Monday, 2017-03-06

« Sunday, 2017-03-05 Index Tuesday, 2017-03-07 »
*** amotoki has joined #openstack-ironic00:27
*** spartacloud has joined #openstack-ironic00:41
*** spartacloud has quit IRC00:46
*** crushil has joined #openstack-ironic00:49
*** SerenaFeng has joined #openstack-ironic01:01
*** hw_wutianwei has joined #openstack-ironic01:02
*** crushil has quit IRC01:07
*** tovin07 has joined #openstack-ironic01:08
*** phuongnh has joined #openstack-ironic01:12
*** rbudden has joined #openstack-ironic01:13
*** zhurong has joined #openstack-ironic01:15
*** bigjools has quit IRC01:29
*** zhangjl has joined #openstack-ironic01:30
*** fragatina has quit IRC01:30
*** fragatina has joined #openstack-ironic01:31
*** fragatin_ has joined #openstack-ironic01:32
openstackgerrityuhui proposed openstack/ironic master: Correct typos in doc files  https://review.openstack.org/43777601:32
*** lambert has joined #openstack-ironic01:33
*** fragatina has quit IRC01:36
*** gcb has joined #openstack-ironic01:48
*** chlong__ has joined #openstack-ironic01:55
*** fragatin_ has quit IRC01:59
*** zhs_ has joined #openstack-ironic02:01
*** Goneri has joined #openstack-ironic02:05
*** Goneri has quit IRC02:14
*** zhurong has quit IRC02:15
*** zhurong has joined #openstack-ironic02:19
*** sasha2 has joined #openstack-ironic02:32
openstackgerritTovin Seven proposed openstack/ironic master: Enable OSProfiler support in Ironic  https://review.openstack.org/35814202:36
*** zhs_ has quit IRC02:54
*** zhs_ has joined #openstack-ironic02:55
*** tiendc has joined #openstack-ironic03:01
*** jaypipes has joined #openstack-ironic03:14
openstackgerritzack chen proposed openstack/ironic-python-agent master: Collected logs is 'bytes' type.  https://review.openstack.org/43890103:27
*** sasha2 has quit IRC03:29
*** stendulker has joined #openstack-ironic03:37
*** pmannidi has quit IRC03:37
*** SerenaFeng has quit IRC03:51
*** jaypipes has quit IRC03:54
*** rbudden has quit IRC04:13
*** udesale has joined #openstack-ironic04:28
*** jaypipes has joined #openstack-ironic04:30
*** SerenaFeng has joined #openstack-ironic04:36
*** spartacloud has joined #openstack-ironic04:43
*** spartacloud has quit IRC04:48
*** nmathew has joined #openstack-ironic04:50
openstackgerritAparna proposed openstack/ironic master: Support to collect log files related to clean step  https://review.openstack.org/44050205:04
*** SerenaFeng has quit IRC05:04
*** aparnav has joined #openstack-ironic05:08
*** SerenaFeng has joined #openstack-ironic05:27
*** SerenaFeng has quit IRC05:28
*** amotoki_ has joined #openstack-ironic05:42
*** vikrant has joined #openstack-ironic05:44
*** amotoki has quit IRC05:46
*** pmannidi has joined #openstack-ironic05:52
*** tiendc has quit IRC05:53
*** jaganathan has joined #openstack-ironic05:59
*** rcernin has joined #openstack-ironic06:08
*** SerenaFeng has joined #openstack-ironic06:22
*** SerenaFeng has quit IRC06:27
*** phuongnh has quit IRC06:27
*** SerenaFeng has joined #openstack-ironic06:38
openstackgerritVarun Gadiraju proposed openstack/ironic master: WIP/DNM: Testing multinode + multitenant + grenade job  https://review.openstack.org/40389506:41
*** dsneddon has joined #openstack-ironic06:48
*** pcaruana has joined #openstack-ironic06:58
*** vmud213 has joined #openstack-ironic07:02
*** pcaruana has quit IRC07:05
*** pcaruana has joined #openstack-ironic07:06
openstackgerritM V P Nitesh proposed openstack/ironic master: ipmi_address is validated  https://review.openstack.org/43694507:07
*** ChubYann has quit IRC07:12
*** tesseract has joined #openstack-ironic07:13
*** fragatina has joined #openstack-ironic07:18
*** jaosorior has joined #openstack-ironic07:30
*** rbartal has joined #openstack-ironic07:30
*** stendulker has quit IRC07:35
*** chlong__ has quit IRC07:36
*** fragatina has quit IRC07:38
*** ralonsoh has joined #openstack-ironic07:40
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Fix README.rst  https://review.openstack.org/44177807:50
*** v3nki has joined #openstack-ironic07:50
v3nkihi ,07:51
v3nkiMust provide Keystone credentials or user-defined endpoint and token, error was: publicURL endpoint for baremetal service in RegionOne region not found07:51
v3nkii'm getting this error, can anyone pls help on this...07:52
v3nkiI have my local.conf file in https://shrib.com/6i6lr9N4dnSrbMj07:55
v3nkiI am beginner, just trying openstack with ironic(using devstack)08:00
v3nkihelp needed08:00
patchbotError: 'supybot.needed' is not a valid configuration variable.08:00
*** racedo has joined #openstack-ironic08:07
*** mjura has joined #openstack-ironic08:09
*** tovin07 has quit IRC08:12
*** tovin07 has joined #openstack-ironic08:12
*** pmannidi has quit IRC08:16
*** karthiks has joined #openstack-ironic08:16
*** Nisha_Agarwal has joined #openstack-ironic08:18
v3nkiI'm getting this error, while installing (devstack with ironic) --> "Must provide Keystone credentials or user-defined endpoint and token, error was: publicURL endpoint for baremetal service in RegionOne region not found"08:29
v3nkianyone pls help...08:29
v3nkii'm trying this with stable release mitaka08:30
*** amarao has joined #openstack-ironic08:30
*** moshele has joined #openstack-ironic08:33
*** mgoddard has joined #openstack-ironic08:34
*** nmathew- has joined #openstack-ironic08:35
*** nmathew has quit IRC08:38
*** venki has joined #openstack-ironic08:40
openstackgerritOksana Voshchana proposed openstack/ironic master: Adds wsgi to ironic-api in devstack  https://review.openstack.org/43085108:41
*** v3nki has quit IRC08:43
*** fragatina has joined #openstack-ironic08:46
*** ccamacho has joined #openstack-ironic08:47
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Use virtualbmc for virtual BM nodes  https://review.openstack.org/42737608:50
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Use JSON baremetal data in testing  https://review.openstack.org/43089808:50
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Use Ansible to create test VMs  https://review.openstack.org/42707108:50
*** zzzeek has quit IRC09:00
*** zzzeek has joined #openstack-ironic09:00
*** moshele has quit IRC09:04
*** amoralej|off is now known as amoralej09:04
openstackgerritMerged openstack/python-ironicclient master: Updated from global requirements  https://review.openstack.org/44087809:10
*** e0ne has joined #openstack-ironic09:13
*** d0ugal has quit IRC09:14
*** nmathew- has quit IRC09:15
*** mgoddard has quit IRC09:16
*** lambert has quit IRC09:17
*** lambert has joined #openstack-ironic09:17
*** milan has joined #openstack-ironic09:23
openstackgerritMerged openstack/ironic-lib master: Updated from global requirements  https://review.openstack.org/44087109:23
milanmorning Ironic! :)09:23
pas-hamorning ironicers :)09:24
pas-hamorning milan09:24
milanmorning pas-ha! :)09:24
*** d0ugal has joined #openstack-ironic09:26
*** d0ugal has quit IRC09:26
*** d0ugal has joined #openstack-ironic09:26
vdrokmorning ironic, milan pas-ha !09:28
milanmorning vdrok! :)09:29
*** udesale__ has joined #openstack-ironic09:29
*** hshiina has joined #openstack-ironic09:29
*** udesale has quit IRC09:32
Nisha_AgarwalHi Ironic'iers09:36
*** athomas has joined #openstack-ironic09:37
*** SerenaFeng has quit IRC09:39
*** derekh has joined #openstack-ironic09:42
*** lucas-afk is now known as lucasagomes09:43
lucasagomesmorning all09:43
vmud213Hi Nisha, lucasagomes, vdrok, milan, pas-ha and Ironicers09:49
*** mgoddard has joined #openstack-ironic09:50
*** udesale has joined #openstack-ironic09:51
*** jmccarthy has joined #openstack-ironic09:52
*** udesale__ has quit IRC09:53
vdrokmorning Nisha_Agarwal lucasagomes vmud21309:57
*** moshele has joined #openstack-ironic09:58
*** vills has joined #openstack-ironic09:59
*** vills_ has joined #openstack-ironic10:00
vdrokvenki: it seems like ironic was not able to register the endpoint in the keystone catalog. Also, I think you don't have to specify stable/mitaka for all projects, you can just checkout stable/mitaka in devstack and run ./stack.sh, all the projects should be on mitaka after that. another thing is, we seem to have issues with requirements in mitaka, so not10:02
vdroksure what's the exact problem there, unless you provide some logs :)10:02
*** MattMan has quit IRC10:02
*** MattMan has joined #openstack-ironic10:03
milan10:04
milanmorning lucasagomes, Nisha_Agarwal, vmud213! :)10:04
aparnavmorning all10:07
*** aparnav has quit IRC10:10
*** stendulker has joined #openstack-ironic10:10
*** athomas has quit IRC10:11
galynamorning :)10:13
lucasagomes:D10:13
vdrokmorning galyna10:13
*** moshele has quit IRC10:15
*** moshele has joined #openstack-ironic10:15
*** aparnav has joined #openstack-ironic10:15
*** athomas has joined #openstack-ironic10:18
aparnavhey vdrok, related to the RFE https://bugs.launchpad.net/ironic/+bug/166973710:21
openstackLaunchpad bug 1669737 in Ironic "[RFE] Collecting log files related to a clean step" [Wishlist,In progress] - Assigned to Aparna (aparnavtce)10:21
vdrokaparnav: morning10:21
aparnavmorning vdrok10:22
aparnavvdrok, all the clean step may not have the log file associated with it, isn't it better to collect log after the particular clean step gets completed?10:22
aparnavvdrok, I have small change in place for the above RFE https://review.openstack.org/#/c/440495/, I am looking for suggestions10:24
patchbotpatch 440495 - ironic-python-agent - Support to collect logs using file name10:24
vdrokaparnav: ok, do we collect the logs in case of cleaning failure now?10:25
vdrokI mean, not in your patches, but at all10:25
*** Haaibo has joined #openstack-ironic10:26
aparnavvdrok, Currenly we do collect ramdisk logs, but not log files specific for clean step10:26
*** Nisha_Agarwal has quit IRC10:27
vdrokaparnav: do you expect the clean step in the ramdisk to expose the list of files somehow, or the user will be providing the ones (s)he wants?10:28
vdrokaparnav: all of that has to be provided in the rfe :) or in spec, I'm not sure how much changes it requires off the top of my head. I thought we don't collect cleaning logs at all, only during deployment. anyway, please provide more info on the rfe, and make sure that patches you do appear in the bug, so that it is trackable10:30
*** mgoddard has quit IRC10:31
aparnavvdrok, I am considering as part of clean step we are aware of the log files and will be able to collect it as part of post clean step hook10:31
*** openstackgerrit has quit IRC10:33
aparnavvdrok, you are right. we will collect ramdisk logs only during deployment. Currently we wont collect cleaning logs.10:34
vdrokaparnav: yeah, ok, that seems fine, please add this info to the bug10:35
aparnavvdrok, sure, thank you10:36
*** amotoki_ has quit IRC10:40
*** openstackgerrit has joined #openstack-ironic10:45
openstackgerritM V P Nitesh proposed openstack/python-ironicclient master: OSC 'node list' recognizes provision state 'available'  https://review.openstack.org/43155610:45
*** zhangjl has left #openstack-ironic10:46
*** dtantsur|afk is now known as dtantsur10:46
*** mgould_ is now known as mgould10:47
mgouldmorning dtantsur lucasagomes milan aparnav galyna vdrok vmud213 pas-ha and other Ironicers!10:48
vdrokmorning mgould :)10:48
*** vills has quit IRC10:48
openstackgerritM V P Nitesh proposed openstack/ironic master: Now displays maintenance and maintenance_reason  https://review.openstack.org/43302910:48
