Thursday, 2015-04-16

*** yamahata has quit IRC00:10
*** daneyon has quit IRC00:33
*** daneyon has joined #tripleo00:34
*** bhunter has quit IRC00:35
*** dasm|afk has quit IRC00:35
*** dasm|afk has joined #tripleo00:36
*** bhunter has joined #tripleo00:36
*** thrash is now known as thrash|g0ne01:09
*** julim has joined #tripleo01:18
openstackgerritMerged stackforge/kolla: Update developer documentation  https://review.openstack.org/17299801:31
*** eghobo has quit IRC01:34
*** jerryz has joined #tripleo01:35
openstackgerritMerged stackforge/kolla: Adds a Status Script  https://review.openstack.org/17351101:36
openstackgerritMerged stackforge/kolla: Updates Docker Client in Heat Template  https://review.openstack.org/17364801:50
*** weiyu-001 has joined #tripleo01:55
openstackgerritMerged stackforge/kolla: Adds OpenStack Clients to Heat Template  https://review.openstack.org/17363801:55
*** weiyu-001 has quit IRC02:01
*** rhallisey has quit IRC02:09
*** tzumainn has quit IRC02:11
*** weshay has quit IRC02:12
*** sdake_ has joined #tripleo02:16
*** sdake__ has joined #tripleo02:19
openstackgerritSteven Dake proposed stackforge/kolla: Add workaround to work with cirros image  https://review.openstack.org/17415802:19
*** sdake has quit IRC02:19
*** sdake_ has quit IRC02:22
*** weiyu-001 has joined #tripleo02:31
*** pradeep has quit IRC02:39
*** jerryz has quit IRC03:00
*** spzala has quit IRC03:06
*** Marga_ has joined #tripleo03:49
*** Marga_ has quit IRC03:49
*** Marga_ has joined #tripleo03:50
*** eghobo has joined #tripleo04:02
openstackgerritMartin André proposed stackforge/kolla: Introduce `tools/kolla` to interact with kolla  https://review.openstack.org/17419204:16
openstackgerritMartin André proposed stackforge/kolla: Add remove command to `tools/kolla`  https://review.openstack.org/17419304:16
*** Guest16106 has quit IRC04:25
*** noslzzp has quit IRC04:34
*** pradeep has joined #tripleo04:54
*** eghobo_ has joined #tripleo05:01
*** eghobo has quit IRC05:05
*** Marga_ has quit IRC05:07
*** jtomasek has joined #tripleo05:20
*** yog__ has joined #tripleo05:23
*** jtomasek has quit IRC05:43
openstackgerritMartin André proposed stackforge/kolla: Add remove command to `tools/kolla`  https://review.openstack.org/17419305:58
openstackgerritMartin André proposed stackforge/kolla: Introduce `tools/kolla` to interact with kolla  https://review.openstack.org/17419205:58
*** jtomasek has joined #tripleo05:58
*** dasm|afk is now known as dasm06:17
*** bnemec has quit IRC06:18
*** jcoufal has joined #tripleo06:18
*** eghobo_ has quit IRC06:22
*** jcoufal has quit IRC06:30
*** eghobo has joined #tripleo06:38
*** ukalifon1 has joined #tripleo06:39
*** jprovazn has joined #tripleo06:40
*** eghobo has quit IRC06:42
*** bnemec has joined #tripleo07:00
*** athomas has joined #tripleo07:19
*** dtantsur|afk is now known as dtantsur07:19
*** Goneri has joined #tripleo07:28
*** ifarkas has joined #tripleo07:39
*** chlong has quit IRC07:40
*** Goneri has quit IRC07:43
*** jistr has joined #tripleo07:43
*** Goneri has joined #tripleo07:50
*** aufi has joined #tripleo07:57
*** mmagr has joined #tripleo08:00
*** regebro has quit IRC08:08
*** bvandenh has joined #tripleo08:10
*** gfidente has joined #tripleo08:12
*** regebro has joined #tripleo08:13
*** lucasagomes has joined #tripleo08:14
gfidentejistr, morning08:15
jistro/08:15
*** hewbrocca has joined #tripleo08:15
*** mmagr is now known as mmagr|afk08:19
*** derekh has joined #tripleo08:19
gfidentederekh, I have a couple of changes which got the +A but didn't get merged as I suppose that was yesterday during gerrit/zuul issues08:21
gfidentederekh, shall I "reverify" ?08:21
derekhpoint me at them, I'll see if a second a+ will do it08:21
gfidentehttps://review.openstack.org/#/c/172572/ and https://review.openstack.org/#/c/173454/08:22
gfidenteI read on the wiki there is some magic keyword which triggers a new 'verify' after they got the +A08:22
openstackgerritMerged openstack/tripleo-heat-templates: Perform basic setup of pacemaker cluster on controllers  https://review.openstack.org/17257208:22
gfidentebut can't find it anymore08:22
gfidenteoh wow, thanks for that08:22
openstackgerritMerged openstack/tripleo-heat-templates: Do not downcase IP addresses  https://review.openstack.org/17345408:23
derekhgfidente: no prob08:23
derekhgfidente: ZUUL probably just missed the first A+ in the stream it follows, but got the second I just added08:24
*** mmagr|afk is now known as mmagr08:24
gfidenteso what I was looking for used to be in the gerritworkflow wiki, but that is now migrated into a doc and can't find it there08:26
derekhgfidente: reverify? it used to be used if your patch failed gate checks but I'm not sure if its still in use, gotta pop out again bbiab08:30
*** derekh is now known as derekh_afk08:30
gfidentettyl thanks08:30
*** sdake__ has quit IRC08:36
*** sdake has joined #tripleo08:39
*** mestery has joined #tripleo08:40
*** mestery_ has quit IRC08:43
*** mcornea has joined #tripleo08:50
openstackgerritBen Kero proposed openstack/diskimage-builder: Fixes path for distros lacking :/bin  https://review.openstack.org/17410709:01
