| *** yamahata has quit IRC | 00:04 | |
| *** markmcclain1 has quit IRC | 00:14 | |
| *** yamahata has joined #openstack-oslo | 01:23 | |
| *** amotoki has joined #openstack-oslo | 01:36 | |
| *** yamahata has quit IRC | 01:54 | |
| *** yamahata has joined #openstack-oslo | 01:54 | |
| *** mriedem has left #openstack-oslo | 02:03 | |
| *** mriedem has quit IRC | 02:03 | |
| *** mgagne1 is now known as mgagne | 02:30 | |
| *** zhiyan_ is now known as zhiyan | 02:35 | |
| *** gtt116 has joined #openstack-oslo | 03:03 | |
| gtt116 | any body online? | 03:03 |
|---|---|---|
| *** jecarey has quit IRC | 03:03 | |
| *** lbragstad has joined #openstack-oslo | 03:07 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Increase usefulness of the retry component compile errors https://review.openstack.org/98681 | 03:57 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Switch to a restructuredtext README file https://review.openstack.org/98623 | 03:57 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Denote that other projects can use this library https://review.openstack.org/98624 | 03:59 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Increase usefulness of the retry component compile errors https://review.openstack.org/98681 | 04:06 |
| *** praneshp has quit IRC | 04:42 | |
| *** praneshp has joined #openstack-oslo | 04:42 | |
| *** lbragstad has quit IRC | 05:01 | |
| *** zhiyan is now known as zhiyan_ | 05:04 | |
| *** flashgordon is now known as jogo | 05:46 | |
| *** gtt116 has quit IRC | 05:49 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Address concurrent mutation of sqlalchemy backend https://review.openstack.org/98689 | 05:53 |
| *** zhiyan_ is now known as zhiyan | 06:08 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Clarify locked decorator is for instance methods https://review.openstack.org/98691 | 06:10 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Sync our version interprocess lock https://review.openstack.org/98696 | 06:40 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Sync our version of the interprocess lock https://review.openstack.org/98696 | 06:45 |
| openstackgerrit | lifeless proposed a change to openstack/oslo-specs: Semantic version support for pbr https://review.openstack.org/96608 | 06:46 |
| *** jaosorior has joined #openstack-oslo | 07:14 | |
| *** ihrachyshka has joined #openstack-oslo | 07:27 | |
| *** tkelsey has joined #openstack-oslo | 07:34 | |
| *** markmc has joined #openstack-oslo | 07:53 | |
| *** flaper87|afk is now known as flaper87 | 08:01 | |
| *** russell_h has quit IRC | 08:01 | |
| *** russell_h has joined #openstack-oslo | 08:02 | |
| *** openstackgerrit has quit IRC | 08:02 | |
| *** gilliard_ has quit IRC | 08:04 | |
| *** openstackgerrit has joined #openstack-oslo | 08:07 | |
| *** praneshp has quit IRC | 08:09 | |
| *** anteaya has quit IRC | 08:10 | |
| openstackgerrit | Christian Berendt proposed a change to openstack/oslo.vmware: replace iterator.next() with next(iterator) https://review.openstack.org/98710 | 08:11 |
| *** russell_h has quit IRC | 08:13 | |
| *** russell_h has joined #openstack-oslo | 08:16 | |
| *** russell_h has quit IRC | 08:21 | |
| *** russell_h has joined #openstack-oslo | 08:24 | |
| openstackgerrit | Ilya Pekelny proposed a change to openstack/oslo.db: oslo.db.api module documentation https://review.openstack.org/93434 | 08:28 |
| *** russell_h has quit IRC | 08:31 | |
| *** russell_h has joined #openstack-oslo | 08:33 | |
| *** russell_h has quit IRC | 08:38 | |
| *** yamahata has quit IRC | 08:41 | |
| *** russell_h has joined #openstack-oslo | 08:42 | |
| openstackgerrit | Ilya Pekelny proposed a change to openstack/oslo.db: oslo.db.options module documentation https://review.openstack.org/93436 | 08:47 |
| haypo | dhellmann: hi. i hesitate to vote -1 or even -2 for https://review.openstack.org/#/c/98429/ | 08:55 |
| haypo | dhellmann: i don't like long lines using *tricks* like "var = var or ..." | 08:56 |
| haypo | dhellmann: and i prefer an explicit loop instead of a lambda function with a call to list.extend(...) | 08:56 |
| haypo | i don't think that it adds any value, some changes make the code less readable IMO | 08:56 |
