarendil https://review.openstack.org/#/c/90106/
13:59 mihgen holser_: did you test standard installation and TestVM run ?
13:59 mihgen holser_: with this patchset? ^^
13:59 holser_ I did
14:00 holser_ standard installation
14:00 holser_ then tested nova.cfg
14:00 holser_ and started a couple of VMs
14:01 holser_ honestly, to do proper test we need to try to start more than server can handle and see distribution across the nodes - but it’s very complex scenario
14:04 mihgen rvyalov: are you sure that we didn't touch kernel?
14:04 mihgen what about nbd kernel module?
14:04 mihgen holser_: ok. thing is that it fails to start even TestVM..
14:04 mihgen which is very tiny one
14:05 rvyalov mihgen: patch for kernel-lt apply yesterday
14:05 mihgen rvyalov: nurla do we have qemu-nbd installed?
14:06 mihgen anything with libguestfs ? did we play with it today?
14:06 akasatkin left #fuel-dev
14:06 alex_didenko guys, I'm looking through release notes for 5.0 and see no mention about puppet upgrade from 2.7 to 3.4. Shouldn't we mention about it?
14:07 alex_didenko also nothing about upgrading ruby to 2.1 on Fuel master node
14:14 e0ne joined #fuel-dev
14:19 e0ne_ joined #fuel-dev
14:55 aglarendil it looks like that we lost python-guestfs library dependency for some packages and that's why smoke test is failing
14:56 aglarendil the optimal varinant is to add python-guestfs to nova-compute requirements
15:03 IlyaE joined #fuel-dev
15:05 BillTheKat joined #fuel-dev
15:16 e0ne joined #fuel-dev
15:25 tzn joined #fuel-dev
15:28 b-zone joined #fuel-dev
15:32 EeeBotu New bug #1312790 in fuel: "error in nova puppet scripts" [High, Confirmed] https://bugs.launchpad.net/bugs/1312790
15:41 jaypipes joined #fuel-dev
15:41 xarses joined #fuel-dev
15:42 warpc joined #fuel-dev
15:43 IvanBerezovskiy left #fuel-dev
15:46 asledzinskiy left #fuel-dev
15:48 vkramskikh_ joined #fuel-dev
15:49 vkramskikh joined #fuel-dev
15:49 vkramskikh joined #fuel-dev
15:51 vkramskikh joined #fuel-dev
15:52 vkramskikh joined #fuel-dev
15:54 vkramskikh_ joined #fuel-dev
15:54 vkramskikh joined #fuel-dev
15:56 rmoe joined #fuel-dev
15:56 vkramskikh joined #fuel-dev
15:58 angdraug joined #fuel-dev
16:03 tzn joined #fuel-dev
16:25 BillTheKat joined #fuel-dev
16:28 mihgen aglarendil: > we lost python-guestfs library dependency
16:28 mihgen aglarendil: how that could happen?
16:29 mihgen aglarendil: so did we rebuild nova-compute with this dependency?
16:30 aglarendil mihgen: I  have just returned from a job interview. I need 5 minutes to understand what is the current status
16:43 aglarendil it looks like python-libguestfs is missing fuse-libs dependency
16:43 aglarendil and thus nova is not able to load guestfs library
16:47 mihgen did we update python-libguestfs?
16:47 angdraug xarses: https://review.openstack.org/90272
16:47 mihgen aglarendil: so what was the issue in centos?
16:47 MiroslavAnashkin mihgen: What service should we use to deliver updated Fuel ISO to customers who needs it? Google drive or download.mirantis.com?
16:47 aglarendil I need to recheck it for ubuntu
16:48 aglarendil mihgen: but I think it is also related to ubuntu
16:50 mihgen aglarendil: ok, please let me know what was the root cause .. pretty interesting
16:53 aglarendil I will
16:58 bookwar there are two more tests we had running on the same iso as the smoke test. You can check them here http://jenkins-product.srt.mirantis.net:8080/job/fuel_5_0_iso/142/downstreambuildview/  One with centos and one with Ubuntu. Maybe they'll help you with analysis
16:59 aglarendil there are also some problems with scheduler
17:01 aglarendil our fix for nova scheduler that improves core filtering is failing because we have corefilter set to 1.0 and we are trying to run 2 virtual machines in parallel on the same compute host
17:06 aglarendil finally.
17:06 aglarendil got it
17:08 openstackstatus joined #fuel-dev
17:14 aglarendil https://review.openstack.org/#/c/90426/
17:15 mihgen aglarendil: so the smoke test issue was caused by scheduler changes only?
