*** Kennan_Vacation is now known as Kennan | 00:03 | |
*** dims has quit IRC | 00:04 | |
*** ndipanov has quit IRC | 00:04 | |
*** dims has joined #openstack-oslo | 00:11 | |
*** jecarey has joined #openstack-oslo | 00:12 | |
*** jecarey has quit IRC | 00:17 | |
*** ozamiatin_ has joined #openstack-oslo | 00:24 | |
*** pradk has quit IRC | 00:30 | |
*** EinstCra_ has quit IRC | 00:31 | |
*** salv-orl_ has quit IRC | 00:38 | |
*** yamamoto has joined #openstack-oslo | 00:56 | |
*** yamamoto has quit IRC | 01:03 | |
*** dims has quit IRC | 01:04 | |
*** mtanino has quit IRC | 01:06 | |
*** dims has joined #openstack-oslo | 01:08 | |
*** achanda has quit IRC | 01:13 | |
*** achanda has joined #openstack-oslo | 01:17 | |
*** achanda has quit IRC | 01:20 | |
*** david-lyle has joined #openstack-oslo | 01:23 | |
*** jaypipes has quit IRC | 01:36 | |
*** yamamoto has joined #openstack-oslo | 01:48 | |
*** edmondsw_ has quit IRC | 01:56 | |
*** yamamoto has quit IRC | 02:00 | |
*** zz_dimtruck is now known as dimtruck | 02:00 | |
*** gcb has joined #openstack-oslo | 02:03 | |
openstackgerrit | Angus Lees proposed openstack/oslo.privsep: Use eventlet greenthreads if eventlet is installed https://review.openstack.org/280471 | 02:10 |
---|---|---|
*** mss-4 has joined #openstack-oslo | 02:16 | |
*** itisha has quit IRC | 02:37 | |
gus | Turns out that^ is the exact opposite of what I need. Apparently I *need* to use native threads, because eventlet doesn't wrap io.open(os.pipe()[0]) | 02:38 |
* gus counts to 20 and then says something unpleasant about eventlet. | 02:40 | |
*** 18VAAAIFD has joined #openstack-oslo | 02:47 | |
*** salv-orlando has joined #openstack-oslo | 03:16 | |
dims | gus :) | 03:16 |
dims | lifeless : so i'll leave it to you to propose fixing oslotest or testtools. it's your call :) | 03:17 |
lifeless | dims: sure | 03:23 |
lifeless | dims: I can see about a patch later in the week, if its urgent, you'll need to do it or find a volunteer | 03:24 |
lifeless | dims: right now, I've documented it in both the oslotest bug and the testtools issue | 03:24 |
lifeless | gus: oh, there is a usual pattern | 03:25 |
lifeless | gus: you should register with the eventlet event loop and have it call your code as needed | 03:25 |
lifeless | gus: if being called from a non-eventlet thing is a use case ( I don't think it is yet ) we can design something for that at that time. | 03:26 |
gus | .. which means requiring eventlet. So that's ok for an oslo library? | 03:26 |
dims | lifeless : ack. thanks | 03:26 |
dims | lifeless : later in the week is ok | 03:26 |
lifeless | gus: I think if you look in oslo.messaging which has the same broad issue - e.g. one connection to amqp from a process | 03:26 |
lifeless | gus: you'll see some example code. IMBW though. | 03:26 |
lifeless | gus: yes, requiring the things we use everywhere is ok :) | 03:26 |
lifeless | gus: I'm not suggesting that the privileged mode should use eventlet though | 03:27 |
lifeless | dims: it should be really easy to do | 03:27 |
gus | I just recall the conversations that occurred when I tried to get an eventlet test case into oslo.db | 03:27 |
lifeless | dims: just delete the method, move the constant assignment to class scope, move the cleanup call to the near-top of setUp | 03:28 |
lifeless | dims: my work setup is all over the place right now is the issue :) | 03:28 |
*** salv-orlando has quit IRC | 03:28 | |
dims | lifeless : base.py? | 03:29 |
lifeless | dims: yes | 03:29 |
lifeless | gus: heh | 03:29 |
lifeless | ok, -> pick up daughter | 03:29 |
dims | lifeless : am signing off, that's exactly what dhellmann was trying last friday https://review.openstack.org/#/c/279728/2/oslotest/base.py | 03:33 |
dims | talk to you later :) | 03:33 |
*** dims has quit IRC | 03:37 | |
openstackgerrit | Angus Lees proposed openstack/oslo.privsep: Ensure fdopen uses greenio object under eventlet https://review.openstack.org/280471 | 03:48 |
*** ozamiatin_ has quit IRC | 03:51 | |
*** jecarey has joined #openstack-oslo | 03:53 | |
*** amotoki has joined #openstack-oslo | 03:56 | |
*** amotoki has quit IRC | 03:56 | |
*** david-lyle has quit IRC | 04:03 | |
openstackgerrit | Merged openstack/pycadf: Add docstring validation https://review.openstack.org/230257 | 04:11 |
*** dimtruck is now known as zz_dimtruck | 04:23 | |
*** sputnik13 has joined #openstack-oslo | 04:27 | |
*** vilobhmm11 has joined #openstack-oslo | 04:31 | |
*** sputnik13 has quit IRC | 04:31 | |
*** jeckersb is now known as jeckersb_gone | 04:39 | |
*** yottatsa has joined #openstack-oslo | 04:46 | |
*** deray has joined #openstack-oslo | 04:46 | |
*** achanda has joined #openstack-oslo | 04:48 | |
*** david-lyle has joined #openstack-oslo | 04:55 | |
*** yottatsa has quit IRC | 05:06 | |
*** subscope has joined #openstack-oslo | 05:09 | |
*** yottatsa has joined #openstack-oslo | 05:47 | |
*** yottatsa has quit IRC | 06:06 | |
*** yottatsa has joined #openstack-oslo | 06:19 | |
*** vilobhmm11 has quit IRC | 06:33 | |
*** subscope has quit IRC | 06:34 | |
openstackgerrit | javeme proposed openstack/oslo.vmware: Implement FileHandle.get_size() method https://review.openstack.org/280004 | 06:34 |
