Monday, 2013-12-09

« Sunday, 2013-12-08 Index Tuesday, 2013-12-10 »
*** rongze has quit IRC00:03
*** jcooley_ has joined #openstack-ironic00:48
*** jcooley_ has quit IRC00:50
*** jcooley_ has joined #openstack-ironic00:52
*** jcooley_ has quit IRC01:08
*** docaedo has quit IRC01:08
*** jcooley_ has joined #openstack-ironic01:11
*** nosnos has joined #openstack-ironic01:48
*** nosnos has quit IRC01:48
*** nosnos has joined #openstack-ironic01:49
*** jcooley_ has quit IRC01:55
*** jcooley_ has joined #openstack-ironic02:03
*** docaedo has joined #openstack-ironic02:14
*** rongze has joined #openstack-ironic02:26
*** jcooley_ has quit IRC02:56
*** jcooley_ has joined #openstack-ironic03:04
*** mwagner_lap has quit IRC03:27
*** mwagner_lap has joined #openstack-ironic04:09
*** nosnos_ has joined #openstack-ironic04:28
*** nosnos has quit IRC04:31
*** rongze has quit IRC04:54
*** datajerk has joined #openstack-ironic04:58
*** prekarat has joined #openstack-ironic05:00
*** Haomeng has joined #openstack-ironic05:16
*** rongze has joined #openstack-ironic05:37
*** rongze has quit IRC05:42
*** nosnos_ has quit IRC05:57
*** nosnos has joined #openstack-ironic05:58
openstackgerritJenkins proposed a change to openstack/ironic: Imported Translations from Transifex  https://review.openstack.org/5962706:00
*** rongze has joined #openstack-ironic06:39
*** arata has joined #openstack-ironic06:44
*** rongze has quit IRC06:45
*** rongze has joined #openstack-ironic06:59
*** nosnos has quit IRC07:01
*** nosnos has joined #openstack-ironic07:01
*** jcooley_ has quit IRC07:18
*** mdurnosvistov has joined #openstack-ironic07:27
*** jcooley_ has joined #openstack-ironic07:29
*** jcooley_ has quit IRC07:41
*** jcooley_ has joined #openstack-ironic07:42
*** jcooley_ has quit IRC07:46
*** nick_long has joined #openstack-ironic08:06
*** viktors has joined #openstack-ironic08:08
*** nick_long is now known as ndipanov08:09
*** ndipanov has quit IRC08:09
*** hughsaunders has quit IRC08:09
*** olukanina has joined #openstack-ironic08:17
*** mdurnosvistov has quit IRC08:18
*** olukanina has quit IRC08:23
*** hughsaunders has joined #openstack-ironic08:25
*** jimjiang_ has joined #openstack-ironic08:31
*** diaron has joined #openstack-ironic08:32
openstackgerritYuriy Zveryanskyy proposed a change to openstack/ironic: Strip '/' from api_url string for PXE driver  https://review.openstack.org/6020508:33
openstackgerritFengqian.gao proposed a change to stackforge/pyghmi: Add BMC bridge request extension  https://review.openstack.org/6076808:34
*** ivanoww has joined #openstack-ironic08:44
*** ivanoww has quit IRC08:45
*** ifarkas has joined #openstack-ironic08:51
*** jistr has joined #openstack-ironic09:03
*** vkozhukalov has joined #openstack-ironic09:14
*** derekh has joined #openstack-ironic09:25
*** tatyana has joined #openstack-ironic09:25
*** lucasagomes has joined #openstack-ironic09:25
*** hughsaunders has quit IRC09:28
*** ndipanov has joined #openstack-ironic09:30
*** hughsaunders has joined #openstack-ironic09:32
*** hughsaunders has quit IRC09:52
*** linuxmonk has joined #openstack-ironic10:00
*** hughsaunders has joined #openstack-ironic10:02
*** max_lobur_afk is now known as max_lobur10:02
*** nosnos has quit IRC10:07
*** nosnos has joined #openstack-ironic10:07
*** slepoj has joined #openstack-ironic10:07
*** slepoj has quit IRC10:08
*** gadddd has joined #openstack-ironic10:11
*** gadddd has quit IRC10:15
*** titanic has joined #openstack-ironic10:18
*** viktors has quit IRC10:22
*** titanic has quit IRC10:41
*** yfujioka has joined #openstack-ironic10:43
*** viktors has joined #openstack-ironic10:59
*** rongze has quit IRC11:17
*** arata has left #openstack-ironic11:19
*** rongze has joined #openstack-ironic11:26
*** mdurnosvistov has joined #openstack-ironic11:35
*** urulama has joined #openstack-ironic11:40
*** linuxmonk has quit IRC11:41
*** rongze has quit IRC11:58
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: Port StringType class from WSME trunk  https://review.openstack.org/6079812:00
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: Add PatchType class  https://review.openstack.org/6019412:09
