*** morazi has quit IRC | 00:05 | |
*** Goneri has joined #tripleo | 00:13 | |
*** shivrao has quit IRC | 00:14 | |
*** alop has quit IRC | 00:17 | |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/213942 | 00:22 |
---|---|---|
*** shadower has quit IRC | 00:23 | |
*** shadower has joined #tripleo | 00:23 | |
openstackgerrit | Merged openstack-infra/tripleo-ci: Test fix for ironic ephemeral drive problem https://review.openstack.org/205273 | 00:28 |
*** rwsu has quit IRC | 00:36 | |
openstackgerrit | Merged openstack/tripleo-heat-templates: Remove hardcoded bridge name in bonded compute NIC config https://review.openstack.org/213861 | 00:37 |
*** penick has quit IRC | 01:27 | |
openstackgerrit | Graeme Gillies proposed openstack/os-cloud-config: Add wol ironic driver support https://review.openstack.org/213956 | 01:34 |
*** sthillma has quit IRC | 01:46 | |
*** yamahata has quit IRC | 01:51 | |
*** rbrady has quit IRC | 02:02 | |
*** olaph has joined #tripleo | 02:13 | |
*** w_ has joined #tripleo | 02:21 | |
*** olaph has quit IRC | 02:23 | |
*** yamahata has joined #tripleo | 02:30 | |
*** yamahata has quit IRC | 02:32 | |
*** yamahata has joined #tripleo | 02:32 | |
*** lazy_prince has joined #tripleo | 02:45 | |
*** untriaged-bot has joined #tripleo | 03:00 | |
untriaged-bot | Untriaged bugs so far: | 03:00 |
untriaged-bot | https://bugs.launchpad.net/os-collect-config/+bug/1482510 | 03:00 |
openstack | Launchpad bug 1482510 in heat "OS::Heat::SoftwareDeployment failed due SSL certificate verification error" [Medium,Triaged] - Assigned to Rico Lin (rico-lin) | 03:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1466037 | 03:00 |
openstack | Launchpad bug 1466037 in diskimage-builder "Signed Fedora and Ubuntu user image built by DIB can`t boot on HP DL380 Gen8 server for lack of mpt2sas driver" [Undecided,Incomplete] | 03:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1483385 | 03:00 |
openstack | Launchpad bug 1483385 in diskimage-builder "install_grub failing for centos7" [Undecided,In progress] - Assigned to Abel Lopez (al592b) | 03:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1471802 | 03:00 |
openstack | Launchpad bug 1471802 in diskimage-builder "ironic-agent element hardcodes interfaces names for DHCP." [Undecided,Fix committed] - Assigned to Om Kumar (om-kumar) | 03:00 |
*** untriaged-bot has quit IRC | 03:00 | |
*** sdake has joined #tripleo | 03:32 | |
*** lazy_prince has quit IRC | 03:33 | |
*** sdake_ has quit IRC | 03:36 | |
*** sdake_ has joined #tripleo | 03:42 | |
*** Goneri has quit IRC | 03:45 | |
*** sdake has quit IRC | 03:45 | |
*** sdake_ is now known as sdake | 04:22 | |
*** adrianopetrich has quit IRC | 04:59 | |
*** mestery has joined #tripleo | 04:59 | |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow customization of Ceph pool names https://review.openstack.org/212664 | 05:02 |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow customization of Ceph client user https://review.openstack.org/212686 | 05:02 |
*** mestery has quit IRC | 05:09 | |
*** adrianopetrich has joined #tripleo | 05:13 | |
*** tzumainn has quit IRC | 05:16 | |
*** rlandy has quit IRC | 05:35 | |
*** lsmola has joined #tripleo | 05:38 | |
*** masco has joined #tripleo | 06:02 | |
*** adrianopetrich has quit IRC | 06:03 | |
*** adrianopetrich has joined #tripleo | 06:04 | |
*** Marga_ has joined #tripleo | 06:08 | |
*** Marga_ has quit IRC | 06:09 | |
*** panda has quit IRC | 06:09 | |
*** Marga_ has joined #tripleo | 06:09 | |
*** panda has joined #tripleo | 06:09 | |
*** adrianopetrich has quit IRC | 06:12 | |
*** adrianopetrich has joined #tripleo | 06:12 | |
*** adrianopetrich has quit IRC | 06:23 | |
*** adrianopetrich has joined #tripleo | 06:23 | |
*** adrianopetrich has joined #tripleo | 06:24 | |
*** bvandenh has joined #tripleo | 06:27 | |
*** jtomasek has joined #tripleo | 06:28 | |
*** adrianopetrich has quit IRC | 06:33 | |
*** sthillma has joined #tripleo | 06:34 | |
*** sthillma_ has joined #tripleo | 06:37 | |
*** sthillma has quit IRC | 06:39 | |
*** sthillma_ is now known as sthillma | 06:39 | |
*** jprovazn has joined #tripleo | 06:40 | |
*** openstackgerrit_ has joined #tripleo | 06:43 | |
*** adrianopetrich has joined #tripleo | 06:45 | |
*** ifarkas has joined #tripleo | 07:09 | |
*** hewbrocca has quit IRC | 07:12 | |
*** sthillma has quit IRC | 07:16 | |
*** shardy has joined #tripleo | 07:36 | |
openstackgerrit | Yanis Guenane proposed openstack-infra/tripleo-ci: Adding more CPU in the HA scenario https://review.openstack.org/213885 | 07:43 |
*** aufi has joined #tripleo | 07:43 | |
*** paramite has joined #tripleo | 07:55 | |
*** jistr has joined #tripleo | 07:55 | |
*** mbound has joined #tripleo | 07:55 | |
*** jcoufal has joined #tripleo | 07:56 | |
*** devvesa has joined #tripleo | 07:56 | |
*** athomas has joined #tripleo | 08:01 | |
*** gfidente has joined #tripleo | 08:01 | |
*** lucasagomes has joined #tripleo | 08:02 | |
*** hewbrocca has joined #tripleo | 08:02 | |
*** mcornea has joined #tripleo | 08:10 | |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/111011 | 08:11 |
*** derekh has joined #tripleo | 08:12 | |
*** derekh has quit IRC | 08:12 | |
*** derekh has joined #tripleo | 08:13 | |
spredzy | derekh, morning. ping https://review.openstack.org/#/c/213885/ | 08:29 |
spredzy | do you think it is possible to do it over the week end or at time lowest number of contribution are made | 08:29 |
