Monday, 2015-08-10

« Sunday, 2015-08-09 Index Tuesday, 2015-08-11 »
*** dimtruck is now known as zz_dimtruck00:09
*** vivek-ebay has quit IRC00:44
pksinghHi alee , are you around?01:04
openstackgerritOpenStack Proposal Bot proposed openstack/barbican: Updated from global requirements  https://review.openstack.org/21088201:04
*** everjeje has quit IRC01:32
*** zz_dimtruck is now known as dimtruck01:34
*** edtubill has joined #openstack-barbican01:41
openstackgerritPradeep Kumar Singh proposed openstack/barbican: Make tests in test_hrefs, test_quota py3 compatible  https://review.openstack.org/21094601:44
*** vivek-ebay has joined #openstack-barbican01:56
*** dimtruck is now known as zz_dimtruck02:06
openstackgerritPradeep Kumar Singh proposed openstack/barbican: Make tests in test_hrefs, test_quota py3 compatible  https://review.openstack.org/21094602:12
openstackgerritPradeep Kumar Singh proposed openstack/barbican: Make tests in test_hrefs, test_quota py3 compatible  https://review.openstack.org/21094602:12
*** vivek-ebay has quit IRC02:18
openstackgerritZhenyu Zheng proposed openstack/barbican: Drop downgrade field in alembic script.py.mako and version  https://review.openstack.org/20932302:24
*** vivek-ebay has joined #openstack-barbican02:59
*** zz_dimtruck is now known as dimtruck03:39
*** vivek-ebay has quit IRC03:54
*** dimtruck is now known as zz_dimtruck04:27
*** edtubill has quit IRC05:12
*** Nirupama has joined #openstack-barbican05:46
*** Nirupama has quit IRC06:00
*** shohel has joined #openstack-barbican06:06
pksinghHi folks, i have submitted some patches related to py3 support in barbican, can you please review so that i can complete the py3 complete porting soon, these patches are currently kind of blocker for me  link is "https://review.openstack.org/#/q/status:open+project:openstack/barbican+branch:master+topic:bp/barbican-py3,n,z"06:08
*** Nirupama has joined #openstack-barbican06:28
*** nickrmc83 has joined #openstack-barbican06:29
*** Nirupama has quit IRC06:45
*** everjeje has joined #openstack-barbican07:15
*** Nirupama has joined #openstack-barbican07:29
*** Nirupama has quit IRC07:29
*** Nirupama has joined #openstack-barbican07:32
*** shohel has quit IRC08:31
*** shohel1 has joined #openstack-barbican08:31
*** shohel1 has quit IRC09:06
*** shohel has joined #openstack-barbican09:06
*** tkelsey has joined #openstack-barbican09:12
*** Nirupama has quit IRC10:22
*** nickrmc84 has joined #openstack-barbican11:08
openstackgerritZhenyu Zheng proposed openstack/barbican: Drop downgrade field in alembic script.py.mako and version  https://review.openstack.org/20932311:09
*** nickrmc83 has quit IRC11:09
openstackgerritTim Kelsey proposed openstack/barbican: Fixing service endpoints for use with devstack  https://review.openstack.org/21111411:40
*** dave-mccowan has joined #openstack-barbican12:01
*** cbader has quit IRC12:52
*** elmiko has joined #openstack-barbican13:03
openstackgerritJoel Coffman proposed openstack/castellan: Remove copy_key operation  https://review.openstack.org/20612613:04
*** xek_ has joined #openstack-barbican13:06
*** lisaclark1 has joined #openstack-barbican13:06
*** diazjf has joined #openstack-barbican13:07
*** xek_ has quit IRC13:11
*** nkinder has joined #openstack-barbican13:16
*** alee_ has joined #openstack-barbican13:19
*** xek has quit IRC13:24
*** xek has joined #openstack-barbican13:25
*** xek has quit IRC13:26
*** xek has joined #openstack-barbican13:26
*** lisaclark1 has quit IRC13:31
*** kfarr has joined #openstack-barbican13:50
tkelseykfarr: are you about?13:56
kfarrtkelsey I was just about to step away for a meeting13:56
kfarrWill be back in an hour?13:56
tkelseyah kk, just wnated to let you know you were right about that Barbican endpoint :)13:57
tkelseynp13:57
kfarrOk!  Sounds like that bug needs to be fixed, but temporary workaround is setting the barbican endpoint in the config file?13:59