*** yfujioka has quit IRC12:13
*** lucasagomes is now known as lucas-hungry12:34
*** rongze has joined #openstack-ironic12:40
*** max_lobur is now known as max_lobur_afk12:52
*** chillroom has joined #openstack-ironic12:59
*** chillroom has joined #openstack-ironic13:00
*** urulama has quit IRC13:09
*** chillroom has quit IRC13:10
*** mwagner_lap has quit IRC13:11
*** nosnos has quit IRC13:13
*** ekarlso- has quit IRC13:23
*** ekarlso has joined #openstack-ironic13:23
*** rongze has quit IRC13:25
*** lucas-hungry is now known as lucasagomes13:31
*** rongze has joined #openstack-ironic13:41
*** peropavlina has joined #openstack-ironic13:41
*** jdob has joined #openstack-ironic13:46
*** rloo has joined #openstack-ironic13:47
*** max_lobur_afk is now known as max_lobur14:06
*** matty_dubs|gone is now known as matty_dubs14:21
*** prekarat has quit IRC14:22
*** prekarat has joined #openstack-ironic14:25
NobodyCamgood morning Ironic14:54
dkehnNobodyCam: morning14:57
NobodyCammorning dkehn :) have a good weekend?14:58
dkehnNobodyCam: yes, very cold here, so hunkered down14:59
NobodyCamya.. I'll bet... cold here. gota be in the 40's :_p15:00
dkehnNobodyCam: whats normal in the winter15:00
NobodyCamabout that.. I'd say it just a bit below normal here15:01
max_loburmorning Ironic :)15:01
*** shadower has quit IRC15:01
NobodyCammornig max_lobur :)15:02
dkehnNobodyCam: way below normal here, even the ski slopes are too cold, but the snow up there is good15:02
*** rongze has quit IRC15:02
NobodyCam:) hehhe snow is too cold15:03
*** rongze has joined #openstack-ironic15:03
dkehnsnow isn't bad , but clear no clouds the temp just drops, re-thinking heading south for the winter, soon15:04
NobodyCam:) thats what I did15:04
*** jdob has quit IRC15:05
dkehnUsually doesn't get this cold , here but it does flood like it did either, I think the climate is changing15:06
NobodyCamgore was right :-p15:06
NobodyCamoh wait that would mean it would be warmer now :-p15:07
dkehnso if there are no more review issues I'm going to get on the generate_sample again15:07
dkehnso you might hear from me today15:07
*** shadower has joined #openstack-ironic15:07
NobodyCam:)15:07
dkehnI'm thinking its something I'm not doing right, because when it parses the .pyc (note you must compile it 1st) it doesn't see my cftStr, where woould be a good place(wiki) to document, so I don't have to relearn this again15:09
dkehnnote that anyone else would do this15:09
NobodyCamdkehn: prob jst a a section to https://wiki.openstack.org/wiki/Ironic15:14
dkehnNobodyCam: ok, great15:14
NobodyCambrb15:17
*** rustlebee is now known as russellb15:17
*** jdob has joined #openstack-ironic15:22
*** rongze has quit IRC15:30
*** rongze has joined #openstack-ironic15:30
*** ifarkas has quit IRC15:50
*** mdurnosvistov has quit IRC15:56
*** vkozhukalov has quit IRC16:20
NobodyCambrb again :-p16:32
*** romcheg has joined #openstack-ironic16:33
NobodyCamromcheg: fyi Just ate the last if the chocolate lastnght :)16:38
NobodyCamit was very yummy... Thank you :)16:39
romchegMorning NobodyCam :)16:39
NobodyCam:)16:39
romchegMy pleasure16:39
*** gurkafa has joined #openstack-ironic16:40
*** gurkafa has quit IRC16:40
romchegNobodyCam: Was it pepper, salt of cherry?16:41
NobodyCamsalt :)16:41
NobodyCamwent great with night time coffee16:41
romchegSince this Lviv Chocolatier opened in Kharkiv, I go there every week to get more different chocolate16:43
romchegNobodyCam: If you go to Pycon in Lviv, I will show you the original factory :)16:44
NobodyCam:) very cool...16:44
NobodyCamI don't think I'll make pycon. :(16:44
NobodyCamhow are things over there? still have the unrest? nothing on our news anymore16:45
romchegNobodyCam: Yesterday in Kyiv http://cs14109.vk.me/c540101/v540101434/17320/0P97YhucC0A.jpg16:46
*** rongze has quit IRC16:46
NobodyCamomg wow16:47
*** prekarat has quit IRC16:48
NobodyCamthere most be 1000's of people there if not 10's of 1000's16:48
*** prekarat has joined #openstack-ironic16:48
romchegNobodyCam: More than 1.5 millions16:49
romchegPeople from all the Ukraine go to Kyiv to take an action16:49
devanandamorning, all16:51
romchegMorning devananda16:51
NobodyCamgood morning devananda16:51
NobodyCamromcheg: wow16:51
*** hemna has joined #openstack-ironic16:51
*** matty_dubs is now known as matty_dubs|lunch16:54
lucasagomesmorning all16:56