derekh | spredzy: ya, that would be the best time to do it, the window todo it should be about 3 or 4 hours | 08:31 |
derekh | spredzy: I'm wondering how sure you are? its not just a simple as bumping up the CPU count, if we change it we also have to consider that we are now overcommiting on CPU on each host and that may also cause problems | 08:32 |
derekh | spredzy: we we may also need to reduce capacity aswell | 08:32 |
spredzy | derekh, well thanks to the platform you provided I saw that there was just one CPU following on jistr review (https://review.openstack.org/#/c/213737/) I realized that on my setup I had the 2.3.5 version but 2 CPUs | 08:36 |
spredzy | derekh, I can rerun devtest on my local setup with a single CPU if you want me to confirm. But based on the logs I have pasted on #clusterlabs they were also pointing to the CPU being overloaded | 08:37 |
derekh | spredzy: OK, I'll take a look at the numbers later to see how much doubling the CPU per instance would cause us to overcommit on CPU. | 08:39 |
jistr | any news re what could be the possible cause? could it be just slow machines? (based on what i'm reading here now) | 08:39 |
jistr | i'm wondering if it makes sense to try update my corosnyc pin patches | 08:40 |
jistr | (seems like 2.3.5 is still being installed in my patch, i'll probably have to do yum downgrade later in the build instead of yum install early in the build) | 08:40 |
spredzy | jistr, cpu overloaded corosync can't pass the message correctly from one node to another etc.... | 08:41 |
spredzy | jistr, on your local setup do you have 1 or 2 cpu per node? | 08:41 |
gfidente | I used to do 1 cpu per node | 08:42 |
gfidente | but CI seems to be particularily slow | 08:42 |
jistr | yeah i just checked, i have 1 cpu too | 08:42 |
gfidente | having 2 cpus would increase memory footprint though | 08:42 |
jistr | i wonder how come it wasn't an issue until now | 08:42 |
gfidente | also true | 08:43 |
jistr | and now it seems to be an issue on every run | 08:43 |
*** karume has joined #tripleo | 08:43 | |
spredzy | jistr, not sure when is the last time puppet-ha passed. But apparently in last version 2.3.5 corosync consumes more cpu that in previous release | 08:43 |
spredzy | way more based on feedback I have had yesterday | 08:44 |
*** adrianopetrich has quit IRC | 08:44 | |
jistr | ah ok, that could be it then | 08:44 |
spredzy | jistr, gfidente could you replicate the flapping issue on your local setup? | 08:44 |
gfidente | was just checking the version of corosync, 2.3.5 | 08:45 |
gfidente | but I don't think I've hit this | 08:46 |
openstackgerrit | Jiri Stransky proposed openstack/diskimage-builder: Fedora: install older corosync https://review.openstack.org/213736 | 08:46 |
gfidente | oh, maybe the CI guests are configured to run in qemu emulation instead of hvm? | 08:47 |
gfidente | do we have output from lscpu or something? | 08:47 |
spredzy | derekh, ^ | 08:49 |
spredzy | for the emulation part | 08:49 |
spredzy | gfidente, apparently not is hostt_info | 08:49 |
gfidente | derekh, every time you post something with Nothing to see I get neuro linguistic programmed and go straight to check what that is! :) | 08:49 |
spredzy | gfidente, http://fpaste.org/256015/98483971/ | 08:49 |
derekh | gfidente: I reckon your not the only one | 08:50 |
gfidente | ahaha | 08:50 |
spredzy | that was taken from one of the CI machine | 08:50 |
derekh | spredzy: they should be using hvm, I'll check now | 08:50 |
*** yamahata has quit IRC | 08:51 | |
openstackgerrit | Jiri Stransky proposed openstack-infra/tripleo-ci: Fedora: install older corosync https://review.openstack.org/213737 | 08:51 |
derekh | spredzy: HA job passed twice yesterday http://goodsquishy.com/downloads/tripleo-jobs.html | 08:51 |
gfidente | yeah they seem to be using hvm | 08:51 |
jistr | oh that's the corosync pin patch i just updated.... i didn't notice it succeeded | 08:52 |
derekh | qemu command : /usr/bin/qemu-system-x86_64 -machine accel=kvm ........ | 08:52 |
jistr | i checked logs yesterday evening and it still seemed to install 2.3.5 though, so i'm a bit puzzled how come it passed | 08:53 |
* jistr checks again | 08:53 | |
jistr | 2015-08-17 21:22:45.205 | Installing : corosync-2.3.5-1.fc21.x86_64 31/170 | 08:53 |
jistr | interesting | 08:54 |
gfidente | fwiw, the CI host CPU seems to be a little faster than what I have here | 08:54 |
gfidente | but I get frequent _monitor actions to Time Out | 08:55 |
derekh | jistr, if your patch didn't work to pin the corosync version this might help : https://review.openstack.org/#/c/141779/ | 08:55 |
gfidente | so I think we're somehow close to the limit anyway | 08:55 |
gfidente | spredzy, jistr, do you get the _monitor timeouts on local env? | 08:56 |
jistr | derekh: thx, i'll update it like this. I did `yum -y downgrade` but some of the jobs might not install corosync on the images i think, and then the downgrade could fail i guess. | 08:57 |
spredzy | gfidente, just teared down my setup let me respawn one | 08:57 |
jistr | previously i did just `yum -y install` early in the run, and that was overriden later by 01-package-installs | 08:57 |
derekh | jistr: yup, the versionlock part should sort that out | 08:58 |
*** untriaged-bot has joined #tripleo | 09:00 | |
untriaged-bot | Untriaged bugs so far: | 09:00 |
untriaged-bot | https://bugs.launchpad.net/os-collect-config/+bug/1482510 | 09:00 |
