Wednesday, 2014-09-03

* jroll heads home00:00
jrollhave a good night everybody :)00:01
NobodyCamnight jroll :)00:01
*** yuanying has quit IRC00:01
rloobye jroll00:02
rloomrda: not sure I understand. 111223 has 3 +2's, why do you want us to +1 it?00:03
mrdarloo: well, not that one, from 111425 onwards00:04
mrdathere's no +2s there, so I just want to bring it to Nova's attention00:05
rloomrda: ok00:05
rloomrda: qq, I think I might be too tired to review. https://review.openstack.org/#/c/111425/19/nova/virt/ironic/driver.py00:11
rloomrda: why is line 43 not py_logging.getLogger(...)?00:12
rloomrda: yes, I'm tired. the answer is cuz logging is from something else, not from logging. I'll look at it tomorrow.00:14
mrdarloo: if you're too tired, take an early mark :)00:14
NobodyCam115540 has -1's :(00:14
mrdarloo: always appreciate your help00:15
rloomrda: I didn't (or forgot) realize that we should be reviewing these too. They weren't on my radar. Sorry.00:15
NobodyCamoh wow are thouse from rev 3/4?00:16
mrdaNobodyCam: that's only because Daniel B and Dan S haven't re-reviewed since the proposed solution to their objections was posted00:16
mrdaNobodyCam: yep00:16
mrdathanks Labor Day :(00:16
NobodyCamhehehe00:17
*** dlaube has joined #openstack-ironic00:17
openstackgerritKyle Stevenson proposed a change to openstack/ironic: Reduce redundancy in conductor manager docstrings  https://review.openstack.org/11850500:17
mrdarloo: I hadn't asked, but we're now getting to the point where we keep them in front of nova cores if at all possible00:17
rloomrda: does nova have something set so that reviewers' scores get carried forward to new revisions?00:18
mrdarloo: yeah, I think this is as a result of the last gerrit update (where the CI toggle came in)00:18
mrdaIt's new to me00:18
* mrda is bugging nova cores privately too00:18
rloomrda. that doesn't happen in Ironic. interesting. ok, for sure, will look at the nova driver patches tomorrow.00:19
mrdarloo: thanks, much appreciated!00:19
*** chuckC has quit IRC00:20
*** dlaube has quit IRC00:20
*** penick has quit IRC00:30
*** penick has joined #openstack-ironic00:37
JoshNanghorizon api wrapper should be done. i'd love some eyeballs on it to make sure its not completely terrible. https://review.openstack.org/#/c/117376/00:40
*** penick has quit IRC00:51
*** penick has joined #openstack-ironic00:57
*** yuanying has joined #openstack-ironic00:59
*** r-daneel has quit IRC01:00
*** lnxnut has joined #openstack-ironic01:06
*** yuanying has quit IRC01:07
*** lnxnut has quit IRC01:16
*** eguz has joined #openstack-ironic01:26
*** eghobo has quit IRC01:26
*** rushiagr_away is now known as rushiagr01:29
openstackgerritA change was merged to openstack/ironic: Fix typo in PXE driver docstrings  https://review.openstack.org/11849501:36
*** bandicot has quit IRC01:41
*** rushiagr is now known as rushiagr_away01:45
*** penick has quit IRC01:46
openstackgerritHaomeng,Wang proposed a change to openstack/ironic: Remove ironic.conf.sample  https://review.openstack.org/10649301:48
*** penick has joined #openstack-ironic01:49
*** penick has quit IRC01:49
*** pcrews has quit IRC01:50
*** eguz has quit IRC01:50
*** pcrews has joined #openstack-ironic01:52
*** nosnos has joined #openstack-ironic01:52
mrdahey, NobodyCam?01:52
mrdaYou still here?01:52
NobodyCamno01:59
NobodyCam:-p01:59
NobodyCammrda: whatsup?01:59
*** yongli has quit IRC02:02
*** yongli has joined #openstack-ironic02:03
*** yuanying has joined #openstack-ironic02:04
mrdaNobodyCam: thanks for responding.  It's ok, I'll leave it for now.  Have a good night!02:07
*** rushiagr_away is now known as rushiagr02:08
NobodyCamlol02:12
NobodyCamok :)02:12
jrollmrda: I'm around for a bit if you need something02:12
NobodyCamI am here02:12
*** yuanying has quit IRC02:12
NobodyCamI was kidding about the no02:12
jrollI have to be here :P02:12
mrdasure, I was getting some private questions over irc on the ironic driver, but I need to qualify what the question actually _is_ before asking you guys :)02:13
NobodyCamlol I have game open in other window02:13
jrollheh, ok02:13
jrollmake them talk in channel :P02:13
mrda:)02:13
NobodyCam:)02:13
jrollI'm mostly serious02:14
jrollthat's really lame02:14
mrdajroll: it's all good02:15
jrollok, if you say so02:15
*** nosnos has quit IRC02:16
*** chuckC has joined #openstack-ironic02:19
*** rloo has quit IRC02:25
*** pcrews has quit IRC02:28
*** harlowja is now known as harlowja_away02:52
*** rushiagr is now known as rushiagr_away03:00
*** yuanying has joined #openstack-ironic03:09
*** yuanying has quit IRC03:18
*** zz_naotok is now known as naotok03:22
*** eghobo has joined #openstack-ironic03:32
*** Poornima has joined #openstack-ironic03:42
*** eghobo has quit IRC03:49
openstackgerritHaomeng,Wang proposed a change to openstack/ironic: Add send-data-to-ceilometer support for pxe_ipminative driver  https://review.openstack.org/11248604:06
*** rushiagr_away is now known as rushiagr04:07
*** eghobo has joined #openstack-ironic04:08
*** yuanying has joined #openstack-ironic04:14
*** jogo has joined #openstack-ironic04:16
jogoso I am reviewing the nova patches04:17
jogoand have a few questions if anyone is around04:17
*** yuanying has quit IRC04:22
*** killer_prince has quit IRC04:24
*** killer_prince has joined #openstack-ironic04:38
*** killer_prince is now known as lazy_prince04:39
*** Nisha has joined #openstack-ironic04:40
*** wanyen has quit IRC04:48
*** nosnos has joined #openstack-ironic04:50
*** jogo has left #openstack-ironic04:53
*** rameshg87 has joined #openstack-ironic04:59
*** rakesh_hs has joined #openstack-ironic05:02
*** yuanying has joined #openstack-ironic05:19
*** enikanorov__ has quit IRC05:23
openstackgerritSyed Ismail Faizan Barmawer proposed a change to openstack/ironic: Add uefi boot mode support in IloVirtualMediaIscsiDeploy  https://review.openstack.org/11656105:26
*** yuanying has quit IRC05:27
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386505:28
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386505:33
*** bmahalakshmi has joined #openstack-ironic05:45
openstackgerritOpenStack Proposal Bot proposed a change to openstack/ironic: Imported Translations from Transifex  https://review.openstack.org/11854806:09
*** k4n0 has joined #openstack-ironic06:18
*** eghobo has quit IRC06:21
*** yuanying has joined #openstack-ironic06:24
*** ndipanov_gone has quit IRC06:26
*** eghobo has joined #openstack-ironic06:27
*** lazy_prince is now known as killer_prince06:28
*** ndipanov has joined #openstack-ironic06:29
*** killer_prince has quit IRC06:31
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver  https://review.openstack.org/11588506:31
*** lazy_prince has joined #openstack-ironic06:31
*** yuanying has quit IRC06:32
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver  https://review.openstack.org/11588506:37
*** rameshg87 has quit IRC06:40
*** rameshg87 has joined #openstack-ironic06:40
*** jcoufal has joined #openstack-ironic06:43
*** eghobo has quit IRC06:45
*** lazy_prince has quit IRC06:53
*** lazy_prince has joined #openstack-ironic06:53
*** foexle has joined #openstack-ironic06:56
*** lazy_prince is now known as killer_prince07:03
*** r-mibu has quit IRC07:03
*** bluex-pl has joined #openstack-ironic07:04
*** lsmola has quit IRC07:16
*** yuanying has joined #openstack-ironic07:29
*** wendar_ is now known as wendar07:36
*** yuanying has quit IRC07:38
*** jistr has joined #openstack-ironic07:53
*** killer_prince is now known as lazy_prince08:04
*** lsmola has joined #openstack-ironic08:07
*** MattMan has quit IRC08:09
*** MattMan has joined #openstack-ironic08:10
*** vdrok_afk is now known as vdrok08:10
*** romcheg1 has joined #openstack-ironic08:12
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386508:14
*** athomas has joined #openstack-ironic08:17
*** bluex-pl has quit IRC08:20
*** bluex-pl has joined #openstack-ironic08:25
*** derekh has joined #openstack-ironic08:25
*** bluex-pl has quit IRC08:25
*** lucasagomes has joined #openstack-ironic08:30
*** yuanying has joined #openstack-ironic08:34
*** yuanying has quit IRC08:43
*** bluex-pl has joined #openstack-ironic08:48
*** stelfer has joined #openstack-ironic08:52
*** jistr has quit IRC09:06
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386509:07
*** dtantsur|afk is now known as dtantsur09:07
dtantsurMorning Ironic09:07
romcheg1Good morning dtantsur!09:08
dtantsurromcheg1, morning09:09
*** jistr has joined #openstack-ironic09:10
openstackgerritRamakrishnan G proposed a change to openstack/ironic: Support for setting boot mode in pxe_ilo driver  https://review.openstack.org/11857809:28
*** rameshg87 has quit IRC09:37
*** yuanying has joined #openstack-ironic09:39
*** rameshg87 has joined #openstack-ironic09:44
*** rameshg87 has left #openstack-ironic09:46
stelferHi all - I have a question about ManagementInterface.  Our power driver can only control power outlets.  Boot device is not something we can control.  In our system, we use PXE for controlling deployment/boot parameters.  Isn't there a default ManagementInterface instance we can use when we have no management capability?09:48
*** yuanying has quit IRC09:48
*** yuriyz has quit IRC09:53
dtantsurstelfer, hi! seems like the only option is having management=None09:54
romcheg1dtantsur: will it work?09:54
dtantsurlemme check09:54
stelferAre there side effects to that?09:54
romcheg1dtantsur: I'm going to check the manager09:55
dtantsurit will do nothing if management is None, I checked now09:56
dtantsurstelfer, if boot device for a machine is not PXE, deploy will fail09:56
stelferThat sounds absolutely fine (at least for our config).  With our driver, nothing else is available.09:57
romcheg1yup, otherwise it should work09:57
stelferThank you09:57
*** viktors|afk is now known as viktors09:57
lucasagomesyup as dtantsur said when setting the boot device we check if the management is present, and if not it does nothing (mgmt is not a core interface so it can be None)09:58
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 1  https://review.openstack.org/11630310:04
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 2  https://review.openstack.org/11858710:06
romcheg1lucasagomes: Is it defined in our dev-doc?10:08
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 3  https://review.openstack.org/11858810:08
romcheg1lucasagomes: it === which interfaces are core ones and which are not10:09
lucasagomesromcheg1, maybe not so explicitly http://docs.openstack.org/developer/ironic/dev/architecture.html#drivers10:09
lucasagomeswould be good to say which one are cores and which one are not10:09
romcheg1lucasagomes: yes, I think that may be clarified better :)10:10
*** romcheg1 has quit IRC10:11
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 4  https://review.openstack.org/11859110:11
lucasagomesromcheg, +110:12
dtantsurso translator much cleanup very patches :D10:12
lucasagomesouch10:12
lucasagomeswe better wait til we get the stuff for J3 in... to avoid merge errors etc10:13
dtantsuryeah, clean up does not need FFE I guess10:13
*** mgoddard has joined #openstack-ironic10:15
dtantsurlucasagomes, I was thinking of approving https://review.openstack.org/#/c/107629/8/ironic/tests/base.py but should not we reraise DbMigrationError in case of SQLite here?10:28
lucasagomesdtantsur, hmmmm I think it's a good point to have an else: raise there10:29
lucasagomesit makes sense to me10:29
dtantsurlucasagomes, mind having a look at https://review.openstack.org/#/c/115974/ ? I'd like it to get into J3 so that we can gather more feedback10:34
lucasagomesdtantsur, will do... just finishing something quick here about the drac stuff10:34
vdrokafternoon Ironic!10:41
*** naotok is now known as zz_naotok10:41
dtantsurvdrok, hi10:43
vdrokhi dtantsur10:43
vdrokdtantsur, lucasagomes, this cleanup stuff gets bigger and bigger :)10:44
dtantsuryeah...10:44
*** yuanying has joined #openstack-ironic10:44
*** Nisha has quit IRC10:47
lucasagomesvdrok, yup heh, morning10:47
dtantsurviktors, around? may I reupload https://review.openstack.org/#/c/107629/ ?10:51
*** yuanying has quit IRC10:53
viktorsdtantsur: hi! Sorry, can we discuss it a bit later?10:53
viktorsI'm a bit busy at the moment10:53
*** Poornima has quit IRC10:54
dtantsurviktors, I just wanted to land it before J3. my only point is to reraise exception10:54
dtantsurbut yeah, if you're busy, let us postpone10:54
vdrokdtantsur, so, do you suggest changing % to , only in places that are modified during cleanup, or should I change all of them?10:54