lucasagomesdevananda, ping... should we normalize the mac address on the API side?16:56
lucasagomesor at lease make the address column in the db case insensitive16:57
*** jcooley_ has joined #openstack-ironic16:57
devanandalucasagomes: morning16:57
devanandalucasagomes: we should not rely on the db case-sensitivity.16:57
lucasagomesHaomeng, raised that in one of my patches (actually 2 of them)16:58
devanandamysql defaults to case-insensitive, but can be changed. IIRC< pgsql defaults to sensitive.16:58
lucasagomesright16:58
lucasagomesI thought there would be something on sqlalchmey that would take care of diff backends16:58
openstackgerritA change was merged to openstack/python-ironicclient: Replace chassis_id with chassis_uuid on Nodes  https://review.openstack.org/5831416:59
devanandalucasagomes: i'm not aware of any setting that automatically munges the case of all inputs17:01
lucasagomesdevananda, right17:02
lucasagomesso I think we can validate (correct format and colon-separated) and normalize it to all lower or upper (pref is lower) case in the API17:04
openstackgerritYuriy Zveryanskyy proposed a change to openstack/ironic: Check connectivity in SSH driver 'validate' method  https://review.openstack.org/6087617:05
devanandalucasagomes: ++17:06
*** ifarkas has joined #openstack-ironic17:10
lucasagomescool will do it :)17:12
NobodyCamfyi: https://review.openstack.org/#/c/60879/17:13
lucasagomesNobodyCam, ha! gogo ironic :)17:14
NobodyCam:-17:15
NobodyCambbt... brb17:16
*** rongze has joined #openstack-ironic17:17
*** rongze has quit IRC17:28
dkehnanteaya: are you the resident generate_sample.sh person here, I see your name https://github.com/openstack/ironic/blob/master/tools/conf/generate_sample.sh17:34
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: API: Validate and normalize address  https://review.openstack.org/6089017:43
NobodyCampost bbt walkies... brb17:49
anteayadkehn: I did the latest commit on that file, yes17:52
anteayadkehn: that came from oslo and I needed to adjut the path so it would work for ironic conf generation17:52
dkehnanteaya: ok, are you familiar with its workings I've got changes into the pxe.py, adding a config parameter, but it doesn't seem to generate the sample, and I wondering what I'm missing17:53
*** rongze has joined #openstack-ironic17:54
*** jcooley_ has quit IRC17:54
dkehnanteaya: so the assumption is that you add or chnage the cfg.StrOPt, in this case  and run the generate_sample.sh from the root, is there more to it than that?17:54
anteayadkehn: do you have a url to a patch?17:59
devanandadkehn: are you inside the venv when calling generate_sample.sh?17:59
dkehndevananda: no I'm not18:00
dkehndevananda: is that a requirement18:00
anteayaand yeah, that is basically what I did, ensure the script can find an imput file and specifiy the output path18:00
devanandadkehn: afaik, yes18:00
devanandadkehn: generate-sample.sh needs to import all the modules, so they need to be in your path, which is part of what being in the venv does18:01
dkehnok, guys maybe thats it, was assuming becuase tox now workins without venv this would too18:01
devanandaanteaya: also, you may be interested in https://review.openstack.org/#/c/60639/18:01
devanandaanteaya: i think this solves our perpetualy-out-of-date-sample-config problem18:01
*** jistr has quit IRC18:02
* anteaya clicks18:02
devanandaNobodyCam: ^18:03
* NobodyCam looks18:03
dkehnyepper that was it, thx , keep forgetting that one18:03
devanandadkehn: :)18:03
*** derekh has quit IRC18:05
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: API: Validate and normalize address  https://review.openstack.org/6089018:12
openstackgerritA change was merged to openstack/ironic: Add tools/conf/check_uptodate to tox.ini  https://review.openstack.org/6063918:14
NobodyCamdevananda: ^^^18:17
devananda:)18:19
*** vkozhukalov has joined #openstack-ironic18:20
*** ifarkas has quit IRC18:21
*** lucasagomes is now known as lucas-afk18:22
lucas-afkbe back for the meeting :)18:22
*** jcooley_ has joined #openstack-ironic18:22
*** max_lobur is now known as max_lobur_afk18:23
*** rongze has quit IRC18:37
*** matty_dubs|lunch is now known as matty_dubs18:38
openstackgerritdekehn proposed a change to openstack/ironic: PXE boot information that is necessary for Neutron  https://review.openstack.org/6090418:41