openstack | Launchpad bug 1482510 in heat "OS::Heat::SoftwareDeployment failed due SSL certificate verification error" [Medium,Triaged] - Assigned to Rico Lin (rico-lin) | 09:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1466037 | 09:00 |
openstack | Launchpad bug 1466037 in diskimage-builder "Signed Fedora and Ubuntu user image built by DIB can`t boot on HP DL380 Gen8 server for lack of mpt2sas driver" [Undecided,Incomplete] | 09:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1483385 | 09:00 |
openstack | Launchpad bug 1483385 in diskimage-builder "install_grub failing for centos7" [Undecided,In progress] - Assigned to Abel Lopez (al592b) | 09:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1471802 | 09:00 |
openstack | Launchpad bug 1471802 in diskimage-builder "ironic-agent element hardcodes interfaces names for DHCP." [Undecided,Fix committed] - Assigned to Om Kumar (om-kumar) | 09:00 |
*** untriaged-bot has quit IRC | 09:00 | |
jistr | derekh: hmm now i wonder -- won't "yum install -y --nogpg yum-versionlock <koji pkgs>" say "yum-versionlock installed, but the <koji pkgs> are already present in newer version, i won't do anything about those"? | 09:00 |
* jistr checks logs | 09:01 | |
jistr | nope the logs are already gone, too old CI run | 09:02 |
jistr | well if it worked back then, i hope it will work now as well :) | 09:02 |
* spredzy afk 15min | 09:02 | |
derekh | jistr: I foget but I'm guessing thats why I put it in /pre-install.d | 09:02 |
derekh | jistr: so it happened b4 mariadb was installed | 09:03 |
jistr | derekh: oh pre-install, i didn't notice that. Thx :) | 09:03 |
jistr | now it makes sense :) | 09:03 |
*** paramite is now known as paramite|afk | 09:08 | |
*** pelix has joined #tripleo | 09:09 | |
*** paramite|afk is now known as paramite | 09:13 | |
openstackgerrit | Jiri Stransky proposed openstack/diskimage-builder: Fedora: install older corosync https://review.openstack.org/213736 | 09:19 |
openstackgerrit | Jiri Stransky proposed openstack-infra/tripleo-ci: Fedora: install older corosync https://review.openstack.org/213737 | 09:20 |
*** dasm_ is now known as dasm | 09:36 | |
*** athomas has quit IRC | 09:44 | |
*** athomas has joined #tripleo | 09:48 | |
gfidente | vx9V681O! | 09:53 |
gfidente | vx9V681O! | 09:53 |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Provide a param to map block devices to Ceph OSDs https://review.openstack.org/200338 | 09:53 |
gfidente | I can type it another time FWIW :) | 09:53 |
hewbrocca | again! again! | 09:54 |
*** jrist has quit IRC | 09:54 | |
* spredzy back by the way | 10:00 | |
*** adrianopetrich has joined #tripleo | 10:01 | |
*** jprovazn has quit IRC | 10:01 | |
*** adrianopetrich has quit IRC | 10:02 | |
*** adrianopetrich has joined #tripleo | 10:02 | |
*** pbourke has quit IRC | 10:11 | |
*** akrivoka has joined #tripleo | 10:14 | |
*** pdb has joined #tripleo | 10:14 | |
*** sdake has quit IRC | 10:15 | |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Pin heat and neutron https://review.openstack.org/214112 | 10:19 |
shardy | derekh: Re the heat pin, does upstream CI use stack domain users yet? | 10:21 |
shardy | last time I looked the devtest undercloud didn't but I've not looked recently | 10:21 |
shardy | context https://bugs.launchpad.net/heat/+bug/1484086 | 10:21 |
openstack | Launchpad bug 1484086 in Keystone "ec2tokens authentication is failing during Heat tests" [Undecided,Incomplete] | 10:21 |
shardy | keystone has broken our ec2tokens auth for non-default domains unless you specify a v3 keystone endpoint in the heat conf | 10:22 |
derekh | shardy: not sure, lemme check | 10:22 |
* shardy doesn't have a working devtest env handy atm | 10:23 | |
shardy | thanks! | 10:23 |
*** jprovazn has joined #tripleo | 10:23 | |
derekh | shardy: doesn't look like it, everything in the keystone user table is in the default domain | 10:24 |
shardy | derekh: Ah, Ok, probably not the issue then | 10:25 |
shardy | derekh: do we have any other clues about why heat may be breaking things? | 10:25 |
derekh | shardy: I reproduced it locally and noticed a bunch of the software deployments metadata was missing, | 10:27 |
derekh | shardy: I think I know why we're not getting overcloud node logs in ci, I'll fix that now which should make debugging easier | 10:27 |
shardy | derekh: Ok, https://bugs.launchpad.net/heat/+bug/1477329 was related to races which meant deployments metadata got lost | 10:28 |
openstack | Launchpad bug 1477329 in heat "Potential race in software deployments db write to server rsrc_metadata" [High,Fix committed] - Assigned to Steve Baker (steve-stevebaker) | 10:28 |
shardy | I wonder if that could be related | 10:29 |
shardy | https://review.openstack.org/#/c/206256/ makes a pretty major change to the deployments metadata logic | 10:30 |
shardy | intended to solve the lost-metadata issue, but... | 10:30 |
derekh | shardy: I tried devtest with that patch reverted locally and it didn't work but I'm not satisfied that the reason wasn't an additional neutron regression | 10:31 |
shardy | derekh: Ok cool, will be interesting to see what the logs say | 10:32 |
derekh | did we change the heat-admin user to "fedora" at some stage ? | 10:32 |
shardy | derekh: Yes there was a change in defaults IIRC which means you just get the cloud image/cloud-init default user | 10:33 |
* shardy looks for commit | 10:33 | |
shardy | https://review.openstack.org/#/c/202614/ | 10:34 |