*** achanda has quit IRC | 06:39 | |
*** david-lyle has quit IRC | 06:39 | |
*** david-lyle has joined #openstack-oslo | 06:40 | |
*** salv-orlando has joined #openstack-oslo | 06:43 | |
*** zzzeek has quit IRC | 06:44 | |
*** mss-4 has quit IRC | 06:45 | |
*** salv-orlando has quit IRC | 06:46 | |
*** subscope has joined #openstack-oslo | 06:52 | |
*** subscope has quit IRC | 06:55 | |
*** achanda has joined #openstack-oslo | 06:59 | |
*** yottatsa has quit IRC | 07:02 | |
*** e0ne has joined #openstack-oslo | 07:02 | |
*** rcernin has joined #openstack-oslo | 07:37 | |
*** mss-4 has joined #openstack-oslo | 07:41 | |
*** ekarlso has quit IRC | 07:44 | |
*** eezhova has quit IRC | 07:44 | |
*** eezhova has joined #openstack-oslo | 07:45 | |
*** Anticimex has quit IRC | 07:45 | |
*** kragniz has quit IRC | 07:46 | |
*** mss-4 has quit IRC | 07:46 | |
*** kragniz has joined #openstack-oslo | 07:46 | |
*** Anticimex has joined #openstack-oslo | 07:46 | |
*** ekarlso has joined #openstack-oslo | 07:57 | |
*** sbelous_ has quit IRC | 08:09 | |
*** e0ne has quit IRC | 08:13 | |
*** shardy has joined #openstack-oslo | 08:16 | |
*** yottatsa has joined #openstack-oslo | 08:21 | |
*** subscope has joined #openstack-oslo | 08:21 | |
openstackgerrit | javeme proposed openstack/oslo.vmware: Implement FileHandle.get_size() method https://review.openstack.org/280004 | 08:22 |
*** ihrachys has joined #openstack-oslo | 08:25 | |
*** salv-orlando has joined #openstack-oslo | 08:30 | |
*** zakora has joined #openstack-oslo | 08:32 | |
*** salv-orlando has quit IRC | 08:32 | |
*** ihrachys has quit IRC | 08:39 | |
*** mss-4 has joined #openstack-oslo | 08:42 | |
*** mss-4 has quit IRC | 08:48 | |
openstackgerrit | Chaozhe Chen(ccz) proposed openstack/oslo.config: Fix wrong check with non-None value when format group https://review.openstack.org/280552 | 08:49 |
*** subscope has quit IRC | 08:50 | |
*** achanda has quit IRC | 09:03 | |
*** ihrachys has joined #openstack-oslo | 09:03 | |
*** subscope has joined #openstack-oslo | 09:04 | |
*** achanda has joined #openstack-oslo | 09:06 | |
*** salv-orlando has joined #openstack-oslo | 09:08 | |
*** mhickey has joined #openstack-oslo | 09:11 | |
*** e0ne has joined #openstack-oslo | 09:16 | |
*** d0ugal has quit IRC | 09:21 | |
*** d0ugal has joined #openstack-oslo | 09:21 | |
*** yamahata has quit IRC | 09:22 | |
*** lucas-dinner is now known as lucasagomes | 09:24 | |
*** yassine__ has joined #openstack-oslo | 09:25 | |
*** achanda has quit IRC | 09:28 | |
*** ndipanov has joined #openstack-oslo | 09:33 | |
*** szaher_ has joined #openstack-oslo | 09:37 | |
*** ndipanov has quit IRC | 09:38 | |
*** ndipanov has joined #openstack-oslo | 09:38 | |
szaher_ | Hi Folks, Can I get instance of the base args parsers namespace using oslo_config ? | 09:39 |
szaher_ | s/parsers/parser/ | 09:39 |
*** ihrachys has quit IRC | 09:39 | |
*** subscope has quit IRC | 09:40 | |
*** binarin has quit IRC | 09:44 | |
*** mss-4 has joined #openstack-oslo | 09:44 | |
*** subscope has joined #openstack-oslo | 09:47 | |
*** ihrachys has joined #openstack-oslo | 09:48 | |
*** mss-4 has quit IRC | 09:48 | |
*** binarin has joined #openstack-oslo | 09:55 | |
*** ozamiatin_ has joined #openstack-oslo | 09:56 | |
openstackgerrit | Merged openstack/devstack-plugin-zmq: [zmq] Setup url for ceilometer gate with zmq https://review.openstack.org/280269 | 10:00 |
*** salv-orl_ has joined #openstack-oslo | 10:02 | |
*** salv-orlando has quit IRC | 10:05 | |
openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Support transport URL https://review.openstack.org/272583 | 10:06 |
*** andymaier has joined #openstack-oslo | 10:09 | |
*** dukhlov has quit IRC | 10:09 | |
*** dukhlov has joined #openstack-oslo | 10:21 | |
*** andymaier has quit IRC | 10:23 | |
haypo | jd__, sileht: hi. i propose to backport to liberty my second important fix on signal handling (oslo.service): https://review.openstack.org/#/c/280590/ can you please take a look? | 10:23 |
*** mss-4 has joined #openstack-oslo | 10:45 | |
*** subscope has quit IRC | 10:47 | |
*** mss-4 has quit IRC | 10:50 | |
*** dims has joined #openstack-oslo | 10:51 | |
*** cdent has joined #openstack-oslo | 10:56 | |
*** dims has quit IRC | 11:12 | |
*** deray has quit IRC | 11:12 | |
*** subscope has joined #openstack-oslo | 11:18 | |
*** dims has joined #openstack-oslo | 11:24 | |
*** sdague has joined #openstack-oslo | 11:27 | |
*** achanda has joined #openstack-oslo | 11:28 | |
*** achanda has quit IRC | 11:34 | |
*** d0ugal has quit IRC | 11:36 | |
*** dims has quit IRC | 11:38 | |
*** gibi has quit IRC | 11:39 | |
*** gibi has joined #openstack-oslo | 11:39 | |
*** sdague has quit IRC | 11:39 | |
*** 18VAAAIFD has quit IRC | 11:40 | |
*** sdague has joined #openstack-oslo | 11:44 | |
*** d0ugal has joined #openstack-oslo | 12:00 | |
*** d0ugal has quit IRC | 12:00 | |
*** d0ugal has joined #openstack-oslo | 12:00 | |
*** ericksonsantos has joined #openstack-oslo | 12:13 | |
*** gcb has quit IRC | 12:13 | |
*** gcb has joined #openstack-oslo | 12:14 | |
*** pcaruana has joined #openstack-oslo | 12:18 | |
*** yamamoto has joined #openstack-oslo | 12:30 | |
*** achanda has joined #openstack-oslo | 12:31 | |
*** mss-4 has joined #openstack-oslo | 12:33 | |
*** lucasagomes is now known as lucas-hungry | 12:35 | |
*** achanda has quit IRC | 12:37 | |
*** mss-4 has quit IRC | 12:38 | |
jd__ | haypo: in case you also have an opinion or whatever https://review.openstack.org/#/c/274077/ | 12:45 |