*** edtubill has joined #openstack-barbican14:00
*** pglass has joined #openstack-barbican14:05
*** SheenaG1 has left #openstack-barbican14:06
*** lisaclark1 has joined #openstack-barbican14:07
*** tkelsey has quit IRC14:08
*** tkelsey has joined #openstack-barbican14:09
tkelseykfarr: , yup I put up a patch https://review.openstack.org/#/c/211114/14:09
*** silos has joined #openstack-barbican14:12
*** lisaclark1 has quit IRC14:16
*** silos is now known as silos_away14:20
*** silos_away is now known as silos14:20
dave-mccowanreaperhulk ping14:20
*** lisaclark1 has joined #openstack-barbican14:22
*** lisaclark1 has quit IRC14:30
*** lisaclark1 has joined #openstack-barbican14:33
*** lisaclark2 has joined #openstack-barbican14:35
openstackgerritMichael McCune proposed openstack/castellan: refactoring castellan configuration  https://review.openstack.org/20618014:35
openstackgerritMichael McCune proposed openstack/castellan: refactoring castellan configuration  https://review.openstack.org/20618014:36
elmikoreaperhulk, kfarr ^^  =)14:37
*** lisaclark1 has quit IRC14:38
*** lisaclark1 has joined #openstack-barbican14:38
*** lisaclark2 has quit IRC14:39
*** spotz_zzz is now known as spotz14:45
*** dave-mcc_ has joined #openstack-barbican15:00
*** shohel has quit IRC15:00
*** dave-mccowan has quit IRC15:04
*** dave-mccowan has joined #openstack-barbican15:07
*** dave-mcc_ has quit IRC15:09
openstackgerritDave McCowan proposed openstack/barbican: Catch any exception from base64.b64decode during validation  https://review.openstack.org/21122415:09
*** lisaclark1 has quit IRC15:10
*** lisaclark1 has joined #openstack-barbican15:19
*** darrenmoffat has quit IRC15:26
*** shakamunyi has joined #openstack-barbican15:27
reaperhulkdave-mccowan: what's up15:27
*** arunkant has joined #openstack-barbican15:27
dave-mccowanreaperhulk when working on py3 compatibility issues, have you run across the API changes for base64.b64decode?  i was wondering if you have an endorsed solution (throw exception when input is not valid base64) that works for py2 and py3.15:27
reaperhulkpy3 might care more that it be a byte string? I don't remember15:29
reaperhulkbut for base64 itself python (and most other impls) are super liberal15:29
reaperhulkthey just throw away crap that's not valid15:29
reaperhulksilently.15:29
reaperhulkwhich leads to: https://github.com/pyca/cryptography/issues/199415:29
dave-mccowanpy3 does that (discards silently), with additional param (validate=True) if you want an exception.  py2 throws an exception always, and does not accept the validate= parameter.15:30
dave-mccowanyour example is on point.  barbican validators will start accepting/discarding crap in payload when using py3, which would be a change in behavior.15:32
reaperhulkpy2 actually accepts crap too as long as it's after a valid pad point15:32
reaperhulke.g. base64.b64decode("AAA=hello")15:32
reaperhulkthat will decode as \x00\x0015:32
reaperhulkwhich is obviously BS15:33
*** zz_dimtruck is now known as dimtruck15:34
*** darrenmoffat has joined #openstack-barbican15:34
dave-mccowanso, is the right answer for barbican to create a wrapper in barbican/common for b64encode, with "if py3: else:" to ensure the behavior we want?15:36
*** rellerreller has joined #openstack-barbican15:42
*** jorge_munoz has joined #openstack-barbican15:53
reaperhulkdave-mccowan: probably yes if you want strict decoding15:56
reaperhulkhow you implement validation in py2 is an interesting question though15:56
reaperhulkI guess you could look at the python source code to see if it's done in python or C ;)15:56
*** nickrmc84 has quit IRC16:02
*** silos1 has joined #openstack-barbican16:05
dave-mccowanreaperhulk i'm most interested in keeping barbican's behavior consistent between py2 and py3.  ideally, they'll accept and reject the same variations of garbage.16:08
*** silos has quit IRC16:08
*** lisaclark1 has quit IRC16:10
reaperhulkdave-mccowan: I'd say that strict decoding errors in py3 represents a significant improvement so trying to make py2 do that is a worthy goal16:20