bkeroSpamapS: ^ :) thanks for the feedback09:02
*** dtantsur is now known as dtantsur|brb09:06
openstackgerritThomas Bechtold proposed openstack/diskimage-builder: Fix partition table creation  https://review.openstack.org/17427209:22
*** jang has joined #tripleo09:22
*** jang1 has joined #tripleo09:22
*** jang has quit IRC09:23
*** jang1 has quit IRC09:23
*** athomas has quit IRC09:32
*** bvandenh has quit IRC09:39
*** pradeep has quit IRC09:42
*** athomas has joined #tripleo09:43
*** ishant has joined #tripleo09:45
*** athomas_ has joined #tripleo09:48
*** athomas has quit IRC09:50
*** akrivoka has joined #tripleo09:51
*** bvandenh has joined #tripleo09:55
*** Marga_ has joined #tripleo09:55
openstackgerritGiulio Fidente proposed openstack/tripleo-incubator: Cleanup logic which enable NeutronDVR  https://review.openstack.org/17428209:57
gfidentethis guy has 2x +2 but needs +A https://review.openstack.org/#/c/173248/09:57
openstackgerritGiulio Fidente proposed openstack/tripleo-incubator: Cleanup logic which enables NeutronDVR  https://review.openstack.org/17428210:00
*** mmagr is now known as mmagr|afk10:02
openstackgerritMerged openstack/tripleo-incubator: Switch default COMPUTESCALE setting down to 1 from 2  https://review.openstack.org/17324810:04
bkerogfidente: heya, not sure I understand your comment on that path bug10:08
bkero/usr/local/sbin would be nice to have, but scripts in many elements are calling 'sh', which is in /bin/. Since arch doesn't have that in the path, building images from it will be broken unless it is added.10:09
gfidentebkero, so on arch I see /bin and /sbin linking /usr/bin10:10
gfidentebkero, and even /usb/sbin links to /usr/bin10:11
*** aufi has quit IRC10:11
bkerogfidente: correct. On Arch the PATH environment variable lacks /bin/.10:11
bkeroSince the chroot is passing that PATH environment variable in, the 'chroot sh "$cmd"' will fail with 'sh: command not found'.10:12
gfidentebkero, yeah and /sbin as well or /usr/sbin because they all point to same path10:12
*** pradeep has joined #tripleo10:12
bkeroI tried hardcoding 'sh' to '/bin/sh', which did work. But then it complained about mktemp and grep being missing because they are also  in /bin/10:12
bkeroappending :/bin/:/sbin/ solved that problem10:13
gfidentewhich image distro are you building?10:13
bkeroI tried building ubuntu and centos. Both had the same problem.10:13
gfidenteso I think the script is just passing your $PATH10:14
bkeroCorrect10:14
gfidentebut you're trying to build an image which doesn't work well with the PATH inherited from arch10:15
bkeroAlso correct, that is what the patch is solving.10:15
gfidenteok it wasn't clear to me from the commit message10:16
*** spredzy|afk is now known as spredzy10:17
gfidentecan I suggest a slightly harder approach which I think would be nice to have? how about we use pathmunge to make sure $PATH has all those you listed? http://www.cyberciti.biz/faq/redhat-linux-pathmunge-command-in-shell-script/10:18
gfidentebkero, do you think it is worth?10:18
openstackgerritBen Kero proposed openstack/diskimage-builder: Fixes path for distros lacking :/bin  https://review.openstack.org/17410710:19
*** pelix has quit IRC10:19
gfidentebkero, maybe we default to add all the standard paths 'after' to make sure if one customizes his PATH gets his stuff working10:19
*** yuanying has quit IRC10:20
bkerogfidente: are you suggesting including pathmunge inside img-functions somewhere?10:20
gfidentebkero, yeah10:20
gfidentenot sure if it is overkilling10:21
*** yuanying has joined #tripleo10:21
gfidentewhat do you think?10:21
bkeroIf we were setting PATH in more places than just right there I think it might be worth it10:21
bkeroAlso if we could reliably source pathmunge from somewhere else instead of having to include our own it might be worth it10:22
bkerobut failing those probably overkill :)10:22
*** pelix has joined #tripleo10:22
gfidentebkero, ack, I'll post a comment about it, thanks for pinging here :)10:23
bkerogfidente: Thanks!10:23
bkerogfidente: another problem is that the comment right above it tells that it is expected to escape-out the PATH variable (\$PATH) so it gets eval'd later. I bet pathmunge wouldn't like that at all.10:23
* bkero back to debugging exactly why sfdisk broke when trying to reread disk.10:24
bkeroIt's calling ioctl(fd, BLKRRPART), which is returning -1, errno 22, which means Invalid Argument. I see the parted folks patched that to be expected and acceptable behavior though, so I might try to do the same for sfdisk and see if util-linux takes it as a bugfix upstream.10:25
*** yuanying has quit IRC10:32
openstackgerritGiulio Fidente proposed openstack/tripleo-incubator: Cleanup logic which enables NeutronDVR  https://review.openstack.org/17428210:40
*** pdb has quit IRC10:53
*** derekh_afk is now known as derekh10:54
*** dtantsur|brb is now known as dtantsur10:56
*** pradeep has quit IRC10:58
*** Marga_ has quit IRC10:59
bkeroOkay, patch submitted to sfdisk to hopefully fix this nonsense, time for bed.11:00