jd__ | not sure I'll go this path | 12:45 |
jd__ | thought what lifeless says really makes sense to me :) | 12:46 |
openstackgerrit | Merged openstack/oslo.config: Fix wrong check with non-None value when format group https://review.openstack.org/280552 | 12:51 |
*** gordc has joined #openstack-oslo | 12:51 | |
*** dtardivel has joined #openstack-oslo | 12:56 | |
*** boris-42 has joined #openstack-oslo | 12:57 | |
openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Support transport URL https://review.openstack.org/272583 | 13:01 |
*** edmondsw has joined #openstack-oslo | 13:03 | |
*** ihrachys has quit IRC | 13:04 | |
*** ihrachys has joined #openstack-oslo | 13:08 | |
haypo | jd__: i feel unable to review such change | 13:12 |
jd__ | haypo: I didn't really ask for a review | 13:12 |
haypo | jd__: there is a whole mailing list dedicated to distutils & cie: https://mail.python.org/mailman/listinfo/distutils-sig | 13:12 |
jd__ | that's already written in the review | 13:13 |
haypo | jd__: distutils suppports various options: --prefix, --exec-prefix, etc. but not --sysconfig | 13:14 |
jd__ | haypo: that's why I sent this review… | 13:16 |
jd__ | haypo: are you Captain Obvious? | 13:16 |
jd__ | :) | 13:16 |
haypo | jd__: i concur with lifeless, this issue should be fixed in pip, setuptools or even distutils | 13:17 |
jd__ | cool, now let's wait for it to happen | 13:18 |
haypo | jd__: more generally, i don't understand these libraries, so again i have no opinion on your change :) | 13:18 |
jd__ | ok | 13:18 |
jd__ | you write a lot for someone that has no opinion :P | 13:19 |
haypo | jd__: a practical issue is that your change must work "as expected" with binary wheel packages which are basically an unzip operation | 13:19 |
*** takedakn has joined #openstack-oslo | 13:24 | |
*** lucas-hungry is now known as lucasagomes | 13:27 | |
*** salv-orl_ has quit IRC | 13:32 | |
*** mss-4 has joined #openstack-oslo | 13:34 | |
*** salv-orlando has joined #openstack-oslo | 13:36 | |
*** mss-4 has quit IRC | 13:38 | |
*** dukhlov_ has joined #openstack-oslo | 13:40 | |
*** EinstCrazy has joined #openstack-oslo | 13:42 | |
*** jeckersb_gone is now known as jeckersb | 13:44 | |
*** dims has joined #openstack-oslo | 13:45 | |
*** yottatsa has quit IRC | 13:48 | |
*** rlrossit has joined #openstack-oslo | 13:50 | |
*** dims has quit IRC | 13:51 | |
rlrossit | rbradfor: I want to cash out my winnings | 13:52 |
*** subscope has quit IRC | 13:53 | |
lxsli_web | dhellmann: got a min to discuss decorating "notify_oslo_config"? Idea is to decouple non-Oslo projects from knowing which Oslo projects need notification after mutate | 14:00 |
*** kgiusti has joined #openstack-oslo | 14:00 | |
*** jeckersb is now known as jeckersb_gone | 14:00 | |
*** zz_dimtruck is now known as dimtruck | 14:01 | |
*** salv-orlando has quit IRC | 14:08 | |
*** yottatsa has joined #openstack-oslo | 14:09 | |
*** takedakn has quit IRC | 14:10 | |
*** subscope has joined #openstack-oslo | 14:22 | |
openstackgerrit | Andy Smith proposed openstack/devstack-plugin-amqp1: Set oslo_messaging_amqp section for qpid user https://review.openstack.org/279736 | 14:22 |
*** EinstCra_ has joined #openstack-oslo | 14:25 | |
*** itisha has joined #openstack-oslo | 14:25 | |
*** regXboi has joined #openstack-oslo | 14:26 | |
*** EinstCrazy has quit IRC | 14:28 | |
openstackgerrit | Alexis Lee proposed openstack/oslo.log: Allow reload of 'debug' option https://review.openstack.org/254821 | 14:29 |
openstackgerrit | Alexis Lee proposed openstack/oslo.log: Allow reload of 'debug' option https://review.openstack.org/254821 | 14:30 |
*** mriedem_away has quit IRC | 14:32 | |
*** yottatsa has quit IRC | 14:32 | |
*** jecarey has quit IRC | 14:34 | |
*** yottatsa has joined #openstack-oslo | 14:34 | |
*** mss-4 has joined #openstack-oslo | 14:34 | |
*** achanda has joined #openstack-oslo | 14:35 | |
*** zzzeek has joined #openstack-oslo | 14:35 | |
*** yottatsa has quit IRC | 14:35 | |
*** mss-4 has quit IRC | 14:39 | |
*** achanda has quit IRC | 14:40 | |
*** yottatsa has joined #openstack-oslo | 14:40 | |
*** dimtruck is now known as zz_dimtruck | 14:44 | |
*** ihrachys has quit IRC | 14:53 | |
*** pradk has joined #openstack-oslo | 14:53 | |
*** nkrinner has joined #openstack-oslo | 14:54 | |
*** nkrinner has quit IRC | 14:54 | |
*** mriedem has joined #openstack-oslo | 14:55 | |
*** mtanino has joined #openstack-oslo | 14:57 | |
*** subscope has quit IRC | 14:59 | |
*** nkrinner has joined #openstack-oslo | 14:59 | |
*** nkrinner has quit IRC | 15:00 | |
*** nkrinner has joined #openstack-oslo | 15:00 | |
*** EinstCra_ has quit IRC | 15:04 | |
*** EinstCrazy has joined #openstack-oslo | 15:04 | |
*** subscope has joined #openstack-oslo | 15:06 | |
*** jecarey has joined #openstack-oslo | 15:07 | |
*** jecarey has quit IRC | 15:07 | |
*** jecarey has joined #openstack-oslo | 15:07 | |
*** EinstCrazy has quit IRC | 15:12 | |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/oslotest: Hack to get back stopall cleanup behavior feature https://review.openstack.org/280757 | 15:12 |
*** sigmavirus24_awa is now known as sigmavirus24 | 15:13 | |
*** EinstCrazy has joined #openstack-oslo | 15:14 | |
rbradfor | rlrossit, checking lastest batch of releases | 15:14 |
rbradfor | rlrossit, sorry, the winner will be clayton, with delighted. | 15:16 |