shardy | derekh: you can reinstate the previous behavior by setting instance_user=heat-admin in heat.conf | 10:35 |
derekh | shardy: ack, so thats the reason we're not getting logs from the CI overcloud nodes | 10:35 |
shardy | IIRC the ec2-user is swapped out for heat-admin in the undercloud config (at least downstream it is..) | 10:36 |
openstackgerrit | xiexs proposed openstack/os-cloud-config: Refactor _extract_driver_info() to reduce complexity. https://review.openstack.org/214128 | 10:37 |
derekh | Anybody want to just merge https://review.openstack.org/#/c/214112/ now ? its the same patch as I tested here https://review.openstack.org/#/c/111011/28 | 10:40 |
derekh | Ci jobs passed (all but Ha job) | 10:40 |
shardy | derekh: +2'd, although I'd like to understand the actual heat regression - I guess we can test that locally tho | 10:42 |
derekh | shardy: yup ack, once I get the logging fixed, I'l submit 2 patches one to remove each of those pins and we can see what the logs show | 10:43 |
derekh | Anybody else out there to approve https://review.openstack.org/#/c/214112/1 ? | 10:52 |
openstackgerrit | xiexs proposed openstack/os-cloud-config: Refactor _extract_driver_info() to reduce complexity. https://review.openstack.org/214128 | 10:53 |
gfidente | derekh, does it count that I get puppet deployment to work without those? | 10:54 |
derekh | gfidente: nope, they use the packages so are uneffected by the pins | 10:54 |
gfidente | derekh, yeah so what I saying is why we want to pin to 7th? | 10:55 |
gfidente | ah the deloran pinning is probably older than that? | 10:56 |
derekh | gfidente: because I know it worked on the 7th, I'll then work on removing the pin | 10:56 |
derekh | gfidente: yup | 10:56 |
gfidente | so you want to wait for CI jobs for that | 10:57 |
gfidente | or have tested manually and want to merge? | 10:57 |
derekh | gfidente: I think we're good to just merge the same patch was tested here https://review.openstack.org/#/c/111011/ | 10:58 |
derekh | gfidente: I only change the bug numbers referenced and a comment | 10:58 |
gfidente | cool, thanks! | 10:58 |
gfidente | (and sorry the slow down, was trying to get a sense of what is happening) | 10:58 |
derekh | gfidente: no prob | 10:59 |
openstackgerrit | Merged openstack-infra/tripleo-ci: Pin heat and neutron https://review.openstack.org/214112 | 10:59 |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Try alternate usernames to get logs from overcloud https://review.openstack.org/214139 | 11:01 |
*** lucasagomes is now known as lucas-hungry | 11:19 | |
*** trown|outttypeww is now known as trown | 11:23 | |
*** jprovazn has quit IRC | 11:34 | |
*** jprovazn has joined #tripleo | 11:37 | |
*** tchaypo has quit IRC | 11:39 | |
*** jprovazn has quit IRC | 11:39 | |
*** ramishra has quit IRC | 11:40 | |
openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add generic local_mac_map function and ceph_osds class https://review.openstack.org/211935 | 11:51 |
gfidente | nvx9V681O! | 11:52 |
gfidente | huh | 11:52 |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Provide a param to map block devices to Ceph OSDs https://review.openstack.org/200338 | 11:52 |
gfidente | jistr, marios ^^ tested manually :) | 11:53 |
jistr | gfidente: ok now you might need to change that password :D | 11:54 |
gfidente | jistr, ahaah I started using random passwords one per service | 11:54 |
gfidente | because I typed the 'single one' too many times already | 11:54 |
gfidente | oh I meant I tested the submission not the *password* ! :P | 11:55 |
jistr | yeah i'm using http://www.passwordstore.org/ | 11:55 |
jistr | git backed, gpg encrypted, with cmdline client | 11:55 |
gfidente | local gpg | 11:55 |
gfidente | yeah | 11:55 |
gfidente | I am doing that manually | 11:55 |
gfidente | with gpg2 --decrypt and --encrypt | 11:55 |
jistr | yeah i used to do that too | 11:55 |
gfidente | but I have to do the copy/paste manually as well | 11:55 |
gfidente | was unsure about calling this tripleo::ceph::osds | 11:57 |
gfidente | as it basically does the same thing except parsing per-mac address | 11:57 |
gfidente | before create_resource | 11:57 |
gfidente | seems to work :) | 11:57 |
*** bvandenh has quit IRC | 11:57 | |
openstackgerrit | Lennart Regebro proposed openstack/tuskar-ui: Enhance plan validation https://review.openstack.org/196051 | 11:59 |
*** dprince has joined #tripleo | 12:04 | |
jistr | gfidente: puppet lint seems unhappy, so given that the patch will need an update anyway, i commented re the doc string | 12:05 |
*** dprince has quit IRC | 12:07 | |
*** dprince has joined #tripleo | 12:07 | |
*** jayg|g0n3 is now known as jayg | 12:11 | |
gfidente | tks | 12:15 |
*** rhallisey has joined #tripleo | 12:16 | |
gfidente | jistr, WARNING: double quoted string containing no variables on line 16 | 12:19 |
gfidente | what is it talking about? I don't see any empty doubles quotes in the whole file | 12:19 |
gfidente | ah it's just suggesting to remove the double quotes around the strings? | 12:20 |
jistr | gfidente: i'll look at line 16 but yes i think that's it | 12:23 |
jistr | probably wants you to use single quotes? | 12:23 |
jistr | if so, that's very strict :D | 12:24 |
*** jprovazn has joined #tripleo | 12:25 | |
*** BadCub has quit IRC | 12:25 | |
*** BadCub has joined #tripleo | 12:27 | |
trown | jistr: sweet tip on pass utility...I have been using keepass, but pass looks much nicer/more unix-y | 12:28 |