*** pradeep has joined #tripleo11:01
*** mmagr|afk is now known as mmagr11:02
*** weiyu-001 has left #tripleo11:04
*** pdb has joined #tripleo11:05
*** aufi has joined #tripleo11:10
*** aufi has quit IRC11:12
*** aufi has joined #tripleo11:12
*** pradeep has quit IRC11:23
*** rhallisey has joined #tripleo11:23
openstackgerritGiulio Fidente proposed openstack/tripleo-incubator: Add some logic to optionally use L3 agents relocation/ha/dvr  https://review.openstack.org/17430811:25
*** sseago has joined #tripleo11:25
openstackgerritGiulio Fidente proposed openstack/tripleo-incubator: Add logic to select L3 agents configuration  https://review.openstack.org/17430811:31
*** thrash|g0ne is now known as thrash11:38
openstackgerritJiri Tomasek proposed openstack/tuskar-ui: Update roles in Service config  https://review.openstack.org/17431111:40
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Switch VIP management from Keepalived to Pacemaker  https://review.openstack.org/17377311:41
*** bvandenh has quit IRC11:42
*** lucasagomes is now known as lucas-hungry11:45
*** weshay has joined #tripleo11:52
*** ishant has quit IRC11:56
*** jdob has joined #tripleo11:58
*** trown|outttypeww is now known as trown12:02
*** Marga_ has joined #tripleo12:06
*** prad has quit IRC12:10
*** jayg|g0n3 is now known as jayg12:16
*** radez_g0n3 is now known as radez12:19
*** dprince has joined #tripleo12:20
*** julim has quit IRC12:22
openstackgerritGiulio Fidente proposed openstack/tripleo-incubator: Add logic to select L3 agents configuration  https://review.openstack.org/17430812:30
openstackgerritRyan Brady proposed openstack/tuskar-ui: Update roles  https://review.openstack.org/17391412:35
*** panda__ is now known as panda12:38
*** rlandy has joined #tripleo12:43
*** mmagr is now known as mmagr|afk12:45
*** mmagr|afk is now known as mmagr12:46
*** julim has joined #tripleo12:47
*** lucas-hungry is now known as lucasagomes12:49
*** Marga_ has quit IRC12:51
*** Marga_ has joined #tripleo12:58
*** tzumainn has joined #tripleo13:04
*** Marga_ has quit IRC13:15
*** noslzzp has joined #tripleo13:16
*** bvandenh has joined #tripleo13:19
*** spzala has joined #tripleo13:20
openstackgerritYanis Guenane proposed openstack/tripleo-heat-templates: galera: Allow installer to setup galera cluster  https://review.openstack.org/16498313:23
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: Add support for generic controller validations  https://review.openstack.org/17415013:25
*** sdake_ has joined #tripleo13:26
openstackgerritMerged openstack/tuskar-ui: Update roles in Service config  https://review.openstack.org/17431113:26
*** sdake has quit IRC13:29
*** trown is now known as trown|brb13:30
openstackgerritMerged openstack/tuskar-ui: Update roles  https://review.openstack.org/17391413:32
*** lblanchard has joined #tripleo13:32
*** trown|brb is now known as trown13:34
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: puppet: add missing ::glance include on controller  https://review.openstack.org/17438213:34
EmilienMdprince, slagle ^13:35
dprinceEmilienM: fine with the change (because include would only import it once), but what functional issues would this cause?13:36
spredzydprince, on https://review.openstack.org/#/c/167538/, did you get to read my answer ? What are your thought ?13:36
EmilienMdprince: we are (trying) to solve packaging issues...13:37
dprinceEmilienM: okay, packaging what? glance itself? or these templates?13:38
* dprince is still slow this morning13:38
EmilienMdprince: https://bugs.launchpad.net/puppet-glance/+bug/144497413:38
openstackLaunchpad bug 1444974 in puppet-glance "openstack-glance which provides glance-manage not getting installed on red hat platforms" [Critical,In progress] - Assigned to Emilien Macchi (emilienm)13:38
dprinceEmilienM: I see. Sounds like a breaking package change that needs to be reverted.13:39
dprinceEmilienM: regardless I'm fine with this fix to the templates on its own.13:39
EmilienMdprince: they are fixing it in RDO right now.13:39
EmilienMdprince: yeah with/without the packaging issue, we always need to load ::<module>13:40
*** prad has joined #tripleo13:40
dprinceEmilienM: cool. +2 on your patch as well13:40
EmilienMit's a convention13:40
EmilienMdprince: thx.13:40
dprinceEmilienM: sure, appreciate the explanation. Just wanted to know why this came up13:40
*** regebro has quit IRC13:40
dprincespredzy: I didn't see your explanation on 167538. Reading now...13:41
dprincespredzy: I still think I'd like the template and the module that uses to to be in the same project.13:42
dprincespredzy: otherwise, someone might come and think they could remove the template (because it isn't used)13:43
spredzydprince, so you'd rather go with a new class in tripleo that just call the file resource ?13:43
dprincespredzy: I think so. Only because to me a template (in the puppet sense) is a private resource and shouldn't be used externally13:44