clayton | ! | 15:16 |
*** mkoderer__ has quit IRC | 15:16 | |
rbradfor | and just so I didn't rig it, I came in second. | 15:17 |
rbradfor | rlrossit, FYI your jazzed came up, but this new round you picked pleased. :-< | 15:18 |
*** zz_dimtruck is now known as dimtruck | 15:18 | |
rbradfor | rlrossit, but you came in third with pleased, well done. | 15:18 |
*** mkoderer__ has joined #openstack-oslo | 15:18 | |
*** jeckersb_gone is now known as jeckersb | 15:23 | |
*** EinstCrazy has quit IRC | 15:25 | |
openstackgerrit | Alexis Lee proposed openstack/oslo.service: Offer mutate_config_files https://review.openstack.org/263312 | 15:28 |
*** yamahata has joined #openstack-oslo | 15:36 | |
*** EinstCrazy has joined #openstack-oslo | 15:39 | |
*** salv-orl_ has joined #openstack-oslo | 15:40 | |
*** salv-orl_ has quit IRC | 15:40 | |
*** salv-orlando has joined #openstack-oslo | 15:41 | |
*** nkrinner has quit IRC | 15:45 | |
lxsli_web | dhellmann: ping | 15:50 |
lxsli_web | I'm adding a "notify_oslo_config_reload" method to oslo.log which needs to be called after mutate_config_files. Trouble is, Nova (for example) uses oslo.service to call mutate_config_files. How does oslo.service know to call notify_oslo_config_reload without any callback mechanism? | 15:51 |
lxsli_web | tcammann suggested decorating notify_oslo_config_reload, would that be acceptable? | 15:52 |
mriedem | sileht: did anyone ever get any further on sorting out the oslo.messaging regression in 4.1.0 that is causing this messaging timeout on the new side of grenade? http://status.openstack.org/elastic-recheck/gate.html#1545002 | 15:56 |
mriedem | it's our top gate bug | 15:56 |
mriedem | haypo: are you going to request a stable/liberty release of oslo.service to get https://review.openstack.org/#/c/256443/ out? | 15:58 |
*** salv-orl_ has joined #openstack-oslo | 16:02 | |
*** salv-orlando has quit IRC | 16:05 | |
*** mss-4 has joined #openstack-oslo | 16:09 | |
lxsli_web | hey oslo.config gives a coverage report now! | 16:11 |
lxsli_web | nice dhellmann | 16:11 |
*** binarin has quit IRC | 16:14 | |
*** EinstCrazy has quit IRC | 16:15 | |
*** EinstCrazy has joined #openstack-oslo | 16:16 | |
*** pcaruana has quit IRC | 16:18 | |
dhellmann | lxsli_web : sorry, had some errands to deal with this morning, but I'm back now | 16:18 |
lxsli_web | ok hi :) | 16:18 |
dhellmann | lxsli_web : is it possible to use oslo.service without oslo.config and oslo.log? can we build this into the service code directly? | 16:19 |
lxsli_web | I don't think you can use it without oslo.conifg | 16:19 |
dhellmann | lxsli_web : we could also provide hooks in oslo.service (method for a subclass to override, argument to constructor, register_config_mutate_callback() method, etc.) | 16:19 |
*** EinstCrazy has quit IRC | 16:20 | |
lxsli_web | I prefer the decorator idea because that way if we enable oslo.foo for config mutation, nova code doesn't have to change | 16:20 |
dhellmann | lxsli_web : presumably the application itself will have some things to do, too, so yeah, I agree we need some sort of registration mechanism | 16:21 |
*** binarin has joined #openstack-oslo | 16:21 | |
lxsli_web | The decorator would be reserved for oslo (middleware) libraries | 16:21 |
dhellmann | so the decorator would register the callback? that feels a little magic | 16:21 |
lxsli_web | not very magic considering some of the shenanigans involved in OpenStack! | 16:22 |
dhellmann | what owns the callback registry? | 16:22 |
lxsli_web | oslo.config | 16:22 |
dhellmann | yeah, that's specifically what I wanted to avoid | 16:22 |
lxsli_web | I'm adding a method to ConfigOpts which Nova can call | 16:22 |
dhellmann | the issue with a decorator is that it would presumably rely on a global to know what functions to call, but the oslo libs do not use the global config object | 16:23 |
lxsli_web | So if Nova doesn't want to notify Oslo libs, it doesn't have to, but oslo internally manages registering which oslo libs are present and allows Nova to notify them all at once | 16:23 |
dhellmann | so adding a requirement that oslo.log only works with the global config object would be a bit of a regression on that design point | 16:23 |
lxsli_web | I'm adding a separate global, are you concerned people may have multiple ConfigOpts? | 16:23 |
dhellmann | we do support that, yes | 16:24 |
lxsli_web | Ah, I need to rethink then | 16:24 |
dhellmann | we have lots of unit tests that rely on that, for example | 16:24 |
dhellmann | and some applications instantiate their own copy | 16:24 |
dhellmann | what I thought we discussed was having some sort of callback mechanism in oslo.service | 16:25 |
dhellmann | or stand-alone, but triggered by oslo.service I suppose | 16:25 |
lxsli_web | the summit was many sleeps ago now! | 16:25 |
dhellmann | :-) | 16:25 |
dhellmann | your comment about not having to update nova if we add reload support to another lib is good, though | 16:26 |
dhellmann | maybe the callbacks should be owned by the ConfigOpt | 16:26 |
dhellmann | we could register them around the same time as calling register_opts() | 16:26 |
lxsli_web | that has promise | 16:27 |
dhellmann | so we still have explicit registration, don't assume a global, but libraries can register their own | 16:27 |
dhellmann | if we don't support assuming any order, we can just use a set to hold the callables | 16:27 |
