Wednesday, 2015-09-16

*** trown is now known as trown|outttypeww00:10
*** BadCub_ has joined #tripleo00:18
*** BadCub has quit IRC00:21
*** jdob has quit IRC00:21
*** shadower has quit IRC00:23
*** shadower has joined #tripleo00:23
*** yamahata has quit IRC00:24
*** shivrao has quit IRC00:25
*** michchap_ has joined #tripleo00:28
*** persia_ has joined #tripleo00:29
*** persia_ has quit IRC00:29
*** persia_ has joined #tripleo00:29
*** slagle_ has joined #tripleo00:31
*** StevenK_ has joined #tripleo00:31
*** StevenK has quit IRC00:32
*** StevenK_ is now known as StevenK00:33
*** slagle has quit IRC00:35
*** Slower has quit IRC00:35
*** michchap has quit IRC00:36
*** persia has quit IRC00:36
*** Slower has joined #tripleo00:42
*** sthillma has quit IRC00:49
*** rwsu has quit IRC00:57
*** weshay_xchat has joined #tripleo01:07
*** yamahata has joined #tripleo01:08
*** BadCub_ is now known as BadCub01:16
*** lazy_prince has joined #tripleo01:18
*** thrash is now known as thrash|g0ne01:18
openstackgerritMerged openstack/diskimage-builder: doc: migrate from README.rst to Sphinx
*** killer_prince has joined #tripleo01:27
*** lazy_prince has quit IRC01:30
*** lazy_prince has joined #tripleo01:30
*** killer_prince has quit IRC01:31
openstackgerritDan Sneddon proposed openstack/tripleo-heat-templates: Configure ctlplane network with a static IP
*** weshay_xchat has quit IRC01:58
*** yog_ has joined #tripleo02:08
*** panda has quit IRC02:10
*** panda has joined #tripleo02:10
openstackgerritMerged openstack/diskimage-builder: Update redhat-common elements README from free text to table formatting
*** Slower_ has joined #tripleo02:27
*** panda has quit IRC02:28
*** Slower has quit IRC02:28
*** lazy_prince has quit IRC02:34
*** lazy_prince has joined #tripleo02:34
*** lazy_prince has quit IRC02:51
*** panda has joined #tripleo02:54
*** untriaged-bot has joined #tripleo03:00
untriaged-botNo untriaged bugs so far! \o/03:00
*** untriaged-bot has quit IRC03:00
*** tzumainn has quit IRC03:38
*** david-lyle has joined #tripleo03:41
openstackgerritIan Wienand proposed openstack/diskimage-builder: Make README.rst a bit more generic
openstackgerritIan Wienand proposed openstack/diskimage-builder: Add #!/bin/bash to library functions
*** shivrao has joined #tripleo04:01
*** shivrao_ has joined #tripleo04:03
*** shivrao has quit IRC04:05
*** shivrao_ is now known as shivrao04:05
*** sthillma has joined #tripleo04:10
*** Marga_ has quit IRC04:16
*** rwsu has joined #tripleo04:21
*** sdake has joined #tripleo04:33
*** sdake_ has joined #tripleo04:34
*** masco_ has joined #tripleo04:35
*** sdake has quit IRC04:38
openstackgerritIan Wienand proposed openstack/diskimage-builder: Add DIB_KEEP_PACKAGE_CACHE variable
openstackgerritIan Wienand proposed openstack/diskimage-builder: Add DIB_USE_PACKAGE_CACHE variable
*** david-lyle has quit IRC04:42
*** david-lyle has joined #tripleo04:45
*** sdake_ has quit IRC04:56
*** sdake has joined #tripleo04:59
*** sdake_ has joined #tripleo05:10
*** sdake has quit IRC05:14
*** lazy_prince has joined #tripleo05:24
*** lsmola has joined #tripleo05:42
*** aufi has joined #tripleo05:51
*** stendulker has joined #tripleo05:57
*** ukalifon has joined #tripleo05:57
*** matbu has quit IRC06:00
*** matbu has joined #tripleo06:02
*** panda has quit IRC06:09
*** panda__ has joined #tripleo06:10
*** jaosorior has joined #tripleo06:18
*** jprovazn has joined #tripleo06:28
*** sthillma has quit IRC06:47
*** shivrao has quit IRC06:51
*** shardy has joined #tripleo06:59
*** marios has joined #tripleo07:01
*** marios has quit IRC07:01
*** marios has joined #tripleo07:01
*** ifarkas has joined #tripleo07:07
*** yamahata has quit IRC07:14
*** shardy_ has joined #tripleo07:16
*** shardy has quit IRC07:18
*** marios has quit IRC07:19
*** panda__ has quit IRC07:19
*** masco_ has quit IRC07:19
*** shardy_ has quit IRC07:21
*** shardy has joined #tripleo07:22
*** marios has joined #tripleo07:24
*** panda__ has joined #tripleo07:24
*** masco_ has joined #tripleo07:24
*** matbu has quit IRC07:44
*** matbu has joined #tripleo07:45
*** sdake has joined #tripleo08:11
*** sdake_ has quit IRC08:12
*** lucasagomes has joined #tripleo08:15
*** yog_ has quit IRC08:15
*** gfidente has joined #tripleo08:18
*** gfidente has quit IRC08:18
*** gfidente has joined #tripleo08:18
*** jistr has joined #tripleo08:19
openstackgerritMerged openstack/tripleo-heat-templates: Add 4 pacemaker constraints for keystone
*** paramite has joined #tripleo08:39
spredzyHello, still hitting new issues when deploying with instack
*** derekh has joined #tripleo08:42
spredzyThis time it complains about rdomanager_oscplugin.v1.overcloud_deploy.DeployOvercloud There are 5 ironic nodes with no profile that will not be used08:42
spredzydoes that ring a bell to anyone derekh jistr gfidente slagle_ ^08:43
* pino|work suggests to avoid multi-pings...08:43
gfidentespredzy, I think that is incorrectly reported as an error because we don't match the hardware profiles08:44
derekhspredzy: I don't know what that means but as far as I've seen its a warning message that happenb on successful runs also08:44
gfidentebut nova is going to rely on the basic 'greater than flavor' profiling08:44
gfidenteand should work08:44
spredzygfidente, derekh ack I will let it pursue then08:45
*** pelix has joined #tripleo08:46
spredzythank you08:46
*** hashar has joined #tripleo08:50
*** kbyrne has joined #tripleo08:53
*** larstobi has quit IRC08:59
*** untriaged-bot has joined #tripleo09:00
untriaged-botUntriaged bugs so far:09:00
openstackLaunchpad bug 1496248 in diskimage-builder "building ironic-agent ramdisk with fedora element fails" [Undecided,New]09:00
openstackLaunchpad bug 1495860 in diskimage-builder ""disk-image-create ironic-agent fedora iso -o my-deploy-ramdisk" failed" [Undecided,Fix committed]09:00
*** untriaged-bot has quit IRC09:00
*** larstobi has joined #tripleo09:00
jistrderekh: hi, does this error ring a bell? (delorean related)
shardymarios: Hey, would you like to approve - it got rebased with a revised commit message as the previous Depends-On seemed to block the merge when I approved it yesterday09:04
shardyCI is green so I think we're good now, then I'll rebase the AllNodes stuff on top09:04
*** athomas has joined #tripleo09:04
mariosshardy: sure looking09:05
derekhjistr: it just means delorean failed to build the rpm, you need to look at the delorean logs09:06
jistrinterestingly, some jobs progress further than that, and then e.g. fail on missing overcloud image
*** dtantsur|afk is now known as dtantsur09:06
derekhjistr: Cannot open: Skipping.09:07
*** mcornea has joined #tripleo09:08
jistrhmm ok could be a download error maybe?09:09
* jistr looks at other logs09:09
jistrin the other one i linked, the image build fails on Cannot open: Skipping.09:11
jistrso probably some connectivity issue.09:11
derekhjistr: yup09:13
*** hashar has quit IRC09:13
*** hashar has joined #tripleo09:14
openstackgerritMerged openstack/tripleo-heat-templates: Enable Cisco Nexus and UCSM plugins
*** akrivoka has joined #tripleo09:22
mariosderekh: i have been really, really (really, apparently) crap at keeping up with email last few days. thanks a lot for championing the meeting change time09:26
derekhmarios: no problem, looks like the new time should suit a lot more people09:28
openstackgerritYanis Guenane proposed openstack/tripleo-incubator: Do not exit when $TRIPLEO_ROOT is not defined
*** sdake has quit IRC09:30
*** paramite is now known as paramite|afk09:30
*** killer_prince has joined #tripleo09:33
spredzyjistr, I do have now a working deployment09:36
*** lazy_prince has quit IRC09:36
*** shardy_ has joined #tripleo09:47
*** shardy has quit IRC09:49
dtantsurhey folks! who should I torture to get a release of diskimage-builder?09:51
dtantsurwe had a lot of ironic-agent related fixes there, that we can't use with dib installed from PyPI09:52
*** shardy_ has quit IRC09:52
*** shardy has joined #tripleo09:53
openstackgerritSteven Hardy proposed openstack/tripleo-heat-templates: Support new form of servers config for ML2 cisco nexus allnodes config.