spredzydprince, well actually we can put it in the if $redis {} in the loadbalancer, since this will be only used when doing HA13:44
dprincespredzy: okay, could do that13:44
dprincespredzy: so long as the template and the puppet code using that (private) template live in the same repo I think I'm happy13:45
spredzydprince, ack13:46
dprincespredzy: while I've got you lets talk about validations again13:47
dprincespredzy: have you had a chance to review https://review.openstack.org/#/c/174150/ yet?13:47
dprincespredzy: I just fixed an issue I hit with a Heat parameter with my initial patch this morning.13:48
dprincespredzy: get_file didn't like {get_param: ...} for some reason. So I removed that bit13:48
spredzydprince, yes, I've got a review w/ the same idea (https://review.openstack.org/#/c/173758/) but was trying to use EmilienM's serverspec hook13:49
spredzythis was before our talk yesterday13:49
*** mmagr is now known as mmagr|afk13:51
spredzydprince, I think Steven comment https://review.openstack.org/#/c/173758/1/puppet/controller-post-puppet.yaml,cm (the 2nd one) also apply13:52
jistrgfidente: fyi here's a first glance at the pcsd work https://github.com/jistr/puppetlabs-corosync/commit/b1c88d5efbab27530680b500273f0f790a3091b813:52
*** mmagr|afk is now known as mmagr13:53
jistrturns out following the HA pacemaker setup steps from the ref doc precisely clashes with the way puppet-corosync works13:53
jistri *think* the commands in the ref doc probably alter corosync.conf and build it step by step, and then puppet-corosync comes and rewrites it from scratch, so things went bad really fast when i tried to merge the two approaches13:55
gfidentejistr, looking at the change now, do you have it posted against puppet-corosync?13:58
jistrfortunately the pcs commands seem to be pretty smart in a way that e.g. if i don't provide the nodes for "pcs cluster auth", they are taken from corosync.conf. So i let the ::corosync class write the conf first, and then i call "pcs cluster auth" and it reads the nodes from there. This is a bit of a departure from the HA docs, but it seems to be the only way to make it work with puppet-corosync's approach to dealing with corosync.conf.13:59
jistrgfidente: not yet, i just cleaned the code and moved it from the tripleo composition layer to puppet-corosync, now i'll use the code from the composition layer, test it end 2 end and then when i know it still works, i'll submit the PR14:00
gfidentejistr, I *think* we're good as long as we implement same config14:00
dprincespredzy: maybe, I did something a little different because I'm not using depends_on14:01
dprincespredzy: well, I'm using depends_on, but I'm not depending on the validations themselves.14:01
dprincespredzy: if that makes sense. I left an inline note on my review about this subtle behavior14:01
dprincespredzy: essentially I wanted the validations to run immediately after each puppet deployment14:02
dprincespredzy: so I used the 'name' property to order them as such. The effect is the validations run in the same os-collect-config pass14:02
spredzydprince, ack will go through it again14:03
dprincespredzy: depends_on for everything might be simpler but the extra round-robin for each Heat API -> os-collect-config run concerned me.14:04
*** sseago has left #tripleo14:04
*** pradeep has joined #tripleo14:05
openstackgerritYanis Guenane proposed stackforge/puppet-tripleo: Loadbalancer: Add support for Redis  https://review.openstack.org/17112114:07
openstackgerritYanis Guenane proposed stackforge/puppet-tripleo: Loadbalancer: Add support for Redis  https://review.openstack.org/17112114:08
openstackgerritYanis Guenane proposed stackforge/puppet-tripleo: Loadbalancer: Add support for Redis  https://review.openstack.org/17112114:09
gfidentespredzy, I will never vote that anymore14:10
spredzygfidente, ahaha sorry some typos I noticed just after :/14:11
*** HenryG_ is now known as HenryG14:11
openstackgerritYanis Guenane proposed openstack/tripleo-heat-templates: Add support for Redis configuration  https://review.openstack.org/16753814:15
*** yog__ has quit IRC14:17
gfidentespredzy, EmilienM is there a recommendation in between cs_primitive from corosync and pacemaker/service in openstack_extra?14:20
*** jpena has joined #tripleo14:21
gfidentelooks like we can stick with cs_primitive to me?14:24
*** akrivoka has quit IRC14:25
jpenagfidente: using pacemaker/service from openstack_extras gives you some advantages, such as automated service stop/start when you do any action that requires refreshing the service14:25
jpenae.g. update a keystone config file -> restart service, this is handled by the pacemaker provider14:26
spredzygfidente, agree with jpena, worth case scenario you can still use it just to create the cs_primitiv14:26
spredzyuntil we need mofre feature from that class14:26
*** Goneri has quit IRC14:26
spredzydprince, went though the review again, Ok I see where I misundertood it the first time. Looks what we needed :) Thanks.14:27