milanmorning aparnav dtantsur galyna mgould! :)10:49
vmud213morning mgould :)10:49
dtantsurmorning ironic, morning milan, mgould, vmud213, lucasagomes, vdrok, aparnav, galyna, pas-ha10:49
pas-hamorning dtantsur mgould :)10:49
*** aparnav has quit IRC10:50
milanguys, I'm wondering, is it possible to decorate a function so that the decorator creates more functions? /me would love that for micro versions tests ;)10:50
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Fix README.rst  https://review.openstack.org/44177810:51
*** gK-1wm-su has joined #openstack-ironic10:51
*** vills_ has quit IRC10:52
*** gK-1wm-su has quit IRC10:55
*** sambetts|afk is now known as sambetts10:58
vdrokmorning dtantsur :)10:58
vmud213morning dtantsur :)10:59
vdrokmilan: I guess it is a metaclass? :)10:59
*** Nisha_Agarwal has joined #openstack-ironic10:59
*** mgoddard has joined #openstack-ironic11:00
* mgould runs screaming11:00
*** udesale has quit IRC11:00
vdrokmilan: oh, I think ddt does that https://github.com/txels/ddt/blob/master/ddt.py11:01
* mgould was about to suggest ddt :-)11:01
vdrokbut that's a decorator on the class11:01
mgouldit's in global-requirements, but IIRC not yet used in ironic11:01
mgouldmilan: what are you trying to achieve?11:01
* mgould apologises for not making any progress on the hypothesis PoC11:02
*** gcb has quit IRC11:07
*** venki has quit IRC11:08
sambettsmilan: why do you need the dectorator?11:14
sambettsdecorator*11:14
sambettsIMO the nicest way to do it is to use class that inherit all the tests from the parent class, and a class variable that sets the micro-version11:15
*** vills has joined #openstack-ironic11:17
*** fragatina has quit IRC11:22
*** fragatina has joined #openstack-ironic11:23
*** stendulker has quit IRC11:25
mgouldis the micro-version the thing you want to parametrize on?11:25
sambettsmilan, mgould: something like http://paste.openstack.org/show/601519/11:29
sambettsmgould: then all previous tests get retested or overridden on new microversions and all old tests will still exist and run too11:30
Nisha_Agarwallucasagomes, hi11:32
mgouldsambetts: and you can add new tests at each stage?11:32
mgouldyeah, that makes sense11:32
mgouldddt might be a bit less elegant in that case11:33
sambettsmgould: the only problem I can see with this is that it means that everytime you add a new microversion you double the number of tests that we run11:33
mgouldsambetts: it's not quite that bad11:34
mgouldif we have n microversions and add k new tests each time, adding a new microversion will add k*(n+1) tests11:35
mgouldand we'll already have k*(n+1)*n/211:35
mgouldso test-length grows quadratically, not exponentially11:35
mgouldwhich is still not great, of course11:36
sambettsif each new class inherits the parent class then you double right?11:38
mgouldno, you add the size of the parent class11:39
mgouldbut you don't just have the parent class, you have the parent, grandparent, great-grandparent...11:39
mgouldso you have k + 2k + 3k + ... + nk tests in total11:39
sambettsbut both the parent class and the child class run so you run all the tests twice, but on different microversions11:40
mgouldand the next time, you add (n+1)k11:40
mgouldsambetts: yes, but that's not the whole test suite11:40
mgouldyou're also running tests on all the previous microversions11:41
mgould[tests on 1.0] + [tests on 1.1] + [tests on 1.2] + ... + [tests on 1.n]11:41
mgouldyou're adding [tests on 1.n+1], which is initially the same size as [tests on 1.n]11:41
mgouldif you were only testing v1.n and v1.n+1, you'd be doubling, but you're also testing earlier versions11:42
*** jaganathan has quit IRC11:44
sambettsso we double the whole suite each time11:44
sambettsbecause 1.n+2 == new tests + test from 1.n+1 and 1.n11:44
sambettsbut 1.n+1 runs and 1.n runs too11:45
mgouldOK, I think I see the confusion11:45
mgouldone minute...11:45
sambetts:/ just want to make sure we don't double the run time of our tests every time we add a microversion11:46
mgouldyep11:46
mgouldthat would indeed be really bad11:46
mgouldhttp://paste.openstack.org/show/601526/11:50
mgould^^^ worked example, adding 2 new tests each time11:51
mgouldor https://docs.google.com/spreadsheets/d/1lTLluDN7UYjwZh_9RklhXgGtmsUxXc6bWKrKBVKlEes/edit?usp=sharing11:56
mgouldgrowth ratio tends to 111:57
mgouldquadratic growth may still be bad enough that we want to avoid this, of course11:58
*** zhurong has quit IRC11:58
mgouldsambetts: makes sense?12:00
*** amotoki has joined #openstack-ironic12:01
* sambetts is just trying to work out what I'm missing ;) 12:02
mgouldtotal runtime should be proportional to column D12:03
mgouldyou're thinking about column C12:03
jmccarthyHmm I just noticed in bifrost there are some defaults/main.yml files that point to repos on git.openstack.org .. is there a handy way to point these all to internal mirrors (besides sed?)12:03
jmccarthydoc/source/offline-install.rst suggests there may be a way with ANSIBLE_GIT_URL ?12:03
*** xavierr has quit IRC12:05
TheJuliajmccarthy: I don't think we put anything like that in for all of the urls.  It is a great idea, I'm surprised we've not done it yet12:05
mgouldjmccarthy: run your own internal DNS server and shadow git.openstack.org? :-)12:05
TheJuliajmccarthy: global default of git_server or something like that, and then update all of the urls use "{{git_server}}/openstack/blah"12:06
TheJuliamgould: I kind of wonder how many of those are out there at this point....12:06
* TheJulia sits in the metal tube waiting for it to be time to take off12:07
mgouldone per OpenStack install, the cynical part of me suggests12:07
mgouldTheJulia: have a pleasant flight12:07
openstackgerritVasyl Saienko proposed openstack/ironic master: WIP/DNM: Testing multi-tenant multi-node grenade job  https://review.openstack.org/43695712:07
*** xavierr has joined #openstack-ironic12:07
TheJuliamgould: I would think one per every deployment that does massive customizations, so maybe 1 in 20?12:07
TheJuliaThat is just totally a gut feeling12:08
mgouldsounds plausible, certainly12:08
mgouldI'd suspect a little higher; probably some don't do massive customizations, but do want better locality and control over uptime12:09
sambettsmgould: I think my coffee starved brain gets it now :-P12:09
TheJuliajmccarthy: If you could open bugs for all of the issues you find with trying to setup an offline install, that would be awesome.  It has been a while since had someone actively provide feedback on that12:09
mgouldsambetts: \o/12:09
mgouldTheJulia, jmccarthy: I can't find anywhere in Bifrost that reads ANSIBLE_GIT_URL12:10
TheJuliamgould: I guess it all comes down to how savvy the deployers12:10
mgouldbut I don't know much about Ansible; maybe it's being read implicitly?12:10
mgouldTheJulia: yeah12:10
TheJuliaSo ANSIBLE_GIT_URL was for a git based install, we've since gone to pip based install12:10
mgouldaaaah, OK12:10
TheJuliaStale docs :(12:10
mgould:-(12:10
*** Nisha_Agarwal has quit IRC12:13
jmccarthyTheJulia, I'm not sure if it's just me .. in my case offline means use our internal mirror of repo vs github12:13
TheJuliaTime to turn off cellular.  This is basically a 6 hour flight, dtantsur, I should be online for the meeting, but if not I have two announcements in the agenda if I don't have data service in the air.12:13
dtantsurcool, thanks for update12:14
jmccarthyTheJulia I can raise a bug to that effect12:14
*** vmud213 has quit IRC12:14
TheJuliajmccarthy: Yeah, the template updates should be easy to make that happen, if I re-appear in the next hour, I'll switch over to my laptop and push up a patch that should do it for you.12:14
mat128morning12:16
* milan reads scrollback12:17
milanmorning mat128, sambetts, TheJulia! :)12:17
mgouldmorning mat12812:17
milansambetts, vdrok, mgould, thanks for the suggestions, I'll check; basically I just wanted to run the same case w/ multiple versions, but maybe it's nonsense12:18
mgouldmilan: OK, probably use ddt then12:18
mgouldsambetts' suggestion assumes you want to run all tests from previous versions under each new version12:19
milanmgould, right. I'd use this for functional tests12:20
* mgould nods12:21
mgouldFWIW, I've been told "use ddt" in a review, so was planning to propose a patch adding it soon12:22
mgouldiyswim12:22
jmccarthyI rasied a quick bug here: https://bugs.launchpad.net/bifrost/+bug/167034012:24
openstackLaunchpad bug 1670340 in bifrost "bifrost has hardcoded git_url in playbooks" [Undecided,New]12:24
*** mvpnitesh has joined #openstack-ironic12:26
mgouldjmccarthy: it looks like you can override those variables one-by-one: http://docs.ansible.com/ansible/playbooks_variables.html12:29
pas-hajmccarthy: what's the problem actually? do you want the root of url to be easily customizable? currently you can put all those '*_git_url' in a yaml file and just do ansible-playbook ... -e @myvars.yaml12:29
jmccarthypas-ha: Hmm ok I have an idea what you mean here alright, I can try that out12:31
*** jaganathan has joined #openstack-ironic12:32
jmccarthypas-ha: The main issue is by default bifrost fetches stuff from elsewhere, and I need to be able to run it such that it uses sources that are within our corporate setting already (like openstack repos we have mirrored for example) - I hit this same problem with the os_keystone_service.py module which is fetched from github12:33
pas-hajmccarthy: since 2.0 ansible has been more strict in defining the vars precedence rules, and currently the role defaults have the lowest priority, easily overridable by just about anything else. OTOH extra vars passed to ansible-playbook AFAIK always win (have the highest priority)12:34
mgouldjmccarthy: I think pas-ha's suggestion should work for you12:35
pas-hajmccarthy: the os_keystone_service fetching should be fixed, that I agree. but that is happening outside of playbooks, when env for bifrost installation itself is prepped.12:35
pas-haone advise would be to use ansible 2.2.x where you do not need to fetch it :)12:35
* pas-ha remembers we should remove that ANSIBLE_GIT_BRANCH or whatever it is from env-setup.sh...12:36
jmccarthypas-ha: I am working around the os_keystone_service module issue alright, altho I can't just bump to 2.2 (it seems to not work?)  https://review.openstack.org/#/c/440588/12:36
patchbotpatch 440588 - bifrost - Update ansible version in env-setup.sh12:36
*** gcb has joined #openstack-ironic12:37
jmccarthypas-ha: I'll give this a shot (it may take a while) "you can put all those '*_git_url' in a yaml file and just do ansible-playbook ... -e @myvars.yaml"12:38
openstackgerritHao Li proposed openstack/ironic-specs master: Make aggregates/availability zones support ironic node  https://review.openstack.org/44192412:38
pas-haoh, that one - this is a funny one, most probably related to Jinja2 version mismatch between latest 2.2.1 and OpenStack's global requirements12:38
mgouldjmccarthy: `git grep '_git_url:' | sort -u > my_mirror.yml`, open file, search-and-replace12:39
jmccarthymgould: Yea that is my other option which should do the trick also12:40
jmccarthyI guess what I'm doing is a variation of the 'offline' install12:41
*** lucasagomes is now known as lucas-hungry12:41
openstackgerritHao Li proposed openstack/ironic-specs master: Make aggregates/availability zones support ironic nodes  https://review.openstack.org/44192412:42