| *** russell_h has quit IRC | 08:58 | |
| *** russell_h has joined #openstack-oslo | 08:59 | |
| *** russell_h has quit IRC | 09:02 | |
| *** stannie has joined #openstack-oslo | 09:03 | |
| *** russell_h has joined #openstack-oslo | 09:04 | |
| *** russell_h has quit IRC | 09:14 | |
| *** russell_h has joined #openstack-oslo | 09:14 | |
| *** russell_h has quit IRC | 09:18 | |
| *** russell_h has joined #openstack-oslo | 09:20 | |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/oslo.config: Add cfgfilter.ConfigFilter https://review.openstack.org/97233 | 09:24 |
| *** russell_h has quit IRC | 09:26 | |
| *** erecio has quit IRC | 09:42 | |
| *** erecio has joined #openstack-oslo | 09:43 | |
| sileht | markmc, Can you review https://review.openstack.org/#/c/75365 ? (I see your progress on the config filter, I hope ceilometer have been fixed before your works on the filter is finished :p) | 09:49 |
| markmc | sileht, will do | 09:50 |
| sileht | markmc, thx | 09:53 |
| haypo | hum, i should try to review oslo messaging patches | 09:54 |
| jaosorior | haypo, is there a specific reason you don't like the "var = var or..." notation? I add more of that stuff in this patch that depends on dhellmann 's patches, here: https://review.openstack.org/#/c/98429/ | 09:59 |
| jaosorior | and replaced some explicit loops with generators | 09:59 |
| haypo | jaosorior: i prefer an explicit if | 09:59 |
| jaosorior | I think the "var = var or ..." is quite readable, so I don't see why it would be a problem :/ | 10:00 |
| haypo | jaosorior: the "localdir = localedir" instruction is not very useful :) | 10:00 |
| *** russell_h has joined #openstack-oslo | 10:01 | |
| haypo | jaosorior: i don't know how to argue why an explicit if is better | 10:01 |
| jaosorior | There is not much difference really, my only argument is that it makes the code a bit more compact, and is still quite readable | 10:02 |
| haypo | jaosorior: that's why i didn't touch the patch (i didn't leave a comment and i didn't vote) | 10:02 |
| haypo | jaosorior: my general rule is one instruction per line, or at least reduce the number of instructions per line | 10:03 |
| haypo | jaosorior: "var = ... or ..." is 2 instructions for example, assign and the "or" trick | 10:03 |
| haypo | jaosorior: you also replaced "if domain is None:" with "domain or ...", it's not the same, you don't check "is None" | 10:04 |
| jaosorior | well, an empty domain is also invalid | 10:05 |
| haypo | jaosorior: in my opinion, python code must not be compact :) compact means harder to parser, harder to maintain, harder to debug, etc. | 10:05 |
| jaosorior | at least the way it was used | 10:05 |
| jaosorior | so now that's checked also | 10:06 |
| jaosorior | and I would say the compact argument is subjective | 10:07 |
| haypo | jaosorior: i'm fine with "translator = lang.gettext if six.PY3 else lang.ugettext", because here you have two different expressions depending on the test | 10:07 |
| haypo | jaosorior: "var = var or ..." looks inefficient to me | 10:07 |
| jaosorior | I'm flexible to changing the patch, that's no big deal | 10:08 |
| haypo | jaosorior: no coding style is "correct", we just have to agree on a coding style. that's why i would like to discuss it here ;) | 10:09 |
| jaosorior | uhm... maybe a little benchmark would be a good idea | 10:09 |
| haypo | jaosorior: a micro-benchmark is useless :) the code is probably only called once | 10:09 |
| jaosorior | and to hear what people's preferences are | 10:09 |
| haypo | i don't really care of performances here ;) | 10:09 |
| jaosorior | what's your timezone? | 10:10 |
| *** anteaya has joined #openstack-oslo | 10:10 | |
| jaosorior | maybe it would be good to wait until more people are online and discuss these style questions | 10:11 |
| haypo | jaosorior: i live in france | 10:11 |
| jaosorior | oh, ok, so I guess two hours less than here | 10:12 |
| jaosorior | I live in Finland | 10:12 |
| jaosorior | do you have a particular opinion about the generator bits in the patch I did? | 10:13 |
| haypo | jaosorior: good https://review.openstack.org/#/c/98429/2/oslo/i18n/_translate.py,cm | 10:14 |