spredzydprince, do you confirm that if validation fail, the stack will stop and the resource will be in a CREATE_FAILED status ?14:27
dprincespredzy: it should, yes14:28
spredzyok, then yep, it does what's it supposed to. Will deploy an overcloud with your patch and try to stick the rabbitmq test in it to have a meaning full basis14:29
spredzythanks dprince !14:29
*** sdake has joined #tripleo14:29
dprincespredzy: so feel free to adopt this and run with it. I would like shardy to look at it too14:30
*** sdake_ has quit IRC14:33
*** jprovazn has quit IRC14:35
*** regebro has joined #tripleo14:36
*** akrivoka has joined #tripleo14:38
openstackgerritMerged openstack/tripleo-puppet-elements: Implement serverspec elements  https://review.openstack.org/17178514:39
*** regebro has quit IRC14:42
*** noslzzp has quit IRC14:45
*** bvandenh has quit IRC14:48
*** noslzzp has joined #tripleo14:50
*** subscope has quit IRC14:50
*** mmagr is now known as mmagr|afk14:51
*** mmagr|afk is now known as mmagr14:53
*** Marga_ has joined #tripleo14:59
*** untriaged-bot has joined #tripleo15:00
untriaged-botUntriaged bugs so far:15:00
untriaged-bothttps://bugs.launchpad.net/os-collect-config/+bug/144499915:00
openstackLaunchpad bug 1444999 in os-collect-config "oslo.config requirement is not following global requirements" [Undecided,New]15:00
*** untriaged-bot has quit IRC15:00
*** asparks has joined #tripleo15:04
*** ukalifon1 has quit IRC15:07
*** jcoufal has joined #tripleo15:17
*** mmagr has quit IRC15:17
*** thrash is now known as thrash|f00dz15:18
*** pdb has quit IRC15:20
*** pdb has joined #tripleo15:21
*** eghobo has joined #tripleo15:26
*** eghobo_ has joined #tripleo15:27
*** pdb has quit IRC15:29
*** eghobo has quit IRC15:31
*** bvandenh has joined #tripleo15:31
*** pdb has joined #tripleo15:32
*** aufi has quit IRC15:33
*** spzala has quit IRC15:33
*** Marga_ has quit IRC15:39
*** eghobo_ has quit IRC15:41
*** jcoufal has quit IRC15:43
*** dtantsur is now known as dtantsur|afk15:46
*** asparks has quit IRC15:49
*** alop has joined #tripleo15:51
*** alop has quit IRC15:53
*** alop has joined #tripleo15:53
*** ifarkas has quit IRC15:55
jistrcore folks: i've triggered a CI run for this - https://review.openstack.org/#/c/166447/ if you recall later today, please take a look at the result and do a +A15:58
jistrseems like it breaks devstack https://bugs.launchpad.net/os-collect-config/+bug/144499915:58
openstackLaunchpad bug 1444999 in os-collect-config "oslo.config requirement is not following global requirements" [High,Triaged]15:58
* jistr bumped it to critical after second thought16:00
dprincejistr: I'll watch for it16:01
jistrdprince: thx16:01
*** Marga_ has joined #tripleo16:07
*** jistr has quit IRC16:08
openstackgerritYanis Guenane proposed stackforge/puppet-tripleo: Loadbalancer: Add support for Redis  https://review.openstack.org/17112116:08
*** athomas_ has quit IRC16:14
*** Marga_ has quit IRC16:16
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: puppet: install Horizon on overcloud-controller  https://review.openstack.org/17444916:18
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: puppet: install Horizon on overcloud-controller  https://review.openstack.org/16439216:20
*** MasterPiece has joined #tripleo16:24
*** adrianopetrich has quit IRC16:24
openstackgerritgreghaynes proposed openstack/tripleo-specs: Add spec for DIB element testing  https://review.openstack.org/17445616:29
greghaynesSpamapS: ^16:29
*** daneyon has quit IRC16:31
*** alop has quit IRC16:31
*** daneyon has joined #tripleo16:31
greghaynesits a very tripleo-esque proposal16:32
*** Marga_ has joined #tripleo16:32
*** Goneri has joined #tripleo16:33
*** derekh has quit IRC16:39
*** EmilienM is now known as EmilienM|afk16:39
*** rhallisey has quit IRC16:41
dprincegreghaynes: the spec looks fine. I gotta ask though do we *really* need a spec to propose unit test to a project?16:42
dprincegreghaynes: if you just simply proposed some testing patches it seems like you might get better feedback...16:42
*** bvandenh has quit IRC16:43
*** rhallisey has joined #tripleo16:43
dprincegreghaynes: oh, and I'm not sure the existance of these tests supplants the desire to have actual integration tests which use DIB (i.e. TripleO w/ Ironic)16:43
dprincegreghaynes: not sure if it is related but I saw someone added that as a potential summit topic16:44
SpamapSgreghaynes: \o/16:44
greghaynesdprince: heh, you wont get any argument from me on just writing the code already16:44
greghaynesI wasnt expecting it to be that brief16:44
SpamapSdprince: That seems like it would be a good working topic. I don't know that we want fishbowl for it since it will be heavy on internals.16:44
SpamapSdprince: The main reason to have narrow tests is to allow developers to move faster. The main reason to have integration tests is to make sure they don't move TOO fast. ;)16:45
*** pelix has quit IRC16:45
* SpamapS turns that into a tweet.16:45
dprinceThe topic I see on the etherpad is "Is there a path where we can stop 'gating' DIB changes on TripleO tests? Maybe better DIB tests?"16:47