mgouldjmccarthy: find . -name main.yml | xargs grep -h _git_url:  | sed -e 's/^\s*//' | sort -u, rather12:42
jmccarthymgould: I'm with you :)12:42
* mgould recommends ag as a replacement for find | grep, BTW: https://geoff.greer.fm/ag/12:43
pas-hamgould: ++ for ag :)12:43
mgouldpas-ha: it's great, isn't it?12:43
pas-hayep, blazing fast and knows about git/backup files etc12:43
pas-hainstalling it everywhere I develop (and can install stuff) :)  https://github.com/pshchelo/dotfiles/blob/master/ansible/roles/devhost/vars/main.yaml#L3612:45
*** rbudden has joined #openstack-ironic12:45
*** amoralej is now known as amoralej|lunch12:46
*** vsaienko has joined #openstack-ironic12:48
*** dprince has joined #openstack-ironic12:50
*** hw_wutianwei has quit IRC12:54
*** bfournie has quit IRC12:56
*** moshele has quit IRC13:05
amaraoHello everyone. I have a silly question: why there is a pxe_drac, but there isn't a agent_drac driver? Is this some kind of big decision or it's just a fluke?13:06
*** venki has joined #openstack-ironic13:08
venkidid anyone got this error ?13:09
venki+./stack.sh:main:1059                      create_keystone_accounts +lib/keystone:create_keystone_accounts:383  local admin_project ++lib/keystone:create_keystone_accounts:384  openstack project show admin -f value -c id Discovering versions from the identity service failed when creating the password plugin. Attempting to determine version from URL. Could not determine a suitable URL for the plugin +lib/keystone:create_keyst13:09
venkisorry !13:09
venki++lib/keystone:create_keystone_accounts:384  openstack project show admin -f value -c id13:09
venkiDiscovering versions from the identity service failed when creating the password plugin. Attempting to determine version from URL. Could not determine a suitable URL for the plugin13:09
sambettsvenki: please use paste.openstack.org13:10
venkiokey sambetts. sorry for inconvenience13:10
sambettsNP :)13:11
venkihttp://paste.openstack.org/show/601543/13:11
*** jkilpatr has joined #openstack-ironic13:12
*** jaganathan has quit IRC13:12
*** g3ek has quit IRC13:13
sambettsvenki: it looks like something went wrong with keystone during devstack, I would try looking in the screen setup by devstack using "screen -x" and see if there are any errors in the logs there, otherwise try asking in #openstack-keystne13:14
sambetts#openstack-keystone13:14
openstackgerritHao Li proposed openstack/ironic-specs master: Make aggregates/availability zones support ironic nodes  https://review.openstack.org/44192413:15
*** g3ek has joined #openstack-ironic13:16
venkisambetts : screen has no errors... So I will ask this doubt in  #openstack-keystone .. Thanks for the help...13:22
dtantsurpas-ha, hah, similar thing: https://github.com/dtantsur/config/blob/master/configure.yml#L3213:22
*** vsaienk01 has quit IRC13:23
*** vsaienk0 has joined #openstack-ironic13:24
*** jaganathan has joined #openstack-ironic13:26
*** xavierr has quit IRC13:26
*** lambert has quit IRC13:27
*** bfournie has joined #openstack-ironic13:27
*** mnaser has left #openstack-ironic13:27
*** xavierr has joined #openstack-ironic13:28
*** venki has quit IRC13:34
*** xavierr has quit IRC13:34
*** ccamacho is now known as ccamacho|lunch13:35
openstackgerritHironori Shiina proposed openstack/ironic master: DNM/POC Workaround for boot from Cinder volumae  https://review.openstack.org/44194413:36
*** xavierr has joined #openstack-ironic13:37
*** jkilpatr has quit IRC13:39
mgouldamarao: I don't know the answer to that specifically, but our driver naming is currently less consistent than it appears13:41
*** jjohnson2 has joined #openstack-ironic13:41
mgouldwe're in the middle of a big refactor to composable drivers13:41
*** skazi has quit IRC13:41
amaraomgould, It's question more about 'iscsi' vs 'IPA' for dracs, not their names.13:41
*** xavierr has quit IRC13:41
mgouldah, OK13:41
mgoulddunno, then13:41
mgouldprobably just that nobody's written one13:42
amaraoActual question is 'why there is no drac + IPA'?13:42
amaraoOh, thanks.13:42
mgouldthere is, it's called pxe_drac :-)13:42
amaraopxe_drac uses iscsi, aren't it?13:42
mgouldthe old bash ramdisk is now deprecated13:42
jrollmorning y'all13:42
jrollamarao: it does, yes13:42
jrollwat, there's no agent_drac? O_o13:43
amaraojroll, and there are no drac driver which uses swift URLs directly?13:43
jrollamarao: guess not13:43
amaraoYep, there is no agent_drac13:43
jrollthis is news to me :P13:43
jrollI kind of just assumed it was there13:43
*** xavierr has joined #openstack-ironic13:44
*** racedo has quit IRC13:44
mgouldwhat is "agent_" meant to imply?13:45
*** baoli has joined #openstack-ironic13:47
*** vsaienko has quit IRC13:47
jrollmgould: the direct/http/swift/whatever you like to call it deploy method13:47
* jroll finds docs13:47
*** vsaienko has joined #openstack-ironic13:47
*** racedo has joined #openstack-ironic13:47
mgouldjroll: OK, thanks13:48
*** baoli has quit IRC13:48
*** baoli has joined #openstack-ironic13:49
amaraoShould I raise a bug?13:49
jrollugh where is this13:49
jrollamarao: can't hurt13:49
*** amoralej|lunch is now known as amoralej13:50
*** baoli has quit IRC13:50
*** baoli_ has joined #openstack-ironic13:50
jrollmgould: so this first diagram is the "pxe" method (which really means iscsi), and the second is the "agent" method (which really means direct/http) https://docs.openstack.org/developer/ironic/deploy/user-guide.html#example-1-pxe-boot-and-iscsi-deploy-process13:50
*** lucas-hungry is now known as lucasagomes13:51
mgouldjroll: thanks!13:51
*** baoli_ has quit IRC13:51
jrollnp :)13:51
openstackgerritDmitry Tantsur proposed openstack/ironic-specs master: Add priorities for the Pike cycle  https://review.openstack.org/43971013:52
*** skazi has joined #openstack-ironic13:54
dtantsuryay, another update!13:54
dtantsurmorning jroll13:54
jrolloh hey I should review that :P13:55
jrollmorning dtantsur :)13:55
openstackgerritSam Betts proposed openstack/ironic master: Make Ironic devstack simulation less dependent on OVS  https://review.openstack.org/44072113:57
openstackgerritSam Betts proposed openstack/ironic master: Rework Ironic devstack baremetal network simulation  https://review.openstack.org/39295913:57
openstackgerritSam Betts proposed openstack/ironic master: Handle multinode VXLAN tunnel inside ironic devstack plugin  https://review.openstack.org/44071813:57
openstackgerritSam Betts proposed openstack/ironic master: Remove extra bridge created for plumbing BM VMs  https://review.openstack.org/44071913:57
*** Nisha_Agarwal has joined #openstack-ironic13:58
openstackgerritDmitry Tantsur proposed openstack/ironic-specs master: Add priorities for the Pike cycle  https://review.openstack.org/43971013:58
*** baoli has joined #openstack-ironic13:59
*** sasha2 has joined #openstack-ironic14:03
openstackgerritM V P Nitesh proposed openstack/ironic master: Now displays maintenance and maintenance_reason  https://review.openstack.org/43302914:06
*** amotoki has quit IRC14:06
*** vsaienko has quit IRC14:06
*** amotoki has joined #openstack-ironic14:06
mvpniteshHi dtantsur , for this https://review.openstack.org/#/c/433029/ change Yuriy Zveryanskyy is saying "This is REST API change. At least microversion should be bumped, but I'm not sure that this change is needed (not state-related fields in states). You can simple use node-show via cli or any set of fields via API client." Should i change the microversion for that patch??14:08
patchbotpatch 433029 - ironic - Now displays maintenance and maintenance_reason14:08
dtantsurmvpnitesh, hi, sorry I don't have a context on this patch, nor any time to dive into it. Mind asking yuriyz_ directly please?14:09
jrollmvpnitesh: yes, (almost) all REST API changes must change the microversion14:10
jrollthis one should14:10
* jroll doesn't know if this change is needed or not either14:10
*** vsaienko has joined #openstack-ironic14:10
*** rloo has joined #openstack-ironic14:10
mvpniteshjroll: should i change it in "doc/source/dev/webapi.rst" of Ironic14:11
dtantsurvsaienk0, any plans to update https://review.openstack.org/#/c/382476/? it's still on the priorities list14:11
patchbotpatch 382476 - ironic - Tempest tests for portgroups14:11
jrollmvpnitesh: well, it's doc/source/dev/webapi-version-history.rst, but yes you should14:12
vsaienk0dtantsur: yeah, I'm waiting for https://review.openstack.org/#/c/440719 and will rework tests for portgroups14:12
patchbotpatch 440719 - ironic - Remove extra bridge created for plumbing BM VMs14:12
dtantsurk14:13
*** amotoki has quit IRC14:15
*** gcb has quit IRC14:15
*** gcb has joined #openstack-ironic14:16
*** Nisha_Agarwal has quit IRC14:17
jmccarthyIt is just me ? bifrost-ironic-install/defaults/main.yml:dib_git_url: https://git.openstack.org/openstack/diskimage-builder seems to be invalid ? I think some of the others are invalid too ?14:18
jmccarthyAside from the smiley face in the middle :)14:19
sambettsjmccarthy: looks valid to me http://git.openstack.org/cgit/openstack/diskimage-builder14:20
openstackgerritJulia Kreger proposed openstack/bifrost master: POC: Overriding the GIT server  https://review.openstack.org/44195814:20
sambettsjmccarthy: the cgit bit is for the webview BTW14:20
jmccarthyHmm yea I noticed some have cgit and some don't14:20
openstackgerritM V P Nitesh proposed openstack/ironic master: Now displays maintenance and maintenance_reason  https://review.openstack.org/43302914:20
jmccarthysambetts: Ok I'll adjust for cgit, I thought they would come up as is14:22
jmccarthyTheJulia: I thought you were in a plane flying someplace ?!14:22
TheJuliajmccarthy: Checkout the revision I just pushed, it m might make it super easy "-e git_server=https://local.git.server/"14:22
*** fragatin_ has joined #openstack-ironic14:23
TheJuliajmccarthy: Yes, and the inter tubes are satellite based.  Using review is... painful :(14:23
*** fragatina has quit IRC14:23
openstackgerritKyrylo Romanenko proposed openstack/python-ironicclient master: Add functional tests for VIFs with ironicclient commands  https://review.openstack.org/43088114:24
jmccarthyTheJulia: Dedication !14:25
*** jcoufal has joined #openstack-ironic14:25
*** makowals has quit IRC14:28
*** makowals has joined #openstack-ironic14:30
openstackgerritJulia Kreger proposed openstack/bifrost master: Make ironic database use more configurable  https://review.openstack.org/44061614:30
*** Goneri has joined #openstack-ironic14:34
jmccarthyTheJulia: It won't quite work as is - I have to map out the 'openstack/' but as part of the git_server variable - but I notice some have 'openstack/' and some have 'openstack-infra/' so changing how the variables maps in won't quite do it either hrm14:36
*** vsaienko has quit IRC14:36
TheJuliaI really hate airlines seats with how easy it is for someone to recline and snap a laptop :(14:37
*** vikrant has quit IRC14:38
TheJuliaOh.. because your mirroring them into a different namespace?14:38
jmccarthyTheJulia: I mean sort of like:14:38
jmccarthygit_server: "https://git.openstack.org/openstack/"14:38
jmccarthyhttps://git.openstack.org/openstack/ironic14:38
jmccarthy{{ git_server }}openstack-ironic14:38
TheJuliayeah, that should work, we would just need something to delineate openstack and -infra name spaces14:39
mgouldgit_openstack_server and git_infra_server?14:39
jmccarthyTheJulia mgould Yea something like that would do it !14:40