jistr | trown: yeah i've been using keepass too for a while, but this is better. keepass was a pain to sync across computers, because the unit of atomicity was the whole password store. This keeps one password per file, optionally in git, so conflicts are practically non existent, and if there are some, you immediately know which passwords are the conflicting ones. | 12:31 |
* jistr passing the praise upstream to pblaho :) | 12:31 | |
openstackgerrit | Sebastien Badia proposed openstack/puppet-tripleo: Initial msync run for all Puppet OpenStack modules https://review.openstack.org/189880 | 12:32 |
trown | ya, super awesome | 12:32 |
*** bvandenh has joined #tripleo | 12:32 | |
*** lucas-hungry is now known as lucasagomes | 12:34 | |
*** shardy_ has joined #tripleo | 12:39 | |
*** rbrady has joined #tripleo | 12:39 | |
*** shardy has quit IRC | 12:41 | |
*** shardy_ has quit IRC | 12:45 | |
*** shardy has joined #tripleo | 12:46 | |
*** chlong has joined #tripleo | 12:51 | |
*** rhallisey has quit IRC | 12:56 | |
*** rlandy has joined #tripleo | 12:57 | |
*** tzumainn has joined #tripleo | 13:03 | |
*** rhallisey has joined #tripleo | 13:04 | |
*** tzumainn has quit IRC | 13:04 | |
*** tzumainn has joined #tripleo | 13:05 | |
openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add generic local_mac_map function and ceph_osds class https://review.openstack.org/211935 | 13:11 |
gfidente | EmilienM, but puppet-ceph might prefer to match against hostname or some other property | 13:13 |
*** rbrady has quit IRC | 13:14 | |
EmilienM | gfidente: I don't really like how puppet-tripleo is becoming | 13:14 |
*** rbrady has joined #tripleo | 13:14 | |
gfidente | EmilienM, heh so that class is basically a replacement for ceph::osds | 13:14 |
EmilienM | it was supposed to be "the last chance to put some puppet code) | 13:14 |
gfidente | where the hash provided is keyed by mac address so we can extract different disk maps based on local mac | 13:15 |
EmilienM | and the module has no tests at all | 13:15 |
EmilienM | I would rather investigate putting this code in puppet-ceph | 13:15 |
EmilienM | as a feature | 13:15 |
gfidente | so I am not against having it in puppet-ceph | 13:17 |
*** persia_ is now known as persia | 13:19 | |
EmilienM | gfidente: does it make sense? | 13:22 |
*** julim has joined #tripleo | 13:24 | |
*** Goneri has joined #tripleo | 13:28 | |
*** masco has quit IRC | 13:28 | |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Try alternate usernames to get logs from overcloud https://review.openstack.org/214139 | 13:28 |
gfidente | EmilienM, I think you're better equipped to tell :) | 13:30 |
EmilienM | gfidente: maybe not , I'm just wondering here | 13:30 |
gfidente | I know puppet has mechanisms to provide different data to hosts based on their hostname | 13:30 |
EmilienM | gfidente: I tend to think: if you have a good use case and you need to patch puppet-tripleo for that, there is something wrong in puppet-ceph, let's fix it | 13:31 |
*** lblanchard has joined #tripleo | 13:31 | |
gfidente | but our is a weird case though where we can't rely on hostnames | 13:31 |
gfidente | and have to use mac addresses | 13:32 |
gfidente | if this is typical of tripleo or more of a generic problem, I don't know | 13:32 |
*** paramite is now known as paramite|afk | 13:33 | |
derekh | Anybody familiar with this error ? | 13:33 |
derekh | | resource_status | CREATE_FAILED | | 13:33 |
derekh | | resource_status_reason | resources.RedisVirtualIP: Property error: resources.VipPort.properties.fixed_ips: "{}" is not a list | | 13:33 |
derekh | I think I may have something missing to go along with https://review.openstack.org/#/c/195328/17/network/ports/ctlplane_vip.yaml,cm | 13:34 |
gfidente | EmilienM, I'm thinking that in puppet-ceph it might be a nice to have if we could make the matching optionally against either mac OR hostname ? | 13:34 |
EmilienM | gfidente: yeah, right | 13:35 |
EmilienM | like adding an option | 13:35 |
EmilienM | to switch the matching system | 13:35 |
EmilienM | and keep the current one by default for backward compatibility | 13:35 |
EmilienM | this sounds reasonable | 13:35 |
*** prad_ has joined #tripleo | 13:39 | |
*** Goneri has quit IRC | 13:45 | |
*** pdb is now known as pbourke | 13:50 | |
*** pbourke has quit IRC | 13:51 | |
*** pbourke has joined #tripleo | 13:51 | |
*** matbu has joined #tripleo | 13:52 | |
openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add generic local_mac_map function and ceph_osds class https://review.openstack.org/211935 | 13:55 |
*** tchaypo has joined #tripleo | 14:02 | |
*** ramishra has joined #tripleo | 14:03 | |
derekh | shardy: have you a minute for a heat question ? | 14:06 |
derekh | I have a heat stack thats failing to created resources.RedisVirtualIP | 14:07 |
derekh | as far as I can see its failing here | 14:07 |
derekh | https://github.com/openstack/tripleo-heat-templates/blob/master/network/ports/ctlplane_vip.yaml#L36 | 14:07 |
derekh | with | 14:08 |
derekh | | resource_status_reason | resources.RedisVirtualIP: Property error: resources.VipPort.properties.fixed_ips: "{}" is not a list | | 14:08 |
derekh | its like the default for "FixedIPs" is taken as a dict instead of a list /me confused | 14:08 |
*** wgd3 has joined #tripleo | 14:11 | |
*** paramite|afk is now known as paramite | 14:12 | |
shardy | derekh: Hmm, that sounds suspicious, as the heat internal default for json paramters is {}, and we're setting the default to [] | 14:13 |