dhellmann | that way registration is idempotent | 16:27 |
lxsli_web | Sounds good, I'll see how it plays out, thank you | 16:27 |
dhellmann | lxsli_web : when I argue against this design in the review, remind me of the nova thing :-) | 16:28 |
lxsli_web | Will do :) | 16:28 |
dhellmann | was there a spec for this work? if so you could update that, too, but I don't remember if we had one | 16:29 |
lxsli_web | We thought it was well enough discussed at the summit to not need one iirc | 16:29 |
*** sdague has quit IRC | 16:31 | |
*** boris-42 has quit IRC | 16:34 | |
*** sputnik13 has joined #openstack-oslo | 16:34 | |
*** sdague has joined #openstack-oslo | 16:34 | |
dhellmann | lxsli_web : ok. it might be worth writing some of this down in an etherpad so we remember why we've changed our mind about the plan | 16:35 |
lxsli_web | I can put it all in the commit message? | 16:36 |
*** sputnik13 has quit IRC | 16:37 | |
dhellmann | lxsli_web : yeah, that would work. I thought there might be a bunch of patches for this, but maybe not. | 16:37 |
*** rcernin has quit IRC | 16:37 | |
dhellmann | whatever you do doesn't have to be very formal, just enough to remember why we chose this implementation over the other suggestion | 16:37 |
lxsli_web | The main one will be on oslo.config to put the machinery in place, the patches to oslo.log and oslo.service should be pretty simple | 16:37 |
*** achanda has joined #openstack-oslo | 16:38 | |
lxsli_web | I've also dumped this convo to: https://etherpad.openstack.org/p/mitaka-cross-project-dynamic-config-services | 16:38 |
dhellmann | lxsli_web : saving the conversation works | 16:39 |
dhellmann | lxsli_web : at one point we talked about a more generic callback system being pulled out of taskflow. it would be good to sync up with harlowja about that, but I think to start we don't need it. we can go back and update the implementation if we decide we need to or if something like that gets built later | 16:40 |
lxsli_web | Sounds good, I also hope we don't need anything more than bare bones | 16:40 |
dhellmann | lxsli_web : hey, what do you think about having a reload_callback attribute be an aspect of the option definition itself? would that work? | 16:41 |
* dhellmann is just thinking out loud | 16:41 | |
lxsli_web | maaaybe but I can see it getting inefficient sometimes | 16:41 |
dhellmann | we probably should have done a spec now that we have 3 options, but oh well | 16:41 |
lxsli_web | If mysql_port, mysql_host, mysql_password change - you need to remake the connection | 16:41 |
lxsli_web | you don't want to do it 3 times if they all change though | 16:42 |
dhellmann | oh, good point | 16:42 |
dhellmann | yep | 16:42 |
lxsli_web | I suppose you could group things by handler though | 16:42 |
dhellmann | the callback could be on the object group | 16:43 |
dhellmann | "if anything in this group changes, call this function" | 16:43 |
lxsli_web | that seems pretty sensible | 16:43 |
dhellmann | wouldn't necessarily be great for the log stuff, though, since that uses the default group | 16:43 |
*** achanda has quit IRC | 16:44 | |
lxsli_web | mm | 16:44 |
lxsli_web | I expect lots of projects use the default group a bit | 16:44 |
dhellmann | yeah, most apps definitely do | 16:44 |
dhellmann | ok, I'm convinced that a separate registry is the simplest thing | 16:45 |
lxsli_web | cool :) | 16:45 |
dhellmann | do we want to support filtering the callbacks so they are only called when some options change? or just let the callback examine what's different? | 16:45 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/tooz: Raises proper error when unwatching a group https://review.openstack.org/278404 | 16:45 |
lxsli_web | I'm passing the ConfigOpts and 'fresh' to each hook. The latter tells them whether to act and the conf object is in case of dependent options | 16:46 |
lxsli_web | EG mysql_* | 16:46 |
dhellmann | fresh is the dict containing options that changed? | 16:46 |
lxsli_web | yep | 16:46 |
dhellmann | sounds good | 16:47 |
dhellmann | we can always add a convenience function to make a decorator for callbacks to skip invoking the expensive function unless the options match | 16:47 |
lxsli_web | I expect each hook to be composed of "if (grp, opt) in fresh:" clauses so they should be pretty fast unless everything changes | 16:48 |
dhellmann | yeah, I was thinking of making that easier with something like @filtered_config_callback(grp, opt) | 16:49 |
dhellmann | but that's an enhancement | 16:49 |
lxsli_web | I look forward to your patch :D | 16:49 |
dhellmann | we'll see how the low-level check-it-yourself api works out first | 16:49 |
dhellmann | haha | 16:49 |
*** yassine__ has quit IRC | 16:55 | |
*** subscope has quit IRC | 16:56 | |
*** zakora has quit IRC | 16:57 | |
*** e0ne has quit IRC | 17:04 | |
*** rcernin has joined #openstack-oslo | 17:09 | |
*** d0ugal has quit IRC | 17:11 | |
*** gordc has quit IRC | 17:14 | |
*** dims has joined #openstack-oslo | 17:34 | |
*** d0ugal has joined #openstack-oslo | 17:38 | |
*** manjeets has joined #openstack-oslo | 17:43 | |
dims | dhellmann : we gotta ask if we could change the green to softer colors :) | 17:43 |
manjeets | what can be possible values of mysql_sql_mode in this https://github.com/openstack/oslo.db/blob/master/oslo_db/sqlalchemy/engines.py#L104 | 17:45 |