jmccarthyMy example there is mixed up but you get me I think14:40
TheJuliamaybe git_openstack_url git_openstack_infra_url, since it has url components.14:40
sambettsdevstack just has GIT_BASE=https://git.openstack.org14:40
TheJuliabut yeah, just naming14:40
sambettsthe openstack/* is the actual repo name14:40
jmccarthyYea and we have names openstack-$project pretty much like14:41
TheJuliaThat feels like a headache14:41
TheJulia:(14:41
* mgould is starting to think that "just override all the URLs one by one" is the least-bad answer here14:41
*** amotoki has joined #openstack-ironic14:41
jmccarthyWell I still have to try out the ansible precendence angle, that could be the way14:42
TheJuliamgould: that does... exist already, at least14:42
mgouldTheJulia: precisely14:42
mariojvgood morning ironic14:42
mgouldmorning mariojv14:42
mariojvjust a reminder, i may not be back from an errand in time for the meeting, but status report for rescue was up to date as of friday14:42
mgouldTheJulia: also, trying to guess exactly which bits of the URLs should be flexible is getting messy14:43
TheJuliajmccarthy: Let us know, I've spoken to a couple people with downstream mirroring of upstream repos recently, I should prod them to see how they are structured or if they are mirroring exactly the same14:43
TheJuliamgould: indeed14:43
mgouldDoes anyone know why we have "wait = yes" in our example xinetd.d/tftp file at https://docs.openstack.org/project-install-guide/baremetal/draft/setup-drivers.html#pxe-setup ?14:43
jmccarthyTheJulia: Will do ! I will do soon hopefully14:44
mgouldJayF: ^^^14:44
*** crushil has joined #openstack-ironic14:45
TheJuliamgould: I have a feeling there is a necessity with udp and tftp, at least for client compatibility to keep things mostly happy14:48
mgouldah, OK14:50
* mgould has found plenty of blogs and HOWTOS etc saying to use that line, but none explaining why14:51
*** zhs_ has quit IRC14:51
*** goneri_ has joined #openstack-ironic14:51
mgouldI think they're all cargo-culting it from each other14:51
*** makowals has quit IRC14:52
*** zhs_ has joined #openstack-ironic14:52
openstackgerritHironori Shiina proposed openstack/ironic master: DNM/POC Workaround for boot from Cinder volume  https://review.openstack.org/44194414:52
*** spartacloud has joined #openstack-ironic14:53
*** makowals has joined #openstack-ironic14:56
TheJuliamgould: likely :)14:57
*** gcb has quit IRC14:57
*** chlong__ has joined #openstack-ironic14:57
openstackgerritMerged openstack/bifrost master: playbooks: roles: bifrost-create-vm-nodes: Reboot VM if boot fails  https://review.openstack.org/43994614:58
*** mtanino has joined #openstack-ironic15:00
*** udesale has joined #openstack-ironic15:02
jmccarthyTheJulia mgould Ok yea that suggestion will do the trick, I will use this technique ( ansible-playbook ... -e @my-git-repos.yaml) - I'll update the bug15:03
mgouldTheJulia: I'm wondering if removing it would make TFTP scale better, because apparently it limits us to one connection at a time15:03
*** jjohnson2 has quit IRC15:03
*** jjohnson2 has joined #openstack-ironic15:04
jrolldtantsur: I have a meeting conflict today, have an announcement in agenda with details15:04
* jroll checks if he has any subteam updates, doubtful15:04
dtantsurk, hope my wifi stops hating me by that time15:04
jroll:P15:05
openstackgerritMerged openstack/bifrost master: Fix README.rst  https://review.openstack.org/44177815:05
TheJuliamgould: I feel like that discussion requires the RFCs to be at the fingertips along with a glass of scotch.15:05
jmccarthyThank you for all the help ! :)15:05
jrolldtantsur: "talked at PTG about anything missing, path to getting vendor hw types, etc. jroll will update with what is left to do." <- where should we document this? a new RFE for pike?15:05
*** jaganathan has quit IRC15:06
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Use virtualbmc for virtual BM nodes  https://review.openstack.org/42737615:06
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Use JSON baremetal data in testing  https://review.openstack.org/43089815:06
openstackgerritPavlo Shchelokovskyy proposed openstack/bifrost master: Use Ansible to create test VMs  https://review.openstack.org/42707115:06
dtantsurjroll, I guess it won't hurt15:07
rloohi and Happy Monday morning all! pas-ha, jroll, dtantsur, high-above-the-skies TheJulia, jmccarthy, mgould15:07
TheJuliapas-ha: sorry :(15:07
jrolldtantsur: yeah, just having trouble thinking where else that could be written down - another alternative is an email and take each piece from there15:07
pas-haTheJulia: nah, wasn't that hard rebase :)15:08
dtantsurmorning rloo15:08
TheJuliapas-ha: Only reason I even clicked the button on the other revision :)15:08
*** spartacloud is now known as zackf15:08
jrollmorning rloo15:08
rloojroll: i know we moved the spec (I'm guessing you're talking about driver comp. is the rfe still open? if so, put wht's left to do there?15:08
jrollrloo: I'd rather close the RFE and track the remaining things separately15:09
rloojroll: or open separate bugs for them?15:09
rloojroll: 'they' did that wrt metrics & individual vendor drivers15:09
jrollrloo: yeah, that was my thought15:09
*** gcb has joined #openstack-ironic15:10
rloojroll: ok, that makes sense.15:10
*** hurricanerix has quit IRC15:13
* milan relocates15:13
*** milan has quit IRC15:14
*** makowals has quit IRC15:14
*** venki has joined #openstack-ironic15:14
* TheJulia steps away... as impossible as that is on an aircraft15:15
*** baoli has quit IRC15:15
rloovdrok: did you get an answer to your question? I don't see any response to it: https://review.openstack.org/#/c/440292/15:17
patchbotpatch 440292 - ironic - Move eventlet monkey patch code (MERGED)15:17
*** makowals has joined #openstack-ironic15:18
*** david-lyle has quit IRC15:20
*** ccamacho|lunch is now known as ccamacho15:21
mgouldOK, looks like the "wait = yes" line was present without explanation in the initial Git revision of xinetd's sample.conf 14 years ago, and has been cargo-culted onwards ever since: https://github.com/xinetd-org/xinetd/commit/f21e19e5a4d1f26dbdecd341f9a9f26f4539d0c5#diff-f820c17bccb75f03dbe186c9a2b416d8R14215:25
*** Nisha_Agarwal has joined #openstack-ironic15:26
vdrokrloo: morning, nah, that was a wrong testing patch, that actually works perfectly fine :) commented on that on the subsequent patch15:26
vdrokand good morning TheJulia and jroll15:26
rloohey, is/was there issues at the gate. I see rechecks, eg: 'recheck over ram limit' https://review.openstack.org/#/c/441491/15:27
patchbotpatch 441491 - ironic - Deprecate support for glance v115:27
rloovdrok: oh, thx!15:27
jroll"over ram limit" is a global thing15:27
rloojroll: global problem?15:27
jrollrloo: or rather, I would guess it is this bug https://bugs.launchpad.net/nova/+bug/164391115:28
openstackLaunchpad bug 1643911 in OpenStack Compute (nova) "libvirt randomly crashes on xenial nodes with "*** Error in `/usr/sbin/libvirtd': malloc(): memory corruption:"" [Medium,Confirmed]15:28
jrollrloo: yeah, it isn't just us15:28
rloojroll: should we note those down in our etherpad?15:28
openstackgerritMerged openstack/ironic-python-agent master: Python 3.4 (py34) is removed from tox envlist  https://review.openstack.org/44128215:28
jrollrloo: I don't know, I saw it in the mailing list in a few different threads over the last month15:29
mgouldwww.xinetd.org is down15:29
rloojroll: it is affecting us. i'll mention it. otherwise, we may get people wondering why it sometimes fails.15:29
jrollrloo: sure15:30
*** Nisha_Agarwal has quit IRC15:30
*** Nisha_Agarwal has joined #openstack-ironic15:30
mgouldthe xinetd maintainer's website: http://synack.net/15:30
openstackgerritM V P Nitesh proposed openstack/ironic master: ipmi_address is validated  https://review.openstack.org/43694515:33
*** hamzy_ is now known as hamzy15:33
*** amotoki has quit IRC15:33
mgould"It should be noted that udp/dgram services normally expect the value to be yes since udp is not connection oriented, while tcp/stream servers normally expect the value to be no.15:35
patchbotError: No closing quotation15:35
mgould"15:35
patchbotError: No closing quotation15:35
mgouldmake your mind up, patchbot15:35
*** mvpnitesh has quit IRC15:35
mgouldI guess if tftp is a deliberately stupid protocol, it makes sense that a tftp server probably wouldn't have the smarts to handle multiplexing connections itself15:36
mgouldthat might also explain why it scales so horribly15:37
*** hshiina has quit IRC15:37
rlooyay, our release notes have been fixed. 'current' really shows master now! https://docs.openstack.org/releasenotes/ironic/unreleased.html15:38
mgouldrloo: w00t!15:38
TheJulia\o/15:39
*** venki has quit IRC15:40
rlooexcept.. argh. why does current - upgrade show that iBoot etc are removed. they were remove in ocata :-(15:41
lucasagomesrloo, nice!15:41
*** Nisha_Agarwal has quit IRC15:41
*** Nisha_Agarwal has joined #openstack-ironic15:41
*** makowals_ has joined #openstack-ironic15:42
mgouldrloo: was the removal noted in the wrong place?15:43
*** makowals has quit IRC15:43
*** e0ne has quit IRC15:44
rloomgould: dunno, it wasn't something that i wanted to look into. but i'm fairly sure it was removed in ocata.15:44
*** Syed__ has joined #openstack-ironic15:44
*** mgoddard has quit IRC15:47
rloomgould: yeah, we removed those drivers on Dec 16. Sigh.15:47
*** Nisha_Agarwal has quit IRC15:47
*** Nisha_Agarwal has joined #openstack-ironic15:48
*** mgoddard has joined #openstack-ironic15:49
*** rbartal has quit IRC15:51
*** amarao has quit IRC15:53
*** rpioso has joined #openstack-ironic15:55
mgouldrloo: curses15:55
rloomgould: i think i know why but i don't know how to fix. that release note had an issue with it and it was updated after ocata branch was created.15:56
mgouldbah15:56
TheJuliaugh, that would do it15:57
JayFdid that release note patch get backported?15:57
JayFif we backport it I suspect it would fix the issue then, right?15:57
rlooTheJulia: do you know how to fix? or should I ask dougH?15:57
rlooJayF: yeah, it was backported.15:57
JayFwell crap. I'd ask Doug, yeah.15:57
TheJuliaI'm not sure there is a way to fix15:57
rlooJayF: well, the release notes for Ocata show it. But I think because of the date, it (something) thinks it should also be in master.15:58
JayFif it was a less important release note I'd just suggest a *shrug* as the solution15:58
rlooJayF: oh, thx for asking doug. fwiw, i opened a bug for it: https://bugs.launchpad.net/ironic/+bug/167040115:58
openstackLaunchpad bug 1670401 in Ironic "'Unreleased' release notes show iBoot driver removed" [Undecided,New]15:58
JayFrloo: oh, I'll be glad to ask Doug and handle it, that wasn't intentionally a volunteer but I don't mind15:58
JayFlol15:58
rlooJayF: perhaps the right thing to have done was NOT to update/backport from master, but only to update in stable/ocata15:59
rlooJayF: the solution might be to delete that file (or clear the contents) from master... thx for volunteering :)15:59
NobodyCamGood Morning Ironic'ers :)15:59
rloomorning NobodyCam!16:00
NobodyCamMorning rloo :)16:00
JayFrloo: asked him in #openstack-release16:00
rloothx JayF. let me know how it goes (or update that bug)!16:00
jlvillalGood morning Ironic :)16:00
NobodyCammorning jlvillal :)16:01
jlvillalNobodyCam, :)16:01