| haypo | jaosorior: bad https://review.openstack.org/#/c/98429/2/oslo/i18n/gettextutils.py,cm -- i don't like lambda | 10:14 |
| haypo | for this one, IMO the current code is more readable and easier to understand | 10:15 |
| *** zhiyan is now known as zhiyan_ | 10:17 | |
| jaosorior | haypo, I'm not very fond of explicit loops appending to lists, since we have generators available. Would you prefer if I would define a function explicitly instead of using a lambda? | 10:17 |
| haypo | jaosorior: it would be better but i still prefer an explicit loop | 10:18 |
| haypo | i don't like to abuse generators | 10:18 |
| haypo | if you are motivated, you should discuss the coding style on the openstack-dev mailing list ;) | 10:19 |
| haypo | for example, "if var: var = ..." vs "var = var or ...", for ...: if ...: list.append(...) vs list.extend(it for it in ... if ...), etc. | 10:20 |
| jaosorior | sure, that sounds democratic enough | 10:20 |
| haypo | be careful, all coding style discuss starts a troll/flamewar :) | 10:21 |
| jaosorior | yeah, that's what I was fearing | 10:21 |
| haypo | since as i wrote, there is no "good" coding style | 10:21 |
| jaosorior | well, I could get rid of the lambda, and of the other one, for that matter, since there are two in that function | 10:22 |
| jaosorior | but keep using those generators, at least that's some middle ground | 10:22 |
| openstackgerrit | Juan Antonio Osorio Robles proposed a change to openstack/oslo.i18n: Trivial refactors for gettextutils https://review.openstack.org/98429 | 10:37 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/oslo.config: Add cfgfilter.ConfigFilter https://review.openstack.org/97233 | 11:14 |
| *** DuncanT has quit IRC | 11:41 | |
| ihrachyshka | markmc: hey! around? | 12:02 |
| ihrachyshka | markmc: I have a question re your kombu (and amqp) driver in oslo.messaging | 12:03 |
| ihrachyshka | markmc: https://review.openstack.org/#/c/38339/1/oslo/messaging/_drivers/impl_rabbit.py , line 1128. why's that? | 12:03 |
| ihrachyshka | dhellmann: you may also have insight into why that's needed ^^ | 12:04 |
| *** stannie has quit IRC | 12:05 | |
| markmc | ihrachyshka, hey | 12:07 |
| markmc | ihrachyshka, somewhere in the driver code we call to_dict() | 12:07 |
| *** erecio has quit IRC | 12:08 | |
| *** stannie has joined #openstack-oslo | 12:08 | |
| markmc | ihrachyshka, that can be refactored though - remove the call to to_dict() and remove the wrapper Context object | 12:08 |
| markmc | ihrachyshka, as far as oslo.messaging is concerned, context is just a free-form dict | 12:08 |
| ihrachyshka | markmc: I'm currently porting neutron to oslo.messaging, and I have a Context class to store all the relevant data, and it has .to_dict() method. The problem is that this hack enforces me to add .iteritems() method to the class, even though if there wouldn't be such a hack, existing .to_dict() would work | 12:09 |
| ihrachyshka | markmc: does it mean I should translate Context to a generic dict somewhere in serialiser? | 12:10 |
| markmc | ihrachyshka, yes, exactly | 12:10 |
| markmc | ihrachyshka, look in nova.rpc at the request context serializer | 12:10 |
| sileht | ihrachyshka, http://git.openstack.org/cgit/openstack/ceilometer/tree/ceilometer/messaging.py#n36 | 12:11 |
| markmc | yeah, that one :) | 12:11 |
| ihrachyshka | thanks guys! | 12:11 |
| markmc | np | 12:11 |
| ihrachyshka | I've almost attempted to fix the issue the wrong way, adding another layer of hackery ;) | 12:11 |
| markmc | no problem that another layer of hackery can't solve :) | 12:12 |
| ihrachyshka | exactly! | 12:12 |
| ihrachyshka | it's like 'hmmm, it fails because .iteritems() is missing? ok, let's add one! great, it passes further! moving to the next failure' :) | 12:13 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/oslo-incubator: cfgfilter: fix some docstring issues https://review.openstack.org/98771 | 12:27 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/oslo-incubator: cfgfilter: don't re-register options https://review.openstack.org/98772 | 12:27 |
| *** erecio has joined #openstack-oslo | 12:35 | |