manjeets | trying to understand what is it used for ? | 17:45 |
openstackgerrit | Alexis Lee proposed openstack/oslo.config: Hooks around mutate_config_files https://review.openstack.org/280847 | 17:46 |
*** e0ne has joined #openstack-oslo | 17:47 | |
*** shardy has quit IRC | 17:48 | |
lxsli_web | When I make a change to oslo.log that requires a change I'm pushing to oslo.config, do I also need to update the requirements.txt in my oslo.log patch to the next version of oslo.config? IE 3.6.0 right now | 17:49 |
lxsli_web | I'm also putting a Depends-On but it doesn't pass CI | 17:49 |
*** mriedem is now known as mriedem_lunch | 17:49 | |
dhellmann | dims : I think I was commenting on the code, but I do have a local custom style-sheet for our gerrit :-) | 17:49 |
lxsli_web | dhellmann: me too! It's the yellow notes which make me wince | 17:50 |
dhellmann | lxsli_web : depends-on will only work in devstack jobs | 17:50 |
openstackgerrit | Alexis Lee proposed openstack/oslo.log: Allow reload of 'debug' option https://review.openstack.org/254821 | 17:50 |
dhellmann | we don't install things from source for unit tests | 17:50 |
lxsli_web | dhellmann: ah right, nothing I can do then? | 17:50 |
dhellmann | so you'll need to test locally (there's a script in oslotest for that) and then release the dependency and then your other patch can land | 17:50 |
dhellmann | if you work with me and dims this week we should be able to get the release you need done fairly quickly | 17:51 |
manjeets | dhelmann: hi | 17:52 |
dims | ++ | 17:52 |
manjeets | dhellmann: | 17:52 |
haypo | mriedem_lunch: "are you going to request a stable/liberty release of oslo.service to get https://review.openstack.org/#/c/256443/ out?" i also proposed a second backport which was also merged, https://review.openstack.org/#/c/280590/ | 17:52 |
lxsli_web | dhellmann: thanks, no particular rush right now I don't think - I'm very pleased with progress so far in fact | 17:52 |
* dims catching up on various things | 17:53 | |
lxsli_web | wait it's 6 already how did that happen? >.< | 17:53 |
dhellmann | lxsli_web : keep in mind the release freeze for libs is coming up | 17:53 |
dhellmann | lxsli_web : final release for non-client libs is Feb 22-26 | 17:54 |
lxsli_web | Good to know | 17:54 |
*** browne has joined #openstack-oslo | 17:56 | |
*** achanda has joined #openstack-oslo | 17:59 | |
haypo | dims, dhellmann, lifeless, mriedem_lunch : hi. so what's the status of oslo in liberty? are we using the liberty branch or not? | 17:59 |
dhellmann | haypo : we should be, why? | 18:00 |
dhellmann | haypo : unless you mean the constraints settings? | 18:00 |
dims | haypo : we have not released libraries from liberty branch if that's what you are asking | 18:00 |
haypo | in releases, the latest oslo.service release for liberty is 0.9, but in upper-constraints.txt, i see oslo.service 1.1.0 | 18:00 |
haypo | does it mean that the "openstack liberty (releases)" uses oslo.service of "mitaka"? | 18:01 |
dhellmann | haypo : yeah, we screwed up and approved constraints updates from the bot that we shouldn't have | 18:01 |
dhellmann | we haven't unwound from those mistakes | 18:01 |
dims | haypo : long story, mriedem_lunch is trying to fix things | 18:01 |
haypo | oh | 18:01 |
dhellmann | we did, however, turn off the bot in branches so we don't make that mistake again | 18:01 |
haypo | well, to come back to oslo.service, it would like to propose a new release. is 0.9.1 a good version number? (the latest version is 0.9.0 in liberty) | 18:02 |
dims | haypo : yep | 18:02 |
dhellmann | haypo : yes, that would be the right version | 18:02 |
haypo | ok | 18:02 |
dhellmann | haypo : ping mriedem_lunch to review the release request | 18:02 |
dims | haypo : you will need blessing from mriedem_lunch and tonyb | 18:02 |
dims | yep | 18:02 |
*** yamamoto has quit IRC | 18:05 | |
*** achanda has quit IRC | 18:06 | |
haypo | dims, dhellmann, mriedem_lunch : https://review.openstack.org/280854 liberty: release oslo.service 0.9.1 | 18:06 |
*** mriedem_lunch is now known as mriedem | 18:06 | |
dhellmann | haypo : ack | 18:06 |
mriedem | "long story, mriedem_lunch is trying to fix things" | 18:06 |
mriedem | i wouldn't say i'm actively trying to fix things | 18:07 |
mriedem | :) | 18:07 |
*** subscope has joined #openstack-oslo | 18:07 | |
haypo | dhellmann: i read that mriedem "fix" is to remove gates :-D | 18:09 |
*** yottatsa has quit IRC | 18:09 | |
haypo | mriedem ;) | 18:09 |
dims | :) | 18:09 |
mriedem | that was my, this is all busted and i don't have the energy to care right now, fix | 18:09 |
dhellmann | at this point in the cycle it might be best to just make sure we don't make the same mistake again and live with what we have there | 18:10 |
dims | mriedem : i'll help starting next week :) we can do a bunch of releases from branch and fix up requirements and constraints | 18:10 |
*** yamamoto has joined #openstack-oslo | 18:10 | |
dhellmann | if no packagers have complained, they either have working systems using those constraints, or have worked around it without sharing their results | 18:10 |
*** yottatsa has joined #openstack-oslo | 18:10 | |
dims | dhellmann : true | 18:10 |
mriedem | i honestly don't know how many packagers are using constraints and maybe are just getting lucky | 18:10 |