*** moshele has joined #openstack-ironic16:02
*** yuan has joined #openstack-ironic16:03
*** udesale has quit IRC16:05
*** gcb has quit IRC16:08
openstackgerritJay Faulkner proposed openstack/ironic master: Revert "reno 'upgrades' should be 'upgrade'"  https://review.openstack.org/44201016:08
JayF^ DNM, science16:08
*** rcernin has quit IRC16:08
TheJulias/science/for\ science!/ :)16:09
*** crushil has quit IRC16:09
mgouldmorning NobodyCam!16:10
NobodyCammorning mgould :)16:10
*** gcb has joined #openstack-ironic16:10
rloovsaienk0, lucasagomes: wrt gate failures for https://review.openstack.org/#/c/441271/16:13
patchbotpatch 441271 - ironic - Clean up eventlet monkey patch comment and reno16:13
rloovsaienk0, lucasagomes: I see there is a bug https://bugs.launchpad.net/nova/+bug/167031916:13
openstackLaunchpad bug 1670319 in OpenStack Compute (nova) "Reschedule of failed instance doesn't happening when scheduler placed two instances to the same ironic node" [Undecided,New]16:13
* lucasagomes looks16:13
*** e0ne has joined #openstack-ironic16:13
rloovsaienk0, lucasagomes: should that be something to note in our etherpad? do either of you know whether these are sporatic failures?16:13
lucasagomesnot sure if it's sporatic, will need to investigate16:14
* lucasagomes is looking at the bug report 16:15
*** baoli has joined #openstack-ironic16:16
*** david-lyle has joined #openstack-ironic16:17
openstackgerritOksana Voshchana proposed openstack/ironic master: Adds wsgi to ironic-api in devstack  https://review.openstack.org/43085116:18
openstackgerritJay Faulkner proposed openstack/ironic master: Delete release note to fix build  https://review.openstack.org/44201816:19
*** baoli has quit IRC16:20
*** david-lyle has quit IRC16:22
*** crushil has joined #openstack-ironic16:24
*** rpioso has quit IRC16:28
vdrokgood morning JayF NobodyCam and jlvillal !16:30
jlvillalvdrok, o/16:30
*** rpioso has joined #openstack-ironic16:30
JayFo/ gm16:30
NobodyCamGood Morning vdrok :)16:30
TheJuliaGood morning16:30
NobodyCam:)16:31
dtantsurthe 3rd part of the PTG summary for early readers: http://dtantsur.github.io/posts/ironic-ptg-atlanta-2017-3.html16:31
dtantsurmorning NobodyCam, jlvillal, JayF16:32
jlvillaldtantsur, o/ :)16:32
NobodyCammorning dtantsur :)16:32
JayFdtantsur: it's been suggested in the past that blog posts about PTG summaries be cross-posted (with full text) to the mailing list, so it's archived 'forever'16:32
*** ricardoas has joined #openstack-ironic16:32
dtantsurJayF, I'll dump all four parts to the ML once I'm done16:33
dtantsurwith all four links as well16:33
dtantsurstay tuned :)16:33
JayFawesome16:33
*** amotoki has joined #openstack-ironic16:34
NobodyCam:( Deprecating the standalone ironic CLI :(16:34
dtantsuryou will be missing it? ;)16:34
NobodyCamyea, I end up with just ironic in many cases :(16:35
NobodyCamOsc is often more then I need16:36
dtantsurI hear you16:37
dtantsurwell, it's not without some unique benefits, like -f value -c <FIELD>16:38
JayFNobodyCam: fwiw, we mentioned in that session that standalone users need better, explicit support before ironic cli dies16:38
dtantsuryeah, which is something we should bring to OSC folks16:38
JayFNobodyCam: not sure how much traction that idea got, since it sorta got streamrolled with 'deprecate asap so we can do half the work for client changes'16:39
dtantsurI don't think it's hard to have --auth-type none16:39
vsaienk0rloo: I've added info about multinode job random failures to etherpad https://etherpad.openstack.org/p/IronicCI16:39
*** amotoki has quit IRC16:39
NobodyCam:) yea, I will need to see how that is going to impact my use cases.16:40
rloovsaienk0: thx! ^^ lucasagomes16:41
lucasagomesoh thanks vsaienk016:41
*** e0ne has quit IRC16:41
openstackgerritJay Faulkner proposed openstack/ironic master: Revert "reno 'upgrades' should be 'upgrade'"  https://review.openstack.org/44201016:41
*** Nisha_ has joined #openstack-ironic16:42
*** Nisha_Agarwal has quit IRC16:42
*** ricardoas1 has joined #openstack-ironic16:43
*** ricardoas has quit IRC16:43
vsaienk0rloo, dtantsur: we can decrease concurrency to 1 and the problem will gone, but not sure it is a proper way, better to push nova to resolve issue16:43
*** ricardoas1 has quit IRC16:43
*** ricardoas has joined #openstack-ironic16:44
dtantsurI think the latter is better..16:44
*** e0ne has joined #openstack-ironic16:45
rloovsaienk0: i'm going to add that multinode failure to our main etherpad too...16:46
vsaienk0rloo: ok16:46
*** sayantan_ has joined #openstack-ironic16:47
JayFrloo: this fixes the reno build in the way reccomended by doug: https://review.openstack.org/#/c/442018/ -- he's going to look into making reno complain in the future when a release note in broken in the way that one originally way (i.e. a misspelled section header means no recognized release notes in a file, reno should error in that case)16:48
patchbotpatch 442018 - ironic - Delete release note to fix build16:48
JayFrloo: related reno bug to that incoming fix I mentioned: https://bugs.launchpad.net/reno/+bug/167042116:48
openstackLaunchpad bug 1670421 in reno "complain when a notes file contains unknown sections" [Undecided,New]16:48
*** hshiina has joined #openstack-ironic16:48
joannamorning everyone :)16:49
NobodyCamgood morning joanna16:49
rlooJayF: thx!16:49
vdrokmorning joanna16:49
joannamorning NobodyCam, vdrok, rloo, JayF, vsaienk016:49
JayFo/16:49
*** moshele has quit IRC16:49
NobodyCam:)16:49
rlooJayF: I wonder about deleting it, in case someone is grep'ing for it but i think that's probably the best thing. (OR leaving the file and clearing the contents.)16:50
*** e0ne has quit IRC16:50
openstackgerritOksana Voshchana proposed openstack/ironic master: DNM: TEST ironic with wsgi  https://review.openstack.org/43729916:50
rlooJayF: if we could add a comment in that file, that might be best?16:50
JayFrloo: clearing the contents will break our build when that reno bug is fixed16:50
JayFrloo: that's why doug suggested a complete delete -- so later reno can enforce "if you have a release notes file, it needs to contain a valid note in some section"16:50
jlvillalvsaienk0, I wonder if in the future (not now) that the grenade multi-node multi-tenant job could be done with Ironic stand-alone. To save time.16:51
rlooJayF: Oh. I didn't look at the bug, would be good to check that all sections are valid.16:51
vsaienk0jlvillal: we need to test upgrades among projects anyway, but of course we may setup ironic only job in future16:52
openstackgerritOksana Voshchana proposed openstack/ironic master: Adds wsgi to ironic-api in devstack  https://review.openstack.org/43085116:53
jlvillalvsaienk0, I was just seeing how it is taking 2 hours and 20 minutes or so to finish...16:53
vsaienk0jlvillal: because we rung smoke tests 2 times with 1 concurrency16:54
jlvillalvsaienk0, FYI: Maybe update commit message? https://review.openstack.org/#/c/440303/5//COMMIT_MSG16:54
patchbotpatch 440303 - openstack-dev/devstack - Ignore second enable_plugin request16:54
jrolljlvillal: well, we want to make sure upgrading doesn't make nova explode, etc, so probably need to keep it that way16:54
rloovsaienk0, jlvillal: before i forget. the ocata priorities has you two listed for rolling upgrades. are you (either of you) good with keeping tabs on the feature/code? https://review.openstack.org/43971016:55
patchbotpatch 439710 - ironic-specs - Add priorities for the Pike cycle16:55
jlvillaljroll, vsaienk0: Yeah. Just a long job, but I think that happens for us in Ironic with simulating bare-metal in VMs16:55
dtantsurmeeting in 5 mins16:55
jrollyeah16:55
*** baha has joined #openstack-ironic16:57
jlvillalJayF, Did we figure out issue with stable/newton broken?16:58
jlvillalJayF, It sort of reminds me of error that was happening when devstack-gate changes were being made to the local.conf file creation. But that was all fixed and should be for all branches.16:58
jlvillalJayF, Oh yeah look at last line of: http://logs.openstack.org/85/437385/1/check/gate-tempest-dsvm-ironic-ipa-wholedisk-bios-agent_ipmitool-coreos-src-ubuntu-xenial/d291396/logs/local.conf.txt.gz16:59
jlvillalThat doesn't look right...16:59
openstackgerritOksana Voshchana proposed openstack/ironic master: Adds wsgi to ironic-api in devstack  https://review.openstack.org/43085117:00
JayFjlvillal: I didn't dig into it at all other than finding it and pointing it out.17:01
jlvillalJayF, I think it is a very old build. Need to do a recheck17:02
openstackgerritOksana Voshchana proposed openstack/ironic master: Adds wsgi to ironic-api in devstack  https://review.openstack.org/43085117:04
*** mgagne_ has quit IRC17:05
*** mgagne_ has joined #openstack-ironic17:05
*** jaypipes has quit IRC17:05
*** racedo has quit IRC17:05
* jlvillal likes that #undo shows the text of the item and pats himself on the back for it17:08
*** tesseract has quit IRC17:10
*** lee4 has joined #openstack-ironic17:11
openstackgerritOksana Voshchana proposed openstack/ironic master: Adds wsgi to ironic-api in devstack  https://review.openstack.org/43085117:12
openstackgerritOksana Voshchana proposed openstack/ironic master: DNM: TEST ironic with wsgi  https://review.openstack.org/43729917:12
*** sayantan_ has quit IRC17:15
*** sayantan_ has joined #openstack-ironic17:15
*** david-lyle has joined #openstack-ironic17:18
*** yuan has quit IRC17:18
aNuposicJayF: Hi Jay, similar to https://review.openstack.org/#/c/423521/14, .network file can be created for oem/tinyipa which will have static ip, subnets, routers and stuff. And same logic can be used to copy it under systemd as done in finalize_rescue.sh17:21
patchbotpatch 423521 - ironic-python-agent - Rescue extension for CoreOS with DHCP tenant networks17:21
jlvillalvsaienk0, FYI: I did a 'recheck' on the devstack-gate patch17:21
vsaienk0thanks17:21
*** venki has joined #openstack-ironic17:22
aNuposicJay: can't it be done that way?17:22
JayFaNuposic: that's the exact idea we were going to use, aiui17:22
JayFaNuposic: glean is a tool that can read that network file ( the network json file ) and write out those configs17:23
JayFaNuposic: it just doesn't support systemd style configs17:23
*** venki_ has joined #openstack-ironic17:23
aNuposicJayF: yes i see that glean right now has support for other distros like ubuntu, coeros, but not for tiny core linux17:23
*** david-lyle has quit IRC17:24
JayFaNuposic: you mean centos, right?17:24
*** venki has quit IRC17:24
aNuposicJayF: but if we follow the approach as mentioned then there is no need to use glean?17:24
JayFaNuposic: it doesn't support coreos today17:24
aNuposicJayF: oops ya centos17:24
JayFaNuposic: there's no way to avoid using some tool to translate "basic network information" into "network configs"17:24
*** dsneddon has quit IRC17:25
JayFaNuposic: and doing anything other than using the network json standard in the configdrive means we're inventing a brand new mechanism to do it all, instead of using the tools that already exist and handle 90% of the problem17:25
*** vinsh_ has quit IRC17:25
aNuposicJayF: but this patch https://review.openstack.org/#/c/423521/14 i believe do not use glean?17:26
patchbotpatch 423521 - ironic-python-agent - Rescue extension for CoreOS with DHCP tenant networks17:26
JayFaNuposic: that's a, uh, workaround -- basically coreos by default DHCPs all interfaces17:26