shardy | I wonder if someone's done the old if not foo: foo={} | 14:14 |
* shardy looks for evidence in heat commits.. | 14:14 | |
derekh | shardy: ok, I'll see if I can trace it in the code aswell, thanks | 14:14 |
*** prad_ is now known as pradk | 14:26 | |
*** sdake has joined #tripleo | 14:31 | |
*** bvandenh has quit IRC | 14:32 | |
jistr | spredzy, derekh: green HA https://review.openstack.org/213737/ there is a legitimate CI failure on the DIB patch it seems, somehow it seems to be ok for the CI though https://review.openstack.org/213736/ | 14:36 |
jistr | i mean it's ok for TripleO CI | 14:36 |
jistr | not for gate-dib-dsvm-functests-devstack | 14:36 |
*** spzala has joined #tripleo | 14:36 | |
*** sdake_ has joined #tripleo | 14:37 | |
*** jrist has joined #tripleo | 14:38 | |
*** jrist has joined #tripleo | 14:38 | |
spredzy | jistr, do you want to move along with that or wait early next week to see if the CPU +1 solves the issue? | 14:39 |
spredzy | or 1 meanwhile we wait for the other | 14:39 |
*** sdake has quit IRC | 14:40 | |
jistr | spredzy, derekh: my 2 cents would be to get the job green, so that CI is usable again for HA and we don't introduce more issues by merging patches which would fail HA CI legitimately | 14:41 |
jistr | the other options i see are either don't merge anything, or put HA stability at risk (not only in CI, but generally), which i don't like much | 14:42 |
*** spzala has quit IRC | 14:42 | |
jistr | and meanwhile work can continue on a proper solution | 14:42 |
spredzy | ack that works for me | 14:43 |
spredzy | jistr, | 14:43 |
spredzy | voted | 14:44 |
jistr | thx | 14:44 |
jistr | fwiw the DIB functests complain that 'dnf versionlock' is not a valid command | 14:45 |
derekh | jistr: spredzy +2 to merging from me, let get it in and see if the HA job stabilizes, then we can decide where to go from there | 14:45 |
shardy | derekh: I'm still waiting for my devtest env to rebuild, but I did some tests (devstack and unit tests) and I can't see any obvious problems with the heat default handling atm | 14:46 |
jistr | i could do yet another update of that patch to check "if corosync and corosynclib are installed, do yum -y downgrade ..." | 14:46 |
jistr | but it might not be worth it | 14:46 |
jistr | since the patch will never be merged into DIB proper | 14:46 |
jistr | and we'd have to wait for another CI round | 14:46 |
derekh | shardy: ok, no problem, thanks for looking, I'll poke around in the heat code a bit and track it down, will let you know when I find out what stupid thing I did wrong ;-) | 14:47 |
jistr | and i'm still not sure if *that* would make the functests happy | 14:47 |
shardy | derekh: I assume the ControlFixedIPs passed in to the stack-create looks reasonable? | 14:48 |
derekh | shardy: will double check, some of this is hidden down in the "openstack deploy" command so will have to dig into it to find out | 14:49 |
shardy | derekh: aha, sounds good, I wasn't sure if you were using oscplugin atm | 14:50 |
*** paramite is now known as paramite|afk | 14:50 | |
*** julim_ has joined #tripleo | 14:51 | |
*** pbourke_ has joined #tripleo | 14:57 | |
*** michchap_ has joined #tripleo | 14:58 | |
*** rhallisey has quit IRC | 14:59 | |
*** pbourke has quit IRC | 14:59 | |
*** julim has quit IRC | 14:59 | |
*** dprince has quit IRC | 14:59 | |
*** mbound has quit IRC | 14:59 | |
*** jtomasek has quit IRC | 14:59 | |
*** michchap has quit IRC | 14:59 | |
*** cody-somerville- has quit IRC | 14:59 | |
*** rhallisey has joined #tripleo | 14:59 | |
*** untriaged-bot has joined #tripleo | 15:00 | |
untriaged-bot | Untriaged bugs so far: | 15:00 |
untriaged-bot | https://bugs.launchpad.net/os-collect-config/+bug/1482510 | 15:00 |
openstack | Launchpad bug 1482510 in heat "OS::Heat::SoftwareDeployment failed due SSL certificate verification error" [Medium,Triaged] - Assigned to Rico Lin (rico-lin) | 15:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1466037 | 15:00 |
openstack | Launchpad bug 1466037 in diskimage-builder "Signed Fedora and Ubuntu user image built by DIB can`t boot on HP DL380 Gen8 server for lack of mpt2sas driver" [Undecided,Incomplete] | 15:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1483385 | 15:00 |
openstack | Launchpad bug 1483385 in diskimage-builder "install_grub failing for centos7" [Undecided,In progress] - Assigned to Abel Lopez (al592b) | 15:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1471802 | 15:00 |
openstack | Launchpad bug 1471802 in diskimage-builder "ironic-agent element hardcodes interfaces names for DHCP." [Undecided,Fix committed] - Assigned to Om Kumar (om-kumar) | 15:00 |
*** untriaged-bot has quit IRC | 15:00 | |
*** yamahata has joined #tripleo | 15:02 | |
*** julim has joined #tripleo | 15:04 | |
*** wgd3 has quit IRC | 15:04 | |
*** julim_ has quit IRC | 15:06 | |
*** dprince has joined #tripleo | 15:12 | |
*** jtomasek has joined #tripleo | 15:12 | |
*** cody-somerville- has joined #tripleo | 15:12 | |
*** dprince has quit IRC | 15:16 | |
*** noslzzp has joined #tripleo | 15:25 | |
*** Goneri has joined #tripleo | 15:25 | |
*** paramite|afk is now known as paramite | 15:25 | |
*** alop has joined #tripleo | 15:27 | |
*** rgowrishankar has joined #tripleo | 15:31 | |
rgowrishankar | Hello, this is Ramkumar Gowrishankar. I work for Nuage and I am creating a set of extra templates to be used for deploying compute nodes. I see the architecture for adding extra config to controller nodes. Can I follow the same pattern and add the scaffolding to create extraconfig templates for compute nodes as well? I will be creating a compute folder in | 15:35 |