*** lisaclark1 has joined #openstack-barbican16:27
*** vivek-ebay has joined #openstack-barbican16:33
*** diazjf has quit IRC16:52
kfarrjvrbanac, was that your camera we used to take the group photos for the midcycle?  Can you please forward them to me?16:53
*** pglass has quit IRC16:56
*** xaeth_afk is now known as xaeth17:07
*** jamielennox is now known as jamielennox|away17:11
*** dave-mccowan has quit IRC17:26
rellerrellerkfarr I was just about to send the same message to jvrbanac :)17:28
*** diazjf has joined #openstack-barbican17:31
*** pglass has joined #openstack-barbican17:35
elmikorellerreller, kfarr, i updated the config review to take the latest functional test config changes17:35
kfarrok elmiko, will take a look :)17:35
elmikothanks =)17:35
diazjfHey everyone, my patch is ready!! https://review.openstack.org/#/c/196876/ Also do you have the pictures from the mid-cycle meetup :)17:36
*** dave-mccowan has joined #openstack-barbican17:56
*** dave-mcc_ has joined #openstack-barbican17:58
jvrbanackfarr, rellerreller, yes I did. I just downloaded the photos yesterday. I'll send them when I get back home tomorrow18:00
jvrbanac^night18:00
jvrbanac^tonight18:00
* jvrbanac can't spell today18:00
kfarrthanks jvrbanac!!18:00
*** dave-mccowan has quit IRC18:01
*** rellerreller has quit IRC18:01
*** vivek-ebay has quit IRC18:04
diazjf^^ awesome18:06
*** pglbutt has joined #openstack-barbican18:09
*** lisaclark1 has quit IRC18:09
*** everjeje has quit IRC18:12
*** lisaclark1 has joined #openstack-barbican18:12
*** pglass has quit IRC18:12
*** vivek-ebay has joined #openstack-barbican18:23
*** lisaclark1 has quit IRC18:28
*** everjeje has joined #openstack-barbican18:29
*** ngupta has joined #openstack-barbican18:34
*** rm_work|away is now known as rm_work18:36
*** lisaclark1 has joined #openstack-barbican18:36
*** lisaclark1 has quit IRC18:43
*** SheenaG1 has joined #openstack-barbican18:51
*** SheenaG1 has left #openstack-barbican18:51
*** dave-mcc_ has quit IRC18:55
*** crc32 has joined #openstack-barbican19:05
*** dave-mccowan has joined #openstack-barbican19:07
*** gyee has joined #openstack-barbican19:16
*** ngupta_ has joined #openstack-barbican19:20
*** ngupta has quit IRC19:21
*** atiwari has joined #openstack-barbican19:28
*** rellerreller has joined #openstack-barbican19:29
*** jhfeng has joined #openstack-barbican19:50
redrobotWeekly meeting is starting in 1 min on #openstack-meeting-alt19:59
*** igueths has joined #openstack-barbican20:01
*** lisaclark1 has joined #openstack-barbican20:02
*** woodster_ has joined #openstack-barbican20:03
*** lisaclark1 has quit IRC20:03
*** lisaclark1 has joined #openstack-barbican20:04
*** lisaclark1 has quit IRC20:04
*** lisaclark1 has joined #openstack-barbican20:04
*** vivek-ebay has quit IRC20:17
*** tkelsey has quit IRC20:19
openstackgerritKaitlin Farr proposed openstack/castellan: Add unit tests for managed objects  https://review.openstack.org/20664920:25
*** diazjf has left #openstack-barbican20:37
*** lisaclark1 has quit IRC20:37
*** rellerreller has quit IRC20:41
*** chlong has joined #openstack-barbican20:49
*** lisaclark1 has joined #openstack-barbican20:55
*** silos1 has left #openstack-barbican21:00
rm_workthe sooner we can get all those Castellan CRs merged, the better21:00
rm_workI saw a couple go in, but there's still quite a few pending21:00
aleerm_work, woodster_ but yes, potentially it could be helpful in providing some migration options21:00
rm_workyeah21:00
aleeso I'd want to evaluate whats in there in light of said migration21:01
rm_workhow is THIS still not merged: https://review.openstack.org/#/c/191884/21:01
aleeand make it easier on ourselves21:01
rm_workah -1 from doug21:01
kfarrrm_work, it would be helpful to have the changes you are proposing posted somewhere for comments.  I realize you have working code in other projects, but it will look differently in castellan as the key manager interface has already changed from what was in cinder21:02
rm_workyeah21:02
rm_worki know, that is why i have been pushing to get your CRs merged21:03