dprinceI'm all for better DIB tests. But I'm not sure the goal of adding them should be to supplant TripleO as a testing tool. That is my point16:47
dprinceIt should be help developers (who can more easily run DIB tests locally) propose better quality DIB patches, which would still get integration tested to ensure something hasn't slipped in16:48
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Configures haproxy as a Pacemaker resource when EnablePacemaker  https://review.openstack.org/17446516:48
SpamapSdprince: Oh no I don't think thats a good topic. No there's not a point and I don't want to do that. :)16:49
greghaynesYes, so I think adding unit tests and even smaller integration tests is a thing we all should want reguardless16:49
*** thrash|f00dz is now known as thrash16:49
greghaynesits kind of a no brainer win16:49
greghaynesmy real hope here is that I can do some hotpathing for container building and maybe some improvements to caching and we all of a sudden have a way to develop elements that we can iterate really fast on16:50
dprincegreghaynes: sure16:51
dprinceSpamapS: not sure who proposed the topic, only mentioned it because it seemed related to the spec16:51
openstackgerritRyan Brady proposed openstack/tuskar-ui: Show required service configuration parameters as required  https://review.openstack.org/16623516:51
greghaynesYea, so I proposed that, just want to make sure they are separate things16:52
greghaynesmy issue is - what value do we provide for infra's use case with those tests16:52
SpamapSdprince: it's possible I did. I may sometimes go crazy and reverse position. ;)16:52
greghaynesor troves16:52
SpamapSdprince: I don't remember proposing it though. :-P16:53
dprincegreghaynes: the value is Infra isn't the only user of DIB16:53
greghaynesyes16:54
greghaynesand youre right - we need to make sure ironic and such still integrates16:54
greghaynesmaybe I made it sound more dire than I thought - I dont want to rip out tripleo CI from dib now, I want us to get comfortable without the goal of cogating with every user16:55
dprincegreghaynes: cool. I'm all for better tests. with or without a Spec16:55
dprincegreghaynes: if spec'ing helps then we can approve it I think...16:56
*** pradeep has quit IRC16:56
SpamapSmordred: https://review.openstack.org/#/c/166650/ is in merge conflict16:56
SpamapSAlso, it would be most appreciated by the infra team if https://review.openstack.org/#/q/status:open+project:openstack/diskimage-builder+branch:master+topic:dib-nodepool,n,z would get priority the next time people do a cycle of reviews on diskimage-builder. Thanks.16:57
mordredSpamapS: yay!16:57
greghaynesYes, its much less LOC now :)16:57
greghaynesok, so a different change that came up in that patch set - we dont have a way to opt-in to caching16:58
greghaynesand some caching is pretty heavy handed16:58
openstackgerritMonty Taylor proposed openstack/diskimage-builder: Add a yum-minimal element that just uses yum  https://review.openstack.org/16664016:58
openstackgerritMonty Taylor proposed openstack/diskimage-builder: Port centos-minimal to yum-minimal  https://review.openstack.org/16664116:58
openstackgerritMonty Taylor proposed openstack/diskimage-builder: Support VHD output format  https://review.openstack.org/15643316:58
openstackgerritMonty Taylor proposed openstack/diskimage-builder: Break install-types out of base  https://review.openstack.org/17304816:58
openstackgerritMonty Taylor proposed openstack/diskimage-builder: Split the debootstrap functions into an element  https://review.openstack.org/16665016:58
openstackgerritMonty Taylor proposed openstack/diskimage-builder: Add element to process config-drive network info  https://review.openstack.org/16700316:58
openstackgerritRyan Brady proposed openstack/tuskar-ui: Show required service configuration parameters as required  https://review.openstack.org/16623516:58
greghaynesthe example being that we tar up the rootfs after a debootstrap since its a 1:1 with a cached cloudimg16:58
*** Marga__ has joined #tripleo17:02
*** Marga_ has quit IRC17:02
mordredgreghaynes, SpamapS: ^^ that should be all rebased up and whatnot17:08
mordredgreghaynes: except we never consume that tarred rootfs by default17:09
SpamapSmordred: sweet I'll rebase-review them now (which means I'll make sure you didn't slip in any Superman 3 hacks and then +2)17:09
greghaynesmordred: yes. Im just thinking that since we should either use it by default or not make it by default17:10
mordredSpamapS: ignore the superman 3 hacks. they are not the droids you are looking for17:10
greghaynessince only doing one of those two things mean by default its not very useful17:10
greghaynesand using it by default is hard, so the not making it by default might be the way to go17:11
SpamapSgreghaynes: --offline opts out of cache refresh. We could also add a --refresh-cache which does the opposite.. ONLY refreshes cache.17:11
SpamapSI'm not actually certain what it is you're trying to achieve though. :-P17:12