rgowrishankar | /puppet/extraconfig/pre_deploy and add the templates there. | 15:35 |
rgowrishankar | Do I need to file a blueprint for this? | 15:35 |
*** ramishra has quit IRC | 15:36 | |
*** tchaypo has quit IRC | 15:36 | |
*** aufi has quit IRC | 15:37 | |
shardy | rgowrishankar: hi! | 15:37 |
shardy | rgowrishankar: I think the answer is yes, you can add a similar OS::TripleO::ComputeExtraConfigPre interface | 15:38 |
shardy | let me check, I thought I actually started some patches for that | 15:38 |
shardy | rgowrishankar: we've not been strictly enforcing blueprints lately, so it's not mandatory, but you may raise one if you wish | 15:38 |
rgowrishankar | shardy: okie dokie... if you already have some patches for it, I will just piggy back on those :) if not, I can start one.. let me know. | 15:40 |
shardy | rgowrishankar: Hmm, looks like I didn't actually get to it, I was thinking of https://review.openstack.org/#/c/196059/4 which allows hieradata overrides for NovaCompute | 15:42 |
shardy | rgowrishankar: it should just be replicating a subset of https://review.openstack.org/#/c/193143/, let me know if you need any help | 15:45 |
*** spzala has joined #tripleo | 15:46 | |
rgowrishankar | shardy: yep, will do.. | 15:46 |
*** mestery has joined #tripleo | 15:54 | |
*** rwsu has joined #tripleo | 15:54 | |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Enable Keystone notifications https://review.openstack.org/211341 | 15:55 |
*** sthillma has joined #tripleo | 15:56 | |
*** Marga_ has quit IRC | 15:57 | |
*** sthillma has quit IRC | 15:57 | |
*** lsmola has quit IRC | 16:00 | |
*** jistr has quit IRC | 16:08 | |
*** tchaypo has joined #tripleo | 16:11 | |
*** ramishra has joined #tripleo | 16:12 | |
*** panda is now known as panda|bbl | 16:20 | |
*** jcoufal has quit IRC | 16:20 | |
*** pbourke_ is now known as pbourke | 16:21 | |
*** sthillma has joined #tripleo | 16:30 | |
*** karume has quit IRC | 16:31 | |
*** sthillma_ has joined #tripleo | 16:31 | |
*** trown is now known as trown|lunch | 16:33 | |
*** devvesa has quit IRC | 16:34 | |
*** sdake_ is now known as sdake | 16:34 | |
*** sthillma has quit IRC | 16:35 | |
*** sthillma_ is now known as sthillma | 16:35 | |
*** noslzzp has quit IRC | 16:37 | |
*** dprince has joined #tripleo | 16:40 | |
*** gothicmi1dfood has joined #tripleo | 16:47 | |
*** gothicmi1dfood has quit IRC | 16:48 | |
*** gothicmindfood has quit IRC | 16:48 | |
*** gothicmindfood has joined #tripleo | 16:48 | |
*** shardy_ has joined #tripleo | 16:55 | |
*** shardy has quit IRC | 16:55 | |
*** ifarkas has quit IRC | 16:58 | |
*** shardy_ has quit IRC | 17:00 | |
*** shardy has joined #tripleo | 17:01 | |
*** penick has joined #tripleo | 17:04 | |
*** penick_ has joined #tripleo | 17:07 | |
*** mestery has quit IRC | 17:07 | |
*** mcornea has quit IRC | 17:07 | |
*** rhallisey has quit IRC | 17:08 | |
*** penick has quit IRC | 17:10 | |
*** penick_ is now known as penick | 17:10 | |
*** rhallisey has joined #tripleo | 17:11 | |
*** sthillma has quit IRC | 17:12 | |
*** lucasagomes is now known as lucas-dinner | 17:14 | |
*** sthillma has joined #tripleo | 17:16 | |
*** yamahata has quit IRC | 17:17 | |
*** absubram has joined #tripleo | 17:20 | |
*** pblaho has quit IRC | 17:23 | |
*** trown|lunch is now known as trown|mtg | 17:33 | |
*** jprovazn has quit IRC | 17:36 | |
*** alop has quit IRC | 17:41 | |
*** sthillma has quit IRC | 17:43 | |
*** yamahata has joined #tripleo | 17:45 | |
*** shivrao has joined #tripleo | 17:47 | |
*** fedora has joined #tripleo | 18:01 | |
*** gfidente has quit IRC | 18:05 | |
*** chlong has quit IRC | 18:06 | |
*** fedora has quit IRC | 18:06 | |
*** Goneri has quit IRC | 18:07 | |
*** paramite is now known as paramite|afk | 18:07 | |
*** chlong has joined #tripleo | 18:09 | |
*** panda|bbl has quit IRC | 18:09 | |
*** panda|bbl has joined #tripleo | 18:10 | |
*** fedora has joined #tripleo | 18:11 | |
*** openstackgerrit_ has quit IRC | 18:11 | |
shardy | Are we still having the bi weekly meeting? | 18:28 |
*** athomas has quit IRC | 18:28 | |
shardy | logs look like not recently, wasn't sure if it's been discontinued, or if not which week is the meeting week | 18:28 |
*** athomas has joined #tripleo | 18:34 | |
*** pelix has quit IRC | 18:42 | |
dprince | shardy: not sure | 18:49 |
dprince | rhallisey, Slower: how goes it, any progress? Do we need to chat again today on docker compute stuff | 18:50 |
rhallisey | dprince, things are progressing. Just messing with neutron now | 18:51 |
rhallisey | fixed libvirt this morning. So stepping closer | 18:51 |
*** paramite|afk is now known as paramite | 18:52 | |
dprince | rhallisey: nice | 18:52 |
*** sdake_ has joined #tripleo | 18:56 | |
slagle | shardy: i cancelled last weeks meeting due to folks being out b/c of travel | 18:58 |
slagle | shardy: the meeting before that, no one showed. other than jdob :) | 18:59 |
slagle | so we called that one off | 18:59 |
*** sdake has quit IRC | 18:59 | |
*** jang has quit IRC | 19:01 | |
*** jang1 has quit IRC | 19:01 | |
*** penick has quit IRC | 19:01 | |
*** jang has joined #tripleo | 19:02 | |
*** jang1 has joined #tripleo | 19:03 | |
shardy | slagle: Ok no worries, thanks for clarifying! | 19:04 |