| *** stannie has quit IRC | 12:43 | |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/oslo.messaging: Use cfgfilter to hide transport opts from API https://review.openstack.org/98779 | 12:56 |
| dhellmann | haypo: you can leave that comment on the patch, it's easier to discuss there -- some of the changes seem OK, some I was indifferent about | 13:12 |
| dhellmann | haypo, jaosorior: I'm still catching up on the scrollback, but it looks like good discussion on style. :-) | 13:13 |
| *** mriedem has joined #openstack-oslo | 13:14 | |
| haypo | dhellmann: hi | 13:16 |
| dhellmann | haypo: hi! | 13:26 |
| *** pafuent has joined #openstack-oslo | 13:39 | |
| jaosorior | haypo, dhellman, well, now that the timezone is more adequate for everyone, do you have any insights on what was discussed? | 13:40 |
| pafuent | Hi. I want to update generate_sample.sh/check_uptodate.sh in the Blazar repository. Should I get the files from the repo and update Blazar manually? or There is another way to do it? | 13:42 |
| *** jecarey has joined #openstack-oslo | 13:44 | |
| dhellmann | jaosorior: I don't mind "var = var or default" in the case where a default is being ensured, esp. when using a mutable value as a default. I do worry in some cases that removing explicit if statements makes test coverage harder to calculate accurately, though I think the tools are getting better at supporting branch coverage. | 13:46 |
| dhellmann | pafuent: you might find these instructions helpful: https://wiki.openstack.org/wiki/Oslo#Syncing_Code_from_Incubator | 13:46 |
| dhellmann | pafuent: we have a tool to help copy modules and scripts from the incubator, since part of copying involves rewriting some defaults in the code to use the project name instead of "oslo" | 13:47 |
| pafuent | dhellmann: Thanks. I'll check that link | 13:48 |
| dhellmann | pafuent: let us know if you run into trouble | 13:48 |
| pafuent | dhellmann: OK | 13:48 |
| ihrachyshka | dhellmann: hey, can we consider latest jsonutils changes in incubator to add support for simplejson for stable back port? | 14:01 |
| jaosorior | dhellman, haypo, well, I see your point, but isn't that case covered by the coverage tool setting a "partially covered" flag on that line? | 14:01 |
| *** beekneemech is now known as bnemec | 14:02 | |
| *** tsekiyama has joined #openstack-oslo | 14:02 | |
| dhellmann | ihrachyshka: I'm not sure if that counts as a "feature" or not | 14:03 |
| dhellmann | jaosorior: does the tool say which part of the line is covered? My knowledge of that aspect of coverage.py is thin. | 14:03 |
| boris-42 | dhellmann one step closer https://github.com/stackforge/osprofiler | 14:04 |
| boris-42 | dhellmann going to make huge read me | 14:04 |
| ihrachyshka | dhellmann: no API changes, just internal performance improvement. I think we should consider it in the same way as any other performance improvements that do not change visible behaviour. | 14:04 |
| jaosorior | dhellman, no, it doesn't say, as far as I've noticed | 14:04 |
| dhellmann | jaosorior: so that would make it more difficult to get useful information from coverage in that case | 14:05 |
| jaosorior | OK, that makes sense | 14:05 |
| jaosorior | should I then revert all the occurrences of the "var = var or ..."? | 14:05 |
| dhellmann | jaosorior: another case where "var = something or somethingelse" can go wrong is with exceptions raised in the expressions something or somethingelse -- depending on how close the expressions are, it's not always clear which one is causing the error (I don't think that came up in your changes, I'm just pointing it out) | 14:06 |
| jaosorior | alright | 14:07 |
| jaosorior | I'll revert those in some hours (gotta head home now) | 14:07 |
| dhellmann | jaosorior: I don't mind using those for ensuring a valid default, but I would have to look at your patch again to know if there are any cases I would prefer you to remove | 14:07 |
| jaosorior | can you point it out as comments on the patch? | 14:07 |
| dhellmann | jaosorior: thanks, I appreciate the work you're doing on this | 14:07 |
| dhellmann | jaosorior: sure, I'll try to get back to it today or tomorrow | 14:08 |
| *** ildikov has joined #openstack-oslo | 14:11 | |
| *** zzzeek has joined #openstack-oslo | 14:29 | |