mordredSpamapS: so - currently, the tarball is created and cached and never used17:12
mordredSpamapS: which means everytime you build an image using debootstrap, you're creating a tarball which is then never consumed17:13
SpamapSAh right, because only --offline disables refreshing it17:13
*** aufi has joined #tripleo17:13
SpamapSso yeah, we need --online :)17:13
greghaynesyerp, and we dont have an invalidation mechanism for it17:13
mordredyah. and if you ARE doing the workflow where you want to run debootstrap every time - there's no way to say "yeah, no need to copy tons of data kthbai"17:14
*** MasterPiece has quit IRC17:14
SpamapSthat we can do now17:14
*** pradeep has joined #tripleo17:15
SpamapShm no17:15
mordredof course, what I really want is "reuse the debootstrap tarball if it exists, but run all the other elements online"17:15
mordredbecause honestly, I do not care if debootstrap has changed its output since 20 minutes ago17:15
*** lucasagomes is now known as lucas-dinner17:15
greghaynesmaybe we just mtime the tarball?17:16
greghaynesif its over a day stale, dont use it unless offline17:16
mordred++17:16
SpamapSOk so let's just make a DIB_ONLINE which has precedence over DIB_OFFLINE, and basically means "You must always fetch even if you have a cache."17:16
SpamapSHm17:16
greghaynesboth those things17:16
*** radez is now known as radez_g0n317:16
SpamapSReally yeah.17:16
SpamapSDIB_CACHE_EXPIRE=<an int number of seconds> ? default to 86400 ?17:17
greghaynes+217:17
SpamapSwe could even get clever and use that in our fetches for cloud images so we don't ever pull two down in one day even if they were updated.17:17
SpamapSsince right now we IMS based on mtime.17:17
mordred++17:19
mordredagree17:19
mordredbecause sometimes, you just don't care17:19
mordredand DIB_CACHE_EXPIRE=0 means "never expire" perhaps17:19
greghaynesdib offline can just set that17:19
mordredor never cache17:19
mordredyeah17:19
mordredsomething17:19
mordredchicken17:19
greghaynesimplementation left as exercise for the reader17:20
* SpamapS is poking at a patch now17:21
SpamapStired of having good ideas and not seeing them immediately implemented. ;)17:21
greghaynesspeaking of actually implementing - annoying thing about the unit testing spec is that we dont have a way to test for failures17:22
SpamapSgreghaynes: we don't?17:22
SpamapSgreghaynes: we have break=after-error .. seems like you can coopt that.17:22
greghaynesah!17:22
*** jpena is now known as jpena|away17:28
*** Marga__ has quit IRC17:30
*** Marga_ has joined #tripleo17:38
*** EmilienM|afk is now known as EmilienM17:43
*** mcornea has quit IRC17:44
*** jprovazn has joined #tripleo17:44
*** yamahata has joined #tripleo17:45
greghaynesSpamapS: You wouldnt have slides from a dib intro-ey talk, would you?17:46
*** vipul has quit IRC17:50
*** EmilienM is now known as EmilienM|afk17:55
openstackgerritRyan Brady proposed openstack/tuskar-ui: Show required service configuration parameters as required  https://review.openstack.org/16623517:56
*** spzala has joined #tripleo17:57
*** trown is now known as trown|lunch17:58
*** sdake_ has joined #tripleo17:59
*** vipul has joined #tripleo17:59
*** Marga_ has quit IRC18:02
*** sdake has quit IRC18:02
*** Marga_ has joined #tripleo18:06
*** jpena|away has quit IRC18:10
*** sdake has joined #tripleo18:13
*** sdake_ has quit IRC18:17
openstackgerritOpenStack Proposal Bot proposed openstack/diskimage-builder: Updated from global requirements  https://review.openstack.org/16643118:18
openstackgerritBen Kero proposed openstack/diskimage-builder: Append full path to img-functions:run_in_target  https://review.openstack.org/17410718:23
*** radez_g0n3 is now known as radez18:24
*** Goneri has quit IRC18:29
SpamapSgreghaynes: no18:32
SpamapSgreghaynes: but lifeless does18:32
*** EmilienM|afk is now known as EmilienM18:33
*** dprince has quit IRC18:35
*** yamahata has quit IRC18:42
greghayneslifeless: hey there18:42
*** mcornea has joined #tripleo18:43
openstackgerritAbel Lopez proposed openstack/diskimage-builder: Use environment setfiles  https://review.openstack.org/17456118:48
*** Marga_ has quit IRC18:49
*** Marga_ has joined #tripleo18:49
*** alop has joined #tripleo18:50
*** Marga_ has quit IRC18:54
*** Marga_ has joined #tripleo18:54
*** trown|lunch is now known as trown18:54
*** eghobo has joined #tripleo19:04
*** pradeep has quit IRC19:09
openstackgerritYanis Guenane proposed stackforge/puppet-tripleo: Loadbalancer: Add support for Redis  https://review.openstack.org/17112119:13
*** Goneri has joined #tripleo19:19
*** radez is now known as radez_g0n319:19
*** pradeep has joined #tripleo19:24
*** noslzzp has quit IRC19:30
*** noslzzp has joined #tripleo19:30
*** spzala has quit IRC19:32
lifelessgreghaynes: ?19:32
*** Marga_ has quit IRC19:35
*** lucas-dinner has quit IRC19:41
*** zbitter is now known as zaneb19:42
*** julim has quit IRC19:48
-openstackstatus- NOTICE: gerrit has been restarted to clear a problem with its event stream. any gerrit changes updated or approved between 19:14 and 19:46 utc will need to be rechecked or have their approval reapplied for zuul to pick them up19:48