rm_workso I can really get down to updating my code21:03
rm_workbased on something more stable and clean :P21:03
rm_workI was hoping that would have happened during the midcycle21:04
woodster_rm_work: you could just depend on a stack of CRs :) Don't want to spend hours rebasing things? :)21:04
rm_workbut i guess focus was elsewhere21:04
rm_workwoodster_: heh. heh heh. heh. T_T21:04
rm_workI will probably just have to do that, though there were some that aren't on top of each other21:05
rm_worklike https://review.openstack.org/#/c/206180/21:05
kfarrrm_work, can you please post links to your implementations?  I want to know more about how you've done it so maybe we can find a solution that will work for everyone21:05
kfarryeah, elmiko just rebased that one earlier today, it had been out of sunc21:06
kfarrsync21:06
rm_workhttps://github.com/openstack/octavia/tree/master/octavia/certificates21:06
rm_workthat is current, but it has diverged a bit from the original stuff21:06
rm_workand been packaged up in a way that more closely fits octavia21:07
rm_workI want to split it back out again to more closely match the Castellan packages21:07
rm_workbut it's honestly going to be pretty trivial i think21:07
elmikorm_work, do you think i should base that config change on top of some others? (i didn't want to create a crazy dep chain)21:08
rm_workelmiko: eh, i am hoping it won't be an issue if we can just MERGE THAT CHAIN21:08
rm_worklol21:08
elmikohehe21:08
elmikook, was just curious21:09
rm_workbut it might be useful to put it on the end of kfarr's chain of CRs if we really don't think that's going to happen soon <_<21:09
elmikothat's a fair point21:09
kfarrelmiko, I think yours is fine as is21:09
elmikoone way or the other, there will need to be changes21:09
elmikokfarr, ok, cool21:09
*** vivek-ebay has joined #openstack-barbican21:09
rm_workkfarr: are you going to update yours for doug's comment today?21:10
kfarrrm_work, that was the same one that was brought up earlier in the meeting21:10
kfarrWe decided it is fine as-is21:10
rm_workok, i may have missed that21:10
rm_workso that means it can merge presently? :P21:10
kfarrYeah, I think so!21:11
rm_workredrobot: https://review.openstack.org/#/c/191884/ GOGOGOGO21:11
*** tkelsey has joined #openstack-barbican21:12
rm_workthat is the parent CR, right? kfarr21:15
kfarrrm_work yes21:15
rm_worki am having trouble tracking this chain, is it all one CHAIN or is it really more of a tree? >_>21:15
kfarrOof, sort of a tree21:15
rm_workyeah... :/21:15
kfarrThe important ones: managed objects -> api changes -> updates in plugins21:16
rm_workwhat is redrobot doing <_< just click link -> click reply -> click +2 +121:16
rm_workit's so easy! :P21:16
*** tkelsey has quit IRC21:16
*** lisaclark1 has quit IRC21:20
elmikolol21:20
*** edtubill has quit IRC21:29
*** ngupta_ has quit IRC21:29
openstackgerritKaitlin Farr proposed openstack/castellan: Update the key manager API  https://review.openstack.org/20322721:31
openstackgerritKaitlin Farr proposed openstack/castellan: Update mock key manager  https://review.openstack.org/20587021:35
*** lisaclark1 has joined #openstack-barbican21:40
openstackgerritMerged openstack/barbican: Updated from global requirements  https://review.openstack.org/21088221:41
rm_workkfarr: just one question on https://review.openstack.org/#/c/203227/5 (commented)21:42
rm_worknot a big deal either, just curious21:42
rm_workdid redrobot go home?21:57
rm_workanyone else on with +2 powers? lol21:57
redrobotrm_work nope, still here... but I haven't been paying attention to IRC21:57
rm_workah21:57
redrobotrm_work what's up?21:57
rm_workredrobot: https://review.openstack.org/#/c/191884/1121:57
rm_workredrobot: +2/+A gogogo21:57
rm_workaugh and now woodster_ has commented21:58
rm_workbut his comment seems to be contrary to the decision made at the meeting just an hour or two ago?21:58
rm_workwell at the least i suppose another comment from you would be useful redrobot21:59
rm_workand woodster_ commented MORE just now lol22:00