| *** dims has joined #openstack-oslo | 14:36 | |
| *** erecio has quit IRC | 14:48 | |
| *** erecio has joined #openstack-oslo | 14:49 | |
| *** erecio has quit IRC | 14:49 | |
| *** ihrachyshka_ has joined #openstack-oslo | 14:54 | |
| *** ihrachyshka_ has quit IRC | 14:54 | |
| *** lbragstad has joined #openstack-oslo | 14:54 | |
| *** ihrachyshka has quit IRC | 14:57 | |
| bnemec | Argh: https://bugs.launchpad.net/oslo/+bug/1327946 | 15:00 |
| *** mriedem has quit IRC | 15:07 | |
| *** mriedem has joined #openstack-oslo | 15:11 | |
| *** erecio has joined #openstack-oslo | 15:12 | |
| *** arnaud__ has joined #openstack-oslo | 15:15 | |
| *** morganfainberg_Z is now known as morganfainberg | 15:18 | |
| *** arnaud__ has quit IRC | 15:20 | |
| zzzeek | dhellmann: what’s the typical process by which we can upgrade SQLAlchemy versions? by which I mean how much can I refer to SQLAlchemy 0.9 features for oslo.db improvements, as it seems like we’re still on 0.8 (or 0.7? I see both in the setup )? | 15:29 |
| dhellmann | zzzeek: https://wiki.openstack.org/wiki/Requirements describes, somewhat, the requirements management process | 15:31 |
| dhellmann | because of our backwards-compatibility requirements it may take some time to be able to update to a new version | 15:32 |
| dhellmann | knowing what the distros actually ship in their packages would help, too | 15:32 |
| *** russell_h has quit IRC | 15:32 | |
| *** russell_h has joined #openstack-oslo | 15:32 | |
| zzzeek | dhellmann: basically, all of this wrap DB exception stuff at the ORM level / explicilt everywhere makes me sad | 15:34 |
| dhellmann | zzzeek: looking across all projects, I see mostly >=0.7.8,<=0.9.99 -- http://paste.openstack.org/show/83418/ | 15:34 |
| zzzeek | dhellmann: and this should be internalized into SQLAlhcemy hooks | 15:35 |
| dhellmann | zzzeek: I'm not sure what you mean | 15:35 |
| zzzeek | for example https://bugs.launchpad.net/oslo/+bug/1214341 | 15:35 |
| dhellmann | so there's a way to make sqlalchemy convert exceptions for us? | 15:35 |
| zzzeek | there are some hooks right now, they would need some improvement to support | 15:36 |
| dhellmann | nice | 15:36 |
| dhellmann | we might want different exceptions for the same error on different ables -- NoSuchInstance vs. NoSuchNetwork or something like that | 15:37 |
| dhellmann | zzzeek: I suggest you start a wiki page or etherpad or something with notes about these sorts of changes. Some are going to be long-term, and we'll need to enlist people to help with them. Some will be easier, and can be isolated in oslo.db. | 15:38 |
| openstackgerrit | Brad Pokorny proposed a change to openstack/oslo-incubator: Mask passwords included without quotes at the ends of commands https://review.openstack.org/97305 | 15:39 |
| dhellmann | zzzeek: eventually the notes can become specs with more details so we can split up the work and you don't end up having to do it all :-) | 15:40 |
| zzzeek | well there’s really just an assortment of patterns I’d add to oslo.db that I think improve upon what is there, it’s not that much work to put them in oslo.db codewise, the work is getting everything to work in all cases, deal with backwards compat, migration paths etc. | 15:41 |
| dhellmann | right | 15:42 |
| dhellmann | having them in oslo.db doesn't ensure adoption, either | 15:42 |
| zzzeek | well that’s somehting else | 15:43 |
| dhellmann | making pattern changes across all of the projects tends to take a lot of education and patience | 15:43 |
| zzzeek | if a project has their own SQLAclhemy code out of nowhere, that’s sort of their own problem | 15:43 |
| zzzeek | but within oslo.db I’d like to make it such that a projecvt that uses it doesn’t have to make many decisions as far as DB integration, there will be a few well-understood paths for things and that’s it | 15:44 |
| zzzeek | oslo.db is framework code, it needs to show the way for those who are using it | 15:44 |
| dhellmann | true, I meant more things like setting up the exception translation where it would be project-specific | 15:44 |