*** yamahata has joined #tripleo19:52
*** tzumainn has quit IRC19:52
*** lucas-dinner has joined #tripleo19:54
*** jprovazn has quit IRC19:55
*** radez_g0n3 is now known as radez19:56
*** jehb has joined #tripleo19:59
*** pradeep has quit IRC20:03
*** rlandy has quit IRC20:03
*** rlandy has joined #tripleo20:04
*** Marga_ has joined #tripleo20:06
*** Marga_ has quit IRC20:11
*** gfidente has quit IRC20:12
*** pradeep has joined #tripleo20:18
*** spzala has joined #tripleo20:19
*** aufi has quit IRC20:20
greghayneslifeless: I hear you have slides from an intro to diskimage-builder talk20:20
*** dprince has joined #tripleo20:24
*** dprince has quit IRC20:54
*** untriaged-bot has joined #tripleo21:00
untriaged-botUntriaged bugs so far:21:00
untriaged-bothttps://bugs.launchpad.net/diskimage-builder/+bug/144516121:00
openstackLaunchpad bug 1445161 in diskimage-builder "setfiles doesn't work on all distros" [Undecided,New]21:00
*** untriaged-bot has quit IRC21:00
openstackgerritMerged stackforge/puppet-tripleo: Loadbalancer: Add support for Redis  https://review.openstack.org/17112121:02
*** jayg is now known as jayg|g0n321:03
openstackgerritMerged openstack/os-collect-config: Updated from global requirements  https://review.openstack.org/16644721:03
*** MasterPiece has joined #tripleo21:04
openstackgerritDan Sneddon proposed openstack/tripleo-heat-templates: Separate the network config by family and add a "node_index" for each group.  https://review.openstack.org/17362421:06
*** Marga_ has joined #tripleo21:07
*** lblanchard has quit IRC21:11
*** Marga_ has quit IRC21:12
*** trown is now known as trown|outttypeww21:13
*** EmilienM is now known as EmilienM|afk21:13
*** Marga_ has joined #tripleo21:18
*** Marga_ has quit IRC21:23
*** Marga_ has joined #tripleo21:24
*** alop has quit IRC21:26
*** Marga_ has quit IRC21:34
*** Marga_ has joined #tripleo21:35
*** yamahata has quit IRC21:36
*** yamahata has joined #tripleo21:36
*** lucas-dinner has quit IRC21:38
*** pradeep has quit IRC21:40
lifelessgreghaynes: are you greg or gregory in OWA?21:48
greghayneslifeless: unsure, im greghaynes@hp21:49
lifelessgreghaynes: greg then.21:50
greghaynesgot it, ty21:50
*** regebro has joined #tripleo21:55
*** EmilienM|afk is now known as EmilienM21:55
*** rhallisey has quit IRC22:00
*** rlandy has quit IRC22:00
*** yamahata has quit IRC22:02
openstackgerritDaneyon Hansen proposed stackforge/kolla: Fixes a bug with stop script for neutron server  https://review.openstack.org/17462422:09
*** thrash is now known as thrash|g0ne22:10
*** BadCub is now known as BadCub_Hiding22:13
*** regebro has quit IRC22:18
*** yamahata has joined #tripleo22:21
*** weshay has quit IRC22:24
openstackgerritDaneyon Hansen proposed stackforge/kolla: Disables firewalld for Heat-deployed Fedora instances  https://review.openstack.org/17462822:26
*** sdake_ has joined #tripleo22:28
*** Marga_ has quit IRC22:29
*** Marga_ has joined #tripleo22:30
*** Marga_ has quit IRC22:30
*** Marga_ has joined #tripleo22:31
openstackgerritDaneyon Hansen proposed stackforge/kolla: Adds Mariadb-client to the Heat template  https://review.openstack.org/17462922:31
*** sdake has quit IRC22:31
*** Marga_ has quit IRC22:32
*** Marga_ has joined #tripleo22:32
*** sdake_ has quit IRC22:33
*** EmilienM is now known as EmilienM|afk22:40
openstackgerritDan Prince proposed openstack/tripleo-incubator: build-images: drive DIB via YAML config file  https://review.openstack.org/17364422:50
openstackgerritDan Prince proposed openstack/tripleo-incubator: devtest_overcloud.sh: split out image building  https://review.openstack.org/17364522:50
openstackgerritDan Sneddon proposed openstack/tripleo-heat-templates: Separate the network config by family and add a "node_index" for each group.  https://review.openstack.org/17362422:52
*** Marga_ has quit IRC22:54
*** Marga_ has joined #tripleo22:55
*** yuanying has joined #tripleo23:01
*** chlong has joined #tripleo23:01
stevebakerlifeless: hey, can you point me to where untriaged-bot is configured?23:02
lifelessstevebaker: uhm, ask rpodolyaka IIRC.23:03
lifelessstevebaker: I have zero clue23:03
stevebakerok23:03
*** Marga_ has quit IRC23:05
*** Goneri has quit IRC23:06
*** EmilienM|afk is now known as EmilienM23:10
*** regebro has joined #tripleo23:12
*** regebro has quit IRC23:21
*** lblanchard has joined #tripleo23:21
*** dguerri is now known as _dguerri23:23
*** _dguerri is now known as dguerri23:24
*** regebro has joined #tripleo23:30
*** sdake has joined #tripleo23:34
*** regebro has quit IRC23:35
*** Marga_ has joined #tripleo23:36
*** mcornea has quit IRC23:37
*** panda has quit IRC23:37
*** regebro has joined #tripleo23:37
*** panda has joined #tripleo23:38
*** lblanchard has quit IRC23:39
*** Marga_ has quit IRC23:40
*** regebro has quit IRC23:48
*** sdake_ has joined #tripleo23:55
*** sdake has quit IRC23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!