openstackgerritSteven Hardy proposed openstack/tripleo-heat-templates: Add "AllNodes" ExtraConfig interface
openstackgerritSteven Hardy proposed openstack/tripleo-heat-templates: Port Cisco Nexus/UCSM ExtraConfig to AllNodes
shardymarios: thanks, just rebased those ^^09:54
openstackgerritDerek Higgins proposed openstack/diskimage-builder: Don't create a centos yum repository
openstackgerritDerek Higgins proposed openstack/diskimage-builder: Add centos7 support for DIB_DISTRIBUTION_MIRROR
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Use a specific CentOS mirror on the jenkins node
derekhI'm hoping those will lessen the amount of problems we've been having download centos stuff, I still have to cover the case of what happens inside the delorean containers10:01
derekhIf they don't improve thins then hosting our own mirror will be next10:01
mariosshardy: voted on - the other one my vote stuck anyway :)10:05
*** panda__ is now known as panda10:05
shardymarios: Cool, thanks, hoping we can finally land that series while CI is behaving ;)10:06
*** jkraj has joined #tripleo10:19
*** jaosorior has quit IRC10:33
*** thrash|g0ne is now known as thrash10:33
*** adrianopetrich has quit IRC10:35
*** pblaho has joined #tripleo10:47
*** morazi has joined #tripleo10:47
*** paramite|afk is now known as paramite10:48
*** regebro has joined #tripleo10:50
*** adrianopetrich has joined #tripleo10:53
*** hashar has quit IRC10:55
*** morazi has quit IRC11:00
*** b3nt_pin has quit IRC11:01
*** beagles has joined #tripleo11:02
*** beagles is now known as Guest8682511:02
*** lazy_prince has joined #tripleo11:13
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Delorean go faster stripes
*** killer_prince has quit IRC11:16
*** regebro has quit IRC11:17
*** regebro has joined #tripleo11:18
*** lucasagomes is now known as lucas-hungry11:23
*** Guest86825 has quit IRC11:23
*** beagles__ has joined #tripleo11:24
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Stop creating id_rsa_virt_power
*** julim has joined #tripleo11:28
jistrspredzy: lucky you :)) i still can't get it going. The latest error i have is this
jistrhave you seen that?11:29
spredzyjistr, yep11:29
spredzythere is another repo you need to use for delorean-deps11:29
spredzyYou need some packages that are there, I don't know if its temporary - or meant to be that way (ie. needs update on the docs)11:30
jistrwhaaat. i had no idea we have this repo. Thanks! :)11:30
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Delorean go faster stripes
* jistr didn't notice it in docs either11:31
spredzyjistr, it isn't. This is what apevec told me to add to by pass the issue you've been seeing and some other ones11:31
spredzywhat I am not sure is that if it is just a temp repo/solution or is meant to stick11:32
mariosjistr: is this about using puppet based overcloud using upstream setup?11:33
jistrmarios: yes11:33
*** beagles__ has quit IRC11:33
mariosjistr: i am running devtest for the first time in a while for that purpose... i am using f21 for virt host cos it worked last time i did this, still the case?11:33
jistrmarios: hmm i think we're sort of getting rid of devtest11:34
jistrmarios: what we're discussing is indeed upstream + puppet, but on instack11:35
* jistr re-running with delorean-deps11:35
spredzyjistr, remove the cache11:36
mariosjistr: oh. we still have for example (was actually looking to instack-build-iomages for the actual overcloud-full)11:36
spredzywe never know :)11:36
mariosjistr: any good docs/pointers, or is it non existant yet?11:36
mariosjistr: oh
*** weshay_xchat has joined #tripleo11:46
* marios never realised these docs went upstream already11:46
*** regebro has quit IRC11:46
*** stendulker has quit IRC11:49
*** trown|outttypeww is now known as trown11:51
*** pelix has quit IRC11:51
*** pelix has joined #tripleo11:53
*** jayg|g0n3 is now known as jayg11:53
*** hashar has joined #tripleo11:54
mariosjistr: spredzy either of you tried the centos as virt host?11:56
spredzymarios, I am running instack on a CentOS host virualizing CentOS nodes11:58
spredzyThings are fine once all the repos in place :)11:58
mariosspredzy: cool thanks12:00
jistrmarios: same for me, trying all-centos setup12:03
*** noslzzp has joined #tripleo12:04
mariosjistr: great thanks12:05
*** rhallisey has joined #tripleo12:09
*** jaosorior has joined #tripleo12:14
*** dprince has joined #tripleo12:16
*** beagles has joined #tripleo12:16
*** beagles has quit IRC12:21
*** lucas-hungry is now known as lucasagomes12:23
*** noslzzp has quit IRC12:28
openstackgerritJiri Stransky proposed openstack/tripleo-heat-templates: Big Switch Neutron ML2 plugin integration
mariosjistr: just rebase? (I see the keystone stuff in diff 7..8)12:34
jistrmarios: yeah just rebase because of merge conflict with Cisco Neutron plugin12:35
mariosjistr: thanks yeah just poking to vote again12:35
jistrmarios: the BSN/Cisco patches are both purely additive, but they happen on the exact same spots in the files, so git considered it a conflict12:36
*** jdob has joined #tripleo12:36
jistrs/patches/changes in the conflicting spots/12:36
*** jprovazn has quit IRC12:36
*** paramite is now known as paramite|afk12:37
mariosjistr: doesn't everybody appreciate a good session with meld though?12:37
jistr(had to google it :) ) i don't use meld but it looks interesting12:38
mariosjistr: vimdiff?!12:38
mariosjistr: i mean, what do you use for merge conflicts12:39
jistrnot even vimdiff12:39
mariosjistr: or have you not had much of that fun (especially porting things downstream say tht)12:39
* jistr is an emacs fella now12:39
mariosoh i see12:39
mariosits like that is it12:39
*** slagle_ is now known as slagle12:39
slaglejistr is hardcore12:40
mariosjistr: is this one of those 'yes but emacs can rebase all your things whilst making you coffee'12:40
jistrmarios: i'm sure it can, but i just do it manually :) E.g. in this case it just ate a closing bracket '}', because it was added by both patches and git thought it's the same.12:41
dtantsurslagle, hi! could you please make a release of DIB soonish? ironic-agent does not build with the current one on Fedora, and is still unreasonably bloated on centos, and does not support agent from packages12:42
jistrmarios: does meld have something smart to deal with that?12:43
mariosjistr: yeah with meld you'd get three versions of the file, local, remote and merge and you can fix as you like12:43
mariosjistr: 'get three versions' i mean is like a text editor with split screens12:44
*** adam_g has quit IRC12:44
jistri see. nice12:44
*** paramite|afk is now known as paramite12:46
*** wendar has quit IRC12:47
*** wendar has joined #tripleo12:47
*** rhallisey has quit IRC12:48
*** dsneddon has quit IRC12:48
*** rainya has quit IRC12:48
*** persia_ has quit IRC12:48
*** rhallisey has joined #tripleo12:48
*** dsneddon has joined #tripleo12:49
gfidentejistr, where is your init.el>12:50
*** persia has joined #tripleo12:50
*** pelix has quit IRC12:50
*** rainya has joined #tripleo12:50
*** adam_g has joined #tripleo12:51
*** adam_g has joined #tripleo12:51
jistrgfidente: i don't usually publish dotfiles and stuff... there's probably nothing secret but also nothing revolutionary :)12:51
openstackgerritMerged openstack/diskimage-builder: Make README.rst a bit more generic
gfidentejistr, was going to suggest show-paren-mode 112:53
*** jaosorior has quit IRC12:54
jistrgfidente: indeed! thx, adding right now12:54
openstackgerritLennart Regebro proposed openstack/python-tuskarclient: Add a parameter to load plan parameters from a file