NobodyCamdkehn: w00t!!!18:41
*** tatyana has quit IRC18:42
devanandahrm. something odd18:43
devanandadkehn: why do you have a new file, ironic.sample.conf?18:43
dkehndevananda: my bad will delete18:44
* NobodyCam wanders off to refil coffee and what not b4 meeting18:46
openstackgerritdekehn proposed a change to openstack/ironic: PXE boot information that is necessary for Neutron  https://review.openstack.org/6090418:47
dkehndevananda: mo better18:47
devanandaya :)18:47
* devananda needs to polish up the consistent hashing patches and add the takeover-node stuff18:48
NobodyCamw00t RE: James Slagle comment on https://review.openstack.org/#/c/60879 :)18:53
*** lucas-afk is now known as lucasagomes18:55
* NobodyCam heads over to setup chairs and coffee for the meeting :-P18:58
*** romcheg1 has joined #openstack-ironic18:59
*** romcheg1 has quit IRC18:59
*** romcheg1 has joined #openstack-ironic19:01
NobodyCammeeting time19:11
*** jcooley_ has quit IRC19:11
*** mdurnosvistov has joined #openstack-ironic19:14
*** jcooley_ has joined #openstack-ironic19:17
devanandavkozhukalov: hi! this https://blueprints.launchpad.net/nova/+spec/libvirt-empty-vm-boot-pxe looks a bit odd to me -- it has a target milestone and series, but is not accepted yet and has no approver set.19:19
devanandavkozhukalov: have you talked with nova ptl about it?19:19
vkozhukalovdevananda, I was not sure about this feature (actually I still isn't). So I just tried to ask them in #openstack-nova but no one was interested. I just decided to concentrate on making workin functional testing scheme for ironic. Actually I think that this initiative https://etherpad.openstack.org/p/tripleo-test-cluster is much more interesting.19:24
vkozhukalovdevananda, I mean it is much more interesting than making nova able to boot nodes via PXE19:26
vkozhukalovdevananda, at least at this moment19:27
devanandayea, tripleo-test-cluster is very interesting indeed, but won't cover ironic directly for a while19:27
*** romcheg1 has quit IRC19:38
*** romcheg1 has joined #openstack-ironic19:38
lifelessmaking Nova be able to have PXE bootin VM's won't help infra for many many months19:39
lifelessbecause to use it infra needs it deployed by the clouds infra test on.19:40
*** jcooley_ has quit IRC19:40
vkozhukalovYes, it nova PXE feature could not help with testing. Actually we decided to start from launching VM manually inside jenkins slave VM which is used to run tests.19:44
*** epim has joined #openstack-ironic20:01
*** yuriyz has joined #openstack-ironic20:02
*** yuriyz has left #openstack-ironic20:02
*** yuriyz has joined #openstack-ironic20:02
lucasagomesrloo, so I don't see any problem in proposing a change, but if you ask before for opinions first you might get things right quicker20:02
lucasagomesI don't I can't think about a methodology for it20:02
*** yuriyz has quit IRC20:02
*** kui has joined #openstack-ironic20:02
rloolucasagomes. yeah. i have been mulling over how to get the reviews 'life span' shorter, and reduce the number of patches.20:03
lucasagomesimo, if you believe ur way is good just propose it and wait for reviews. If there's some doubts might worth asking on IRC for opnions20:03
lucasagomesrloo, yea :(20:03
devanandarloo: part of patch lifespan is we have still a fairly small review team, who are also the most active developers, and so bandwidth of -core folks isn't that high yet20:03
rlooit just seems a shame to dot your is and cross your ts, only to get to the 'end' when the solution isn't _quite right_. I mean, it is good to catch it, would be better to catch these sooner.20:03
rloobut then, there are a lot of moving parts.20:04
NobodyCamahh so we have a meeting following us again20:04
devanandarloo: this happens in other projects too, and from what i've seen, generally speaking, the larger the change, the more it helps to have a clearly thought-out and presented blueprint before hand20:04
lucasagomesrloo, I hear ya, but I think it's part of the process20:04
*** rongze has joined #openstack-ironic20:04
lucasagomesso now we have a meeting after the ironic one?20:05
devanandarloo: but even then, because code is a much clearer communication tool for engineers than verbage, i've also seen code for an approved blueprint get rejected20:05