mriedem | *aren't using constraints | 18:11 |
dhellmann | yeah, that too | 18:11 |
dhellmann | mriedem : we should probably have this discussion in #openstack-release, but it would be interesting to see what would happen if we cap requirements in applications but not libs | 18:11 |
*** yottatsa has quit IRC | 18:18 | |
*** yottatsa has joined #openstack-oslo | 18:19 | |
*** achanda has joined #openstack-oslo | 18:23 | |
*** david-lyle has quit IRC | 18:27 | |
*** vilobhmm11 has joined #openstack-oslo | 18:27 | |
*** d0ugal has quit IRC | 18:28 | |
*** d0ugal has joined #openstack-oslo | 18:29 | |
*** vilobhmm11 has quit IRC | 18:30 | |
*** e0ne has quit IRC | 18:33 | |
*** mriedem has quit IRC | 18:35 | |
*** vilobhmm11 has joined #openstack-oslo | 18:35 | |
*** david-lyle has joined #openstack-oslo | 18:35 | |
openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: Check devstack gate zmq https://review.openstack.org/229779 | 18:36 |
*** vilobhmm11 has quit IRC | 18:40 | |
*** mriedem has joined #openstack-oslo | 18:42 | |
*** yamamoto has quit IRC | 18:49 | |
*** yottatsa has quit IRC | 18:58 | |
*** yottatsa has joined #openstack-oslo | 18:58 | |
*** jecarey_ has joined #openstack-oslo | 19:00 | |
*** jecarey_ has quit IRC | 19:00 | |
*** yottatsa has quit IRC | 19:00 | |
*** yottatsa has joined #openstack-oslo | 19:03 | |
*** jecarey_ has joined #openstack-oslo | 19:03 | |
*** jecarey has quit IRC | 19:04 | |
*** mhickey has quit IRC | 19:05 | |
*** gibi has quit IRC | 19:06 | |
*** gibi has joined #openstack-oslo | 19:07 | |
*** vilobhmm11 has joined #openstack-oslo | 19:07 | |
*** gibi has quit IRC | 19:08 | |
*** mtanino__ has quit IRC | 19:12 | |
*** gibi has joined #openstack-oslo | 19:12 | |
*** yottatsa has quit IRC | 19:12 | |
*** sputnik13 has joined #openstack-oslo | 19:15 | |
*** dimtruck is now known as zz_dimtruck | 19:17 | |
*** zz_dimtruck is now known as dimtruck | 19:21 | |
*** ozamiatin_ has quit IRC | 19:22 | |
*** yamamoto has joined #openstack-oslo | 19:50 | |
*** yamamoto has quit IRC | 19:56 | |
haypo | dims, dhellmann : is "openstack/oslo.service SHA 8b6e2f64e8f82ec736a4c5b792d448ad7faf051f found version 0.9.1 NEW SHA found in descendants" an error? tox-validate failed, http://logs.openstack.org/54/280854/1/check/gate-releases-tox-validate/f06099d/console.html | 20:00 |
*** e0ne has joined #openstack-oslo | 20:02 | |
*** jecarey_ has quit IRC | 20:03 | |
*** e0ne has quit IRC | 20:05 | |
*** jecarey_ has joined #openstack-oslo | 20:11 | |
*** jecarey_ has quit IRC | 20:12 | |
*** jecarey has joined #openstack-oslo | 20:13 | |
*** jecarey has quit IRC | 20:13 | |
dhellmann | haypo : http://logs.openstack.org/54/280854/1/check/gate-releases-tox-validate/f06099d/console.html#_2016-02-16_19_25_23_698 | 20:13 |
*** jecarey has joined #openstack-oslo | 20:13 | |
openstackgerrit | Ronald Bradford proposed openstack/oslo.config: Show namespace version in generator comment https://review.openstack.org/274270 | 20:14 |
*** lucasagomes is now known as lucas-dinner | 20:14 | |
haypo | dhellmann: oh, i just understood | 20:17 |
haypo | dhellmann: ok, i copied the email address from mitaka | 20:17 |
dhellmann | haypo : yeah, we need to make the error output a bit more obvious | 20:17 |
dhellmann | haypo : using the same address should be right | 20:17 |
haypo | dhellmann: yeah, just prefix the line with "ERROR: " ;) | 20:18 |
*** david-lyle has quit IRC | 20:18 | |
dhellmann | haypo : indeed :-) | 20:19 |
lifeless | morning | 20:24 |
lifeless | dhellmann: so, oslotest | 20:24 |
dhellmann | hi, lifeless | 20:24 |
*** dtardivel has quit IRC | 20:28 | |
dims | lifeless, dhellmann : so we have this review for now - https://review.openstack.org/#/c/280757/ | 20:31 |
*** rlrossit has quit IRC | 20:33 | |
haypo | oh, first time that i abandon a patch from someone else. https://review.openstack.org/#/c/186233/ | 20:34 |
haypo | "Adding failure detail to Exceptions on encode", two -1 votes, no activity since ~1 year from the author | 20:34 |
dims | sputnik13 : any feedback? ^^ | 20:35 |
haypo | sputnik13: oh, i didn't know that you was online. see my comment, feel free to reopen the issue, but see also comments explaining negative votes ;) | 20:37 |
lifeless | dhellmann: dims: reviewed | 20:37 |
dhellmann | lifeless : it would be nice if we could get oslotest back into a releasable state, and then decide on the deprecation. That whole class should be deprecated, for example, not just the behavior w.r.t. stopall | 20:39 |
openstackgerrit | Dmitriy Ukhlov proposed openstack/oslo.messaging: test https://review.openstack.org/280920 | 20:39 |
haypo | it's very strange. i'm listing reviews of various oslo objects | 20:39 |
haypo | and most of them have *no* open review at all | 20:40 |
haypo | does it mean that oslo became kind of stable? | 20:40 |
dims | dhellmann : +1 | 20:40 |
haypo | if my rationale is right, oslo message is not stable yet :) | 20:40 |
*** subscope has quit IRC | 20:42 | |
*** ozamiatin_ has joined #openstack-oslo | 20:45 | |
lifeless | dhellmann: I guess | 20:46 |
dhellmann | lifeless : we went through and converted most of its features to fixtures about a year ago, but I think at that point we didn't have a good deprecation alert system like debtcollector | 20:47 |
*** ozamiatin_ has quit IRC | 20:52 | |
*** gordc has joined #openstack-oslo | 20:55 | |
sputnik13 | haypo dims it's fine with me, thanks | 20:58 |