slagledtantsur: sure12:55
*** jaosorior has joined #tripleo12:56
*** Goneri has joined #tripleo12:57
*** rpothier has joined #tripleo12:58
*** dsneddon has quit IRC13:02
openstackgerritMerged openstack/tripleo-incubator: Do not exit when $TRIPLEO_ROOT is not defined
*** masco_ has quit IRC13:09
*** jkraj has quit IRC13:10
*** hashar has quit IRC13:12
*** hashar has joined #tripleo13:12
*** pelix has joined #tripleo13:13
*** jprovazn has joined #tripleo13:14
*** beagles has joined #tripleo13:15
*** dsneddon has joined #tripleo13:16
*** rlandy has joined #tripleo13:18
*** tzumainn has joined #tripleo13:20
pino|workslagle: (and others) can you please review, so gate for python-tripleoclient passes?13:21
*** lblanchard has joined #tripleo13:25
jistrpino|work: on its way13:28
pino|workjistr: merci13:28
mariosjistr: spredzy did you guys have issues with instack-virt-setup, i dont mean but more like (just exits)13:34
derekh5 patches that should help our reliability13:35
derekh1. pin to a specific fedora mirror
derekh2. pin to a specific CentOS mirror , ,
derekh3. pin to a specific CentOS mirror (in delorean containers) and other delorean speedups
derekhSome of these faild ci because of reliability, I'll recheck where appropriate but wouldn't mind reviews in the mean time13:35
derekhI've also increased the size of the sqiude cache, the Cenos cloud image wasn't being cached and we were downloading it a lot13:35
spredzymarios, yup jistr has a patch for that13:36
spredzylet me look it up13:37
mariosspredzy: much appreciated13:37
mariosspredzy: thanks!13:38
spredzymarios, it says abandon but for your situation you need to add this --replacepkgs13:38
jistrmarios: which delorean url do you use?13:39
jistrin newer builds this workaround wouldn't be necessary i think13:39
jistri do apply it in my script still though13:39
jistri'll check how the file looks on my setup13:40
mariosjistr: i used  (
mariosjistr: i also enabled the delorean-deps after your chat with spredzy earlier13:41
marios(nto sure if it was needed at this stage)13:41
jistrmarios: ok... fwiw i've just had a successful deployment with the delorean-deps and this
jistrso the repo i used is different13:42
derekhAlso can somebody take a look at this , we now have a instack-undercloud jobs and can't merge without it
* jistr on it13:42
spredzymarios, for repo I also rely on
mariosspredzy: jistr gents thanks vrey much for all the info13:42
* spredzy erf jistr already posted it :)13:42
*** weshay_xchat has quit IRC13:47
openstackgerritAntoine Musso proposed openstack/diskimage-builder: Prettify 'Developing Elements' documentation
*** weshay_xchat has joined #tripleo13:51
*** weshay_xchat has quit IRC13:54
*** yamahata has joined #tripleo13:54
*** weshay_xchat has joined #tripleo13:55
slaglespredzy: EmilienM : so just wanted to double check if the connect timeout is already set somewhere else for ?13:57
EmilienMspredzy: look my comment on
slaglelooks like it is in the haproxy module13:58
*** pradk has joined #tripleo13:59
spredzyEmilienM, I had simply look at the haproxy doc and those configuration seemed corrrect14:03
spredzyok we need to check why it isn't happening14:03
*** stendulker has joined #tripleo14:04
*** panda has quit IRC14:09
*** pradk has quit IRC14:10
*** panda has joined #tripleo14:10
*** pradk has joined #tripleo14:12
*** julim has quit IRC14:12
*** Marga_ has joined #tripleo14:16
*** yog_ has joined #tripleo14:16
*** lazy_prince has quit IRC14:17
jprovaznryansb, hi14:18
ryansbjprovazn: hi, what's up?14:19
jprovaznryansb, looking at - there might be a problem when scaling down14:19
jprovaznit seems this expects scale up only14:20
ryansbit does - I wasn't aware scaling down existed14:20
jprovaznryansb, kk, yes - scale down exists, just wanted to point this out in case you are planning to add more improvements14:21
ryansbI am planning to add improvements to the validation, so feedback is super-welcome14:21
ryansbI think that particular segment should handle scaling down though14:22
ryansbas it only checks that enough non-maintenance-mode ironic nodes have the required flavor tag14:22
ryansbwhich, if less than the current deployed number, will still pass14:22
*** pradk has quit IRC14:24
*** pradk has joined #tripleo14:24
*** sdake has joined #tripleo14:25
jprovaznryansb, supposing that I update profile matching rules and re-run ahc-match which causes that number of nodes with a profile decreases to lower number than number to which I want to scale, I guess this fails14:25
ryansbyeah, then it would14:26
ryansbbut it should fail in that case14:26
ryansbsince if you did that, your scale would fail14:26
jprovaznah, right14:26
jprovaznok, cool14:27
*** egafford has joined #tripleo14:27
ryansbthanks so much for asking about this stuff - it's super helpful to hear about more cases validations need to handle14:27
*** sdake_ has joined #tripleo14:27
jprovaznryansb, actually hold on14:27
jprovaznryansb, well, probably it's fine, I'm just thinking if *decreasing* of allocated resources should raise this error14:28
*** sdake has quit IRC14:29
jprovaznryansb, also maintenance mode for already allocated nodes is potential issue (I can have running nodes which are part of deployment in maintenance mode)14:29
ryansbI think so, imagine: you start with 3 computes deployed, then make an ahc-match change that reduces the "qualified" nodes for 114:29
ryansb*to 114:29
ryansbthen I scale down to 214:30
ryansbI would want an error, since there's only one node that would be deployable14:30
jprovaznryansb, yes, makes sense14:30
*** openstackgerrit has quit IRC14:31
*** openstackgerrit has joined #tripleo14:31
ryansbbut you are correct, that if I were to have 3 computes14:32
ryansbthen put one in maintenance14:32
ryansbthen do a stack update with the same number of computes14:32
ryansbthen validation warnings would fire14:32
*** ukalifon has quit IRC14:32
ryansbbut I think that might be the "right" thing, since it's likely I don't want to deploy to an instance that's in maintenance mode14:33
jprovaznryansb, well, such instance is already running maybe14:33
ryansbyeah, but won't nova refuse to deploy to a maintenance-mode ironic node?14:34
jprovaznryansb, it will, I meant that you deploy your OC, then find out a temporary issue on a node, so you put it into maintenance mode, but this node is still part of your OC14:34
jprovaznthough question is why you would put it into maintenance mode for other reason than don't allow nova to manage it14:35
ryansbso I think a warning from the validator would be a good thing14:36
ryansbif only to prod you and say "hey, take this out of maintenance mode if you want nova to be able to do stuff"14:36
*** Marga_ has quit IRC14:36
*** julim has joined #tripleo14:37
jprovaznryansb, fair enough14:37
*** Marga_ has joined #tripleo14:37
ryansbat least, I *think* that's what people would want14:37
*** david-lyle has quit IRC14:37
jprovaznryansb, kk, makes sense14:38
jprovaznryansb, the _check_nodes slightly overlaps with - if you in future add support for dealing with omitted scaling/flavor params, it might be replaced14:38
ryansbI see14:39
*** pblaho has quit IRC14:42
*** pblaho has joined #tripleo14:42
*** dprince has quit IRC14:42
derekhjrist: responded to your question
jristare they yellow stripes?14:46
derekhjrist: yellow and red, can't wait to show them off14:46
jristthat's nearly unheard of14:47
*** pblaho has quit IRC14:49
*** pblaho has joined #tripleo14:50
shardyHey folks, all voting CI is green on this series, and all have at least one +2:14:51