woodster_rm_work: I thought the decision was to not pass an arg to the get_encoding or format methods? I'm bike shedding on the name :) If 'format' is indeed a synonym for 'encoding' here though, I think that would be more clear long term22:00
woodster_kfarr: ^^^22:00
rm_workclasses don't need a `pass` i don't think22:00
rm_workthey aren't functions22:00
*** nkinder has quit IRC22:01
redrobotwoodster_ was about to comment on there22:01
redrobotwoodster_ "format" is a property, not a method22:01
woodster_rm_work: that's probably true, just different than I've seen with other empty classes in the the barbican code base...no biggy though22:01
kfarrI think of format and encoding as two different things22:01
redrobotkfarr +1, after thinking about it, they definitely are different things22:01
*** lisaclark1 has quit IRC22:02
woodster_so format describes the data at rest, and encoding is the envelope around it perhaps?22:02
*** igueths has quit IRC22:03
redrobotwoodster_ a concret example would be for RSA keys22:03
*** nkinder has joined #openstack-barbican22:04
woodster_(I feel like I was denied a good content-types discussion at the mid cycle)22:04
kfarrYeah, format would be PKCS#8 or PKCS#1, encoding would be DER or PEM, yeah?22:04
rm_workkfarr: re-commented22:04
rm_workon https://review.openstack.org/#/c/203227/522:04
rm_workwhich is actually similar to woodster's comment on the other CR that touches the same line22:05
rm_workhttps://review.openstack.org/#/c/191884/11/castellan/key_manager/barbican_key_manager.py,cm22:05
redrobotwoodster_ what kfarr said22:05
rm_workwoodster_: if you want to comment on the algorithm thing, see here: https://review.openstack.org/#/c/203227/5/castellan/key_manager/barbican_key_manager.py,cm22:06
rm_workERK i am so sorry22:06
rm_worki left a parens unmatched >_<22:06
rm_workmy own OCD requires me to make another comment22:07
woodster_so encoding is some operation done to data, and the data has an intrinsic format, correct? If so it seems that both format and encoding could be properties then. Certainly knowledge of the current encoding of the data is required to perform a more advanced 'get_encoded('some other encoding')' call later?22:09
*** nkinder has quit IRC22:09
woodster_sorry, just trying to think this through22:09
*** nkinder has joined #openstack-barbican22:11
kfarrwoodster_ thanks for your thoughts!  I see where you are coming from. Right now, we just assume everything is DER encoded22:12
kfarrAnd we assume only one format at the moment, adding more formats in the future could require a more hefty conversion from one to the other22:12
*** xaeth is now known as xaeth_afk22:12
rm_workkfarr: re-commented again22:15
rm_workkfarr: maybe what you're looking for is getattr(managed_object, "algorithm", None) ?22:16
*** crc32 has quit IRC22:16
kfarrrm_work: commented22:17
rm_workdon't see a new comment yet :(22:17
kfarrrm_work sorry, forgot to click a button22:18
rm_workheh22:18
rm_workok so this all changes anyway22:19
rm_worki'll comment once that's rebased i guess?22:20
kfarrOk!22:20
kfarrWas hoping to make reviews easier by splitting things up into smaller patches, but I'm not so sure that's the case anymore since it's so confusing with all the dependencies :(22:21
rm_workheh yeah, dependency chains are rough22:21
rm_workespecially when some of them actually do have to touch the same code22:21
*** dimtruck is now known as zz_dimtruck22:24
kfarrkk I'm audi22:27
*** kfarr has quit IRC22:27
*** jamielennox|away is now known as jamielennox22:38
*** spotz is now known as spotz_zzz22:40
*** pglbutt has quit IRC22:41
*** vivek-ebay has quit IRC22:44
*** lisaclark1 has joined #openstack-barbican22:54
woodster_alee, are you there?23:10
*** rellerreller has joined #openstack-barbican23:17
*** vivek-ebay has joined #openstack-barbican23:26
*** jhfeng has quit IRC23:33
*** lisaclark1 has quit IRC23:43
*** rellerreller has quit IRC23:51
« Sunday, 2015-08-09 Index Tuesday, 2015-08-11 »

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