lucasagomeswe need to be strict with the time again :)20:05
devanandarloo: because the fokls who approved the BP and the folks who wrote it were thinking of different things (or didn't think it through all the way)20:05
rloowell, I suppose there isn't one solution to address this.20:07
rloowe just need more deva's.20:08
lucasagomeshehe +120:08
devanandaaaaaaah! I can already barely handle one of me!20:08
rlooa 24/7 deva :-)20:08
lucasagomesbtw, it's late here and I'm pretty hungry20:08
devanandalucasagomes: have a good night :)20:08
lucasagomesso have a good night everyone! see yall tomorrow20:08
NobodyCamlucasagomes: enjoy dinner... and thank you20:09
rloonight lucasagomes!20:09
lucasagomesdevananda, rloo NobodyCam, thanks! see u20:09
*** lucasagomes has quit IRC20:09
rlooso devananda, do you have a few minutes to discuss https://blueprints.launchpad.net/ironic/+spec/breaking-resource-locks20:10
devanandarloo: sure thing20:10
NobodyCamrloo: I feel it good form to discuss anything major like work on BP's in channel first20:10
rlooNobodyCam: yes, that's what I think I'll do. Right now. :-)20:11
NobodyCamjust reading back20:11
NobodyCamheheheh20:11
rlooI haven't worked on a bp before.20:11
rlooCan anyone enter information into this thing?20:11
NobodyCamanyone can comment on them20:11
rloook, i tried to change 'Assignee' to myself but wasn't sure how.20:12
devanandarloo: the assigned person or the PTL can edit them20:12
rlooIs it in the 'whiteboard' section that people put their comments?20:12
NobodyCamput a review with that bp tagged20:12
devanandarloo: i think only the PTL can change certain fields, target to a release, etc.20:12
rloook.20:12
rlooare you serious? you have to write code first?20:12
NobodyCamdevananda: will Mr Jenkins also do that?20:12
devanandarloo: but anyone should be able to edit the whiteboard20:12
devanandarloo: no.20:13
devanandarloo: if you write code that relates to the bp, put "blueprint your-bp-name" in the commit message on a line by itself20:13
devanandarloo: jenkins will add a comment to the whitebaord linking to that patch set20:13
rlooso I don't really care if I don't have my name down as assignee. but I would care, if because my name wasn't there, and I was working on it, someone else decided to work on it too. That wouldn't be good.20:13
NobodyCamrloo: no but somehting like the wip review that is just a commit message saying your working on this20:14
rlooMy real question though, is trying to understand the scope of this BP.20:14
devanandarloo: right. so anyone can write any code. and you can tag any BP in your commit msg20:14
devanandarloo: the purpose of assigning the BP to a single person is to assign responsibility for tracking completion of the feature20:14
rloobut I don't want to write any code, until i understand the scope and have some sort of OK that the approach is what people are expecting.20:14
devanandarloo: so that as as we approach a milestone or releae, I know who to nag :)20:14
rlooah, I thought the BP assignee was the person resp for implementing.20:14
devanandarloo: it usually is20:15
rlooSo how do you make sure people aren't working on things that other people are working on?20:15
devanandaeg, sometimes a team might implement a BP, and so its assigned to the team lead20:15
devanandarloo: we talk to each other20:15
devanandathat's all20:15
rloobut talking on IRC isn't that effect in a global world.20:15
rlooeffective i mean.20:15
devanandacode is a form of talking20:15
devanandatoo20:15
NobodyCamrloo: the BP's whiteboard can do some of that20:16
devanandalook at the review board. look at the BP's whiteboard. subscribe to it. you'll get an email if someone else comments or posts related code20:16
rlooright, but maybe i'm old fashioned. i don't want to code until I understand what I'm going to code.20:16
devanandarloo: that's not old fashioned. that's good :)20:16
NobodyCamrloo: click the edit icon on the BP20:17
rloook. that's fine about the assignee part then. I'll just add comments to the BP.20:17
NobodyCamunder whiteboard20:17
NobodyCam:)20:17
devanandaso one thing that I've done somewhat backwards as the PTL, which may be causing some confusion, is that I created BPs as a way to communicate what I think should be done20:17