shardyif anyone can +A them that would be great, as they've been pending for a while now14:52
*** pblaho has quit IRC14:52
openstackgerritMerged openstack/diskimage-builder: Fix test cleanup trap to cleanup tmpdir
egaffordderekh: (very brief) ping14:54
*** shardy_ has joined #tripleo14:54
derekhegafford: hi14:55
*** pblaho has joined #tripleo14:55
*** shardy has quit IRC14:55
egaffordderekh: Hi there! I heard that you were working on a script to field an upstream dev env based on instack. Is that actively true, and do you have any broad strokes notions of timelines if so?14:56
egaffordVery much not trying (or even authorized) to try to extract or hold to deliverables dates; just looking for a rough guess if one makes sense.14:56
derekhegafford: All of our ci is now based on instack, but I have no script to set up and env for devs14:57
derekhegafford: best we have for devs would be the docs that slagle has been updating14:57
*** mcornea has quit IRC14:58
egaffordderekh: Ok, there may have been a miscommunication then. Thanks; that's all I need to know.14:58
*** pblaho has quit IRC14:58
slaglethere's no script i'm aware of, but i agree we do need one14:58
*** jistr is now known as jistr|call14:58
egaffordslagle: If there's a functional, documented process, that's adequate.14:59
derekhegafford: no prob, a script would be good and one will probably appear at some stage but I've been focused on the ci job14:59
egaffordYou've got a big flow, it can sensibly decompose into a few steps.14:59
*** pblaho has joined #tripleo14:59
*** pblaho has quit IRC14:59
slaglewe did used to have one (instack-setup-host)14:59
egaffordslagle: I just need an env. :)14:59
slaglethat would prep the undercloud so that all you had to do was run "openstack undercloud install"14:59
*** pblaho has joined #tripleo14:59
*** shardy_ has quit IRC14:59
*** untriaged-bot has joined #tripleo15:00
untriaged-botUntriaged bugs so far:15:00
*** untriaged-bot has quit IRC15:00
openstackLaunchpad bug 1495860 in diskimage-builder ""disk-image-create ironic-agent fedora iso -o my-deploy-ramdisk" failed" [Undecided,Fix committed]15:00
*** shardy has joined #tripleo15:00
*** openstackgerrit has quit IRC15:01
egaffordslagle: Yup. So what's the status of the doc derekh just posted? Functional? Actively in transition?15:01
*** openstackgerrit has joined #tripleo15:01
shardyderekh: Hey is re-instating that port stuff which previously broke TripleO15:01
shardyderekh: gate-tripleo-ironic-overcloud-f21-nonha says green, but is there anything else we can do to de-risk it before landing?15:02
shardyderekh: obviously I can pull and test locally, but was looking more for CI green-ness ;)15:02
derekhshardy: we could merge this and then rerun tripleo ci to make sure15:03
shardyderekh: aha, yeah we missed the bug tag in the check experimental didn't we..15:04
openstackgerritMerged openstack-infra/tripleo-ci: Remove heat temprevert
derekhshardy: yup, best thing to do now would be remove our temprevert and rerun to make sure15:05
derekhshardy: just like that ;-)15:05
shardyderekh: done and done, thanks! :)15:06
*** pblaho has quit IRC15:12
*** pblaho has joined #tripleo15:13
slagleegafford: sorry, had to step away for a quick call15:14
slagleegafford: i think it should be working15:14
slaglethere are some outstanding patches though15:14
slaglespredzy: how did you make out following tripleo-docs?15:14
spredzyslagle, current doc won't work out of the box15:15
spredzymainly packaging issue15:16
slaglei see15:16
slaglewhat is the issue?15:16
spredzyI have followed your patch but need to be added also15:16
slaglehmm, i didnt need that when i installed 2 days ago15:16
spredzyelse we have python-scss and other related packge issue while applying horizon stuff15:16
slagleah, i had disabled tuskar15:16
slaglethat's my bad for going off course15:17
*** pblaho has quit IRC15:17
slagleok, i can add that to my existing patch15:17
derekhspredzy: slagle that delorean-deps.repo was added into delorean a day or so ago15:17
*** pblaho has joined #tripleo15:17
spredzyslagle, also I had to apply this patch as of w/o it it would block
*** pblaho has quit IRC15:19
egaffordslagle: One immediate question: You've got enable RDO trunk as:15:19
egaffordsudo curl -o /etc/yum.repos.d/delorean.repo15:19
*** pblaho has joined #tripleo15:19
egaffordPerhaps better to use there?15:19
slagleegafford: current-passed-ci doesnt mean anything for tripleo, unfortunately15:20
spredzyderekh, ack. I ll need to rebuild from scratch in a new env and see what happens with delorean-deps, w/o it openstack undercloud install fails15:20
slaglei'm not sure how that symlink is getting created, but it's not from any successful tripleo ci run15:20
egaffordslagle: Ah. That's a pity.15:20
slaglepackstack probably15:21
derekhslagle: its another ci job downstream in rdo I think15:21
slaglebut yea, we do need a symlink for tripleo-ci15:21
derekhslagle: we should probably do something similar15:21
egaffordIs there a reasonable process that can be described to find the current link? Hardcoding won't likely do it.15:21
openstackgerritDougal Matthews proposed openstack/tripleo-common: [WIP] Remove Tuskar support
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: [WIP] Remove Tuskar support
openstackgerritDougal Matthews proposed openstack/tripleo-docs: [WIP] Remove Tuskar from the CLI usage
slagleegafford: yea, agreed, the docs are just what i was trying to get up asap. the way to figure out the current in use link is to look in openstack-infra/tripleo-ci, but it's hardcoded there as well15:22
openstackgerritDougal Matthews proposed openstack/tripleo-docs: [WIP] Remove Tuskar from the CLI usage
slaglespredzy: is abandoned saying it's solved by a patch that merged in july15:23
slagleso clearly it isn't :)15:23
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: [WIP] Remove Tuskar support
*** Marga_ has quit IRC15:24
spredzyslagle, yeah I've seen the comment, but someway I got it by that so did marios earlier today15:25
derekhspredzy: is this the patch you were missing ? <--- slagle15:25
spredzyso not sure if it didn't get merge in or what, haven't look further atm, just using jistr|call patch let me went through15:25
jistr|callmaybe our delorean pin is older than that?15:26
spredzyderekh, thing is I am using the packages, so if the review isn't in the package I prob don't have it15:26
spredzyI am still not sure at which frequency the reviews are made part of the packages15:27
derekhslagle: spredzy so the trunk repo referenced in the docs is from 2015-08-26, that patch I linked to merged 2015-09-03 ans Ci is using one from 2015-09-0915:30
derekhspredzy: you shoudl try the same as ci
*** lazy_prince has joined #tripleo15:30
derekhspredzy: if it works then update the docs15:30
*** jistr|call is now known as jistr15:30
slaglei have a patch already to do so15:30
slaglei'm just adding the delorean-deps to it now and will push up a new revision15:31
openstackgerritJames Slagle proposed openstack/tripleo-docs: Document using latest TripleO related packages
*** pblaho has quit IRC15:32
slaglehey, our irc notifications are updated!15:32
*** pblaho has joined #tripleo15:32
slaglederekh: ^ have a look at that one. especially what i'm doing with enabling 2 deloreans, and then using includepkgs15:32
derekhslagle: cool, will do, I'll mimic it in a toci patch and see what happens15:33
slaglei'm trying to make this match what CI is doing as close as possible, without having to document that you need to install delorean itself and build packages locally15:33
derekhslagle: yup, sounds reasonable, if this works we can remove all but one of the package builds from ci15:34