viktorsdtantsur: I'll ping you in a 30-60 minutes, ok?10:54
dtantsurvdrok, my suggestion is to change only what you already change10:54
dtantsurviktors, ack10:55
vdrokdtantsur, ah, ok, will do10:55
*** Poornima has joined #openstack-ironic10:57
*** bmahalakshmi has quit IRC10:58
*** marios has quit IRC10:59
*** bmahalakshmi has joined #openstack-ironic11:00
*** marios has joined #openstack-ironic11:01
*** marios has joined #openstack-ironic11:02
*** comstud has quit IRC11:03
*** comstud has joined #openstack-ironic11:08
*** rakesh_hs has quit IRC11:08
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: Fix minor issues in the DRAC driver  https://review.openstack.org/11839711:10
openstackgerritA change was merged to openstack/ironic: Add send-data-to-ceilometer support for pxe_ipminative driver  https://review.openstack.org/11248611:11
*** Poornima has quit IRC11:14
*** yuriyz has joined #openstack-ironic11:14
openstackgerritA change was merged to openstack/ironic: Reduce redundancy in conductor manager docstrings  https://review.openstack.org/11850511:14
*** Poornima has joined #openstack-ironic11:23
viktorsdtantsur: still around?11:34
dtantsurviktors, yep11:35
viktorsso as for your comment11:35
viktorsdo you really think, that `raise` is required?11:35
viktorsdtantsur: look also at lines 72-75 with the similar case11:36
dtantsurviktors, yes, otherwise devs will have hard time figuring out why their tests do not work11:36
*** pelix has joined #openstack-ironic11:36
* viktors want to drop this class at all11:37
viktorsdtantsur: so what about this logic - https://github.com/openstack/ironic/blob/master/ironic/tests/base.py#L76-L79 ?11:38
dtantsurviktors, what do you mean?11:39
viktorsdtantsur: i mean, than in that case we just check, is db-file exist, if yes - we suppose, that DB was initialized correctly and don't check nothing in this case11:41
dtantsurit also sucks11:42
dtantsurlemme think11:42
*** lucasagomes is now known as lucas-hungry11:42
dtantsurviktors, do you mean that we'll never actually hit this exception?11:43
dtantsurbecause sqlite on file will never get upgraded?11:44
viktorsdtantsur: yes, we are not trying to update sqlite file at all11:46
*** bluex-pl has quit IRC11:48
dtantsurviktors, I see. Your NOTE and if check are misleading then. could you drop the if and leave a comment, that this exception won't be risen for sqlite?11:48
*** yuanying has joined #openstack-ironic11:49
openstackgerritSyed Ismail Faizan Barmawer proposed a change to openstack/ironic: Add UEFI based deployment support in Ironic  https://review.openstack.org/11435711:51
viktorsdtantsur: I see no big sense in it, bit I'm ok with this. Also its would be nice to know also lucas-hungry opinion about it, because yesterday we decided to keep old logic in Database.__init__()11:51
*** bmahalakshmi has quit IRC11:51
*** bluex-pl has joined #openstack-ironic11:52
bluex-pljroll: hi, I would be glad if you could take a look at https://bugs.launchpad.net/ironic/+bug/1314148/comments/211:53
dtantsurviktors, I mean the comment says "If we use SQLite with the old DB schema, we" which is misleading, because this case cannot happen IMO11:53
dtantsuryou won't reach this line11:53
*** yuanying has quit IRC11:58
*** rameshg87 has joined #openstack-ironic11:59
* viktors thinking11:59
*** lsmola has quit IRC12:01
openstackgerritVinay B S proposed a change to openstack/ironic: Adds instructions for deploying instances on real hardware  https://review.openstack.org/11861412:01
*** HenryG is now known as HenryG_afk12:04
openstackgerritVinay B S proposed a change to openstack/ironic: Adds instructions for deploying instances on real hardware  https://review.openstack.org/11861412:04
*** lsmola has joined #openstack-ironic12:10
*** romcheg1 has joined #openstack-ironic12:14
*** nosnos has quit IRC12:16
*** nosnos has joined #openstack-ironic12:17
*** nosnos has quit IRC12:21
*** rameshg87 has left #openstack-ironic12:23
viktorslucas-hungry, dtantsur - folks, what do you think,m if I'll rollback changes in file ironic/tests/base.py at all?12:24
viktorsin patch https://review.openstack.org/#/c/10762912:25
*** bluex-pl has quit IRC12:37
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 1  https://review.openstack.org/11630312:40
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 2  https://review.openstack.org/11858712:41
dtantsurviktors, it may make sense to split into 2 patches and discuss separately12:41
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 3  https://review.openstack.org/11858812:42
openstackgerritVladyslav Drok proposed a change to openstack/ironic: Translator functions cleanup part 4  https://review.openstack.org/11859112:42
*** rameshg87 has joined #openstack-ironic12:43
*** vdrok is now known as vdrok_afk12:43
*** rloo has joined #openstack-ironic12:47
*** rloo has quit IRC12:47
openstackgerritVictor Sergeyev proposed a change to openstack/ironic: Use metadata.create_all() to initialise DB schema  https://review.openstack.org/10762912:52
*** igor has joined #openstack-ironic12:52
*** lucas-hungry is now known as lucasagomes12:52
*** igor has quit IRC12:52
viktorsdtantsur: see ^^12:53
*** igor has joined #openstack-ironic12:53
* dtantsur is looking12:53
*** rameshg87 has quit IRC12:53
*** yuanying has joined #openstack-ironic12:54
dtantsurviktors, I like it. please feel free to raise base.py change on top of it12:55
dtantsurlucasagomes, please have a look/approve too ^^^12:55
lucasagomesdtantsur, ack will do now12:55
*** faizan has joined #openstack-ironic12:56
*** igor has quit IRC12:56
lucasagomesviktors, dtantsur ack yeah if this changes can be separated I think it makes sense to have the two patches and discuss them individually12:56
lucasagomesjust approved the 10762912:57
viktorslucasagomes, dtantsur - ok, I'll try to do something with DB testing :)12:58
lucasagomesviktors, ta much!12:58
viktorslucasagomes: thank you!12:58
*** enikanorov has joined #openstack-ironic13:01
*** rloo has joined #openstack-ironic13:02
viktorsby the way, is there any volunteers to fix documentation for ironic installation and close bug 1347604 ? :)13:02
*** yuanying has quit IRC13:03
openstackgerritYuriy Zveryanskyy proposed a change to openstack/ironic: Control extra space for images conversion in image_cache  https://review.openstack.org/11597413:07
*** rloo has quit IRC13:08
*** rloo has joined #openstack-ironic13:08
*** rloo has quit IRC13:08
*** rloo has joined #openstack-ironic13:09
*** rloo has quit IRC13:09
*** rloo has joined #openstack-ironic13:10
*** rloo_ has joined #openstack-ironic13:11
dtantsurlucasagomes, mind having one more look ^^^?13:13
*** rloo_ has quit IRC13:13
*** linggao has joined #openstack-ironic13:13
dtantsurah I see13:14
*** rloo_ has joined #openstack-ironic13:15
*** rloo has quit IRC13:16
*** rloo_ has quit IRC13:16
*** enikanorov__ has joined #openstack-ironic13:17
*** rloo has joined #openstack-ironic13:19
*** enikanorov has quit IRC13:19
rloohi Ironickers13:20
rloolucasagomes: when you have a minute, I added a comment to 118397 but Shrews already approved. Want to make sure you see it.13:21
Shrewsrloo: oh, was the comment there before i approved?13:21
rlooShrews: no, I'm just too slow. You approved before I added the comment, but I think it is worth getting an answer to my question.13:22
Shrewsrloo: ah. sorry about that13:22
lucasagomesrloo, will do13:22
lucasagomesdtantsur, did already13:23
rlooShrews: no worries. It is great that we have so many people reviewing at the same time :D13:23
dtantsurrloo, hi13:24
rloohi dtantsur13:24
*** rameshg87 has joined #openstack-ironic13:24
lucasagomesrloo, yeah :( as I'm using the default dialect there I think we are safe13:24
lucasagomesbut yeah it would be good to have that catch thre13:24
lucasagomesthere*13:24
rloolucasagomes: ok, do you want me to open a bug about it?13:25
rameshg87lucasagomes: rloo: please have a look at ilo-virtual-media driver review: https://review.openstack.org/11386513:25
lucasagomesrloo, hmm I can toss a patch up for that real quick, idk if we need a bug for it now (as we use the default there, it should never raise that exception)13:26
lucasagomesrloo, that works?13:26
rloolucasagomes: i'm fine if you toss a patch up. Wasn't sure you were going to ;)13:26
lucasagomesrloo, ack, well I've the code already open here cause I was doing that changes13:26
lucasagomes1 sec13:26
rloorameshg87: I've got a whole pile of patches to review. I decided to go on strike and not review any more til kilo. (just joking)13:27
rameshg87rloo: :)13:27
rameshg87rloo: you could just complete the patches that you started reviewing :D13:27
rloorameshg87: no, I don't always do that cuz I don't work every day and I don't want things delayed because I'm not around. Also, it takes me a long time to do a review and I find it hard to iterate too many times on the same patch.13:28
rameshg87rloo: okay :) no problems13:29
rloorameshg87: I suppose I could just select a small number of patches instead and only focus on those.13:29
rameshg87rloo: okay13:30
*** bluex-pl has joined #openstack-ironic13:34
*** jasondotstar has joined #openstack-ironic13:39
jrollmorning lucasagomes dtantsur Shrews rloo rameshg87 bluex-pl and the rest of ironic :)13:41
jroll(busy day?)13:41
lucasagomesjroll, morning13:41
rloomorning jroll13:41
dtantsurjroll, morning13:41
rloorameshg87: wrt 113865. https://review.openstack.org/#/c/113865/23/ironic/drivers/modules/ilo/common.py13:44
rloorameshg87: line 30513:44
rloorameshg87: what I dont' like is the '6'13:44
jrollbluex-pl: read that comment, can you propose that as a new version of your patch? don't worry about tests yet, but I want to see what it looks like13:44
jrollbluex-pl: I expect that might be an ok solution13:44
rloorameshg87: if the string changes from 'swift:' to 'foo:', then it becomes 4, not 6. IF someone remembers to change that number.13:45
rloorameshg87: so if you split boot_iso into key, value, you can check key==swift, and just use value.13:45
*** faizan has quit IRC13:46
jrollfaizan nooooooo don't go :(13:46
*** r-daneel has joined #openstack-ironic13:47
openstackgerritA change was merged to openstack/ironic: Fix minor issues in the DRAC driver  https://review.openstack.org/11839713:47
jrollgah uefi is so ready13:49
Shrewsjroll: how ready is it? ....13:49
jrollrloo: there's one nit here, but will you review this to make sure your concerns are taken care of? https://review.openstack.org/#/c/114357/13:49
jrollother than that lgtm13:49
jrollShrews: needs one little nit fixed, I might just do it13:50
Shrewsjroll: oh, i thought that was the opening line to a uefi joke  :-P13:50
rloojroll: I'm starting to think that if things are almost ready, I shouldn't look.13:50
jrolllol13:50
jroll^ for both of you :P13:50
rameshg87jroll: good morning13:55
jrollheya rameshg87 :)13:55
*** chuckC has quit IRC13:55
rameshg87rloo: just reading your comments13:55
*** dhellmann has quit IRC13:57
dtantsurjroll, left more comments there13:57
rloorameshg87: i started reviewing your patch, but jroll asked me to look at uefi so yours is in the queue again.13:57
rameshg87rloo: i can change if you really insist, but my defense was it was all internally generated within the code. so splitting them into key-value pair was just a overhead :)13:58
*** dhellmann has joined #openstack-ironic13:58
rameshg87rloo: there is no chance of us getting 'foo' in there unless someone writes new code without reading the current code13:58
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: Handle DracInvalidFilterDialect exception  https://review.openstack.org/11863813:59
lucasagomesrloo, ^13:59
rameshg87rloo: no problem.13:59
rloorameshg87: so it doesn't have to be split, that was a suggestion. I just don't like the hardcoded 6. you can even do 'len('swift:') or whatever.13:59
rloolucasagomes: thx13:59
*** romcheg1 has quit IRC13:59
*** yuanying has joined #openstack-ironic14:00
jrolldtantsur: ok, you think I should push a new version or wait for faizan? want to land this today14:00
jrolllucasagomes: ^^14:00
*** romcheg1 has joined #openstack-ironic14:00
rameshg87rloo: ah okay, i get it. i am okay with changing that.14:00
lucasagomesjroll, +1 for landing it today14:00
jrollok14:00
* jroll fixes things14:00
lucasagomesI think it's mostly nits now, so I don't see a problem in pushing a new quick patch-set for it14:01
rameshg87rloo: len('swift:') i will change that ...14:01
lucasagomesjroll, thx!14:01
dtantsurjroll, I think it's ok for you to fix14:01
rameshg87lucasagomes: jroll: is mine ilo-virtual-media-iscsi in your queue ? :)14:01
lucasagomesrameshg87, yes!14:01