devanandawhereas for most projects, fokls who want to implement a specific feature propose a BP20:17
NobodyCamdevananda: that has worked very well for our team20:17
devanandathen naturally, there's already someone who knows what they want to write20:17
rloook, but you're assuming someone knows what they want to write in that case.20:18
devanandasort of20:18
rlooin my case, I'm not sure.20:18
rlooeg, for this blueprint. there are a bunch of ideas.20:18
devanandain the case at hand, I know there's a need for this feature. I'm hoping someone else will see a solution and write it :)20:18
NobodyCamrloo: it he worst case it spawns converstions just like this one20:18
NobodyCamwhich imo is Good20:19
devanandarloo: would it be easier to think of if this were a bug instead of a BP?20:19
rlooone is 'there should be a way for an admin to manually break a lock when necessary'. that's what Yuriy's change does (or will do).20:19
rloono, bug or BP is fine. There are already ideas mentioned in this thing -- so I feel that to work on it, I should understand the background, etc, and see if thoughts have changed since the time of writing.20:20
rlooeg, this bp mentions 'time when lock was taken', ... Is that what we want?20:21
NobodyCambrb20:22
rlooand devananda, in some previous discussion (last month), you had mentioned that you were thinking of something like "it's basically some way to trigger "UPDATE nodes SET instance_uuid=NULL, *_state=NULL, WHERE ...", with a few safeguards around it"20:23
devanandahumm20:24
* devananda gets more coffee and ruminates20:26
kuidevananda, NobodyCam: I want to make contribution and focus on ironic project, is there some suggestion about the first step/task?20:28
*** vkozhukalov has quit IRC20:28
devanandakui: hi! we're happy to point you to getting-started resources. do you have a specific contribution you'd like to make?20:29
devanandarloo: i think we can't have lock-breaking change the instance_uuid20:30
rloodevananda: yes.20:30
devanandarloo: it has to change the reservation. i'm not sure yet if it should clear target_*_state, or if the conductor which takes over should try to complete said state transitions20:30
devanandarloo: this is where it gets tricky20:30
rloodevananda:  I almost wonder whether it is premature to work on this.20:30
kuidevannanda: maybe start from simple one to get familiar with the project.20:31
devanandarloo: it does feel premature, yes20:31
devanandarloo: the original bug -- let me find it -- is pretty essential, though20:31
devanandarloo: https://bugs.launchpad.net/ironic/+bug/125034820:31
rloodevananda: yes, that's the one yuriy is working on though.20:32
rloodevananda: i think that may be enough for now, until we have more experience with conductors dying.20:32
rloodevananda: otoh, if we feel like we want some sort of timestamp for the reservation, and that modifies the db, we may want to put that in before icehouse.20:33
rloodevananda: maybe you can help out kui first, and we can discuss afterwards.20:33
kuirloo: thank you20:34
*** jcooley_ has joined #openstack-ironic20:34
devanandakui: have you gotten a development environment for ironic up and running yet?20:35
devanandarloo: it would be easy to add a timestamp to the reservation data20:36
devanandarloo: though i'm hesitant to add a db column taht we aren't going to use right away, just on principle20:36
kuidevananda: not yet, I just know the steps of devtest.sh in tripleo, and get known about the diskimage-builder.20:36
devanandaahh ok20:36
NobodyCamwas making a bagel ... hi kui20:37
devanandakui: https://wiki.openstack.org/wiki/Ironic#Using_Disk_Image_Builder may help you then20:37
kuiNobodyCam: Good afternoon, NobodayCam20:37
devanandakui: that wiki section has instructions for making tripleo build an environment with ironic in the undercloud node20:37
devanandakui: so you can then test the code there20:37
NobodyCamwe tested that this weekend so I pretty sure its still vaild20:38
devanandaNobodyCam: I'm removing the python-ironicclient from taht wiki, now that it's landed in incubator/scripts/install-tools, ya?20:38
NobodyCamya :)20:38
kuidevannanda: great, I will follow this wiki page.20:39
devanandakui: also a venv for unit tests can be created by: http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html20:39