JayFaNuposic: so by assuming dhcp client networks, we hopped over the entire problem space you're working to help out in now :)17:27
*** rcernin has joined #openstack-ironic17:27
jlvillalvdrok, Is there a bug about vbmc and Python 3?17:27
jlvillalhttp://logs.openstack.org/57/436957/15/experimental/gate-tempest-dsvm-ironic-ipa-partition-pxe_ipmitool-tinyipa-python3-ubuntu-xenial-nv/781d77a/logs/devstacklog.txt.gz#_2017-03-06_12_45_43_81817:27
venki_I'm trying devstack with ironic using this local.conf file, http://paste.openstack.org/show/601620/ . But once stack completed, i am unable to find ironic.conf file17:28
venki_Is anything i'm missing ?17:28
vdrokjlvillal: have not seen it17:28
jlvillalvdrok, Thanks17:28
aNuposicJayF: I see. So that means same thing won't be possible in case of tinyipa and for tinyipa it has to use some kind of tool and that's why glean. Have i got it clear?17:28
jlvillallucasagomes, http://logs.openstack.org/57/436957/15/experimental/gate-tempest-dsvm-ironic-ipa-partition-pxe_ipmitool-tinyipa-python3-ubuntu-xenial-nv/781d77a/logs/devstacklog.txt.gz#_2017-03-06_12_45_43_818 FYI17:29
jlvillallucasagomes, Python 3 job17:29
JayFaNuposic: for the case of *any static networking* we need a tool like that17:29
lucasagomesjlvillal, a-ha, mind opening a bug against vbmc ?17:29
aNuposicJayF: got it17:30
jlvillallucasagomes, On github?17:30
lucasagomesjlvillal, I can look into it tomorrow17:30
JayFaNuposic: all the devstack stuff uses DHCP auto configuration, but probably <50% of ironic users use it, so before we have the static stuff working it's not a very useful feature17:30
*** david-lyle has joined #openstack-ironic17:30
lucasagomesjlvillal, no here at launchpad, it's a openstack project :D17:30
lucasagomesjlvillal, https://bugs.launchpad.net/virtualbmc17:30
jlvillallucasagomes, Oh I thought it was: https://github.com/umago/virtualbmc/17:30
lucasagomesjlvillal, oh that's how it started17:30
lucasagomesbut I've moved onto openstack/virtualbmc later on17:30
jlvillallucasagomes, You might want to upload a patch killing it there :)17:31
lucasagomesjlvillal, btw, it's a project under the ironic umbrealla, so you and all ironic cores can +2+a patches for it as well17:31
jlvillalAnd directing people to the OpenStack one17:31
lucasagomesjlvillal, yeah, I have to good clal17:31
jlvillallucasagomes, Thanks17:31
lucasagomescall*17:31
lucasagomesI will do that (same for sushy)17:31
jlvillallucasagomes, https://bugs.launchpad.net/virtualbmc/+bug/167044117:33
openstackLaunchpad bug 1670441 in VirtualBMC "Error when running under Python 3" [High,New]17:33
aNuposicJayF: ok, so first step first i will add support for tinyipa in glean and then use that tool in ironic-python-agent17:33
lucasagomesjlvillal, o/ thanks!17:33
lucasagomesadded to my list17:33
JayFaNuposic: I'd suggest just trying the configs that the centos/fedora support spits out17:33
JayFaNuposic: from your email it looks like those might just work17:33
*** Nisha_ has quit IRC17:40
*** dsneddon has joined #openstack-ironic17:41
*** vsaienko has joined #openstack-ironic17:42
*** moshele has joined #openstack-ironic17:42
*** baha has quit IRC17:44
*** athomas has quit IRC17:47
openstackgerritJohn L. Villalovos proposed openstack/virtualbmc master: Fix TypeError when writing config file in Python 3  https://review.openstack.org/44209517:50
jlvillallucasagomes, FYI: https://review.openstack.org/#/c/442095/17:50
patchbotpatch 442095 - virtualbmc - Fix TypeError when writing config file in Python 317:50
lucasagomesjlvillal, oh fixed that already ? 1 sec I will look after the ironic meeting17:54
jlvillallucasagomes, Thanks :)17:54
*** ralonsoh has quit IRC17:55
*** beekneemech is now known as bnemec17:56
lucasagomesjlvillal, one-liner is easy to review, added a comment17:58
*** jaosorior has quit IRC17:58
jlvillallucasagomes, I have to fix unit tests...17:58
lucasagomesright on :D17:59
lucasagomesI will keep an eye on the patch17:59
jlvillallucasagomes, Hopefully in a few minutes a new one17:59
lucasagomes++17:59
NobodyCamgah totally missed the meeting :(18:00
jrollJayF: in the past, this list was denoting the lead for the project management junk for a feature, not the development or design18:01
*** mgoddard has quit IRC18:01
rloojroll: maybe we should update that page, enhance 'status' to be 'project management-related tasks'?18:02
lucasagomesNobodyCam, you've been assigned to all the high prio stuff :D18:02
openstackgerritJohn L. Villalovos proposed openstack/virtualbmc master: Fix TypeError when writing config file in Python 3  https://review.openstack.org/44209518:02
jlvillallucasagomes, ^^18:02
NobodyCamieee LOL!!!!18:02
jrollrloo: do I want to know how else the word status might be interpreted?18:02
JayFjroll: still the same thing applies: we should try to involve any contributor who is willing to follow our standards and work on something upstream18:02
rloojroll: we know, others might not if JayF is right18:03
JayFjroll: limitiing that sort of work to folks who are already cores just limits the scale-up of new cores18:03
jrollJayF: if there are people that we trust to track major features, they should almost just be core18:03
JayFeven if it's just status, limitiing it to cores is dangerous for ^ that reason18:03
jrollyou may be right18:03
JayFresponsibility should be given out in little chunks, not in one big chunk via gerrit acl18:03
JayFI don't have a better solution that can be clear cut, which is why I said in the meeting to move on from it18:04
JayFbut we do need to take to heart and consider how this stuff looks to new or up-and-coming contributors, and see if it feels exclusive18:04
*** wanyen has joined #openstack-ironic18:04
dtantsuryeah, if we allow non-core people there (which I'd love to as well), I don't understand how to choose between them for certain popular features18:04
JayFI know before I was core, it felt /very/ exclusive to me with some of this stuff.18:04
jrollI guess I don't care much, I'm just saying we've been doing it this way, please prove it's broken rather than telling me we might hurt someone's feelings18:04
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: Serve boot configs from Ironic API  https://review.openstack.org/36943818:05
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: Implement Driver API for dynamic boot config  https://review.openstack.org/39439818:05
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: Add RPC API to get the boot config  https://review.openstack.org/39439918:05
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: Allow not writing pxe configs to disk  https://review.openstack.org/39439718:05
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: Convert iPXE boot config to proper Jinja template  https://review.openstack.org/39346218:05
jrollif we change it, and seem to slow down, I'll just say I told us so18:05
vgadirajI have no issues per se if we limit the primary contacts to just cores.18:05
JayFjroll: the thing I worry about, frankly, is looking at some of the projects that have dwindling numbers of cores because they didn't do a good job of sheparding new contributors to that level18:05
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: DNM test enabled dynamic iPXE  https://review.openstack.org/39194318:05
JayFjroll: I don't want that to happen to ironic, and being more inclusive of non-cores in all aspects, including here, is one strategy for avoiding that outcome.18:06
rlooJayF: what do you propose then, for this particular list of priorities?18:06
soliosgIMHO, I'm more with jroll's idea. We should aim to simplify and list only one or two main contacts. A name on that list kind of assigns the responsibility for the task, but not _credit_ for the actual work (like git log and stackalytics do)18:06
rlooJayF: I mean, who would you put down next to these priorities?18:06
vgadirajI  initially requested my name to be added because I saw non-cores on the list and I have been working on rolling upgrades and implementing multinode w/grenade since October.18:06
jlvillalpas-ha, JayF, FYI: I think stable/newton is working again. Every test passed on the recheck. https://review.openstack.org/#/c/437385/18:07
patchbotpatch 437385 - ironic-python-agent (stable/newton) - Support psutil 5.0.118:07
* sambetts -> home 18:07
*** sambetts is now known as sambetts|afk18:07
JayFrloo: If we kept the list 100% the same, but didn't say they were "core" contacts, probably most of my concern would be bypassed18:07
jrollidk how much more I want to discuss this in this channel, honestly18:07
sambetts|afkNight all18:07
rloojlvillal: yay!18:07
jlvillal:)18:07
soliosgBut either way we go, looks good to me18:07
JayFrloo: but it doesn't matter that much. It just bugs me, but I don't have anything concrete that's better :/18:07
jrollbut to dtantsur's point, I don't want 6 names to ping to ask a question about a thing18:07
jrollI want one18:07
rlooJayF: i'm good with that. Please propose.18:07
jrollmaybe two18:07
rlooJayF: and what jroll said. maybe we should explicitly mention some number, 2, 3, 4 max people?18:08
wanyeniLO driver team plan to contribute to RedFish driver and boot from cinder volume so we would like to be part of the decsion making18:08
dtantsurJayF, I'm fine with rephrasing, but not with just dropping "core". otherwise people don't understand why they can't be there18:08
jrollrloo: but how will we choose, for the big features? someone is going to get left out!18:08
rlooJayF: i have to say, I'm glad you brought it up, but I wish you had brought it up eg 2 cycles ago :)18:08
*** chlong__ has quit IRC18:08
jrollrloo: somewhat sarcasm there18:08
rloojroll: let the folks themselves choose18:08
rloojroll: presumably they are working together...18:08
dtantsurwanyen, awesome! please check the respective changes (patches for BFV and the spec for redfish)18:08
lucasagomeswanyen, that's great! We still current working on the spec, comments are appreciated there18:08
JayFrloo: well, I learn things over time :) trying to make it better, and making sure the concerns are planted so maybe someone smarter than me will stumble upon a solution in the shower one morning ;)18:09
rloojroll: and if they can't choose, or no one is coordinating, then ... that's what we need to know18:09
wanyenyes. we will.  Nisha already workingon redfish inspect interface18:09
vdrokgood night everyone!18:10
rloodtantsur: i think we want to open it up so that a non-core could be responsible for a feature priority.18:10
*** jmccarthy has left #openstack-ironic18:10
dtantsurg'night vdrok18:10
mjturekseeya vdrok18:10
rloonight vdrok18:10
lucasagomeswanyen, cool, that's not part of the initial spec but I believe it can come after as a RFE18:10
dtantsurrloo, I'm still looking forward to a practical way to do that18:10
jrollrloo: looking back, the previous discussion was about the count, not whether it was cores or not (we had non-cores there in mitaka)18:10
lucasagomesthe spec is the bare-minimum for a driver FWIW18:10
jlvillalpas-ha, Is stable/mitaka still broken?18:10
jrollrloo: context, https://review.openstack.org/#/c/241223/2/priorities/mitaka-priorities.rst@3718:10
* mgould -> home; good night!18:10
patchbotpatch 241223 - ironic-specs - Add Mitaka project priorities (MERGED)18:10
*** mgould is now known as mgould|afk18:10
rloojroll: yeah, i know. am just trying to addres JayF's concerns.18:10
jrollnewton was also not core-exclusive18:10