rameshg87rloo: i will wait for you if you have any more comments.  if only nits, i will make sure, i will fix it in the next patch. is that fine ?14:02
rloorameshg87: you realize, you won't like it if you have 'swift:' in two places. similarly for 'glance:'14:02
dtantsureverything targeted to J3 is a priority today14:02
rloorameshg87: yeah, that's fine. it may take me another hour or 2 to get through yours.14:03
rameshg87rloo: yeah, okay :(.  if it looks too ugly, i will resort to key-value pair.14:03
*** sanek11 has quit IRC14:03
*** lazy_prince is now known as killer_prince14:03
*** pcrews has joined #openstack-ironic14:04
openstackgerritStig Telfer proposed a change to openstack/ironic: Adds SNMP power driver  https://review.openstack.org/11840414:06
dtantsurrloo, lucasagomes, do we really need to catch DracInvalidFilterDialect? the only possible cause is programming error IMO14:07
*** yuanying has quit IRC14:08
lucasagomesdtantsur, yeah, that's true... it's super unlikable to happen... I just put it there cause it's a valid exception that can be raised by that method14:08
lucasagomesbut yeah, dtantsur to be honest I'm happy handling or not handling it really14:09
rloodtantsur, lucasagomes: that's the question I had for you. Wanted to make sure that it had been thought of, instead of forgotten to think about it.14:09
rloodtantsur, lucasagomes: if that made sense...14:09
lucasagomesrloo, oh14:09
lucasagomesrloo, right... hmm yeah as dmitry pointed I don't see how that can be raised if not by an mistake when coding14:10
dtantsurrloo, for me it's just yet another possible programming error, like ValueError14:10
rloodtantsur, lucasagomes: ie, if it can only be via program error, then maybe you want it to be raised14:10
lucasagomesrloo, I will abandon that patch14:10
rloodtantsur: sometimes ValueError needs to be caught. It depends...14:10
rloolucasagomes: that's fine. And I won't look to see if the docstring needs to be updated with "raises..."14:11
lucasagomesrloo, oh there's that too14:11
* lucasagomes thinking14:11
*** stelfer has quit IRC14:11
lucasagomeslet's leave it there and see what other folks think about it14:11
* rloo hates handling errors etc14:12
* lucasagomes brain's not working properly 14:12
NobodyCamgood morning Ironic14:17
rloomorning NobodyCam14:18
openstackgerritJim Rollenhagen proposed a change to openstack/ironic: Add UEFI based deployment support in Ironic  https://review.openstack.org/11435714:18
jrollmorning NobodyCam :)14:18
NobodyCammorning rloo jroll14:18
lucasagomesmorning NobodyCam14:18
NobodyCammorning lucasagomes :)14:21
NobodyCamwhats left to land, and can we land it14:21
NobodyCamlol :)14:21
jrolllucasagomes dtantsur NobodyCam https://review.openstack.org/11435714:22
jrollI think that's ready14:22
dtantsuralready looking :)14:22
jrollstill don't love the bit about setting boot device, but I don't think there's much more we can do :(14:22
romcheg1Morning NobodyCam, jroll, rloo!14:22
jrollhiya romcheg1 :)14:22
rloohi romcheg114:22
rloojroll: wrt 114357, pxe_utils.py, _link_ip_address_pxe_configs(). still want to know what happens if _get_pxe_ip_address_path() raises InvalidIPv4Address.14:23
Shrewslucasagomes: rloo: dtantsur: my view is that exception handling should be done for runtime errors, not programming errors14:23
Shrewsso, i'd leave it as is, but that's only my opinion14:24
openstackgerritStig Telfer proposed a change to openstack/ironic: Adds SNMP power driver  https://review.openstack.org/11840414:24
jrollrloo: oops, I missed that14:24
jrollrloo: what I wanted to do, was validate the ip address when we get it from neutron14:25
jrollrloo: and don't return it in get_ip_addresses if it is not valid14:25
jrollhow does that sound?14:25
rloojroll: that's fine with me. what happens if/when we handle ipv6?14:26
jrolllord.14:26
rloojroll: just update the code in neutron then. forget i asked.14:26
jrollhonestly, I doubt elilo supports it14:26
jrollbut yeah14:26
lucasagomesShrews, thanks, yeah I'm grand leaving as is14:26
lucasagomesI will abandon that patch, the more I think the less sense it makes14:26
rloolucasagomes, Shrews: yeah, I'm fine with that ;)14:26
dtantsurjroll, once rloo's concern is fixed, I'm ok with the code14:26
jrollok14:27
jrollsec14:27
lucasagomesrloo, Shrews dtantsur thanks14:27
openstackgerritSzymon Wróblewski proposed a change to openstack/ironic-python-agent: Remove unnecessary argument injection in agent sync and async decorators.  https://review.openstack.org/11752914:27
*** rushiagr is now known as rushiagr_away14:28
toabctlGheRivero: hey. friendly reminder for another review of https://review.openstack.org/#/c/117704/14:32
*** jcoufal_ has joined #openstack-ironic14:32
*** jcoufal has quit IRC14:34
bluex-pljroll: finally found some time to push it, take a look when you can14:34
jrollbluex-pl: thanks, will look later on14:35
jrollrussell_h: can you look at this as well? https://review.openstack.org/11752914:35
*** HenryG_afk is now known as HenryG14:36
jrollbluex-pl: it is kind of duplicating the name between command_map and the decorator, but idk, might be good14:36
rloojroll: any idea if uefi would be used with ipminative?14:37
*** chuckC has joined #openstack-ironic14:38
jrollrloo: good question, does it matter? :)14:38
jrollrloo: there's the whole thing with... nobody is using ipminative14:39
rloojroll: yes, but I decided that someone can change the comments/msg in that case.14:39
rloojroll: asking cuz 'ipmitool' is mentioned in the patch. but forget about it.14:39
jrollmmm yeah, I would assume ipminative behaves the same way14:39
jrollso maybe that's bad...14:40
rlooit is bad, when/if it happens. so better to try to address now, but i know we're in a hurry.14:40
jrollI mean, I can push more14:40
jrollabout to push another anyhow for the invalid ip thing14:40
jrollI can just s/ipmitool/ipmi/ there I think?14:41
rloojroll: ok, if you don't mind.14:41
*** rameshg87 has quit IRC14:41
rloojroll: yeah. if some other mgt interface is used in the future, someone else can update it further. but at least it'll be ok for ipmi.14:41
openstackgerritJim Rollenhagen proposed a change to openstack/ironic: Add UEFI based deployment support in Ironic  https://review.openstack.org/11435714:41
jrollyeah14:41
jrollI would expect that other management interfaces will work ok14:42
rloojroll: and that's it for my comments. will wait for your next patch.14:42
jrollI don't love having references to ipmi in the pxe driver code, but if only ipmi fails, I think it makes sense14:42
jrollrloo: just pushed :P14:42
rloojroll: you're fast!14:42
jrollI was looking at it after you asked :P14:43
jrollwas waiting for tests to run14:43
*** romcheg1 has left #openstack-ironic14:43
*** romcheg1 has joined #openstack-ironic14:44
rloojroll: I don't actually like raising ConfigInvalid exception from a validate() because I'm afraid of the ramifications (eg, code that handles Invalid/Missing might now have to be updated to handle ConfigInvalid).14:45
rloojroll: sorry, I should just shut up.14:45
jrolllol14:45
jrollI mean... that's a good point14:45
jrollwhat would you prefer? invalidparameter whatever its called?14:46
jrollI'm almost wondering if that's the right thing to do14:46
rloojroll: I think Invalid would be easier to deal with.14:46
jrollif one is running with ipxe enabled, they shouldn't configure a node that way14:46
rloojroll:  or at least an exception that is subclassed from Invalid.14:46
jrollhrm14:47
rloojroll: it does bother me that ipxe is enabled 'globally', but uefi is enabled node by node only.14:47
jrollright... I was just thinking about that14:47
jrollbut ipxe is more of an environmental thing, I think14:48
jrollI can't imagine why one would want half ipxe, half not-ipxe14:48
rloowhich is what made me think that a node's settings maybe overrides an environmental setting.14:48
jroll(actually, can't imagine why one would want anything non-ipxe, but that's a different topic)14:48
jrollmmm14:48
rloobut that could be a separate patch.14:49
jrollso drop to "regular pxe" if the node is configured for uefi?14:49
rlooyeah.14:49
jrollinteresting14:49
jrollthat should definitely be a separate patch14:49
rlooif that does happen, then we wouldn't want to raise any exception cuz there wouldn't be a conflict, so seems safer to use InvalidP.. instead of ConfigInvalid14:50
jrollyeah; agree14:51
*** rameshg87 has joined #openstack-ironic14:51
jrollthere we are14:52
openstackgerritJim Rollenhagen proposed a change to openstack/ironic: Add UEFI based deployment support in Ironic  https://review.openstack.org/11435714:52
* jroll brb14:52
*** jcoufal_ has quit IRC14:55
*** rushiagr_away is now known as rushiagr14:59
*** yuanying has joined #openstack-ironic15:04
bluex-pljroll: yeah, I think decorators could update command_map automatically (http://paste.openstack.org/show/105253/)15:05
*** bluex-pl has quit IRC15:11
jrollneat15:11
*** yuanying has quit IRC15:14
*** jcoufal has joined #openstack-ironic15:22
*** rloo has quit IRC15:29
*** rloo has joined #openstack-ironic15:29
NobodyCambrb15:31
*** rameshg871 has joined #openstack-ironic15:38
rameshg871rloo: i am just fixing the issue straightaway along with nits from dtantsur15:38
*** bmahalakshmi has joined #openstack-ironic15:40
*** rameshg87 has quit IRC15:40
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386515:41
*** rameshg87 has joined #openstack-ironic15:41
*** rameshg871 has quit IRC15:43
*** chuckC has quit IRC15:45
*** bandicot has joined #openstack-ironic15:45
*** Poornima has quit IRC15:49
*** chuckC has joined #openstack-ironic15:53
*** igordcard has joined #openstack-ironic15:53
*** foexle has quit IRC15:56
*** eghobo has joined #openstack-ironic15:56
dtantsurcalling it a day, see you15:57
rloobye dtantsur15:58
Shrewsgoodnight dtantsur15:58
romcheg1Bye dtantsur!15:58
*** dtantsur is now known as dtantsur|afk15:58
NobodyCamnight dtantsur|afk15:59
*** sirushti_ has joined #openstack-ironic16:03
*** pelix1 has joined #openstack-ironic16:04
*** ndipanov has quit IRC16:05
*** victor_lowther_ has joined #openstack-ironic16:07
*** NobodyCa1 has joined #openstack-ironic16:07
*** sirushti has quit IRC16:08
*** pelix has quit IRC16:08
*** MattMan has quit IRC16:08
*** bandicot has quit IRC16:08
*** jasondotstar has quit IRC16:08
*** k4n0 has quit IRC16:08
*** dividebin has quit IRC16:08
*** rushiagr has quit IRC16:08
*** victor_lowther has quit IRC16:08
*** NobodyCam has quit IRC16:08
*** krtaylor has quit IRC16:08
*** GheRivero has quit IRC16:08
*** dividehex has joined #openstack-ironic16:08
*** Ugallu has joined #openstack-ironic16:08
*** sirushti_ is now known as sirushti16:08
*** k4n0_ has joined #openstack-ironic16:09
*** rushiagr has joined #openstack-ironic16:10
*** yuanying has joined #openstack-ironic16:10
*** MattMan has joined #openstack-ironic16:10
*** krtaylor has joined #openstack-ironic16:10
*** igordcard has quit IRC16:10
*** krtaylor has quit IRC16:11
*** GheRivero has joined #openstack-ironic16:13
*** scubacuda has joined #openstack-ironic16:15
*** victor_lowther_ is now known as victor_lowther16:15
*** romcheg2 has joined #openstack-ironic16:16
*** romcheg1 has quit IRC16:17
*** yuanying has quit IRC16:18
NobodyCa1so 114357 is good to land after mr approves?16:20
*** dlaube has joined #openstack-ironic16:21
JayFI think so, it's been through quite the review gamut :)16:21
NobodyCa1lol and 4 +2s at this point :)16:23
*** NobodyCa1 is now known as NobodyCam16:24
NobodyCamjust the tempest test left to go16:26
*** igordcard has joined #openstack-ironic16:32
NobodyCambrb16:33
*** romcheg1 has joined #openstack-ironic16:33
*** romcheg2 has quit IRC16:35
*** jistr has quit IRC16:35
*** krtaylor has joined #openstack-ironic16:37
*** igordcard has quit IRC16:40
*** romcheg1 has left #openstack-ironic16:40
NobodyCamromcheg: still fix'n up packages?16:40
*** romcheg1 has joined #openstack-ironic16:42
NobodyCamrameshg87: you around?16:42
rameshg87NobodyCam: yes16:44
rameshg87NobodyCam: i was about to ping you :-)16:44
NobodyCamhehehe :) rameshg87 up for a rebase on https://review.openstack.org/#/c/115885?16:44
rameshg87NobodyCam: yes, but waiting for the parent to get approved so that i can avoid more rebasing :)16:45
rameshg87NobodyCam: shall i rebase it again now ?16:45
NobodyCamlet /me take a look at dep now16:46
rameshg87NobodyCam: okay16:47
rameshg87JayF: would you also like to take a look at https://review.openstack.org/#/c/113865/16:49
JayFrameshg87: I have a vote on it :)16:49
rameshg87JayF: for a +2 :)16:49
JayFI am not a core on Ironic16:49