slaglei think if we can figure that bit out, we should work on a developer script that toci can use, and maybe generate the docs from that script as well (like we were doing for devtest)15:34
spredzyderekh, actually you are right, using the above ^ I don't have the issue with /dev/loop (this issue happens only with current doc), but I still need the delorean-deps.repo15:34
jistrmarios: all jobs have passed on this one, just not in the same round due to CI instability
jistrmarios: we could merge i guess?15:34
*** david-ly_ has joined #tripleo15:35
* jistr has a bunch of patches landed downstream but not upstream because CI doesn't like me15:35
mariosjistr: reading back15:35
mariosjistr: yeah at least it passed first time yesterday15:36
mariosjistr: shall we try one more?15:36
jistrhehe we can :) but in aggregate all is green on that patch15:37
openstackgerritDerek Higgins proposed openstack/diskimage-builder: Download a compressed centos cloud image
mariosjistr: k, sec15:37
* jistr rechecked a few others where the aggregate hasn't been reached yet15:37
mariosjistr: (looking to see what actually failed on the f21puppethat)15:37
jistralso this one has an aggregate green, and is good now as we switched from Fedora to centos, just needs more +2s
openstackgerritMerged openstack/tripleo-heat-templates: Set pacemaker default resource-stickiness
*** david-ly_ is now known as david-lyle15:41
jistrmarios: the HA job failed on openstack Required file "./overcloud-full.qcow2" does not exist15:41
jistrmarios: we hit that elsewhere too i think, due to network trouble images fail to build sometimes15:42
mariosjistr: yeah see it now (was looking for timeout :/)15:42
*** shivrao has joined #tripleo15:46
mariosjistr: so i definitely needed thanks !15:47
jistrnp :)15:48
marios(otherwise couldn't import instackenv.json to register nodes, like ERROR: openstack No valid host was found. Reason: No conductor service registered which supports ... etc15:48
jistryeah i've seen that before too15:48
jistrmarios: thx for the merge15:50
* jistr has just 4 patches out of upstream/downstream order now :D15:51
openstackgerritPino Toscano proposed openstack-infra/tripleo-ci: [WIP] Generate overcloud image outside the undercloud
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Pull tripleo related packages from delorean/current
openstackgerritMerged openstack-infra/tripleo-ci: Delorean go faster stripes
egaffordslagle: Started from a clean CentOS env; following exactly. First thing I'm seeing is ERROR: Cannot find $TRIPLEO_ROOT/tripleo-incubator/scripts (which is valid; I haven't created TRIPLEO_ROOT or cloned tripleo into it.) Is this expected?15:53
derekhslagle: ^ I think that should be a ci rep-representation of your doc patch15:53
*** jistr has quit IRC15:54
mariosegafford: yeah things still being worked out15:54
mariosegafford: and fixed hopefully15:54
*** alop has joined #tripleo15:59
*** olaph has quit IRC15:59
*** olaph has joined #tripleo16:00
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Remove code to pull clone repositories
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Stop building tuskar-ui-extras
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Stop building ironic-discoverd
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Remove support for mgt-master branches
*** w_ has joined #tripleo16:01
slaglederekh: ok, thanks will take a look16:02
*** olaph has quit IRC16:04
*** ifarkas has quit IRC16:08
derekhslagle: Its quit likely to fail, I havn't tried it, I have yet to set up a env outside of CI, I'll follow your doc in the morning (including your patch) and add feedback16:08
*** aufi has quit IRC16:08
*** Marga_ has joined #tripleo16:09
*** marios has quit IRC16:09
*** marios has joined #tripleo16:10
*** dprince has joined #tripleo16:13
*** hashar_ has joined #tripleo16:14
*** hashar has quit IRC16:15
*** sdake_ is now known as sdake16:23
*** w_ is now known as olaph16:23
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Pull tripleo related packages from delorean/current
*** hashar has joined #tripleo16:28
*** beagles is now known as beagles-brb16:30
*** hashar_ has quit IRC16:31
*** lazy_prince has quit IRC16:35
*** paramite has quit IRC16:40
*** trown is now known as trown|lunch16:44
*** hashar has quit IRC16:46
bnemecderekh: CI is still failing pretty hard on delorean builds. :-(16:47
derekhbnemec: in the last hour ?16:47
bnemecderekh: Yeah16:47
bnemecPretty much everything in the last hour:
bnemecWondering if something changed between when that delorean ci change passed and when it merged.16:48
derekhbnemec: some of those a legitimate fails, but this looks like a problem16:50
derekhbnemec: <openstackgerrit> Merged openstack-infra/tripleo-ci: Delorean go faster stripes
derekhbnemec: opps16:50
derekhbnemec: 2015-09-16 16:41:00.482 | --2015-09-16 16:41:00--
derekh2015-09-16 16:41:00.841 | HTTP request sent, awaiting response... 403 Forbidden16:50
bnemecSeeing the same 403 here.16:50
derekhmoving to #rdo16:51
egaffordCool; it's not just me.16:51
derekhbnemec: so looks like there is a problem with the trunk server, they are talking about switch DNS to the backup server, so I guess wait and see16:53
bnemecderekh: Yeah, I saw that.  Thanks.16:54
derekhbnemec: I gotta run, some of them errors are legitimate fails so I guess we'll see what happens when dns is switched,16:56
derekhbnemec: I'll check back in a bit16:57
bnemecderekh: Cool, thanks.  No sense beating our heads against the wall when we know it won't work.16:57
derekhbnemec: in the mean time I opened this
derekha bot will be spamming this channel every hour about it until its moved16:58
*** dtantsur is now known as dtantsur|afk16:58
*** derekh is now known as derekh_afk16:59
bnemec<3 trellooobot :-)16:59
*** yamahata has quit IRC17:00
*** pradk has quit IRC17:00
*** beagles-brb is now known as beagles17:01
*** pradk has joined #tripleo17:02
slaglehmm, delorean is not building latest os-collect-config17:03
*** athomas has quit IRC17:04
slagleoh, nm, must have goobered my images somehow17:05
*** yog_ has quit IRC17:06
bnemecslagle: rdo is having issues right now, so if you have any dependencies on that it's likely to break.17:07
slagleyea, i know the repo is down17:07
slaglebut i have an old version on these images i built a week or so ago17:07
slaglei probably just messed up the build17:07
*** trellooobot has joined #tripleo17:10
trellooobot| Title                               |              URL              | Members           | Last Active |17:10
trellooobot| trunk server is down, all tripleo c | | **NEEDS MEMBERS** |    11 min   |17:10
*** trellooobot has quit IRC17:10
*** lucasagomes is now known as lucas-dinner17:11
*** lsmola has quit IRC17:13
*** yog_ has joined #tripleo17:19
*** stendulker has quit IRC17:21
*** stendulker has joined #tripleo17:22
*** shivrao has quit IRC17:23
*** stendulker_ has joined #tripleo17:24
*** stendulker has quit IRC17:24
*** yamahata has joined #tripleo17:28
*** shivrao has joined #tripleo17:29
*** stendulker_ has quit IRC17:33
*** akrivoka has quit IRC17:36
*** yog_ has quit IRC17:43
*** yog_ has joined #tripleo17:46
*** trown|lunch is now known as trown17:46
*** pelix has quit IRC17:47
*** pelix has joined #tripleo17:49
*** pradk has quit IRC17:50
*** pelix has quit IRC17:51
*** beagles has quit IRC17:53
*** sdake_ has joined #tripleo17:54
*** sdake has quit IRC17:58
dprinceslagle: I need to land this before this network validation patch can land
dprincederekh_afk: I think we (#tripleo) should consider running our own Delorean servers18:00