jrollhuh, I guess I take back what I said18:11
JayFso it's just a wording thing in this document. Not a policy/historical thing. That makes it an even smaller problem.18:11
jrollocata only had cores but I don't know if that was a thing18:11
lucasagomesjlvillal, btw, I'm almost calling it a day but I wanted to +2 that vbmc patch18:11
dtantsurJayF, I'm open to better wording, please propose18:11
lucasagomesjlvillal, you did test it with the py3 job locally or something ?18:11
jrolldtantsur: something wrong with the Ocata wording?18:11
jlvillallucasagomes, No. I need the gate to do that.18:12
dtantsurjroll, everything :) people assumed they all have to be added18:12
*** xavierr has quit IRC18:12
rlooJay, jroll, dtantsur: so if the concern was the list of many names, let's cap it and explicitly mention that cap. 3?18:12
dtantsurthat's how I ended up with the current proposal18:12
jrolldtantsur: I think "primary contacts" handles that18:12
lucasagomesjlvillal, oh right, ok so I will leave the +1 for now tomorrow I will re-look18:12
lucasagomesthanks for the fix btw18:12
dtantsurjroll, it proved not ot handle it already18:12
jrolldtantsur: we just need to continue teaching people the limit is 2-3 people :)18:12
jlvillallucasagomes, If I do a depends-on will it get pulled in or is a PyPi dependency?18:12
JayFI put a comment on the spec; basically suggesting we drop "core" from the wording, and make a hard limit on numcontacts/feature18:12
dtantsurhow do I pick these 3 people out of 2 cores and 4 non-cores?18:12
lucasagomesjlvillal, depends-on pulls it yeah18:12
jlvillallucasagomes, If I made a test openstack/ironic patch.18:13
dtantsurJayF, please amend it with a practical way to do that18:13
rloodtantsur: you don't have to pick. you ask all those people to pick.18:13
jrolldtantsur: pick the 2 cores or the two people working most on it :P18:13
jrolldtantsur: I'll leave words in a comment18:13
jlvillallucasagomes, Okay. I will try that :)18:13
lucasagomesjlvillal, oh wait... hmmm hang on, not sure if we install vbmc from sourc ein devstack18:13
jlvillallucasagomes, I don't know either18:13
JayFdtantsur: if the output is all cores, that's OK, as long as it's not all cores just because they're cores18:13
* dtantsur rage quits18:13
lucasagomesperhaps the patch to ironic can do that (just for testing purpose)18:13
* lucasagomes checks18:13
lucasagomesjlvillal, because we have a job in VBMC that builds it from source, but I don't think it runs against ironic18:14
jlvillallucasagomes, Oh. I guess I could try to update requirements.txt with a git:// URL thingy...18:14
dtantsurJayF, rloo, I'm not changing anything until somebody becomes realistic18:15
lucasagomesjlvillal, yeah18:15
dtantsurand explains how it works in practice, not in the pink ideal world18:15
jrolldtantsur: suggested some wording in a comment18:15
openstackgerritMilan Kováčik proposed openstack/ironic-inspector master: Adds node state to the API response data  https://review.openstack.org/43641518:15
jlvillallucasagomes, But I think we could still be +2 on that patch :)18:15
*** xavierr has joined #openstack-ironic18:15
* rloo just wants confirmation from jlvillal and vsaienk0 as to whether one of them will be responsible for rolling upgrades feature. if not, then i need to put my name down there in the priorities thing.18:15
dtantsurjroll, it addresses none of my concerns18:15
jrolldtantsur: are your concerns "too many people are commenting on this patch", or something different?18:15
lucasagomesjlvillal, bumped18:15
lucasagomesjlvillal, anyway, tomorrow we can get that sorted :D18:16
lucasagomes(or today if one more core +a it)18:16
dtantsurjroll, <dtantsur> how do I pick these 3 people out of 2 cores and 4 non-cores?18:16
jlvillalI was down for rolling upgrade CI work. But I haven't touched the rolling upgrade non-CI work.18:16
wanyenLet's the folks to pick what features they want to participate18:16
jrolldtantsur: just pick18:16
dtantsura practical example, btw18:16
dtantsurso, I did pick. problems solved?18:16
jrolldtantsur: it's a weird decision that you need to make as PTL, it isn't easy, but you just have to do it18:16
rloodtantsur: i don't think you should pick. i think you ask those folks to pick.18:17
* jlvillal is still in a haze that rolling upgrade CI might be done soon. It is a Christmas Miracle :)18:17
dtantsurrloo, you mean, responsible cores?18:17
dtantsurI don't think asking 4 unrelated non-cores to pick one of them for presumably favorable position is going to work smoothly18:17
jrollfwiw, I just limited it and chose people myself, it turned out fine18:17
rloodtantsur: no, there's a group A, B, C, D, E that say they want to be primary contacts. We only want max of 3 (say). Ask them to pick the three.18:18
dtantsurjroll, that's what I did, and here we are18:18
rloodtantsur: i did that with console work; i sent email to all the console folks and said i wanted 2 (or 3 i don't recall) as primary contact for the subteam.18:18
jrolldtantsur: people have a problem with the paragraph above excluding cores. they don't have a problem with who you picked18:19
jroll(afaict)18:19
rloojroll: it turned out fine as far as you were concerned. maybe there were quite JayF out there that didn't say anything but we're happy or whatever18:19
JayFdtantsur: 99.99% of my concern is simply the word "core" between primary and contacts multiple places on that page. I just don't want non cores reading that document to think there's a requirement of being core to be a primary contact18:19
dtantsurjroll, they did, before I wrote that note18:19
rloos/quite/quiet/18:19
jrollrloo: I can't help people if they don't tell me that something is a problem18:19
jrolldtantsur: can I borrow the table flip emoji, please18:19
JayFdtantsur: it is all likely to end up being cores, because there aren't many senior folks on the project who /aren't/ core today18:19
rloojroll: i know, i know. but JayF is bringing up something so we're trying to address it, right?18:19
dtantsurjroll, sure, and get one for me18:20
*** vills has quit IRC18:20
jrollrloo: yes, his problem is the words, not the people, from what he has said here18:20
jrolldtantsur: who had a problem with the list of people?18:20
rloojroll: right, so let's change the words, and put  a max number.18:20
*** lucasagomes is now known as lucas-afk18:20
rlooi have a problem with the list of people. and jroll, we did too before.18:20
dtantsurjroll, people who asked to add them18:21
* lucas-afk calling it a day, have a great evening all (I will check the conclusion about the names later on) :D18:21
* dtantsur updates the spec with "any people max of 2"18:21
jrolldtantsur: it's not that they had a problem, seems they were just volunteering18:21
*** rpioso has quit IRC18:21
jrollrloo: I think we're talking past each other18:21
rloojroll: quite possible :)18:22
jrollprobably everyone is18:22
*** wanyen has quit IRC18:22
dtantsurjroll, re your "Link?" comments, do we need details on something obvious (redfish driver) or openstack-defined (python 3.5 support)18:23
dtantsur?18:23
jrolldtantsur: I think it makes sense to, like rolling upgrades here: http://specs.openstack.org/openstack/ironic-specs/priorities/ocata-priorities.html#rolling-upgrades18:24
dtantsurlooks like useless work for me, but dunno18:24
dtantsurI'll update this bit tomorrow, the coming update is only about names and small nits18:25
jroll"useless work for me", welcome to PTL18:25
jrollfine with me18:25
* dtantsur already knows who is not going to run the next cycle :/18:25
jrollI won't -1 for not having the description, just curious why18:25
jrollit gets better, promise :)18:26
*** athomas has joined #openstack-ironic18:27
*** rpioso has joined #openstack-ironic18:27
openstackgerritMerged openstack/ironic-inspector master: Python 3.4 (py34) is removed from tox envlist  https://review.openstack.org/44127518:29
jlvillaljroll, FYI: A Python 3 related patch https://review.openstack.org/#/c/442095/18:29
patchbotpatch 442095 - virtualbmc - Fix TypeError when writing config file in Python 318:29
*** e0ne has joined #openstack-ironic18:29
dtantsurjlvillal, do you want to be a contact for splitting away tempest plugins? I'm not sure if it was you who added yourself18:29
*** ccamacho has quit IRC18:30
jlvillaldtantsur, Works for me. I am assisting Solio on that.18:30
dtantsurcool, then I'll remove myself in favor of soliosg18:30
soliosgdtantsur: jlvillal is my mentor, I'm fine with adding his name on the priorities list18:30
dtantsurcool18:31
* JayF dubs soliosg the bus driver ;)18:31
openstackgerritDmitry Tantsur proposed openstack/ironic-specs master: Add priorities for the Pike cycle  https://review.openstack.org/43971018:31
dtantsurjroll, JayF, rloo, ok, here it is ^^^18:31
jrolljlvillal: one thing inline18:32
JayFdtantsur: that's a lot better, thanks. I appreciate it, the wording does matter to some folks :)18:32
jroll+118:33
jrolland highlights another good reason to have cores in there :)18:33
*** rpioso has quit IRC18:33
jlvillaljroll, Thanks. I'll ponder on the unit tests :)18:33
dtantsurok :)18:33
*** moshele has quit IRC18:33
jrolljlvillal: a test_add_with_port_as_int() would be the simplest, I think :)18:33
jrollthanks dtantsur18:33
dtantsurnp18:33
dtantsurnow calling it a day, see you tomorrow18:34
jrollhave a good night18:34
* jroll slides over the whiskey bottle18:34
*** baoli has joined #openstack-ironic18:34
dtantsur:)18:34
*** dtantsur is now known as dtantsur|afk18:34
JayFThis is an easy change to review; removing a ocata release note from master that was being accidentally built into unreleased release notes because it was edited post-ocata-cut https://review.openstack.org/#/c/442018/18:36
patchbotpatch 442018 - ironic - Delete release note to fix build18:36
*** amotoki has joined #openstack-ironic18:36
*** mgoddard has joined #openstack-ironic18:38
rlooJayF: so that fixes it? good.18:39
JayFrloo: yep, doug and I confirmed. It looks like it didn't fix it locally for me because of environment issues in my dir that doug also helped me fix :)18:39
*** gfhellma has joined #openstack-ironic18:40
JayF(it always looked right in the gate, it now looks right in the gate + locally)18:40
*** jcoufal has quit IRC18:40
* jroll wonders if we're meant to +1 or +2 the priorities doc, does +218:40
rlooJayF: :)18:40
soliosgOn removing py34 support, now that https://review.openstack.org/#/c/441275/ and https://review.openstack.org/#/c/437414/ have merged. References to "Python 3.4" and py34 in https://docs.openstack.org/developer/ironic/dev/dev-quickstart.html are no longer necessary, right?18:40
patchbotpatch 441275 - ironic-inspector - Python 3.4 (py34) is removed from tox envlist (MERGED)18:40
patchbotpatch 437414 - ironic - Developer guide should not include Python 3.4 (MERGED)18:40
rloojroll: i think +218:40
jrollJayF: +W18:40
openstackgerritJohn L. Villalovos proposed openstack/virtualbmc master: Fix TypeError when writing config file in Python 3  https://review.openstack.org/44209518:40
jrollsoliosg: sounds right to me18:40
*** amotoki has quit IRC18:41
*** rpioso has joined #openstack-ironic18:42
*** ijw has joined #openstack-ironic18:43
*** baha has joined #openstack-ironic18:44
*** diablo_rojo has joined #openstack-ironic18:45
*** jcoufal has joined #openstack-ironic18:46
*** hshiina has quit IRC18:46
*** ccamacho has joined #openstack-ironic18:48
openstackgerritJoanna Taryma proposed openstack/ironic master: Follow up patch of 3428cb74f0a6fe05c23cf00dc66da4c1d9766409  https://review.openstack.org/44086518:49