JayFonly Ironic-specs and Ironic-python-agent16:49
rameshg87JayF: oh okay :)16:50
rameshg87lucasagomes: still around for a review on https://review.openstack.org/#/c/113865/16:51
JayFI think we'll all be looking at that today :)16:52
ShrewsJayF: your comment on 118467 about separating power states... you mean put them in another file? Or separating them visually within the produced documentation?16:52
JayFit's gotta land for j-316:52
JayFShrews: yes.16:52
JayFShrews: one or the other16:52
JayFShrews: just as it is now it's all mixed together16:52
ShrewsJayF: i can't find a way to manipulate docstrings to do it within a single module16:53
JayFwhat's the review # for that?16:53
Shrewsanyway, that's low priority16:53
Shrews11846716:53
JayFShrews: even whitespace, or something would be helpful16:54
JayFShrews: like # NOTE(shrews) All states below here are power states16:54
jrollShrews: I'd argue we should not make any actual code changes there... just to not touch the nova reviews16:54
JayFjroll: I have that comment on there already :)16:55
Shrewsjroll: yeah, well that's why it's WIP'd16:55
jrollalso, I think init may come in useful one day16:55
jrollright16:55
Shrewsit was just bugging me16:55
jrollJayF: right, I think I worded it different :p this shpuld just be docs/comments changes IMO16:56
rameshg87NobodyCam: i am just rebasing 115885 anyway , hoping the parent will go through :)16:56
rameshg87jroll: got some time for a review on https://review.openstack.org/#/c/113865/ ? :)16:57
jrollrameshg87: I'm on mobile right now but will look later today :)16:58
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver  https://review.openstack.org/11588516:58
jrollrameshg87: I want to land all the ilo stuff today, will make it happen16:58
jrollrameshg87: you okay with us fixing nits in those if you aren't around?16:59
*** rameshg871 has joined #openstack-ironic16:59
jrollrameshg87: I'm on mobile right now but will look later today :)17:00
jrollrameshg87: I want to land all the ilo stuff today, will make it happen17:00
jrollrameshg87: you okay with us fixing nits in those if you aren't around?17:00
openstackgerritLucas Alvares Gomes proposed a change to openstack/ironic: Driver merge review comments from 111425  https://review.openstack.org/11869317:01
*** derekh has quit IRC17:01
lucasagomes^^ folks17:01
lucasagomesrameshg87, will do17:01
rameshg871lucasagomes: thanks :)17:01
*** rameshg87 has quit IRC17:01
rloolucasagomes: I'm looking at 111425 now. Already have a nit :-(17:03
*** jcoufal has quit IRC17:03
lucasagomesrloo, cool, post there and I can update that ironic patch17:03
lucasagomesactually if you can please update that patch too17:03
lucasagomesjust put a new patch-set fixing the nit17:04
rloolucasagomes: ok, i only started looking and am eating lunch at the same time so may take a bit of time ...17:04
lucasagomesrloo, no bothers, we may also have to talk to mrda to sync it over17:04
lucasagomesso it's cool17:04
rloolucasagomes: okay17:04
*** yuanying has joined #openstack-ironic17:15
rloohey lucasagomes or anyone, question about the driver: https://review.openstack.org/#/c/111425/19/nova/virt/ironic/driver.py17:15
rlooin _validate_instance_and_node()17:16
rlooit just says that the instance exists if it returns true, doesn't it? I don't see where the 'node' comes in.17:16
*** chuckC has quit IRC17:16
*** MattMan has quit IRC17:17
rloo^^ NobodyCam, do you know?17:17
lucasagomesrloo, you mean in the instance_exists() ?17:18
lucasagomesyeah the docstring is not super well worded17:18
rloolucasagomes: no, I mean the _validate_instance_and_node() method.17:18
lucasagomesrloo, that checks if the instance uuid is already associated with some node in Ironic17:19
rloolucasagomes: it just looks like it validates that there is *a* node associated with that instance.17:19
lucasagomesyup17:19
*** pelix1 is now known as pelix17:19
rloolucasagomes: but the docstring seems to indicate something else. that it is associated with 'this node'?17:19
lucasagomesthat reruns the node object as well17:20
lucasagomesreturns17:20
lucasagomesrloo, oh... I see hmm17:20
*** harlowja_away is now known as harlowja17:20
lucasagomesthis node makes no sense because that's the driver17:21
lucasagomesthe node uuid actually is _in_ the instance object17:21
lucasagomesI mean... urgh...17:22
rloolucasagomes: but nothing here checks that the node's uuid (in the instance object) is the same uuid as the returned node17:22
lucasagomesrloo, yeah... AFAIK instance has an attribute 'node' that is the node uuid17:22
lucasagomesso it may make sense to check whether it's same17:22
lucasagomes:(17:23
lucasagomesand ofc reword that docstring17:23
*** yuanying has quit IRC17:23
rlooi don't really want to change code at this point. let me see who/what calls this method, etc...17:23
rlooso nothing checks that the node's uuid is the same uuid as the instance thinks it should be.17:25
*** stelfer has joined #openstack-ironic17:26
lucasagomesrloo, yeah, I think it makes sense to check that too... maybe that worth open a bug17:26
lucasagomesbecause InstanceNotFound may not be the appropriated error for that case17:27
lucasagomesthe instance exists the node is not correct17:27
lucasagomesgotta take a deeper look into it17:27
rloolucasagomes: so should I add a comment about it in the patch, or pretend I didn't see it?17:27
lucasagomesrloo, hah hmmmm17:29
* lucasagomes thinking 17:29
lucasagomesrloo, I really dunno... well I think we should be transparent17:29
lucasagomesmaybe put a comment and say that we will open a bug about it17:29
rlooor should i wait and ask mrda? i can put a comment and then we can put another comment saying it'll be fixed in another patch?17:29
lucasagomesI don't even know if that case is possible without investigating it17:29
lucasagomesbut we have to be transparent in the reviews and put our concerns there IMO17:30
rlooor we can update the docstring to reflect what it is really doing.17:30
rloosec, let me see if i can figure out who wrote that method. maybe they will know...17:30
lucasagomesrloo, yeah also17:30
lucasagomesrloo, you can put the docstring changes on that patch I put up fixing the other problem with 11142517:30
rlooeww, i have to fix the docstring? :-)17:31
lucasagomeslol17:31
lucasagomesrloo, please?17:31
rlooha ha. yeah, after i finish going through the patch.17:32
lucasagomesack17:32
lucasagomesbecause I'm almost calling it a day here17:32
lucasagomesit's late and I gotta go out buy some food for dinner17:32
rloolucasagomes: yeah, go home!17:33
lucasagomesalright yeah... I'm going to call it a day!17:35
lucasagomeshave a good night everybody!17:35
lucasagomessee ye tomorrow17:35
rloogoodnight lucasagomes17:35
*** stelfer has quit IRC17:37
*** lucasagomes is now known as lucas-dinner17:37
NobodyCamnight lucas-dinner17:43
NobodyCamlucas-dinner: I'm going  to land 11386517:44
NobodyCamrameshg871: 113865 has several nit's could / would you be willing to address them in a new patch?17:45
lucas-dinnerNobodyCam, ack17:46
NobodyCam:)17:46
*** pcrews has quit IRC17:47
rameshg871NobodyCam: i can address them ...17:48
rameshg871NobodyCam: alternatively i could address them 115885 as well if you like :)17:48
rameshg871NobodyCam: the patch which is dependent on it17:49
rameshg871lucas-dinner: regarding moving to vendor.py, the intention was to keep the related vendor passthrus together.  currently we have the vendor-passthru for deploy which is in deploy.py17:50
NobodyCamrameshg871: up to you. but I'm ok with landing 113865 as is, if you can address the nit in 115885 or a new patch I'm ok with that too17:51
rameshg871NobodyCam: i will address them in 115885 itself17:52
NobodyCamthen I am landing 113865 now !!!!17:52
rameshg871NobodyCam: thanks ... i will start addressing them in 115885 then ...17:52
NobodyCam+a'd :)17:53
rameshg871NobodyCam: thanks :)17:54
openstackgerritA change was merged to openstack/ironic: Control extra space for images conversion in image_cache  https://review.openstack.org/11597417:55
ShrewsNobodyCam: nooooooo17:55
* Shrews too slow17:55
NobodyCamhuh17:59
NobodyCamwhat17:59
NobodyCamwait17:59
Shrewsrameshg871: i left additional comments on 113865 for you. now i know how rloo felt this morning when I approved a review out from under her  :)17:59
NobodyCam:(17:59
ShrewsNobodyCam: no worries. all stuff that can be addressed later.17:59
NobodyCam:)17:59
NobodyCamphew17:59
NobodyCamjroll: JayF: we (I) will need lots of your help with 115885, for the agent side of things18:02
jrollNobodyCam: hello hello, I can help18:03
NobodyCammorning jroll :) have some coffee18:04
NobodyCamlol18:04
jrollhehe, thank you!18:04
NobodyCamjust pointing out that i am not a IPA expert and would like your eyes on 115885!!18:04
jrollyep :)18:05
jrollwill review18:05
NobodyCam:)18:05
jrolldid we land ilo/iscsi yet/18:05
NobodyCamty18:05
rameshg871Shrews: just looking at it now18:05
rameshg871Shrews: will address them in follow up patch18:05
jrolloh niiiice18:05
jrollwe did18:05
jroll\o/18:05
NobodyCam:)18:05
Shrewsrameshg871++18:05
jrollJayF: going to want your eyes on 115885 too18:05
rlooShrews: still good to add comments, even though it is 'too late' :-)18:06
JayFyup I know :)18:06
rameshg871Shrews: NobodyCam: i have a doubt in general. do we document exceptions for all the functions, even for private internal methods within a file ? :)18:06
JayFwas waiting for the one above it to land18:06
Shrewsrameshg871: everyone else can correct me if i'm wrong, but if a function can raise any exception (even from other functions it calls), it should be documented.18:07
rameshg871Shrews: but what if the function doesn't itself raise, but one of the called function raises ?18:08
NobodyCamShrews: I like to see that.. I have yet to block a patch for not doing it18:08
rameshg871Shrews: do we document the exceptions of the called functions too ?18:08
Shrewsrameshg871: if you aren't handling them, and they can be propogated up, i would think so18:09
NobodyCam++18:09
Shrewsthat way, a programmer hoping to use that function knows they need to handle those exceptions18:09
rlooShrews, NobodyCam, rameshg871: for noninternal functions (ie, those not prefixed with _), should document all exceptions that can be raised by it (or any called functions).18:09
rloothat's what I was told18:10
openstackgerritA change was merged to openstack/ironic: Use metadata.create_all() to initialise DB schema  https://review.openstack.org/10762918:10
rloowrt internal functions. I think they should do the same too, but I was told that they either shouldn't or don't need to.18:10
rloo(I forgot the exact details).18:10
rameshg871rloo: okay ..18:10
Shrewsyeah, i don't think we have it written down anywhere, but i still like to see it18:11
rameshg871rloo: Shrews: just wondering if we document all the exceptions of all the called methods, won't it pile up the list of exceptions in docstring :)18:11
jrollI think for internal functions, it's not required, I still believe they are nice to have18:12
rlooam thinking I should have tried to capture that info. it is probably in IRC somewhere, a discussion I had with deva I think. many months ago.18:12
rloorameshg871: if you were a programmer, would you want to know what kind of exceptions might be raised?18:12
rloorameshg871: sorry, I don't mean to say that you aren't a programmer18:12
rameshg871rloo: yeah, i get it :)18:12
Shrewsrloo: Perhaps we should start compiling an "Ironic Best Dev Practices" wiki?18:12
jrolllol, harsh :P18:12
rameshg871Shrews: ++18:13
rlooShrews, I think that's a great idea.18:13
rloodo we even know when/if we should put docstrings for methods? :-)18:14
Shrewswhen we get -1'd b/c we don't have it?  ;)18:14
rameshg871Shrews: LOL18:14
Shrewsoh yay. that time of day when my laptop starts acting flakey. /me reboots18:17
*** yuanying has joined #openstack-ironic18:20
NobodyCamwoo hoo just added a item to the NOVA agenda: https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting18:20
NobodyCamlooks like I need to be up and awake at 7am tomorrow :-p18:20
jrollNobodyCam: we also have our graduation review on the 9th :o18:20
NobodyCamjroll: ya just figured I'd ask how they want us to attempt to land the proxy api commands stuff18:21
jrollyeah18:22
jrollthey haven't given you a good answer yet?18:22
rloowill we need a feature freeze exception for the driver?18:22
jrollgod I hope not18:22
NobodyCamoh should ask for one just in case or would that take the pressure of the nova cores from reviewing our patches?18:23
jrollshould talk to mrda18:23
rloomaybe mrda would know18:23