trownI think we really just need delorean to be hosted somewhere more reliable18:05
*** pradk has joined #tripleo18:05
trownI mean we dont need 5 9's, but 1 would be nice18:05
slagledprince: why the "if" statement?18:06
*** Goneri has quit IRC18:06
*** Goneri has joined #tripleo18:06
dprinceslagle: because other patches will fail until this lands18:06
dprinceslagle: say someone else has a t-h-t patch in gerrit18:07
slagleright, i get it18:07
slaglehow horrible18:07
dprinceslagle: I know, welcome to package gating issues man18:08
dprinceslagle: ideally, both of these would land in sequence18:08
dprinceslagle: i.e. Depends-On18:08
dprinceslagle: but given we use for package specs ATM that isn't going to happen yet18:08
dprinceslagle: we can remove the conditional after both land however18:09
*** panda has quit IRC18:10
*** trellooobot has joined #tripleo18:10
trellooobot| Title                               |              URL              | Members           | Last Active |18:10
trellooobot| trunk server is down, all tripleo c | | **NEEDS MEMBERS** |    71 min   |18:10
*** trellooobot has quit IRC18:10
slaglewe could also make tht use setup.cfg for all these things that need to be copied18:10
*** panda has joined #tripleo18:10
dprinceslagle: yep, I suspect there will be other fun cases we hit that aren't always easy to deal w/18:11
slagledprince: +2'd, delorean-ci might be down due to the outage18:13
dprinceslagle: cool, yeah I expect the packaging guys may not like these sorts of changes though18:15
dprinceslagle: in this case you are the maintainer I guess though18:15
slagleoh i am?18:16
*** openstackgerrit has quit IRC18:16
slaglehow did i get tricked into that18:16
*** openstackgerrit has joined #tripleo18:16
*** jprovazn has quit IRC18:17
dprincetrown: yeah, more reliability would help too18:24
trowndprince: having run a secondary delorean downstream last cycle...I can say it is not something I would want to do again :p18:26
trownalthough... it was mostly the rebases that were painful18:27
dprincetrown: yeah, well the upstream one just runs on its own mostly18:27
dprincetrown: if we don't solve this developers will run their own anways...18:27
*** david-lyle has quit IRC18:30
EmilienMslagle: I'm still investigating why does not provide the right haproxy config - I ran the manifest locally and I have the right config on my setup18:32
EmilienMwhat's the status of puppet-ha job?18:32
*** sthillma has joined #tripleo18:34
openstackgerritEmilien Macchi proposed openstack/puppet-tripleo: loadbalancer: fix MySQL timeout HAproxy config
*** yog_ has quit IRC18:48
*** nijaba_ has joined #tripleo18:54
*** nijaba_ has quit IRC18:54
*** nijaba_ has joined #tripleo18:54
slagleEmilienM: hasn't passed in a while18:55
*** nijaba has quit IRC18:55
slagleEmilienM: scratch that18:55
slagleleave it to derekh :)18:56
slagleEmilienM: it's been passing on and off since that patch ^ merged18:57
slagledprince: i did float the idea of tripleo maintaining it's own delorean to derek18:57
slaglebut just for the packages we care about18:57
slagleeffectively what $PROJECTS is set to in now18:57
dprinceslagle: yeah, but we pretty much rely on all of them18:58
dprinceslagle: because puppet upstream may require them18:58
*** pradk has quit IRC18:58
EmilienMnice stuff
slaglemy concern about mainting the whole thing (and even just ours), is that the other delorean (offical one) will get outdated quick18:58
slagleand stale18:58
slagleif tripleo only cares about theirs18:58
*** pradk has joined #tripleo19:00
dprinceI wouldn't go as far as saying I don't care about the other one19:00
dprinceBut I do care about uptime19:01
dprinceand clearly the cloud running that has no SLA19:01
dprinceso I'd say lets just go and run a Delorean instance somewhere that does have good uptime19:01
slaglewould we set the expectation that tripleo works with the official delorean? or just the tripleo one?19:03
slagleif both, we'd need CI on both...that's the only way we could set that expectation19:03
dprinceslagle: likely both would work, but if asked the question I'd say just ours19:04
slagleso then we're back to relying on the bad infra19:04
slagleif we say just ours, the other will never work19:04
slaglei'll make that prediction right now :)19:04
dprinceslagle: right, well we can't control that19:05
dprinceslagle: if it matures then I'd be happy to switch back19:05
slaglei dunno, it feels kind of like forking from the community, that community being rdo. which seems wrong19:06
dprinceslagle: first question would be... where is this delorean instance running19:06
slaglemaybe we can figure out another solution, like just getting it moved to better infra19:06
dprinceslagle: yeah, well upstream packages is what I really want19:06
slagledprince: the cloud man, it's in the cloud19:06
dprinceslagle: but we haven't got that yet19:07
dprinceslagle: Yeah, well some clouds have uptime guarantees... and some don't19:07
dprinceslagle: re, the forking the community thing I gotta say I disagree19:08
trownthere was talk a while back about moving it to RAX19:08
dprincecollaboration around a common set of specs is close enough collaboration19:08
dprincetrown: that would be my first choice19:08
trownI think that would be pretty great to support another openstack cloud19:08
slagledprince: forking is probably too harsh19:09
slaglemy point is just that if we're not using that delorean, we're not working with those that are19:09
slaglei could see us patching/hotfixing ours to make us happy19:09
slaglespecs included19:09
dprinceslagle: Call it a mirror? however you need to justify this19:10
dprinceslagle: all I want is uptime along w/ this collaboration19:10
dprincesame specs and everything19:10
*** trellooobot has joined #tripleo19:10
trellooobot| Title                               |              URL              | Members           | Last Active |19:10
trellooobot| trunk server is down, all tripleo c | | **NEEDS MEMBERS** |   131 min   |19:10
*** trellooobot has quit IRC19:10
trownright on queue19:10
dprinceslagle: there really is very little hot fixing w/ Delorean now I think19:10
slaglei know, all done to the same instance19:10
slaglethat's kind of my point :)19:11
slaglevs divergence19:11
dprinceslagle: you run your own local instack right? would you call it divergent too?19:11
dprinceslagle: I call it customized, with better uptime perhaps19:11
dprinceslagle: just depends on how you view it19:12
slagleif i patched it and told people to use that vs the other one, then yes, i would call that divergent19:12
slagleand hosted it in a git repo19:12
dprinceslagle: no patch would be applied to ours that wasn't sent to the other (the sources are shared)19:12
slagleand said it was what "worked"19:12
slagleand no gaurantee that the other "worked"19:12
dprinceslagle: and the spec repos are identical19:12
slagleabsolutely divergent19:12
slagleabsolutely a fork19:12
dprinceslagle: there is not separate repo though19:13
dprinceslagle: I think you are way overstating these things19:13
*** matbu has quit IRC19:13
slagleok maybe so19:13
dprinceslagle: call it an "availability zone" man19:13
slaglemy point is merely, if there's 2, and the content is not the same...only 1 will work, the 1 being we CI19:13
dprinceslagle: seriously, there is no difference in git sources in my view of maintaining this19:13
slagleif we are truly talking a mirror here, then fine19:14
slaglewhy not just use rsync?19:14
slagleno need to run a delorean instance19:14
dprinceslagle: I said a mirror work work19:14