shivrao | dprince: can you please review this if you have some time: https://review.openstack.org/#/c/208655/ | 19:11 |
openstackgerrit | Merged openstack/tripleo-heat-templates: Wire in Compute pre-deployment extraconfig https://review.openstack.org/208655 | 19:14 |
dprince | shivrao: +2 | 19:14 |
dprince | shivrao: thanks | 19:14 |
*** olaph has joined #tripleo | 19:14 | |
shivrao | dprince: thanks | 19:15 |
*** w_ has quit IRC | 19:15 | |
*** sdake_ is now known as sdake | 19:16 | |
*** w_ has joined #tripleo | 19:16 | |
EmilienM | dprince: what do you think about using puppet to configure keystone resources? | 19:18 |
EmilienM | dprince: we did a great job lately in puppet-keystone by supporting v3 API | 19:19 |
EmilienM | we have great features and federation is coming very soon | 19:19 |
*** olaph has quit IRC | 19:19 | |
dprince | EmilienM: I'm a bit mixed. Part of my sort of likes all that sort of "tenant" level configuration to come after installation is finished. | 19:20 |
dprince | EmilienM: however, keystone is fairly crucial to doing pretty much anything, so I think perhaps we can blur the lines a bit and try it | 19:20 |
dprince | EmilienM: remind me, where does Spinal Stack do the keystone bits? | 19:21 |
*** paramite has quit IRC | 19:21 | |
*** Goneri has joined #tripleo | 19:21 | |
*** olaph has joined #tripleo | 19:23 | |
*** w_ has quit IRC | 19:23 | |
*** alop has joined #tripleo | 19:25 | |
*** rhallisey has quit IRC | 19:28 | |
*** w_ has joined #tripleo | 19:40 | |
*** olaph has quit IRC | 19:43 | |
*** mestery has joined #tripleo | 19:44 | |
*** w_ has quit IRC | 19:44 | |
*** Marga_ has joined #tripleo | 19:47 | |
*** Marga_ has quit IRC | 19:52 | |
*** Marga_ has joined #tripleo | 19:52 | |
*** mestery has quit IRC | 20:08 | |
*** mestery has joined #tripleo | 20:10 | |
*** penick has joined #tripleo | 20:16 | |
*** penick has quit IRC | 20:20 | |
*** panda|bbl is now known as panda | 20:21 | |
*** panda is now known as Guest4619 | 20:21 | |
*** akrivoka has quit IRC | 20:27 | |
*** jayg is now known as jayg|g0n3 | 20:28 | |
*** Guest4619 is now known as panda | 20:39 | |
*** trown|mtg is now known as trown|outttypeww | 20:45 | |
*** lblanchard has quit IRC | 20:45 | |
*** lucas-dinner has quit IRC | 20:52 | |
*** derekh has quit IRC | 20:52 | |
*** mbound has joined #tripleo | 21:00 | |
*** julim has quit IRC | 21:00 | |
*** untriaged-bot has joined #tripleo | 21:00 | |
untriaged-bot | Untriaged bugs so far: | 21:00 |
untriaged-bot | https://bugs.launchpad.net/os-collect-config/+bug/1482510 | 21:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1466037 | 21:00 |
*** trown|outttypeww is now known as trown|ptoAug21 | 21:00 | |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1483385 | 21:00 |
untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1471802 | 21:00 |
*** untriaged-bot has quit IRC | 21:00 | |
openstack | Launchpad bug 1482510 in heat "OS::Heat::SoftwareDeployment failed due SSL certificate verification error" [Medium,Triaged] - Assigned to Rico Lin (rico-lin) | 21:00 |
openstack | Launchpad bug 1466037 in diskimage-builder "Signed Fedora and Ubuntu user image built by DIB can`t boot on HP DL380 Gen8 server for lack of mpt2sas driver" [Undecided,Incomplete] | 21:00 |
openstack | Launchpad bug 1483385 in diskimage-builder "install_grub failing for centos7" [Undecided,In progress] - Assigned to Abel Lopez (al592b) | 21:00 |
openstack | Launchpad bug 1471802 in diskimage-builder "ironic-agent element hardcodes interfaces names for DHCP." [Undecided,Fix committed] - Assigned to Om Kumar (om-kumar) | 21:00 |
*** mbound has quit IRC | 21:01 | |
*** Goneri has quit IRC | 21:04 | |
*** lucas-dinner has joined #tripleo | 21:04 | |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Try alternate usernames to get logs from overcloud https://review.openstack.org/214139 | 21:10 |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: [WIP] Remove heat pin https://review.openstack.org/214348 | 21:10 |
*** Marga__ has joined #tripleo | 21:10 | |
openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: [WIP] Remove neutron pin https://review.openstack.org/214349 | 21:10 |
*** Marga_ has quit IRC | 21:13 | |
*** rhallisey has joined #tripleo | 21:22 | |
*** rhallisey has quit IRC | 21:22 | |
*** mestery has quit IRC | 21:25 | |
*** athomas has quit IRC | 21:27 | |
*** penick_ has joined #tripleo | 21:32 | |
*** Marga__ has quit IRC | 21:40 | |
*** Marga_ has joined #tripleo | 21:40 | |
*** olaph has joined #tripleo | 21:43 | |
*** sthillma has joined #tripleo | 21:50 | |
*** penick_ has quit IRC | 21:52 | |
*** absubram has quit IRC | 22:01 | |
*** dprince has quit IRC | 22:02 | |
*** lucas-dinner has quit IRC | 22:23 | |
*** pradk has quit IRC | 22:24 | |
*** penick has joined #tripleo | 22:48 | |
openstackgerrit | Rhys Oxenham proposed openstack/tripleo-heat-templates: Add network templates for multiple NICs with flat and VLAN tagged networks https://review.openstack.org/214372 | 22:49 |
*** rhallisey has joined #tripleo | 22:53 | |
*** dsneddon is now known as dsneddon_brb | 23:00 | |
*** penick_ has joined #tripleo | 23:09 | |
*** penick has quit IRC | 23:09 | |
*** penick_ is now known as penick | 23:09 | |
*** chlong has quit IRC | 23:29 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!