kuidevananda: I spend some time to study the di-b,  skuicloud.wordpress.com, the latest blog record the procedure of generating ramdisk.20:40
devanandakui: with those two things, you can write and test code. then I'd suggest looking at the bug list for things you want to do, or at the review board and commenting on reviews to get to see what's under development20:40
kuidevananda: I am familiar with the venv, since I have committed 30+ patches in Havana.  I come here to dive into Ironic dedicately.20:41
rloodevananda: there are still a few more months til icehouse, so we can punt on the timestamp thing til later. and if on principle, you don't want to put it in, that's fine with me.20:42
devanandarloo: i think timestamp on reservation /might/ be useful. but until we have a clear use for it, i think adding it is premature20:43
devanandakui: awesome20:43
NobodyCamdevananda: is there a use case for reservations expirering?20:44
devanandaNobodyCam: down the road? maybe. do we need that for Icehouse? I dont see why20:44
NobodyCam:)20:44
rloodevananda: ok, so as far as the blueprint goes (and https://etherpad.openstack.org/p/IcehouseIronicNextSteps - "API to break a task_manager lock"), that is addressed by https://bugs.launchpad.net/ironic/+bug/1250348 (for now, til perhaps icehouse)20:45
kuidevananda: Do we have a plan to make Ironic graduating from incubation during Icehouse ?20:46
*** jcooley_ has quit IRC20:46
devanandarloo: with the discussion we had with yuriyz today, I think that his work addresses the immediate need, yes20:46
devanandakui: yes!20:46
*** jcooley_ has joined #openstack-ironic20:47
rloodevananda: ok, I've updated https://etherpad.openstack.org/p/IcehouseIronicNextSteps to reflect that (specifically, link to the review)20:48
devanandarloo: great, thanks much!20:48
kuidevananda: Great, I can contribute my help to approach the target, more or less :)20:48
devanandakui: in general, we are aiming at basic functionality by I-2 milestone (Jan 22)20:49
devanandakui: if we can deliver that, I will go to TC and propose that we gradaute this cycle20:49
kuidevananda: OK, I will catch up, time is tight.20:50
*** jcooley_ has quit IRC20:54
*** jcooley_ has joined #openstack-ironic20:55
*** romcheg1 has quit IRC20:55
NobodyCamdevananda: just thought of something20:56
NobodyCamwe are adding dhcp options to PXE driver20:56
devanandaya?20:57
NobodyCambut all of ironic's drivers will need a dhcp_options_for_instance as that is called from nova/compute/manager.py:                dhcp_options = self.driver.dhcp_options_for_instance(instance)????20:57
devanandaNobodyCam: no20:57
*** jcooley_ has quit IRC20:57
devanandaNobodyCam: that is a flaw in the implementation in Nova. I commented on it, but someone removed my comment20:58
devanandaNobodyCam: nova shouldn't care about the dhcp_opts_for_instance  in the ironic driver20:58
devanandabecause not all deploy drivers will use dhcp/pxe20:58
kuidevananda: I "watched" last two weekly meeting silently, every attendee take charge one aspect of Ironic, I am willing to work with all of you together and get more task assigned.20:58
devanandait actually shouldn't even be there in nova-baremetal either20:58
NobodyCamkui: awesome :)20:59
kuidevananda: I'm going back to sleep,  Have a good day!20:59
NobodyCamhave a good night kui :)20:59
kuiNobodyCam: Have a good day !20:59
NobodyCamand Thank you :)21:00
devanandakui: rather than assign some task to you, i prefer if you find work you want to do and then do it (maybe check with us to see that no one else is already doing it, too)21:00
devanandakui: good night!21:00
kuidevananda: Yes, I agree,  I just want to align with the Ironic directory, from the perspective of PTL.21:01
devanandakui: :)21:01
kuidevananda: It is 5:00 AM here :)21:01
devanandaheh. sleep well )21:01
kuidevananda, NobodyCam: Thank you all, Bye21:01
NobodyCamdevananda: should I look at refactoring https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L103321:03
devanandaNobodyCam: ooh. gah. right.21:04
devanandaNobodyCam: someone moved it /that/ far up the stack. i thought it was just in the driver21:04
NobodyCamya that what I had thought21:04
NobodyCamuntil I looked21:05
devanandanova/nova/virt$ git grep 'def dhcp_options_for_instance'21:05