| zzzeek | OK, well there might be multi-levels to that. at the core level I’d like to have more of an oslo.db specific set of rules that add whatever information is possible. project-specific rules would then have a hook within oslo.db | 15:45 |
| zzzeek | that is, only oslo.db deals with the SQLA hooks | 15:45 |
| dhellmann | I like that approach | 15:46 |
| zzzeek | https://github.com/openstack/oslo.db is the main repo? you folks don’t use tags or anything? | 15:50 |
| * zzzeek reads https://wiki.openstack.org/wiki/BranchModel | 15:51 | |
| zzzeek | ah http://tarballs.openstack.org/oslo.db/ , we have no tags! | 15:54 |
| *** mrodden has quit IRC | 15:54 | |
| dhellmann | zzzeek: http://git.openstack.org/cgit/openstack/oslo.db and there won't be any tags until we have a release | 15:58 |
| openstackgerrit | Gordon Sim proposed a change to openstack/oslo-specs: Added spec for proposed AMQP 1.0 driver https://review.openstack.org/96729 | 16:00 |
| zzzeek | dhellmann: dumb question, what is the relationship between git.openstack.org and github.com/openstack? | 16:03 |
| *** mrodden has joined #openstack-oslo | 16:07 | |
| *** ildikov_ has joined #openstack-oslo | 16:08 | |
| *** yamahata has joined #openstack-oslo | 16:10 | |
| *** ildikov has quit IRC | 16:10 | |
| *** markmcclain has joined #openstack-oslo | 16:21 | |
| *** yamahata has quit IRC | 16:22 | |
| *** yamahata has joined #openstack-oslo | 16:22 | |
| *** praneshp has joined #openstack-oslo | 16:26 | |
| *** yamahata has quit IRC | 16:35 | |
| *** yamahata has joined #openstack-oslo | 16:35 | |
| dhellmann | zzzeek: the github repos are mirrors of git.openstack.org | 16:56 |
| zzzeek | ok | 16:56 |
| dhellmann | zzzeek: the CI system pushes merges up there, and automatically closes pull requests with instructions for how to submit through gerrit | 16:56 |
| zzzeek | CI pushes … from git.openstack to github ? | 17:04 |
| zzzeek | e.g. an approved changeset on gerrit goes to git.openstack first | 17:05 |
| *** harlowja_away is now known as harlowja | 17:09 | |
| dhellmann | zzzeek: yes, that's right (sorry for the lag, was in a meeting) | 17:15 |
| *** tkelsey has quit IRC | 17:15 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/oslo-specs: Graduate oslo.serialization https://review.openstack.org/97315 | 17:22 |
| *** markmcclain has quit IRC | 17:30 | |
| *** markmcclain has joined #openstack-oslo | 17:33 | |
| *** russell_h has quit IRC | 17:36 | |
| *** pafuent has left #openstack-oslo | 17:37 | |
| *** russell_h has joined #openstack-oslo | 17:39 | |
| *** russell_h has quit IRC | 17:41 | |
| *** russell_h has joined #openstack-oslo | 17:41 | |
| *** mkoderer has quit IRC | 18:02 | |
| *** stannie has joined #openstack-oslo | 18:08 | |
| *** stannie has quit IRC | 18:08 | |
| *** mrodden has left #openstack-oslo | 18:16 | |
| *** erecio has quit IRC | 18:19 | |
| *** erecio has joined #openstack-oslo | 18:19 | |
| *** dims has quit IRC | 18:25 | |
| *** dims has joined #openstack-oslo | 18:25 | |
| *** ildikov has joined #openstack-oslo | 18:27 | |
| *** ildikov has quit IRC | 18:29 | |
| *** markmcclain has quit IRC | 18:30 | |
| *** ildikov_ has quit IRC | 18:30 | |
| *** markmcclain has joined #openstack-oslo | 18:30 | |
| *** markmcclain has quit IRC | 18:31 | |
| *** markmcclain has joined #openstack-oslo | 18:31 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Make intentions a tuple (to denote immutability) https://review.openstack.org/98851 | 18:39 |
| openstackgerrit | Juan Antonio Osorio Robles proposed a change to openstack/oslo.i18n: Trivial refactors for gettextutils https://review.openstack.org/98429 | 18:45 |
| *** ildikov|2 has joined #openstack-oslo | 18:45 | |
| *** ildikov|2 is now known as ildikov | 18:54 | |
| *** markmcclain has quit IRC | 19:00 | |
| *** erecio has quit IRC | 19:04 | |
| *** erecio has joined #openstack-oslo | 19:04 | |
| zzzeek | hey dhellmann can I use http://paste.openstack.org/show/83418/ as a guide to all projects that are currently on SQLA? e.g. to look for patterns used anywhere ? | 19:05 |
| *** markmcclain has joined #openstack-oslo | 19:07 | |
| *** praneshp has quit IRC | 19:08 | |
| openstackgerrit | Doug Hellmann proposed a change to openstack/oslotest: Add API documentation https://review.openstack.org/98861 | 19:15 |