jrollI think he's been talking to nova about it18:23
NobodyCamyea I think he has been18:23
NobodyCamwill hit him up today18:24
NobodyCamI can add it a ffe request to their agenda if needed18:24
*** yuanying has quit IRC18:28
*** pcrews has joined #openstack-ironic18:28
jrollthis ipa thing is great18:31
jrollrameshg871: ^^ :)18:31
openstackgerritRishabh Jain proposed a change to openstack/ironic: Spelling and Grammar mistakes fixed  https://review.openstack.org/11873318:32
*** rushiagr is now known as rushiagr_away18:33
openstackgerritlinggao proposed a change to openstack/ironic: Interactive console support for ipminative driver  https://review.openstack.org/9733118:36
rameshg871jroll: thanks :-)18:37
rameshg871NobodyCam: i think i will need a manual rebase after submission of https://review.openstack.org/#/c/115974/18:38
rameshg871NobodyCam: this is for the approved patch ...18:39
rameshg871NobodyCam: shall i just raise a new patchset after rebasing ? i think jenkins will fail anyway eventually18:40
rameshg871jroll: ^^ what's your thought on this18:41
jrollare you sure 115974 will land first?18:41
rameshg871jroll: 115974 has already landed18:42
jrollI mean, merge18:42
jrolloh18:42
jrollit did, hmm18:42
jrollit's up to you if you want to wait and see or just do it18:42
jrollyou could do the rebase locally18:42
rameshg871jroll: i think i will just do it..it's failing for me locally18:42
jrollif you get conflicts, ok18:42
jrollyeah, go for it18:42
openstackgerritlinggao proposed a change to openstack/ironic: Interactive console support for ipminative driver  https://review.openstack.org/9733118:42
rameshg871jroll: okay ..18:42
jrollrameshg871: please rebase agent patch too, I'm about to +2 I think18:43
*** Ugallu has quit IRC18:45
NobodyCamjroll: LOL... and I can't find any nits. I love it.18:50
jroll:D18:50
jrollI mean18:50
jrollI tried18:50
jrollI wanted to be as good as ruby18:50
jrollbut I couldn't find anything I would change18:50
NobodyCam:)18:50
NobodyCamseems I need to do a walkies18:52
rameshg871jroll: NobodyCam: okay18:52
NobodyCamanyone have a free cycle to rebase https://review.openstack.org/#/c/11749918:52
rameshg871NobodyCam: jroll: can i just push the new patch set ? jenkins will stop processing the approved patch set ?18:53
jrollI can in a few, NobodyCam18:53
jrollrameshg871: yep, go ahead18:53
rameshg871jroll: NobodyCam: i am addressing the nits in it as well :)18:53
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386518:54
rameshg871NobodyCam: jroll: ^^^18:54
rameshg871NobodyCam: jroll: please have a look at it18:55
rameshg871NobodyCam: jroll: rebase + nits addressed18:55
* jroll puts it on the end of his 4 review queue18:55
rloojroll: I believe that you can be better than me. Just look a bit more closely ;)18:57
jrollrloo: my nose was touching the screen!18:57
jrollcan't look closer18:57
jroll:P18:57
rloojroll: ohh, sorry, you're one of those. You need to back up instead :D18:58
jrollhaha18:58
*** penick has joined #openstack-ironic19:00
NobodyCamrameshg871: awesone... will review and approve once Mr jenkins is thru with it19:01
*** derekh has joined #openstack-ironic19:03
NobodyCamlol19:04
NobodyCamShrews: (and everyone else too ofc) I'm about to +a 114357 last looks?19:08
*** derekh has quit IRC19:08
NobodyCamgoing .. going ...19:10
rlooNobodyCam, others: question about the driver: https://review.openstack.org/#/c/111425/19/nova/virt/ironic/driver.py19:11
jrollNobodyCam: do it19:11
rloowrt list_instances() and list_instance_uuids(). Why does the latter do set(instance_uuids)?19:11
NobodyCamgone.19:11
JayFnext summit rloo has to give a talk on how to code review19:12
NobodyCamJayF: ++19:12
NobodyCam:)19:12
rlooJayF: it takes me a long time. I make myself read almost every line.19:12
JayFreading is never my problem19:13
JayFit's always the understanding that gets me ;)19:13
rlooha ha. I try to understand most of what I read too ;)19:13
jrollrloo: base class docstring says it should return a set :)19:14
jrolloh, no, I should read code before I look19:14
rloojroll: but it actually ends up returning a list. list_instance_uuids().19:14
jrollthat's a good question19:15
rloothe only thing I can think of, is if you use a set, you want unique, but the uuids should be unique anyway.19:15
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver  https://review.openstack.org/11588519:15
jrollright19:15
rlooanyway, it seems like either they are unique or they aren't. in any case, both methods should be handlign them the same way, no?19:15
rlooand they should be unique unless we're doing it cuz we have cases where the same instance uuid was associated with more than one node.19:16
rameshg871jroll: ^^19:16
jrollrameshg871: thanks19:16
NobodyCamhumm19:16
NobodyCamrloo: I'm not sure19:16
jrollrloo: yeah, idk, I'm not worried about it at this point, I'll leave that up to other people19:16
jrollrloo: I would let it land, file a bug, then fix it :P19:16
rloojroll: what is the fix? to be safe and use set() in both cases? ahh, forget it, I'll not say anything for now. see if someone else notices.19:17
jrollrloo: I tend to think that if the instance uuids are not unique, we have bigger problems19:18
JayFUUID collisions are very rare19:19
JayFand would probably break lots more things than just that :)19:19
rloojroll: true. I just want this code to land!19:19
jrollrloo: although, maybe it's a different case becuase one loops over "nova instances" and one loops over "ironic nodes"19:19
jrollironic could have one instance associated with multiple nodes (due to bug or race condition)19:19
JayFIsn't the idea to use a UUID so you /don't/ ever have to worry about getting a collission? :)19:19
jrollbut nova instances will always have a unique instance uuid19:19
jrolldoes that make sense?19:19
jrollnova's db won't allow dupplicate uuids19:20
rloojroll: they're both looping over the nodes from the call to ironic to get the list of nodes19:20
jrollironic's did allow that in the past, but now it does not19:20
jrolloh.19:20
jrollright19:20
jroll`for i in` tripped me up19:20
jrollidk, it's probably fine for now19:21
jroll(imho)19:21
rlooyeah, so the only thing I can think of, is that Ironic might have had 2+ nodes associated with the same instance uuid. there were issues before wrt deploying an instance I think, that got nodes in some wedges states?19:21
jrollyeah, we made that column unique as a result19:21
jrollrameshg871: looks like it kept my +2 :)19:22
rlooahh. in that case, it should always be unique. so the set() isn't needed. I Don't Think.'19:22
jrollright19:22
JayFhttps://review.openstack.org/#/c/113865/25 has no votes jroll19:22
jrolllooking at that right now19:22
jrollso many reviews19:23
JayFI thought that was the one you indicated kept your +219:23
jrollno, the ilo/agent did19:23
* JayF not feeling well having trouble keeping up19:23
jroll:(19:23
rlooShrews: here's your chance to review 113865 again.19:24
*** yuanying has joined #openstack-ironic19:25
jrollrameshg871: +2'd ilo/iscsi19:25
JayFoh nooooooooo19:26
* JayF found something19:26
JayFbut it's a nit at least19:26
jrolllol19:26
jrolldo it19:26
jrollit can be fixed!19:26
rameshg871jroll: i was wondering why it kept your +2 for new patch19:26
JayFrameshg871: want to fix it real quick?19:26
JayFrameshg871: found an issue in the docstring, will post the comment right meow19:26
jrollrameshg871: because it was a simple rebase for that patch19:27
rameshg871jroll: yeah19:27
rameshg871JayF: sure ..19:27
* jroll mashes f5 waiting for JayF19:27
NobodyCamlol19:27
JayFhttps://review.openstack.org/#/c/113865/25/ironic/drivers/modules/ilo/deploy.py,cm line 8019:28
rameshg871JayF: fixing right away19:28
JayFwoo19:28
NobodyCamoh wow great catch JayF19:28
JayFheh19:28
NobodyCam:-p19:28
JayFrloo tries to get everything in one pass19:29
JayFI try to get one thing on every pass19:29
JayFlol19:29
JayFnot literally19:29
rlooone is better than nothing ;)19:29
NobodyCam:)19:29
jrollwhoa, you and your crazy new view19:29
rloobut yeah, cuz I go crazy if I do too many (> 1) passes.19:29
JayFdid that link you directly to the new view?19:29
JayFhaha19:29
jrollJayF: yeah, ",cm"19:30
NobodyCam:)19:30
* NobodyCam is old school19:31
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386519:31
rameshg871jroll: NobodyCam: ready to go again ^^^ :)19:31
jrollI have to re-+2 this? really?19:31
NobodyCamlol ok this is it for 113865, this is the version that will land!19:31
NobodyCamlol19:31
jrolldone19:32
JayF113865 needs +2a19:32
JayFdooet19:32
jrollalso needs jenkins :|19:32
NobodyCamI will as soon as mr J is done19:32
JayF"I'm NobodyCam, and I approved this driver"19:32
* jroll will rebase 117499 now19:32
ShrewsI +2'd. NobodyCam can +A19:32
JayF:P19:32
jrollso much code today19:32
Shrewsjroll: plz to review all my codez NOW. KTHXBAI19:33
NobodyCamthis is crush day. we tag a release tonight19:33
*** wanyen has joined #openstack-ironic19:33
JayFwhat is left to land?19:33
jrollNobodyCam: wait19:33
JayFilo-ipa19:33
*** yuanying has quit IRC19:33
jrollNobodyCam: you wanted me to rebase https://review.openstack.org/#/c/11749919:33
jrollbut it has a dependency19:33
jrollNobodyCam: so I'm going to leave that alone, I think19:34
JayFjroll: it probably needs to be rebased off the dep, lol19:34
jrollJayF: well, I don't care so much about that, because dep isn't approved or even close19:34
jrollafaik19:34
rlooit is only j3. there's another release after that.19:34
jrollright19:35
jrolland this is a bugfix19:35
openstackgerritRamakrishnan G proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver  https://review.openstack.org/11588519:35
* jroll leaves it alone19:35
* jroll takes a break for lunch, bbiba19:37
JayFilo-iscsi just failed tripleo, checking on it19:38
JayFand the only thing in the console log is it failing to upload logs19:39
JayFnot our fault, it seems19:39
jrollbefore I go...19:39
jrollthis needs to get through: https://review.openstack.org/#/c/118404/19:40
jroll"We will refactor to split the code into submodules. Should have the new patchset tomorrow."19:40
jrolland stig isn't on irc19:40
jrollI'm fine with landing without the split (so is lucas)19:40
jrollwill review after lunch19:40
jroll(also failing tempest, yay.)19:41
openstackgerritA change was merged to openstack/ironic-python-agent: Add vmedia boot support in IPA  https://review.openstack.org/11527519:46
adam_ghas anyone noticed pep8 checks randomly failing on check_uptodate.sh ?19:49
NobodyCamoh happy happy joy joy cable company died19:49
* NobodyCam is now teathered to a phone :(19:49
NobodyCamadam_g: the pep8 error I've seen have all been real19:50
NobodyCamjroll: yep your correct ...19:51
adam_gNobodyCam, im curious more about the check_uptodate.sh config check (whcih runs as part of pep8)  i wonder if the random python hash seed stuff may be causing false negatives there.19:51
*** rameshg871 has quit IRC19:51
NobodyCamoh that I can not speak for.19:52
*** bmahalakshmi has quit IRC19:54
NobodyCamoh nice error: http://logs.openstack.org/65/113865/26/check-tripleo/check-tripleo-ironic-undercloud-precise-nonha/2d3a128/console.html19:55
JayFNobodyCam: is that what I pointed out earlier?19:56
JayF12:39:07 <JayF> and the only thing in the console log is it failing to upload logs19:56
NobodyCamoh sorry I was with out connection then :-p19:56
NobodyCamit was so close to landing19:57
NobodyCam:-p19:57
JayFlet it land, obviously the tripleo failure is unrelated19:58
JayFand that doesn't vote19:58
NobodyCam:-p19:58
*** chuckC has joined #openstack-ironic19:59
NobodyCamlol couldn't clone the devstack-gate repo20:00
NobodyCam:-p20:00
NobodyCamok its got a few. brb20:01
openstackgerritA change was merged to openstack/ironic: Add UEFI based deployment support in Ironic  https://review.openstack.org/11435720:01
rlooanyone know if nova is like us, they want aself.assertEqual(expected, actual)?20:13
NobodyCamcould just add them to https://bugs.launchpad.net/ceilometer/+bug/127710420:17
NobodyCamnova cli is there20:17
rlooNobodyCam: this is our driver test code - 111425.20:20
openstackgerritJosh Gachnang proposed a change to openstack/ironic: Improve IPA client library  https://review.openstack.org/11111820:22
NobodyCamrloo: where20:24
rloohttps://review.openstack.org/#/c/111425/19/nova/tests/virt/ironic/test_driver.py, line 98, 101 (so far)20:24
NobodyCamrloo: did they request that our is that just a bug/error on our part20:26
NobodyCamthat line has not changes throught all 19 reviews20:26
rlooI'm reviewing that patch now and noticed it. that's why I asked.20:26
NobodyCamI think its just our bad20:26