JayFjoanna: ^ when possible, we try to make the commit titles for followups describe what they're doing18:49
JayFjoanna: although keeping that line + the link to the patch you're following up in the body is super helpful18:50
openstackgerritMerged openstack/ironic master: Delete release note to fix build  https://review.openstack.org/44201818:51
JayF^ ty jim+ruby18:51
jrollnp, thanks for the fix18:52
openstackgerritSolio Sarabia proposed openstack/ironic master: Remove references to py34 from developer guide  https://review.openstack.org/44211318:53
rlooyw JayF, thx for chasing/fixing it!18:54
*** sayantan_ has quit IRC18:55
JayFit's no problem. I'll take something straightforward like that any day of the week :)18:55
rlooJayF: :)18:55
openstackgerritSolio Sarabia proposed openstack/ironic master: Remove references to py34 from developer guide  https://review.openstack.org/44211318:55
joannarloo, vsaienk0, vdrok: I updated the followup patch for deadlock18:55
joannaI left a comment in the review after speaking with oslo.db person18:56
rloothx joanna. will look at it later today; i seem to have a backlog of stuff to do...18:56
joannarloo: sure, no worries :)18:56
joannathanks!18:56
*** sbezverk has joined #openstack-ironic18:58
*** baoli has quit IRC18:58
*** racedo has joined #openstack-ironic19:01
*** dprince has quit IRC19:02
*** pcaruana has quit IRC19:04
*** moshele has joined #openstack-ironic19:05
*** crushil has quit IRC19:05
*** baoli has joined #openstack-ironic19:09
openstackgerritVasyl Saienko proposed openstack/ironic master: WIP/DNM: Testing multi-tenant multi-node grenade job  https://review.openstack.org/43695719:10
*** crushil has joined #openstack-ironic19:11
*** racedo has quit IRC19:12
*** mjura has quit IRC19:13
*** vsaienko has quit IRC19:14
*** racedo has joined #openstack-ironic19:15
*** pcaruana has joined #openstack-ironic19:15
*** yee379 has quit IRC19:21
*** yee379 has joined #openstack-ironic19:21
*** moshele has quit IRC19:25
*** racedo has quit IRC19:27
*** dprince has joined #openstack-ironic19:28
*** baoli has quit IRC19:29
*** sayantan_ has joined #openstack-ironic19:30
*** chlong has joined #openstack-ironic19:31
*** vsaienko has joined #openstack-ironic19:33
openstackgerritJohn L. Villalovos proposed openstack/ironic master: WIP/DNM: Testing multi-tenant multi-node grenade job  https://review.openstack.org/42214919:36
*** fragatin_ has quit IRC19:42
*** ChubYann has joined #openstack-ironic19:45
*** sayantan_ has quit IRC19:48
*** baoli has joined #openstack-ironic19:50
*** baoli has quit IRC19:50
*** ElCoyote_ has joined #openstack-ironic19:57
*** goneri_ has quit IRC19:58
*** Goneri has quit IRC19:58
*** Goneri has joined #openstack-ironic19:58
openstackgerritRamamani Yeleswarapu proposed openstack/ironic-inspector-specs master: Collect Bios, Memory and Processor Info from Dmidecode  https://review.openstack.org/40679020:00
*** ijw_ has joined #openstack-ironic20:06
*** ijw has quit IRC20:09
*** sayantan_ has joined #openstack-ironic20:10
*** ijw has joined #openstack-ironic20:11
*** e0ne has quit IRC20:14
*** ijw_ has quit IRC20:14
*** gfhellma has quit IRC20:16
*** vills has joined #openstack-ironic20:26
aNuposicFolks, is tinyIPA the default ramdisk type for ironic nodes in devstack?20:27
openstackgerritSolio Sarabia proposed openstack/ironic master: Remove references to py34 from developer guide  https://review.openstack.org/44211320:27
jrollaNuposic: I believe so, yeah20:28
rloosoliosg: i just +A'd it :-(20:28
*** gfhellma has joined #openstack-ironic20:28
aNuposicokay bcoz i can see IRONIC_RAMDISK_TYPE=tinyipa in devstack/lib/ironic but just to confirm :)20:29
rlooJayF: you want it put back? https://review.openstack.org/#/c/442113/20:29
patchbotpatch 442113 - ironic - Remove references to py34 from developer guide20:29
aNuposicjroll ^20:29
jrollaNuposic: yep, that would be the default20:29
aNuposiccool20:29
soliosgrloo: Sorry, I read the initial comment, and uploaded the patch without being aware of Jay'Fs comment20:31
*** fragatina has joined #openstack-ironic20:31
rloosoliosg: no worries, will see what JayF thinks. Based on his comment, i think it woudl be good to put back20:31
soliosgrloo: anyway, should I revert the change?20:31
rloosoliosg: maybe put it in paren, '(without replacing ..)' or whatever it was.20:32
soliosgrloo: I think it's better if we word it like "This installs Python 3.5 without replacing the system version…"20:32
rloosoliosg: put yourself in the shoes of someone reading those instructions...20:32
soliosgJayF: ^^20:32
JayFsoliosg: ++ I like your rewording20:35
*** gfhellma has quit IRC20:36
*** amotoki has joined #openstack-ironic20:37
*** ijw_ has joined #openstack-ironic20:38
*** amoralej is now known as amoralej|off20:39
*** ijw has quit IRC20:41
*** amotoki has quit IRC20:42
openstackgerritSolio Sarabia proposed openstack/ironic master: Remove references to py34 from developer guide  https://review.openstack.org/44211320:42
*** crushil has quit IRC20:43
openstackgerritRuby Loo proposed openstack/ironic master: [devstack] Use global requirements for virtualbmc  https://review.openstack.org/44071520:43
*** mgoddard has quit IRC20:43
openstackgerritMario Villaplana proposed openstack/python-ironicclient master: Log warning when API version unspecified  https://review.openstack.org/44215320:44
mariojvcc dtantsur|afk jroll ^20:47
mariojvsince y'all helped lead that discussion20:47
mariojvit's a bit difficult to log the warning only when the flag is missing without changing python-osc itself, so i just check whether the default version is used at the time the client is created instead20:48
*** chlong has quit IRC20:48
openstackgerritJohn L. Villalovos proposed openstack/ironic master: WIP/DNM: Don't enable ironic in grenade settings  https://review.openstack.org/44215420:53
*** sayantan_ has quit IRC20:57
*** jcoufal_ has joined #openstack-ironic20:57
*** sayantan_ has joined #openstack-ironic20:58
*** athomas has quit IRC20:58
openstackgerritJohn L. Villalovos proposed openstack/ironic master: WIP/DNM: Testing multi-tenant multi-node grenade job  https://review.openstack.org/42214920:59
*** jcoufal has quit IRC21:00
jrollmariojv: nice, will add to my queue21:03
mariojvthx21:03
*** lucas-afk has quit IRC21:04
*** rook has quit IRC21:05
*** sayanta__ has joined #openstack-ironic21:05
*** rook has joined #openstack-ironic21:06
*** mgoddard has joined #openstack-ironic21:06
*** rook is now known as Guest9483321:06
*** lucasagomes has joined #openstack-ironic21:07
*** sayantan_ has quit IRC21:08
openstackgerritAnnie Lezil proposed openstack/ironic-inspector master: Reapply doesn't update started_at time  https://review.openstack.org/43105321:13
*** pcaruana has quit IRC21:13
jlvillalmariojv, As a note, I think of it as "major" version as opposed to "macro" version. But just a semantic thing.21:15
-openstackstatus- NOTICE: restarting gerrit to address performance problems21:16
*** dprince has quit IRC21:17
*** crushil has joined #openstack-ironic21:22
openstackgerritMichael Turek proposed openstack/ironic master: Add OpenBMC class for accessing nodes using OpenBMC  https://review.openstack.org/42983621:24
*** sayanta__ has quit IRC21:24
openstackgerritMerged openstack/ironic-inspector master: Fix tox test failed because timezone is CST  https://review.openstack.org/44164021:25
openstackgerritMichael Turek proposed openstack/ironic master: Add OpenBMC class for accessing nodes using OpenBMC  https://review.openstack.org/42983621:26
*** Guest94833 is now known as rook-new21:32
*** dyasny has quit IRC21:32
*** vsaienko has quit IRC21:32
*** baoli has joined #openstack-ironic21:33
*** racedo has joined #openstack-ironic21:34
*** racedo has quit IRC21:36
*** vsaienko has joined #openstack-ironic21:39
*** jkilpatr has joined #openstack-ironic21:42
*** dyasny has joined #openstack-ironic21:43
mariojvjlvillal: ah, ok21:47
*** jkilpatr has quit IRC21:48
*** jkilpatr has joined #openstack-ironic21:52
*** vsaienko has quit IRC21:53
*** ijw_ has quit IRC21:55
openstackgerritSolio Sarabia proposed openstack/ironic master: Remove references to py34 from developer guide  https://review.openstack.org/44211321:56
*** crushil has quit IRC21:57
*** gfhellma has joined #openstack-ironic21:58
*** jkilpatr has quit IRC21:58
*** gfhellma has quit IRC21:58
*** gfhellma has joined #openstack-ironic21:58
aNuposicjlvillal: have fixed releasenote issues you had suggested on my ceph patch, could you please take a look at it https://review.openstack.org/#/c/407639/21:58
patchbotpatch 407639 - ironic - Config drive support for Ceph21:58
*** vsaienko has joined #openstack-ironic21:59
*** mgoddard has quit IRC22:00
*** Goneri has quit IRC22:00
*** jkilpatr_ has joined #openstack-ironic22:01
*** mtanino_ has joined #openstack-ironic22:02
*** mgoddard has joined #openstack-ironic22:03
*** baha has quit IRC22:03
*** mtanino has quit IRC22:04
openstackgerritJoanna Taryma proposed openstack/ironic master: Fixed release note for DBDeadLock handling  https://review.openstack.org/44218922:09
*** rook-new is now known as rook22:12
*** jcoufal_ has quit IRC22:14
*** mgoddard has quit IRC22:14
*** vsaienko has quit IRC22:18
*** jjohnson2 has quit IRC22:21
*** vsaienko has joined #openstack-ironic22:22
*** baoli has quit IRC22:23
*** ijw has joined #openstack-ironic22:24
*** jkilpatr_ has quit IRC22:24
*** jkilpatr_ has joined #openstack-ironic22:28
*** ijw has quit IRC22:29
*** vsaienko has quit IRC22:31
*** jkilpatr_ has quit IRC22:34
*** ijw has joined #openstack-ironic22:35
*** vsaienko has joined #openstack-ironic22:37
*** jkilpatr_ has joined #openstack-ironic22:37
*** amotoki has joined #openstack-ironic22:38
*** ijw has quit IRC22:38
*** ijw has joined #openstack-ironic22:38
*** amotoki has quit IRC22:42
*** bfournie has quit IRC22:44
openstackgerritRuby Loo proposed openstack/ironic-specs master: Allow spec to have RFE from any ironic project  https://review.openstack.org/44061022:53
*** ccamacho has quit IRC22:53
*** lee4 has quit IRC22:54
*** vsaienko has quit IRC22:54
*** vsaienko has joined #openstack-ironic22:57
*** derekh has quit IRC23:04
*** jkilpatr_ has quit IRC23:04
*** jkilpatr_ has joined #openstack-ironic23:06
openstackgerritMerged openstack/ironic-lib stable/mitaka: Use constraints for all the things  https://review.openstack.org/44066323:12
*** pmannidi has joined #openstack-ironic23:12
*** vsaienko has quit IRC23:13
*** jkilpatr_ has quit IRC23:18
*** rcernin has quit IRC23:19
*** jkilpatr_ has joined #openstack-ironic23:20
openstackgerritMerged openstack/ironic master: Clean up eventlet monkey patch comment and reno  https://review.openstack.org/44127123:37
openstackgerritMerged openstack/ironic master: Remove references to py34 from developer guide  https://review.openstack.org/44211323:39
openstackgerritMerged openstack/ironic-python-agent master: imagebuild: coreos: Build sha256 checksum files  https://review.openstack.org/44061423:43
*** rpioso has quit IRC23:47
*** jkilpatr_ has quit IRC23:54
*** jkilpatr_ has joined #openstack-ironic23:58
« Sunday, 2017-03-05 Index Tuesday, 2017-03-07 »

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