17:15 aglarendil nope
17:15 aglarendil by both bugs
17:15 aglarendil scheduler changes lead to failure of one test
17:15 aglarendil and libfuse2 issue lead to failure of another test
17:15 aglarendil please +1 my request ^^
17:16 openstackstatus NOTICE: Gerrit will be unavailable for a few hours starting at 1600 UTC on Monday April 28th for an upgrade. See https://wiki.openstack.org/wiki/GerritUpgrade
17:16 aglarendil actually, this scheduler fix passed FUEL CI successfully and I consider this as a miracle
17:17 mihgen aglarendil: do we run OSTF on Fuel CI now?
17:17 aglarendil at least for CentOS + nova-network
17:17 aglarendil more details can provide nurla
17:17 mihgen aglarendil: and I still don't understand another issue, libfuse2
17:17 mihgen what did we change
17:17 mihgen it was working fine in the morning
17:17 aglarendil I do not know, but I know that absence of libfuse made nova use nbd driver which is rather unstable
17:18 mihgen interesting, so might be even lucky that we caught this intermittent issue
17:18 aglarendil yep
17:19 aglarendil it could be some kind of floating bug
17:19 mihgen smokes were failing 3 times in a row though..  might be because of just scheduler
17:20 aglarendil Actually scheduler issue may lead only to failure of the only test that launches 2 vms in parallel
17:21 mihgen aglarendil: and we do it every OSTF run, so it's easy to catch :)
17:22 aglarendil well, Mike, may be scheduler is a little bit racy
17:22 mihgen I put +1 for cpu_allocation_ratio = 8.0 … though wondering if we may need 4 or even lower value
17:22 aglarendil I just used chose a middle between 16 (default) and 1
17:23 aglarendil I actually think that the only problem was scheduler
17:23 aglarendil that was a little bit racy
17:25 aglarendil thus allowing tests to pass sometimes
17:25 aglarendil because it makes decisions based on the current state of database. and database updates are racy
17:25 nurla we have ostf run on Fuel Ci with centos cluster
17:26 aglarendil thus if you even try to boot 2 vms one by one you can face the issue
17:26 aglarendil that you are not able to do so
17:26 aglarendil because scheduler thinks that you are still running 2 vms
17:26 aglarendil it is really funny
17:26 aglarendil *I meant 1 vm
17:27 mihgen interesting
17:27 mihgen all right, I'm happy it's done
17:27 mihgen do we need to sync mirrors first?
17:27 mihgen and then re-run smoke
17:30 aglarendil I think we can rebuild ISO at least now as fuse2 issue may not be related
17:30 IlyaE joined #fuel-dev
17:30 aglarendil I am gonna trigger ISO rebuild
17:33 r0mikiam dteselkin_: https://gerrit.mirantis.com/#/c/15040/ +1 pls
17:42 tzn joined #fuel-dev
17:46 artem_panchenko joined #fuel-dev
17:57 mihgen I see -1 from Jenkins: https://review.openstack.org/#/c/90023/1
17:58 mihgen I'm wondering if openstack ci already down..
18:00 dshulyak_ joined #fuel-dev
18:20 xarses joined #fuel-dev
18:23 bookwar smoke has finished successfully
18:33 jeremydei joined #fuel-dev
18:44 mrasskazov joined #fuel-dev
19:01 xarses joined #fuel-dev
19:05 xarses joined #fuel-dev
19:06 vkozhukalov joined #fuel-dev
19:08 tzn joined #fuel-dev
19:12 vkramskikh joined #fuel-dev
19:13 IlyaE joined #fuel-dev
19:22 tzn joined #fuel-dev
19:40 xarses aglarendil: https://bugs.launchpad.net/fuel/+bug/1310926, bug isn't complete patches to kill processes in namespace are related to https://bugs.launchpad.net/fuel/+bug/1285929
19:41 xarses aglarendil: https://bugs.launchpad.net/fuel/+bug/1310926 needs to be fixed by patch to neutron
19:44 fandi joined #fuel-dev
20:31 BillTheKat joined #fuel-dev
20:38 vkramskikh joined #fuel-dev
20:45 openstackstatus joined #fuel-dev
20:50 openstackstatus joined #fuel-dev
20:52 IlyaE joined #fuel-dev
20:58 openstackstatus NOTICE: Gerrit will be unavailable for a few hours starting at 1600 UTC on Monday April 28th for an upgrade. See https://wiki.openstack.org/wiki/GerritUpgrade
21:15 aleksandr_null joined #fuel-dev
22:17 xarses joined #fuel-dev
22:39 IlyaE joined #fuel-dev
22:57 xarses joined #fuel-dev
23:05 tatyana joined #fuel-dev
23:43 IlyaE joined #fuel-dev

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