dprinceslagle: just uptime is all I want here19:14
*** dujelly has joined #tripleo19:19
*** dujelly has quit IRC19:19
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Use alternative trunk repository
derekh_afkslagle: dprince reading back a bit, I don't we want to start running our own delorean, if anything just mirror the repositories we care about19:22
derekh_afkdprince: slagle e.g. in that patch ^^ I;m using
slagleack. is that what they setup as the failover?19:25
*** matbu has joined #tripleo19:26
derekh_afkslagle: the rdo peopleo? they had a mirror of all the data, switch DNS for the outage and all went fine, they switched back again today and all was fine, until something went wronge on the instance, they didn't know what when I was talking to them19:27
*** hashar has joined #tripleo19:27
dprincederekh_afk: agree, mirror would be best I think19:29
derekh_afkslagle: sorry, now I  know what you meant, no that was my failover incase theirs didn't work19:29
dprincederekh_afk: will monitor this patch and if it passes we can use it temporarily19:30
dprincederekh_afk: +2 from me, I've no issue w/ it19:30
slaglederekh_afk: ok, cool :)19:30
*** matbu has quit IRC19:46
*** matbu has joined #tripleo19:47
*** w_ has joined #tripleo19:48
*** sdake has joined #tripleo19:49
*** olaph has quit IRC19:51
*** olaph has joined #tripleo19:52
*** sdake_ has quit IRC19:52
*** jaosorior has quit IRC19:53
*** w_ has quit IRC19:53
egaffordslagle, et al.: On undercloud install from, I'm seeing "Error: /Stage[main]/Apache::Service/Service[httpd]/ensure: change from stopped to running failed: Could not start Service[httpd]: Execution of '/bin/systemctl start httpd' returned 1: Job for httpd.service failed. See 'systemctl status httpd.service' a20:01
egaffordnd 'journalctl -xn' for details." This occurs just after /Stage[main]/Keystone::Service/Service[keystone]. Is this a known issue, or am I special (and likely wrong about something)?20:01
egaffordI'm doing this on CentOS 7 (host and guest), should it matter.20:02
*** openstackstatus has quit IRC20:02
*** mestery has quit IRC20:02
slagleok, that's probably the scss error20:03
dprinceegafford: I think there is a Horizon bug issue that still causes apache start errors20:03
slagledid you enable the delorean-deps repo?20:03
slagleegafford: i've got a patch out to the docs here:
*** mestery has joined #tripleo20:03
egaffordslagle: I don't believe that I did, on average. I'm pretty strictly following the doc; figure that at least that way I'm serving a purpose while I fail. :)20:03
slagleegafford: alternatively you can set enable_tuskar=false in undercloud.conf20:04
egaffordslagle: Cool; thanks.20:04
slaglealthough, i'm not sure that would clean up what's already been done :/20:04
*** openstackstatus has joined #tripleo20:04
*** ChanServ sets mode: +v openstackstatus20:04
egaffordslagle: Well, I can always respin the whole instack from image.20:04
egaffordslagle: That's not a big sacrifice at this point.20:05
derekh_afkbnemec: slagle dprince my patch has gotten past the stage where it runs delorean which is a start20:05
*** Goneri has quit IRC20:05
derekh_afkbnemec: slagle dprince we still gotta find out if DIB will use the mirror or not, but I wont be around to check it20:05
derekh_afkbnemec: slagle dprince so I'll leave it with ye ;-)20:06
*** derekh_afk is now known as derekh20:06
dprincederekh: cool, thanks20:06
slagleegafford: ok, in the case enable_tuskar=true might be the easiest20:07
slagleerr, sorry20:07
trownslagle: unfortunately that does not work :p20:07
slagleegafford: enable_tuskar=false20:07
slagletrown: why?20:07
trownI have that on my todo list to file a bug20:07
trownthat option and enable_tempest are not fully wired20:08
egaffordtrown: Good to know.20:08
trownthere is nothing in that puts them to the instack_env20:08
trownI fought with it for a good part of yesterday evening20:09
dprincetrown: hmmm, when I run instack I point it to my undercloud.conf20:09
slaglewell enable_tuskar really is about the ui only, so it fixes the horizon issue20:09
dprincetrown: is this not the preferred way to do it?20:09
openstackgerritDerek Higgins proposed openstack/diskimage-builder: Download a compressed centos cloud images
trowndprince: all the docs go through the tripleoclient20:09
dprincetrown: if you do that the option works fine for me... at least I think it was20:10
*** cody-somerville has joined #tripleo20:10
trownso `openstack undercloud install`20:10
*** trellooobot has joined #tripleo20:10
trellooobot| Title                               |              URL              | Members           | Last Active |20:10
trellooobot| trunk server is down, all tripleo c | | **NEEDS MEMBERS** |   191 min   |20:10
*** trellooobot has quit IRC20:10
dprincetrown: right, I do the same thing we do in tripleo-ci ATM20:10
*** derekh has quit IRC20:11
trownI think there is another issue beyond that even20:11
trownI set the env variable ENABLE_TUSKAR='false' and it was reset to true by the time that the puppet-stack-config element ran20:12
trownI ended up with this super hacky hack to not use tuskar or tempest:
trownas tempest also fails the undercloud install on liberty because there is no tempest liberty rpm20:13
dprince may be up again20:15
dprincetrown: trying it again now... pretty sure it works for me20:16
slagletrown: it *should* make it into the env here:
egaffordslagle, et al: Adding delorean-deps got me past that hump, happily. Now on to "novaclient.exceptions.BadRequest: Keypair data is invalid: failed to generate fingerprint" (just after "Completed phase post-configure"). Again, if not known, I'll poke; if known, great.20:16
trownslagle: ah it is probably entirely this secondary issue then where it gets reset sometime before the puppest-stack-config20:18
*** beagles has joined #tripleo20:18
slagleegafford: haven't seen it. it's trying to upload ~/.ssh/ to nova. does that file exist?20:18
egaffordAyup. It actually generates that key. ("Generated new ssh key in ~/.ssh/id_rsa")20:19
egaffordRerunning to see if it just hiccuped.20:20
slaglecheck the nova logs maybe for more info20:20
*** ukalifon has joined #tripleo20:25
trownEmilienM: I think the unit tests will now pass on helps to know how to run them locally :)20:28
EmilienMtrown: I'll review it asap20:29
trownEmilienM: I am not sure what to do for dtantsur|afk comment though. I guess I do not really understand the intended audience of the puppet-ironic module20:29
*** weshay_xchat has quit IRC20:31
trownEmilienM: my patch is pretty opionated towards tripleo, but I can see the point to just make it like upstream defaults for ironic-inspector devstack plugin and change the other stuff in instack-undercloud20:31
dprincetrown: Just ran the installer w/ cat /home/stack/undercloud.conf | grep enable_tuskar20:32
dprinceenable_tuskar = false20:32
*** jayg is now known as jayg|g0n320:32
*** weshay_xchat has joined #tripleo20:32
trowndprince: with delorean liberty packages?20:33
dprincetrown: and apache is started, but there is no tuskar, tuskar-ui20:33
dprincetrown: Delorean master20:33
dprincetrown: trunk...20:33
dprincetrown: instack master -> trunk Delorean20:33
*** weshay_xchat has quit IRC20:34
openstackgerritJames Slagle proposed openstack/tripleo-heat-templates: Use already Deployed/Installed servers
*** weshay_xchat has joined #tripleo20:34
trowndprince: hmm... these are my exact steps
trowndprince: I will give it another go, and see if the problem happens again20:37
slagletrown: you're adding those options under the [auth] section20:37
slagleneed to be under [DEFAULT]20:37