| *** ihrachyshka has joined #openstack-oslo | 19:15 | |
| dhellmann | zzzeek: those are just the integrated projects, but it should give you all of those | 19:16 |
| zzzeek | ok... | 19:17 |
| dhellmann | zzzeek: here's a thing for cloning all of the git repos: http://lists.openstack.org/pipermail/openstack-dev/2013-October/017532.html | 19:18 |
| zzzeek | ah thanks | 19:18 |
| *** markmcclain1 has joined #openstack-oslo | 19:19 | |
| dhellmann | zzzeek: my copy takes about 8 GB, but I may have some extra stuff in there | 19:19 |
| zzzeek | ive checked them all out manually | 19:19 |
| *** markmcclain1 has quit IRC | 19:20 | |
| *** markmcclain1 has joined #openstack-oslo | 19:20 | |
| dhellmann | ok, that script will give you *ALL* of them, including the attic and stackforge | 19:20 |
| zzzeek | it does | 19:23 |
| *** markmcclain has quit IRC | 19:23 | |
| zzzeek | so I am surprised that these reqs all state 0.7.8 - 0.9.9, that is very wide open ? | 19:23 |
| dhellmann | we're trying to take into account what the different distros include, I believe | 19:23 |
| zzzeek | I’m trying to find where tests are being run to see what versions are really used, im not seeing quite on jenkins.openstack.org where things are ? | 19:23 |
| zzzeek | I see | 19:23 |
| dhellmann | you should probably do some digging in the openstack-dev archives; I know there were some discussions about sqlalchemy versions in the last year | 19:24 |
| dhellmann | yeah, tests are weird | 19:24 |
| dhellmann | zuul manages the test jobs, and passes them out to one of several jenkins servers to run | 19:24 |
| zzzeek | so how do I get a sense of that, 0.7.8 is very old, there’s more than a few pretty awful connection pool bugs in early 0.8 versions | 19:25 |
| dhellmann | I believe we're running on precise right now, although we do have some centos servers for python 2.6 I think | 19:25 |
| zzzeek | and here I am looking to things that will work best with new features I want to add in 0.9.5 | 19:25 |
| dhellmann | you know, I *believe* we actually install sqla via pip, so we should get the most current version available in our tests | 19:25 |
| zzzeek | yeah | 19:25 |
| zzzeek | im looking at RH and it seems to be 0.8.4 based on the current RPM that redhat 6 has | 19:26 |
| dhellmann | the folks in #openstack-infra would be able to give you a more definitive answer, but I think the tests use pip but we want to support the distro version of the package | 19:26 |
| zzzeek | OK so then…what distros do we support and when ? :) sorry am i just leaning on you too hard here | 19:26 |
| zzzeek | this is a lot to ask | 19:27 |
| dhellmann | nah, this is all good bootstrapping | 19:27 |
| dhellmann | we support long term releases, so whatever RHEL is at now and the LTS from Canonical (12.04, moving to 14.04 (I think)) | 19:27 |
| dhellmann | I'm not sure what the deal is with debian or suse | 19:28 |
| dhellmann | either ttx or the folks in infra should be able to help you find the people who are our interfaces for the distros | 19:28 |
| zzzeek | shoudlnt this be in the main openstack docs | 19:29 |
| dhellmann | I think we leave it up to the distros to claim whatever *they* support, since we don't actually do the support ourselves | 19:29 |
| dhellmann | and there is probably a release document somewhere that says some of this | 19:29 |
| zzzeek | yeah i am poking around... | 19:29 |
| zzzeek | um well what do you think of the notion of new sqlalhcemy features that oslo.db can use when version > 0.9.X is detected, then fallback to known semi-public API technique for existing versions ? | 19:30 |
| dhellmann | that seems like a reasonable approach | 19:30 |
| zzzeek | OK, so folks are OK with a little bit of sqla._underscore here and there if I pin it to released versions | 19:31 |
| *** praneshp has joined #openstack-oslo | 19:31 | |
| dhellmann | it also seems like something to make sure we release note, so deployers know their options | 19:31 |
| zzzeek | OK so, i like this wiki idea | 19:31 |
| dhellmann | I would rather not rely on private APIs | 19:31 |
| dhellmann | but if we only have it in oslo.db, and it comes with a big warning comment, it might be ok | 19:32 |