devanandabaremetal/driver.py:    def dhcp_options_for_instance(self, instance):21:05
devanandadriver.py:    def dhcp_options_for_instance(self, instance):21:05
devanandait's not used by any other driver21:05
devanandaI would say, don't implement that method in the ironic driver at all21:06
devanandathe parent definition just calls "pass"21:06
devanandahttps://github.com/openstack/nova/blob/master/nova/virt/driver.py#L90821:06
NobodyCamok :)21:06
devanandanov-ironic driver doesn't need that21:07
NobodyCamwas looking at the log and saw it was getting called.. and not by the driver21:08
NobodyCambrb21:14
*** mdurnosvistov has quit IRC21:34
*** openstackgerrit has quit IRC21:37
*** openstackgerrit has joined #openstack-ironic21:37
*** rongze has quit IRC21:52
*** openstackgerrit has quit IRC21:57
*** openstackgerrit has joined #openstack-ironic21:57
*** jdob has quit IRC22:00
*** mdurnosvistov has joined #openstack-ironic22:14
*** kui has quit IRC22:18
*** rongze has joined #openstack-ironic22:18
openstackgerritDevananda van der Veen proposed a change to openstack/ironic: Fix auto-generated REST API formatting  https://review.openstack.org/6095422:25
openstackgerritDevananda van der Veen proposed a change to openstack/ironic: Fix auto-generated REST API formatting  https://review.openstack.org/6095422:26
devanandawoot! formatting is good now!22:33
devanandahttp://docs-draft.openstack.org/54/60954/2/check/gate-ironic-docs/f67944b/doc/build/html/webapi/v1.html22:33
* NobodyCam looks22:33
devanandacompared to docs.openstack.org/developer/ironic/webapi/v1.html22:33
devanandahttp://docs.openstack.org/developer/ironic/webapi/v1.html22:33
NobodyCamdevananda: where is The ‘rest-controller’ directive of this resource seems to be crashing sphinxcontrib-pecanwsme with “CRITICAL ironic [-] list index out of range” errors. comming from22:35
NobodyCamits in both22:35
devanandayea, i know22:35
devanandaI've sent an email to dhellman about that22:35
devanandai think it's a bug in the sphinx extension he wrote22:35
NobodyCamahh22:36
NobodyCam:)22:36
devanandai am not familiar enough with sphinx to debug it in any reasonable amount of time :(22:36
NobodyCambut looks great22:36
openstackgerritDevananda van der Veen proposed a change to openstack/ironic: Developer doc index page points to correct API docs  https://review.openstack.org/6096022:39
NobodyCamwe'll need to keep an eye on doc strings as things like *_id: UUID of a * get updated22:39
devanandayea. I think lucas has some patches that correct some of that nomenclature22:39
NobodyCamya22:39
devanandawe should do a cleanup of it22:39
devanandaespecially since it's quite public in our API docs now :)22:40
devanandabrb22:40
NobodyCamk22:40
*** matty_dubs is now known as matty_dubs|gone22:45
devanandaabck22:47
devanandaback22:47
NobodyCambw22:47
NobodyCamwb22:47
*** rongze has quit IRC22:51
*** epim has quit IRC22:54
*** epim has joined #openstack-ironic22:54
*** mdurnosvistov has quit IRC23:11
NobodyCamdevananda: question23:13
NobodyCamlook into starting on some of the tests for the nova driver23:14
openstackgerritDevananda van der Veen proposed a change to openstack/ironic: Add prepare, cleanup, takeover methods to deploy  https://review.openstack.org/5889423:14
NobodyCamare going to be testing things like adding a node. or should all of the ironic "work" be mocked?23:14
NobodyCamthere really is no nova db to do any testing23:15
*** epim has quit IRC23:15
openstackgerritDevananda van der Veen proposed a change to openstack/ironic: Add prepare, cleanup, takeover methods to deploy  https://review.openstack.org/5889423:17
devanandaNobodyCam: i dont quite understand your question23:17
devanandaNobodyCam: also, i need to get food. bbiaf or call me?23:17
NobodyCamcall while you get food23:17
openstackgerritDevananda van der Veen proposed a change to openstack/ironic: Add config option for # of conductor replicas  https://review.openstack.org/5979423:20
*** rongze has joined #openstack-ironic23:51
NobodyCamdevananda: when you get back. I can not change tags on bugs :-p23:53
devanandaNobodyCam: back23:54
NobodyCamlol23:54
devanandaNobodyCam: join the bug team?23:54
NobodyCamhttps://bugs.launchpad.net/ironic/+bug/125934623:54
« Sunday, 2013-12-08 Index Tuesday, 2013-12-10 »

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