slaglesince it's just going at the end of the file20:37
dprincetrown: yeah, I would recommend scp'ing your own instack file into the VM perhaps20:37
trownslagle: ah, that would explain the optiongs not working20:37
dprincetrown: or use sed20:37
trownslagle: but not the setting of the env variables directly20:38
*** weshay_xchat has quit IRC20:38
trownbut if the options work, I do not much care if the env variables work20:38
slaglesince they're not set, the defaults are used20:39
slaglesetting env vars externally isn't meant to work20:39
slagleif that's what you mean20:39
trownonly for legacy stuff20:39
trownsweet thanks20:40
*** lucas-dinner has quit IRC20:40
slagleyea, i guess20:40
slaglemaybe that was intended, maybe not :)20:40
*** weshay_xchat has joined #tripleo20:40
*** shardy has quit IRC20:44
egaffordslagle: Have undercloud, will travel, hopefully. :) Thanks for the help.20:45
slaglenp. thanks for sticking with it20:45
slaglewe're trying to make it better :)20:45
egaffordslagle: Eh, it's my job at the moment. Oh, I know, seriously.20:45
egaffordI *really* hope my admiration for what you guys are doing is coming through consistently, even as I try to shine a bit of light on the new user UX.20:46
slagleit is :). and we take all feedback, good and bad20:47
*** dprince has quit IRC20:47
egaffordEh, yeah, but you guys have been working really hard; the last thing anyone needs is for feedback to be pointlessly negative.20:48
egaffordLeave that to customers. :)20:48
* egafford loves customers; sorry customers! egafford kids.20:49
trownthey can be pretty harsh though :p20:50
egaffordtrown: From time to time, yes, absolutely.20:50
trownI came from a support job, so I have experienced it first hand20:51
egaffordAnd in the end, the customer is always right, but it's really nice to always be decent to humans.20:51
egaffordtrown: Yeah, I've done my share of customer support from dev (and had a previous career in nursing), so yeah, customes.20:51
egaffordr in there.20:51
trownwhoa...nursing would be a whole other level20:52
trownI know I am probably at my worst if I am in a hospital for any reason...excepting maybe labor and delivery20:52
egaffordIt's a whole nother thing to have someone actively throw a used diaper on the floor in pent-up frustration during a conversation and then have to clean it up while they watch, yeah.20:52
egaffordWe do have that in common with monkeys if pressed, I guess.20:53
*** untriaged-bot has joined #tripleo21:00
untriaged-botNo untriaged bugs so far! \o/21:00
*** untriaged-bot has quit IRC21:00
egaffordAw, bot is so happy!21:00
egaffordWay to have a positive attitude, bot.21:01
*** ukalifon has quit IRC21:04
*** cody-somerville has quit IRC21:09
*** trellooobot has joined #tripleo21:10
trellooobot| Title                               |              URL              | Members           | Last Active |21:10
trellooobot| trunk server is down, all tripleo c | | **NEEDS MEMBERS** |   251 min   |21:10
*** trellooobot has quit IRC21:10
*** lblanchard has quit IRC21:11
slaglei am going to silence that since appears back21:13
*** yamahata has quit IRC21:14
egaffordslagle, et al: Overcloud images built and registered. However:21:19
egaffordopenstack baremetal import --json instackenv.json21:19
egaffordWARNING: ironicclient.common.http Request returned failure status.21:19
egaffordERROR: openstack No valid host was found. Reason: No conductor service registered which supports driver pxe_ssh. (HTTP 400)21:19
slagleegafford: yea, edit /etc/ironic/ironic.conf and set enabled=false under [inspector]21:20
slaglerestart openstack-ironic-conductor21:20
slaglei haven't had time to dig into this21:21
slaglei think it's just a matter of the package missing from delorean21:21
egaffordIs that a known issue w/ a bug?21:21
slagleegafford: you can check journalctl -u openstack-ironic-conductor to confirm that's the issue21:21
slaglenot sure21:21
slagletrown: is there a bug open on this  ^^21:21
trownit is missing ironic-inspector from delorean21:21
egaffordDriverLoadError: python-ironic-inspector-client Python module not found21:21
trownwhich I have been working on...but delorean failing for the whole day kind of slowed that down21:22
trownegafford: installing will fix it21:25
trownhopefully that will be in delorean by the end of the week21:25
egaffordtrown: Cool. I'm filing a tripleo bug for tracking, just so it's not lost: any specific component you'd like me to log it against?21:26
trownhmm...I think it would be an RDO bug21:26
trownsince it is just missing packages from liberty delorean21:26
egaffordOK... if we're depending on RDO as the canonical development environment setup tool, though, it seems like we're backed into a circular dependency there.21:27
trownreally it is the fact that we got the client in without the main package21:27
egafford(Unless there are plans to create a purely upstream dev env creation tool that doesn't depend on RDO packages and does everything involved from source.)21:28
trownwell for packages ya... I do not know how to handle that for tripleo... the 'real' fix would not involve any patches to tripleo repos though21:29
egaffordWhile that's true, I'd caution against any move that makes bugs in your canonical dev env setup flow a downstream concern. Still, that's not mine to decide; it's your project. I'll hold off on filing if you're already in motion, if it's not clear where the bug should go. Might be worth talking about the ramifications of using midstream to deploy upstream in terms of bug tracking and triage though, within your teams.21:31
trownegafford: ya, you bring up a good point21:33
trownthe delorean packages are from master, but it is still midstream of a sort21:33
trownas changes to RDO can break tripleo as this shows21:34
egaffordQuite so.21:34
trownI gotta run...have a good evening21:35
egaffordYou've sort of built a dam beneath RDO and turned the whole upper 2/3 of your river into a pleasant lake.21:35
egaffordtrown: See you!21:35
*** julim has quit IRC21:35
*** cody-somerville has joined #tripleo21:35
*** cody-somerville has joined #tripleo21:35
*** trown is now known as trown|outttypeww21:35
*** hashar has quit IRC21:38
*** rpothier has quit IRC21:48
*** beagles has quit IRC21:54
*** david-lyle has joined #tripleo21:56
*** pradk has quit IRC21:59
*** MasterPiece has joined #tripleo22:03
*** panda has quit IRC22:09
*** panda has joined #tripleo22:10
*** david-lyle has quit IRC22:29
openstackgerritDerek Higgins proposed openstack/diskimage-builder: Download a compressed centos cloud images
*** jpeeler has quit IRC22:32
EmilienMtrown|outttypeww: fails before of RDO outage, don't worry about jenkins -122:32
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: external loadbalancer: Add net_vip_map_external
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Make load balancer deployment optional via template param
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the Controller nodes
gfidente^^ rebased22:52
*** cody-somerville has quit IRC22:53
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Make load balancer deployment optional via template param
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the Controller nodes
*** egafford has quit IRC23:07
openstackgerritShiva Prasad Rao proposed openstack/tripleo-heat-templates: Enable Cisco N1KV driver
openstackgerritMerged openstack/diskimage-builder: Add #!/bin/bash to library functions
*** shivrao has quit IRC23:41
*** chlong has quit IRC23:41
*** yamahata has joined #tripleo23:51

Generated by 2.14.0 by Marius Gedminas - find it at!