| zzzeek | right but the reason we don’t like to do that is beacuse, they are unstable. my proposal is, we rely on those APIs only for existing, released versions of SQLA - new versions will offer public API that oslo.db will be set up to accommodate at the same time | 19:32 |
| dhellmann | that makes sense | 19:32 |
| zzzeek | im not sure of the CI environemnt yet but in theory we shoudl ahve oslo.db tested against multiple SQLA versions | 19:33 |
| zzzeek | unsure meaning, I don’t know where I can see CI logs of actual oslo.db test runs... | 19:33 |
| zzzeek | geuss it’s not these https://jenkins.openstack.org/view/Devstack%20Nodes/search/?q=oslo.db | 19:34 |
| dhellmann | links to the logs are saved in gerrit on the patch that triggered the build, so you could look at an existing patch | 19:34 |
| zzzeek | good idea | 19:34 |
| dhellmann | there's also the logstash UI, but I don't know if that's public yet | 19:34 |
| dhellmann | that's really more for looking for patterns across builds anyway | 19:34 |
| dhellmann | hmm, internal issues coming up, I need to step away for a bit | 19:35 |
| zzzeek | ok | 19:35 |
| *** yfujioka has joined #openstack-oslo | 19:40 | |
| *** markmcclain1 has quit IRC | 19:42 | |
| *** yfujioka has quit IRC | 19:51 | |
| openstackgerrit | Bertrand Lallau proposed a change to openstack/oslo-incubator: Improve amqp rpc logging format for debug purpose https://review.openstack.org/98873 | 19:58 |
| *** jecarey has quit IRC | 20:05 | |
| *** morganfainberg is now known as morganfainberg_Z | 20:06 | |
| *** boris-42 has quit IRC | 20:08 | |
| openstackgerrit | Tomoki Sekiyama proposed a change to openstack/oslo-specs: Add ChainingRegExpFilter into rootwrap https://review.openstack.org/98536 | 20:11 |
| *** boris-42 has joined #openstack-oslo | 20:11 | |
| *** jecarey has joined #openstack-oslo | 20:19 | |
| *** arnaud__ has joined #openstack-oslo | 20:30 | |
| *** markmcclain has joined #openstack-oslo | 20:34 | |
| bnemec | zzzeek: There's a pip-freeze file that gets created for some of the test jobs: http://logs.openstack.org/67/96467/14/check/check-tempest-dsvm-full/310a9b5/logs/pip-freeze.txt.gz | 20:52 |
| openstackgerrit | lifeless proposed a change to openstack/oslo-specs: Semantic version support for pbr https://review.openstack.org/96608 | 20:52 |
| *** morganfainberg_Z is now known as morganfainberg | 21:10 | |
| *** praneshp has quit IRC | 21:11 | |
| *** arnaud__ has quit IRC | 21:14 | |
| *** arnaud__ has joined #openstack-oslo | 21:26 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Clarify locked decorator is for instance methods https://review.openstack.org/98691 | 21:27 |
| *** mrda-longweekend is now known as mrda | 21:30 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Rename additional to general/higher-level https://review.openstack.org/98617 | 21:31 |
| *** tsekiyama has quit IRC | 21:32 | |
| *** arnaud__ has quit IRC | 21:37 | |
| *** mriedem has quit IRC | 21:45 | |
| *** praneshp has joined #openstack-oslo | 21:50 | |
| *** mestery has joined #openstack-oslo | 21:57 | |
| mestery | dhellmann: FYI, we have an approved spec for the oslo.messaging port to neutron, and we've started landed patches to that effect: https://blueprints.launchpad.net/neutron/+spec/oslo-messaging | 21:58 |
| *** dims_ has joined #openstack-oslo | 22:00 | |
| *** markmcclain has quit IRC | 22:01 | |
| *** dims has quit IRC | 22:02 | |
| *** ihrachyshka has quit IRC | 22:10 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Address concurrent mutation of sqlalchemy backend https://review.openstack.org/98689 | 22:17 |
| *** flaper87 is now known as flaper87|afk | 22:23 | |
| *** krotscheck has joined #openstack-oslo | 22:25 | |
| krotscheck | Does anyone out there have simple ‘getting started’ example code for Oslo.Messaging? | 22:26 |
| *** zzzeek has quit IRC | 22:55 | |
| *** zzzeek has joined #openstack-oslo | 23:01 | |
| *** morganfainberg is now known as morganfainberg_Z | 23:25 | |
| *** jaosorior has quit IRC | 23:32 | |
| *** morganfainberg_Z is now known as morganfainberg | 23:42 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!