NobodyCamyea I would tag it20:27
rlooyeah, lucas-dinner has a patch up to fix the issues in this patch, so I'll just fix it there.20:27
NobodyCamlol :)20:27
rloomight as well make it pretty :-)20:27
NobodyCam++ dont block that patch20:27
NobodyCamand correct20:27
rlooyeah, and that!20:28
*** yuanying has joined #openstack-ironic20:29
openstackgerritJeremy Stanley proposed a change to openstack/python-ironicclient: Work toward Python 3.4 support and testing  https://review.openstack.org/11880120:30
openstackgerritJeremy Stanley proposed a change to stackforge/pyghmi: Work toward Python 3.4 support and testing  https://review.openstack.org/11882920:32
jrolladam_g: have an example?20:32
lifelesserm wut "py27 runtests: commands[2] | bash -c cat .testrepository/1 >>.testrepository/020:33
lifeless"20:33
lifelessthat is 100% crack20:33
jrolllol20:33
lifelessseriously, wtf, its going to be building a huge 0 file over time and have no impact on the reporting after the first time its run20:34
jrolllifeless: is that in jenkins jobs? if so, wouldn't .testrepository get thrown away each time, anyway?20:35
lifelessits also in local jobs20:36
jrollwelp.20:36
jrollreporting seems to work for ironic for me, fwiw20:36
lifelessdo me a favour20:36
lifelessls -lh .testrepository/020:37
jrollwow20:37
jroll46M20:37
lifelessand cat .testrepository/next-stream20:37
jroll13520:37
jrollwhich seems to be correct20:37
lifelessls -lh .testrepository/220:37
lifelessand 320:37
jroll58k20:38
jroll671k20:38
lifelessyeah20:38
lifelessso I'm right :)20:38
lifelessfixing20:38
lifelessnext time folk, talk to me20:38
jrollare you thinking it's inflating?20:38
*** yuanying has quit IRC20:38
lifelessevery time you run it20:38
lifelessit being tox20:38
jrollbecause /4 is back to 59k20:38
lifelessits concatenating 1 onto 020:39
jrollI mean, 0 clearly is inflating like mad20:39
lifeless- read tox.ini20:39
jrollright20:39
jrollI see the problem, I'm just curious why /3 is important20:39
lifelessbaselining20:39
lifelessthere are two testsuites being run as separate things20:39
jrollright, ok20:39
lifelessone will be bigger than the other20:39
jrollah, yes20:39
jrollforgot there's two suites20:40
lifelessif either of 2 or 3 was 46M I might have been wrong20:40
jrollon the "next time, talk to me" thing... idk what testr docs look like, but maybe there's something to note here20:40
jroll(depending why this was done in the first place)20:40
jrollshouldn't need to ask a library author how to use the library L|20:40
lifelesswell20:40
jrolls/L|/:|20:40
lifelessthe issue is upstream discover20:40
lifelessthe talk to me bit is more if there is a limitation in a tool, lets fix the tool rather than workaround20:41
jrollok, sure :)20:41
lifelessand honestly, concatting two files in a stateful database .... how is that not a workaround (And in this case not even a good one)20:43
jrollagree20:43
* lifeless stabs pip depends a bit harder20:48
lifelessI should upgrade this dev container :)20:48
lifelessI got onto this from the hash ring thread20:49
lifelesstried to run tests and my eyes bled :)20:49
jrolllol20:49
lifelessadventures in yak shaving20:49
jrolllifeless: bug related to that: https://bugs.launchpad.net/ironic/+bug/135551020:50
lifelessjroll: ah no thats the update-based-on-ring thing20:52
lifelessjroll: the thread was saying that the ring isn't stable20:52
jrolllifeless: right... I would just turn that bug into "our hash ring sucks in so many ways"20:52
lifelessoh great20:52
lifelessDuplicateOptError: duplicate option: rpc_backend20:52
rloojroll: oh, did you say that there's room for lots of improvement with our hash ring?20:54
jrolluh oh20:54
* jroll hides20:54
rlooha ha. where were you when we needed it?20:55
jrolloh wait, you're just nitpicking my grammar again!20:55
jrollI was building an agent for you :P20:55
rlooand a lovely agent it is!20:55
jrollthat's questionable, but thanks :)20:55
rlooI know, there's room for improvement there too :D20:55
lifelessany thoughts on this duplicate option thing ?20:56
lifelessah, I think I found it20:57
jrolllifeless: check your config?20:57
lifelessits because I'm working on fixing this two-test-suite-thing20:57
lifelesswe need to unregister that option from nova20:58
NobodyCamuggh 113865 failed python 2720:59
jrollwat21:00
lifelessjroll: was that way to NobodyCam  or me ?21:00
jrollto NobodyCam21:00
NobodyCamcouldn't get some dep21:00
jrollidk anything about that option, and not going down that rabbit hole right now21:00
adam_glifeless, that will all go away when the ironic.nova.* goes away21:01
lifelessadam_g: yes, but its broken now.21:01
adam_glifeless, the dual test suites, cat'ing of test database, oslo errors21:01
lifelessadam_g: and I can't ignore the egregiously broken stuff21:02
NobodyCamjroll: ERROR:   py27: could not install deps21:03
NobodyCami'll recheck as soon as i can21:03
jrollsilly jenkins21:03
jrollhttps://review.openstack.org/#/c/118404/3/doc/source/deploy/drivers.rst21:04
jrollline 5821:04
jroll2c is actually a thing? that's weird21:04
adam_gjroll, https://bugs.launchpad.net/oslo-incubator/+bug/1365136 was mostly wondering if its showing up anywhere outside of tempest21:04
jrollah, I see21:04
jrolladam_g: so things just move around, I guess?21:05
mrdaMorning Ironic!21:05
jrollmrda!21:05
jrollwe have a million things for you to do!21:05
mrdad'oh21:05
mrda:)21:05
jrollhehehe21:05
adam_glifeless, test discovery attempts to import everyting under ironic.*.  ironic.nova.* is importing nova config at some point, causing the conflicts on the oslo resources21:05
jrollI'm kidding, ignore me21:05
lifelessadam_g: I know, I'm fixing.21:05
NobodyCammorning mrda21:06
jrollI'd rather just land the nova driver21:06
mrdajroll: +121:06
jrollbut... working software is cool too21:06
NobodyCammrda: will we need a ffe for nova driver?21:06
jrollmaybe if we all pounce nova-core at the same time, we won't21:06
jrollNobodyCam: is devanand1 back tomorrow, I guess?21:07
NobodyCamjroll: maybe, I supect he is looking for pants21:07
NobodyCamlol21:07
NobodyCami'll be at nova meeting tomorrow21:08
jrolllol21:08
jrollyou're supposed to burn the giant man, not your pants21:08
NobodyCamlol21:09
NobodyCamohh thouse just get lost from what I hear21:09
jrollheh21:09
jrollalways keep extra21:09
jrollit's like towels21:09
jrollnever leave without one21:09
jrollalways have spares21:10
jrolletcetc21:10
NobodyCam+4221:10
jroll:D21:10
NobodyCamlol all out errors should start with "Don't worry"21:10
NobodyCams/out/our/21:11
jrolllol21:11
jrollI have so many minor nits I'm trying to hold back21:11
jrollon this snmp driver21:11
jrolltoo much code21:11
mrdaso FYI, I believe we'll need to get a Feature Freeze Exception (FFE) for the Ironic driver from Nova now.  Because not all the code has +2.  mikal gets back from the USA today, and I'll approach him about it once he lands.21:14
rloojroll: if there are that many, snmp isn't ready to go21:14
rlooadam_g: question for you about a change you made: https://github.com/openstack/ironic/commit/4d4aeab4e57d623ca06bdfdceddbee2791851d03#diff-9d1bcbeec6093089dc050d1d913a601e21:15
mrdaunless it all lands today :-/21:15
NobodyCammrda: want to add us to the ffe section of: https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting21:15
rlooadam_g: do you recall why you used list(set(...)) in list_instance_uuids()?21:15
mrdayup21:15
jrollrloo: I'm thinking like 2 spaces instead of one in a comment21:15
jrollerr, in a string21:16
rloomrda, you're going to hate me. I -1'd one of the driver patches.21:16
mrdaI'll rustle up some cores21:16
mrdarloo: saw that :)21:16
rloojroll: does it happen a lot? is that the only thing?21:16
mrdarloo: thank you for your review though21:16
jrollrloo: just something I noticed21:16
mrdait's ok21:16
mrdaeverybody hates me :)21:16
rloomrda: lucas-dinner started a patch to address them: https://review.openstack.org/#/c/118693/21:16
rloomrda: and he wanted me to update it.21:17
* mrda needs to rustle up nova cores right now21:17
rloomrda: I, for one, am glad you are doing what you are doing ;)21:17
mrdarloo: thanks, just gald I can help21:17
mrdas/gald/glad/21:17
rloomrda: so I think we need to put up another revision for patch 111425 before asking the nova cores to look (again)21:18
adam_grloo, one sec21:18
mrdarloo: extra_specs21:18
mrdarloo: not needed anymore?21:18
rloomrda: lucas-dinner doesn't think so and I don't either but I don't know much. he took it out so I'm guessing we don't.21:18
rloomrda: I can't figure out why in the nova patch, the LOG string (line 92 of driver.py) doesn't have _LW(), yet we have it in our ironic tree.21:19
jrollrloo: are we sure nova's convention is to use _LW?21:19
jroll(etc)21:20
jrollI'd be willing to bet they aren't conforming to that yet, or something21:20
rloojroll: there are other _LW() in that file. we should, at the very least, be consistent. and it is _LW() in our version of it.21:20
jrolloh21:20
jrollok21:20
jrollshould fix thart21:20
jroll-r21:20
rlooI'm not sure how/what to update for _validate_instance_and_node() though.21:20
rloothe docstring is incorrect based on the code.21:21
adam_grloo, it looks like, at the time, there was no unique constraint on instance_uuid in the database21:21
mrdaNova use _LW21:21
rlooadam_g: ok, that's what I thought. So I can change it now -- remove the set() part, right?21:21
jrollrloo: I wonder if the docstring should read 'ensure this instance is still associated with a node'21:22
adam_grloo, see lucas-dinner's comments @  https://review.openstack.org/#/c/98268/1/ironic/nova/virt/ironic/driver.py21:22
jrollrloo: but I may be wrong21:22
mrdaThe code was changing in this area during proposal, so it's entirely possible that a few were missed.21:22
NobodyCamwill a recheck restarjobs that are already running?21:22
jrollNobodyCam: idk21:22
rlooadam_g: thx, that comment from lucas-dinner helped. So yes, it can be removed now.21:22
adam_grloo, yeah, it should be okay to use just a list straight from node.list21:23
mrdaSo it's getting serious.  It has to land by tomorrow week.  There's no exception beyond Friday week.  So I would be arguing only necessary changes21:23
lifelessI thought we switched to oslo.messaging21:23
lifelessso why do we still have the rpc module21:23
adam_grloo, im not sure thats worth the churn at this point tho? unless its noted as a blocker by nova reviews?21:23
rloojroll: yeah, I think that docstring is good. My concern is deleting the stuff that is there now, although we can get it back later.21:23
lifelessah...21:23
lifelessgit fail - pyc files21:24
rloojroll: I am wondering whether there was meant to be some code that checks node.uuid == instance[node] or something like that.21:24
NobodyCamlooks like nope :(21:24
jrollrloo: not sure what you mean?21:24
jrollrloo: oh, right21:24
jrollrloo: not sure21:24
rlooadam_g: I noticed it so I should have shut up. but the code there and list_instances is inconsistent -- which is what made me wonder why they were different.21:24
adam_grloo, :)21:25
rloowe have to update that patch anyway, so might as well do a good job on it now (as long as we don't break anything)21:25
rloojroll: exactly, not sure, and adding that code *could* break something. which is why changing the docstring is safer at this point.21:26
jrollrloo: on the other hand, the more changes we have in there, the harder it is for nova to review21:26
rloojroll: you mean if they do a diff between revisions?21:26
lifelesswow, ironics test suite writes to stderr a /tonne/21:26
jrollyes21:26
lifeless    result = function(*args, **kwargs)21:26
lifelessTypeError: <lambda>() takes no arguments (1 given)21:26
lifelessetc21:26
jrollrloo: and they likely will, who would read all the code again?21:27
jrolllifeless: that looks broken, I don't get that in stdout21:27
mrdaSo we really do want minimal changes to the driver21:27
rloojroll: true. now you tell me. I wouldn't have put my comments in there then.21:27
jrollmrda: agree21:27
mrdaonly if something is broken21:27
lifelessjroll: its being eaten by the stdout eater, but if thats not enabled...21:27
lifelessjroll: try this21:27
lifelessOS_STDOUT_CAPTURE=0 OS_STDERR_CAPTURE=0 testr run21:28
jrollrloo: sigh21:28
*** dhellmann is now known as dhellmann_21:28
mrdaFor example, the ironicclient caching has come up in review a few times.  Nova things it's an easy fix, but they are happy to bump that to a "must fix in K".  So, please, let's focus on critical for J release bugs21:28
jrolllifeless: /me runs21:28
mrdaIs https://review.openstack.org/#/c/118693 critical?21:28
lifelessoh, sorry .testr.conf is buggy there21:28
lifelesslet me paste a fix21:28
mrda(for J release in the ironic driver)21:28
lifelessjroll: in .testr.conf change21:29
jrollmrda: I think there was a -1 for that21:29
lifelessOS_STDOUT_CAPTURE=1 to OS_STDOUT_CAPTURE=${OS_STDOUT_CAPTURE:-1}21:29
lifelessjroll: and similar for ERR, then do what I said above21:29
rloomrda: joe -1'd the extra-specs thing.21:29
mrdarloo: cool, thanks for pointing out jogo's -121:30
jrolllifeless: oh, well, I think that's e.g. LOG.exception21:31
rloomrda: any way for me to undo my comments in 111425? I can remove my -1 from that patch if you want.21:31
mrdaIn that case, can we get eyeballs on https://review.openstack.org/#/c/118693 ?21:31
mrdaI will get the change up today21:31
jrolllifeless: also eventlet being a piece of something21:31
mrdabut I need 2x+2's for that :)21:31
*** linggao has quit IRC21:31
jrollmrda: just have you 121:31
jrolls/have/gave21:31
openstackgerritlifeless proposed a change to openstack/ironic: Unbreak debugging via testr  https://review.openstack.org/11889621:31
lifelessjroll: indicates we haven't setup logging properly in the tests, at minimum :)21:32
jrollrloo: too late now, comments are forever21:32
jrolllifeless: indeed21:32
mrdathanks jroll21:32
*** krtaylor has quit IRC21:32
rloojroll, mrda: so what should I do if anything? address the comments in 118693 or not?21:32
jrollrloo: let's not do that unless they are absolutely necessary21:33
mrdawhat comments in 118693?21:33
rloojroll: so who decides which are necessary?21:33
jrollrloo: is something horribly broken?21:34
mrdaoh, the review comments in 111425 that aren't yet in 118693?21:34
rloomrda: my comments in 111425, that 118693 was meant to address.21:34
mrdaright21:34
rloojroll: no nothing horribly broken.21:34
jrollright... I would leave it21:34
mrdaso the docstrings, go ahead and address21:34
jrollalthough I think nova might pick up on those and ask for the same :|21:34
rloojroll: but the comments are in 111425. will other reviewers ask why they weren't addressed?21:34
mrdarloo: yes21:34
mrdathey will ask21:34
jrollright21:34
jrollsadface21:35
*** yuanying has joined #openstack-ironic21:35
* rloo has sadder face21:35
mrdaso I wouldn't touch the log levels21:35
mrdawe can live with that21:35
rloomrda: maybe that's why the log level is diff in ironic tree than nova patch.21:35
mrdawe will need to handle InvalidArchitectureName21:35
mrdarloo: I think that's my pebcak actually21:36
rloomrda: yeah. Does anyone know how?21:36
rloomrda: what is a pebcak?21:36
jrollI had it handled once upon a time21:36
jrolland people hated it21:36
jrollyay.21:36
rloomrda: I'm thinking payback, or pebble-cake.21:36
jrollrloo: Problem Exists Between Chair And Keyboard21:36
mrdaproblem exists between computer and keyboard21:36
mrdaclose :)21:36
jrollheh21:36
rlooha ha21:36
jrollI mean... the bluetooth between my keyboard and computer isn't always happy, either21:37
mrdaok, so rloo would you be able to update 118693 addressing the nits and the exception?21:37
mrdabut leaving logging alone?21:37
rloook, if InvalidArchitectureName, I'm going to set the cpu_arch to None.21:37
rloomrda: yeah, there was only one logging nit anyway I think.21:37
mrdarloo: I think that's fair, because that's what canonicalize() does internally for None provided21:38
mrdaand then if we can get eyeballs on that, I can get it merged into Nova this morning, and off we go.21:38
NobodyCamgah the queue is so slow!21:39
mrdaOnly 8 calendar days until drop dead :(21:39
mrdaIt's that time of the year NobodyCam21:39
NobodyCamyep21:39
mrdaMore sad is that I only have one Nova core signed up21:39
jrollI didn't expect it to be jogo21:40
mrdaStrike that, two nova cores21:40
mrdaOne more to go21:40
mrdaand dansmith21:40
jrollniiiice21:40
NobodyCammrda: we are here if you us!21:40
NobodyCamgah21:41
mrdajust get 118693 readied :P21:41
NobodyCamif you *NEED* us.. :-p21:41
mrdaNova meeting page updated21:43
rloomrda: fyi, there's a -1 by joe wrt CONF.ironic.client_log_level.21:43
*** yuanying has quit IRC21:43
NobodyCam:-p21:44
mrdarloo: looks like a nit to me21:44
mrdaJoe comments "yeah, I think its fine to leave as is. I would like to see a low priority bug opened against this though"21:45
rloomrda: sorry, that's for the use of 'icli'.21:46
rloomrda: see line 13921:46
rloomrda: that reminds me, are you opening bugs when they ask for them?21:46
mrdarloo: I will, but haven't.  Because that only started with Joe overnight21:47
mrdaI can do that21:47
rloothx mrda. didn't know that was a new thing.21:47
mrdarloo: but you're right re: Joe's logging conf21:48
mrdaIt appears he wants this changed21:48
mrdarloo: can you do that?21:49
rloomrda: our version of driver.py doesn't have the arch.canonicalize() stuff?21:49
rloomrda: do what?21:49
mrdarloo: logging changes21:49
rloomrda: I'm NOT undoing Chris behren's changes.21:49
jrollhe does want it changed? :/21:49
rloojroll: Joe thinks we can set the ironicclient logging via nova's logging.conf stuff.21:50
mrdarloo: there is a review for the canonicalize change, but I haven't updated it yet.  We did this one Nova-first-then-ironic since we were finding it hard to get agreement on the way forward21:50
rloojroll: I have no idea how, but it seems to me that Chris would have known if there was an 'easy' way?21:50
mrdaso I'll backport that change into ironic21:50
jrollcomstud: ^^21:51
rloomrda:  should I wait for the backport, to get it into 118693?21:51
jrollrloo: if that's doable, then let's do it, but that's a nit imo21:51
jrollwell, maybe not, config options are important21:51
rloocomstud: more specifically, see line 139: https://review.openstack.org/#/c/111425/19/nova/virt/ironic/driver.py21:52
mrdaJust do it independently 116772 is the patch for canonicalize21:52
mrdaoh I see21:52
mrdaOk, should we rebase 118693 on top of 116772?21:52
rloomrda: that's fine. lucas' patch can be rebased after 116772 lands.21:53
mrdarloo: does that work for you? (if I update 116772 immediately)21:53
rloomrda: or we could rebase on top of...21:53
mrdaI'll update that in 5 minutes (need to grab breakfast first ;)21:53
jrolllet me know if I can help at all, bt21:54
jrollw21:54
comstudhm21:58
mrdajroll: can you look at comstud's comment on L124 of https://review.openstack.org/#/c/116772/5/ironic/nova/virt/ironic/driver.py to see if we need to do anything?21:58
mrdacomstud: oh hai21:58
jrollmrda: he took that back in irc but not in gerrit21:59
mrdaahh21:59
jrollmrda: the code is fine21:59
mrda\o/21:59
* mrda stops arm flailing21:59
comstudmrda: my comment is wrong22:00
comstudsorry, i should have replied there22:00
comstudbut i replied in the main review comment area22:00
* comstud adds a reply there22:01
comstudbut from my same reply earlier22:02
comstud"So, I guess I also have concerns that this may potentially make the image_props_filter not match anymore for certain people's configs..."22:02
mrdaright22:02
openstackgerritRuby Loo proposed a change to openstack/ironic: Driver merge review comments from 111425  https://review.openstack.org/11869322:04
openstackgerritKyle Stevenson proposed a change to openstack/ironic: Improve IPA client library  https://review.openstack.org/11111822:04
kylestevJoshNang: ^22:04
rloomrda, jroll: I pushed up another revision. it doesn't have the fix to address the arch exception (needs to be rebased first), and joe has some comments in test_driver.py that aren't addressed.22:05
jrollrloo: what comments aren't addressed? (and why?)22:06
rloomrda, jroll: the dinner bell has rung so I need to make an appearance. I'll check in about an hour or so.22:06
jrollrloo: also, can you go ahead and rebase?22:06
jrolloh22:06
jrollok22:06
rloojroll: he wanted to combine two tests some how.22:06
mrdathanks rloo !22:06
jrollhuh.22:06
jrollrloo: so it definitely needs work?22:06
mrdaI'll have 116772 updated by then22:06
jrollI can try to finish it22:06
rloohttps://review.openstack.org/#/c/111425/19/nova/tests/virt/ironic/test_driver.py22:06
jrollonce mrda is done mucking aroudn22:06
mrdalol!22:06
rloooh, there were two comments joe had, see the bottom of the page.22:06
NobodyCamrloo: Get a node associated with an instance. that implys that a instance can have more then one node? would not Get node associated with instance. be more correct?22:06
rlooNobodyCam: 'Get the node associated with the instance'? Sorry, I was in a hurry. jroll will update it, won't you jroll? Thx!22:07
jrollsuuuuuuuuuuurrrrrrrrrrrre22:07
mrdarloo: go eat dinner!22:08
jrollwhatever you say22:08
openstackgerritKyle Stevenson proposed a change to openstack/ironic: Improve IPA client library  https://review.openstack.org/11111822:08
NobodyCamlol22:08
NobodyCamenjoy your dinner rloo :)22:09
*** openstack has joined #openstack-ironic22:11
NobodyCamjroll: your reworking 118693?22:13
jrollNobodyCam: not yet, feel free to take it22:13
jrollI'm looking at snmp22:13
jrollstill22:14
jrolltrying not to get distracted :|22:14
mrdaNobodyCam: I'm just testing 116772 which should probably be a dep for 11869322:15
NobodyCammrda: I have time to go somke before you push up 116772?22:24
mrdaya22:24
NobodyCam:) brb22:24
NobodyCam:-p22:24
jrolllast call for -1's on https://review.openstack.org/#/c/118404/22:25
jrollI'm ready to land it22:25
jrollstepping away for a few real quick22:25
jrollNobodyCam etc ^^22:25
jrollactually, I'm just going to land it, y'all have from now until gate jobs pass to strike it down :)22:25
mrdajroll: I used to work at a company that launched a product without a way to bill for it, knowing that they had a month to write the billing interface, before the 1st customer needed to be billed.22:27
mrdaIt's a bit like that :)22:27
jrollheh22:27
openstackgerritSyed Ismail Faizan Barmawer proposed a change to openstack/ironic: Add uefi boot mode support in IloVirtualMediaIscsiDeploy  https://review.openstack.org/11656122:28
* mrda thinks we need to parallelise pep822:30
kylestevmrda: +1... it's really slow :(22:32
NobodyCamgah tempest test still over an hour away22:33
NobodyCamfor 11386522:33
kylestevmrda: alternatively, you could use an OnMetal IO instance :P22:37
mrdakylestev: +122:37
mrdaI'm actually using a 30Gb Perf222:37
kylestevoooh, nice choice!22:37
jrollperf cloud ftw22:37
jrollwell22:38
mrdabut pep8 is still serial22:38
jrolloops, I'm not supposed to say that out loud :P22:38
kylestevha22:38
* jroll brb22:38
kylestevbeing a core must so sooo inconvenient22:38
*** krtaylor has joined #openstack-ironic22:39
*** yuanying has joined #openstack-ironic22:40
mrdakylestev: but at least my py27 runs are quick22:41
openstackgerritMichael Davies proposed a change to openstack/ironic: Nova review updates for _node_resource  https://review.openstack.org/11677222:41
mrdaNobodyCam: ^^ feel free to rebase on top of this22:42
NobodyCammrda:22:42
NobodyCammrda: ack22:42
lifelesssuccess22:42
NobodyCam:)22:42
lifelessRan 1365 tests in 67.274s22:42
lifelessOK22:42
lifelessjroll: adam_g:^22:42
NobodyCamlifeless: :)22:43
lifelessone run22:44
adam_glifeless, cool. what'd you need change?22:46
*** yuanying has quit IRC22:48
lifelessadam_g: incoming22:51
openstackgerritlifeless proposed a change to openstack/ironic: Run the whole test suite as one run  https://review.openstack.org/11892022:51
jrollkylestev: I laughed, but still not sure what you mean :P22:53
lifelessmrda: pep8 is parallelised in the next release22:54
lifelessmrda: though why it needs it is terrible :)22:54
NobodyCammrda: is this right? http://paste.openstack.org/show/8u4fMymo4sHXBPLoi5cv/22:55
mrdalifeless: did we discuss this last week?22:55
lifelessmrda: it was discussed in #openstack-infra the other day22:55
mrdalifeless: ahh ok22:55
openstackgerritlifeless proposed a change to openstack/ironic: Ignore backup files  https://review.openstack.org/11892222:56
mrdaNobodyCam: wow22:56
mrdaNow I'm not so sure22:56
NobodyCamya22:56
NobodyCamI'm going to not do that22:56
mrdagood plan22:56
NobodyCam:-p22:56
mrda:)22:56
kylestevjroll: not being able to talk up your employer's products22:57
mrdaOk, I'll bbs.  School run time.22:57
jrollkylestev: nah, as an onmetal dev I'm supposed to say "virtualization is dead"22:58
jrollor something22:58
kylestevahhh23:00
lifelessadam_g: thoughts appreciated on that patch; if its too ugly to bear I can just use it locally to let me be sane in the interim23:00
jrollkylestev: (long live virtualization) :P23:01
openstackgerritChris Krelle proposed a change to openstack/ironic: Driver merge review comments from 111425  https://review.openstack.org/11869323:01
jrolldid we actually land all the j3 blueprints?23:01
jrollI think we did23:01
jrollholy cow23:01
NobodyCamif we can them thru the gate23:02
jrollright23:02
jrolls/land/approve/23:02
jroll:)23:02
NobodyCam:)23:03
* NobodyCam checks zuul and goes out for a smoke23:03
jrollNobodyCam: were you going to rebase that on top of the other thing or?23:03
NobodyCamjroll: no ... 4 RE: see ttp://paste.openstack.org/show/8u4fMymo4sHXBPLoi5cv/23:04
* jroll tries to find the other one23:04
jrollright, I saw that23:04
NobodyCam:-p23:04
jrollI can try to rebase if you'd like23:04
NobodyCamyou welcome to :)23:04
jrollI don't think it matters, though23:05
NobodyCamI just review -d 116772 and then git rebase review/michael_davies/11677223:05
jrolllet's just land both23:05
jrolloh23:05
jrollI usually do review -X 11677223:05
jrollit's fine23:05
jrollthey don't need to depend, just both need to land23:05
* NobodyCam points angry finger ar mr J23:06
NobodyCams/ar/at/23:06
wanyenjroll: ilo/uefi boot still need review23:06
wanyenjroll, ilo/ipa and ilo/iscsi driver have not yet merged. right?23:07
jrollwanyen: was ilo in the uefi spec?23:07
jrollwanyen: they have not merged, they have been approved, I think23:07
* jroll double checks23:07
NobodyCamilo/iscsi is in gate now... will need recheck23:08
NobodyCambut it trying to land23:08
jrollI don't see uefi for ilo in a spec23:08
wanyenjroll: uefi spec cover both pxe and vendor drivers23:08
* jroll looks23:08
wanyenjroll: Add uefi boot mode support in IloVirtualMediaIscsiDeploy  https://review.openstack.org/11656123:09
jrollthanks23:09
jrollthat's a lot of code23:09
jrolland failing tests23:09
jrolland we have... 51 minutes by my clock23:09
jroll(my clock may be off by a few hours23:09
jrollbut there's only two cores around by now, as far as I can tell23:09
openstackgerritKyle Stevenson proposed a change to openstack/ironic: Improve IPA client library  https://review.openstack.org/11111823:10
jrollthis wasn't on the priority list for j3; we can try to push it through but I can't make a promise23:10
wanyenjroll: I thought the last patch was fine23:10
jrolloh, jenkins is still running23:10
jrollwanyen: I'm still not going to promise that this will land23:11
mrda.23:12
wanyenjroll: whateve you can help will be highly appreciated.  Doe silo/ipa get approved.  Last time I check it only got one +223:12
jrollwanyen: oops, we'll need to bug NobodyCam to land it23:13
wanyenjroll: the uefi boot was on teh J3 list and the blue print and design spec cover both pxe and vendor drivers23:13
adam_glifeless, added one note, but i'm okay with adding it for now23:14
jrollwanyen: again, we will try. but working code was not uploaded until 22:28 on feature freeze day, and we still don't know if it's working23:15
jrollwanyen: so you understand where I'm coming from23:15
jrollwanyen: have you reviewed the code?23:15
wanyenjroll,  yes.  I understand.23:16
jrollwanyen: all the reviews we can get will help23:16
* mrda just reviewed 118693. Others who care about the Ironic driver landing in Nova should likewise.23:16
mrda:-)23:16
jrollso many distractions today23:17
jroll"Returns True if unacceptable."23:17
jrollshould this be :returns: True if unacceptable?23:18
jrollI don't actually care much23:18
* NobodyCam thinks 113865 will the que for at least 10 hours 23:18
NobodyCamI saw that too23:18
NobodyCambut nothing else in that file follows that23:18
jrollgreat23:19
jrollI would assume that would give Ruby a heart attack23:19
jrollimbw23:19
jrollanyway, +2'd23:19
*** openstackstatus has quit IRC23:19
jrollNobodyCam: if the other ironic cores are gone for the day, I would say we should just approve that23:19
* jroll is probably a bad person for saying that23:19
*** openstackstatus has joined #openstack-ironic23:20
*** ChanServ sets mode: +v openstackstatus23:20
NobodyCamno it needs to land23:20
jrollok23:20
jrollnot sure if rloo is coming back23:20
jrollalthough she also pushed a patch23:20
NobodyCambut going to be in jenkins hands for a while23:20
jrollwant me to just land it?23:20
*** yuanying has joined #openstack-ironic23:20
jrolloh, right, jenkins23:21
mrdajroll: rloo said she'd be back in an hour23:21
jrollpoor zuul23:21
mrdaafter dinner23:21
jrollmrda: right, but I realized she also touched that23:21
mrday'all should just work in my timezone23:21
mrdait would everything *so* much easier23:21
mrda:)23:21
jrollwe should all just work on utc :P23:21
jrollso, when is j3 cut? end of day utc?23:22
wanyenjroll, tx.  ilo-iscsi uefi boot will remove teh need for manual setting of boot device and boot mode so it will really help user's uefi boot expereicne.23:22
NobodyCamjroll: by that its 11 pm and I should be in bed23:22
jrollwanyen: (for HP hardware)23:22
wanyenjroll, yes.23:22
jrollNobodyCam: by that it's 11pm and we still have hours of sunlight to party!23:22
mrdaNobodyCam: ouch.  9am-ish here23:22
jrollwanyen: I udnerstand the benefits, not everyone has hp hardware :(23:23
wanyenjroll, is it possible to apply for exception in case it does not land in time for J3.23:23
NobodyCamWe should be able to recheck running jobs23:23
* NobodyCam pops over to infra23:24
jrollwanyen: I think only devanand1 can approve exceptions23:24
jrollwanyen: like I said, we want this code, we will try to get it through23:24
jrollwanyen: I just am not going to make a promise23:24
wanyenjroll, I see.  Your andotehr reviewes help will be much appreciated.23:25
*** lucas-dinner has quit IRC23:25
wanyenjroll, tx.23:25
openstackgerritA change was merged to openstack/ironic: Adds SNMP power driver  https://review.openstack.org/11840423:26
jrollwoo ^23:26
*** coolsvap has quit IRC23:29
mrdaNobodyCam: just wondering if you could sanity check 11869323:30
mrdabut if it's too late, then I understand23:30
*** chuckC has quit IRC23:30
* mrda is a morning person23:30
NobodyCammrda: where?23:30
mrdaNobodyCam: https://review.openstack.org/#/c/118693/323:30
wanyenNobodyCam: can you review and approve ilo-ipa  driver https://review.openstack.org/#/c/115885/ ?23:30
mrdaforget me, you uploaded it :)23:31
* mrda needs to wait for rloo to return23:31
NobodyCamhuh23:31
jrollmrda, rloo also touched that patch23:32
mrdalol23:32
jrollmrda: I'm ok with approving it myself, without another reviewer23:32
jrollspecial circumstances (tm)23:32
jrollmrda: but need jenkins first :|23:32
mrdait's just with the check queue so long...23:32
jrolland then have to wait hours23:32
jrollyeah23:32
jrollI mean, I can push the button, I don't care much for our non-voting jobs23:33
NobodyCamwe can try and change 113865's commit message that will re queue it23:33
jroll(mostly joking)23:33
mrdaAnd once I've got check and a +A I need to get it through Nova's check23:33
jrollNobodyCam: go for it23:33
mrdabefore nove cores look at it23:33
mrdale sigh23:33
* mrda is preparing for Paris23:33
jrollmrda: you don't *have* to wait for an ironic +A to put it in the nova patch23:33
mrdawell, Deva would beat me up if you guys didn't approave anything non-trivial and I proposed it to Nova23:34
mrdayou'd run me out of town!23:34
*** coolsvap has joined #openstack-ironic23:34
jrollgimme a break23:34
jrollit's gotta go23:35
*** romcheg1 has quit IRC23:35
openstackgerritChris Krelle proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386523:36
NobodyCamjroll: ^^^ only change was commit message in web interface23:36
jrollwhee23:36
* jroll doublechecks23:36
NobodyCamat least this way we're not waiting for it to finish jsut to recheck it23:38
mrdaok, I'm going to merge 118693 back into Nova and upload a new revision of 11142523:40
jrollmrda: we still need the canonicalize() change +2'd yeah?23:40
mrdathat's back into Ironic, so less urgent23:41
mrda(ref https://review.openstack.org/#/c/116772/6)23:41
jrolloh, is that in nova already?23:41
mrdayeah, I proposed it over there as we're trying to get Nova consensus on it.23:42
mrdaNo point us all agreeing on a way forward and Nova rejecting it23:42
jrollright, ok23:42
mrdaSo I reversed the ordering on that23:42
jrollso that's... yeah23:42
Shrewsgah. 113865 still not thru?23:42
jrolllol23:42
jrollNobodyCam: OH MY GOD23:42
jrollNobodyCam: 113865 needs a rebase now23:43
jrollffs23:43
NobodyCamShrews: I just checnged the commit message to retrigger the job23:43
NobodyCamno23:43
mrdaShrews: >8 hour check gate23:43
jrollyes23:43
jrolllook again23:43
Shrewsmrda: that is clearly your fault  :-P23:43
mrdalol23:43
mrdaI wish I had so much power23:43
jrollthe snmp driver landed, maybe related23:43
jrolllikely related, actually23:43
jrollughhhhhhhh23:43
jrollNobodyCam: want me to get that?23:43
mrdaCan we have another 1000 nodes added to nodepool?23:44
jrolloh wait, Shrews is here :o23:44
jrollmrda: talk to whoever does quota at rax23:44
jrollI vote yes23:44
* Shrews is not the core you are looking for </waving hands>23:44
mrdaOH, SHREWS IS HERE23:44
NobodyCamjroll: I'm on phone bandwidth could you23:44
Shrewsjroll: srsly though, need something?23:44
jrollShrews: can you look at https://review.openstack.org/11869323:44
mrdajroll: +123:44
jrollNobodyCam and rloo both touched it, so they won't +223:45
jrollNobodyCam: np23:45
* Shrews puts beer down... looks23:45
mrdaand we're a bit short on cores in this TZ :)23:45
Shrewsaye23:45
jrollnonono keep the beer up23:45
openstackgerritJim Rollenhagen proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver  https://review.openstack.org/11386523:47
jrollNobodyCam: ^23:47
jrollNobodyCam: only setup.cfg conflicted, I'm going to +2 due to that23:47
NobodyCamTY jroll :)23:47
NobodyCamya23:48
rloohi, am I back just in time not to need to review anything?23:48
NobodyCamthis was to land at like 9:30 this morning23:48
jrollrloo: lol23:49
Shrewsmrda: ironic_driver.... was this ever used?23:49
NobodyCamlol23:49
NobodyCamoh no we have things to review23:49
Shrewsi mean, i guess it was, but i suppose it's an ancient artifact now23:49
mrdaShrews: no-one seems to think so.  It's a baremetal vestige23:49
jrollso what's actually left to review..23:49
mrdacopied over and modified23:49
Shrewsyeah23:49
jrollrloo: you should also look at https://review.openstack.org/118693 even though you touched it23:50
rloolooking...23:50
NobodyCamrloo: are you ok with my changes to https://review.openstack.org/#/c/11869323:50
jrollneed one more +2 on https://review.openstack.org/#/c/115885/23:50
rlooso wrt 118693, we decided not to address Joe's comments in test_driver.py?23:50
NobodyCamjroll: it too needs rebase now23:51
jrolland need eyes on this, but not as urgent: https://review.openstack.org/11656123:51
jrollNobodyCam: right, that23:51
* jroll does it23:51
mrdayay team!23:51
NobodyCammrda: ???23:51
openstackgerritJim Rollenhagen proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver  https://review.openstack.org/11588523:52
jrollNobodyCam: clean rebase ^^23:52
rloojroll: wrt 118693 and your nit in driver.py, the reason why I didn't do a ":returns:' is cuz that would mean doing a :param: and I wasn't sure if I should be making so many changes.23:52
jrollrloo: fair enough, thanks for the update :)23:52
rloothat driver.py is inconsistent wrt docstrings23:52
jrollyeah23:53
rlooso 118693 wasn't rebased on top of that other one with the arch.canonicalize() stuff.23:53
rloobut we need to make a change to the arch.canonicalize() stuff, unless one of you already did that. let me find/look at the other patch23:53
jrollright, the rebase went crazy23:54
jrollrloo: I think its already in nova23:54
mrdaFWIW, I added in the exception handler for canonicalize back into ironic via 11677223:55
mrdajroll: rloo: correct, the canonicalize stuff is already proposed in nova.23:55
rloomrda. great, that's what i was hoping.23:55
jrollcool23:55
mrdaI did that one backwards to try and get nova core signoff before we modified our tree23:55
NobodyCamjroll: 115885 now has two +2's23:55
jrollNobodyCam: neat, thanks :D23:55
mrda...because everyone was indecisive on that issue23:56
Shrewsmrda: 118693 lgtm. can be +A'd now, or do you just need the +2?23:56
*** chuckC has joined #openstack-ironic23:58
Shrewsor, jroll, even23:58
jrollShrews: needs jenkins23:58
Shrewspfft. what does he know23:58
mrdaShrews: Just a +2 for now, when Jenkins happens, then we can +A23:58
NobodyCamShrews: do we need 116772 first?23:58
Shrewsit's +2'd23:58
mrdaShrews: thanks23:58
rloomrda: so there's still the issue of that CONF.ironic.client_log_level. Do you have to ask Joe if he is -1 against it?23:59
jrollShrews: he's a big jerk23:59
jrollNobodyCam: 116772 is already in nova23:59
mrdarloo: I'll see if I can awaken him23:59
ShrewsNobodyCam: ugh. haven't seen that one.23:59

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