haypo | sputnik13: ok :) | 20:59 |
*** ndipanov has quit IRC | 21:03 | |
*** haypo has left #openstack-oslo | 21:03 | |
*** sputnik13 has quit IRC | 21:06 | |
*** sputnik13 has joined #openstack-oslo | 21:07 | |
*** sdague has quit IRC | 21:07 | |
*** sputnik13 has quit IRC | 21:08 | |
*** mhickey has joined #openstack-oslo | 21:10 | |
*** kgiusti has left #openstack-oslo | 21:11 | |
openstackgerrit | Merged openstack/pycadf: Updated from global requirements https://review.openstack.org/280864 | 21:25 |
*** rlrossit has joined #openstack-oslo | 21:25 | |
*** mhickey has quit IRC | 21:34 | |
*** dims has quit IRC | 21:37 | |
*** jeckersb is now known as jeckersb_gone | 21:38 | |
openstackgerrit | Ian Cordasco proposed openstack-dev/pbr: Include wsgi_scripts in generated wheels https://review.openstack.org/280270 | 21:43 |
*** dims has joined #openstack-oslo | 21:50 | |
openstackgerrit | Sabari proposed openstack/oslo.vmware: Update NFC lease progress in a new thread https://review.openstack.org/280942 | 21:51 |
stevemar | rpodolyaka: ping | 21:52 |
*** mriedem has quit IRC | 21:55 | |
stevemar | dims: also a ping for you | 21:58 |
dims | stevemar : pong | 21:58 |
stevemar | dims: https://review.openstack.org/#/c/279828/3 | 21:58 |
stevemar | dims: the problem is: both keystone_middleware and whatever service using it (say nova), can both use oslo.messaging | 21:59 |
stevemar | dims: in this case, the author wants one to go his message queue, the other to go to log | 21:59 |
stevemar | dims: keystonemiddleware uses this for their audit events | 22:00 |
stevemar | dims: it just seems like the two will clobber each other... | 22:02 |
*** salv-orlando has joined #openstack-oslo | 22:03 | |
dims | stevemar : we'll need to get sileht into this one. y, it's always been the service having control over the config and the lifecycle | 22:03 |
dims | stevemar : we got as far as splitting rpc and notifications in mitaka, this will have to wait till newton | 22:04 |
*** salv-orl_ has quit IRC | 22:05 | |
*** cdent has quit IRC | 22:06 | |
stevemar | dims: okay, please comment as you see fit | 22:12 |
dims | stevemar : one feedback there is for someone to try the scenario and tell us what works and what doesn't :) | 22:12 |
*** jeckersb_gone is now known as jeckersb | 22:14 | |
*** sputnik13 has joined #openstack-oslo | 22:16 | |
stevemar | dims: the other question i have is related to https://bugs.launchpad.net/keystone/+bug/1490571 | 22:37 |
openstack | Launchpad bug 1490571 in OpenStack Identity (keystone) "Refactor use of oslo.db.sqlalchemy.session.EngineFacade because it's deprecated" [Medium,In progress] | 22:37 |
stevemar | dims: when will the deprecating bits be removed? | 22:37 |
dims | 'N' i believe, rpodolyaka will know for sure | 22:39 |
*** gordc has quit IRC | 22:43 | |
*** dims has quit IRC | 22:47 | |
stevemar | dimtruck: been trying to poke him :) | 22:48 |
stevemar | dimtruck: okay, any harm in leaving it in? seems like a lot of work | 22:48 |
dimtruck | dims ^ | 22:49 |
dimtruck | ;) | 22:49 |
*** Nakato has quit IRC | 22:50 | |
*** Nakato has joined #openstack-oslo | 22:50 | |
*** regXboi has quit IRC | 22:52 | |
*** pradk has quit IRC | 22:55 | |
*** sigmavirus24 is now known as sigmavirus24_awa | 23:00 | |
*** rlrossit has quit IRC | 23:02 | |
*** jecarey has quit IRC | 23:14 | |
*** dukhlov_ has quit IRC | 23:16 | |
*** rcernin has quit IRC | 23:18 | |
*** flwang has quit IRC | 23:19 | |
*** jecarey has joined #openstack-oslo | 23:26 | |
*** DuncanT_ has joined #openstack-oslo | 23:30 | |
*** jecarey has quit IRC | 23:30 | |
*** tpeoples_ has joined #openstack-oslo | 23:31 | |
*** gibi_ has joined #openstack-oslo | 23:32 | |
*** flwang has joined #openstack-oslo | 23:33 | |
*** dimtruck is now known as zz_dimtruck | 23:35 | |
*** cburgess has joined #openstack-oslo | 23:35 | |
*** EmilienM_ has joined #openstack-oslo | 23:35 | |
*** clayton_ has joined #openstack-oslo | 23:36 | |
*** jamespag` has joined #openstack-oslo | 23:36 | |
*** gibi has quit IRC | 23:37 | |
*** ekarlso has quit IRC | 23:37 | |
*** clayton has quit IRC | 23:37 | |
*** ativelkov has quit IRC | 23:37 | |
*** tpeoples has quit IRC | 23:37 | |
*** Horgix has quit IRC | 23:37 | |
*** EmilienM has quit IRC | 23:37 | |
*** mdavidson has quit IRC | 23:37 | |
*** cburgess_ has quit IRC | 23:37 | |
*** jamespage has quit IRC | 23:37 | |
*** DuncanT has quit IRC | 23:37 | |
*** clayton_ is now known as clayton | 23:37 | |
*** EmilienM_ is now known as EmilienM | 23:37 | |
*** jamespag` has quit IRC | 23:37 | |
*** jamespag` has joined #openstack-oslo | 23:37 | |
*** EmilienM is now known as Guest28836 | 23:37 | |
*** tpeoples_ is now known as tpeoples | 23:40 | |
*** Guest28836 has quit IRC | 23:41 | |
*** Guest28836 has joined #openstack-oslo | 23:41 | |
*** ativelkov has joined #openstack-oslo | 23:41 | |
*** Guest28836 is now known as EmilienM | 23:42 | |
manjeets | one of my test that need mysql backend works fine on my mahcine but on jenkins gate it gives this issue http://paste.openstack.org/show/487199/ | 23:42 |
manjeets | any workaround to fix this ? | 23:43 |
*** mdavidson has joined #openstack-oslo | 23:43 | |
*** DuncanT_ is now known as DuncanT | 23:43 | |
*** Horgix has joined #openstack-oslo | 23:43 | |
*** dims_ has joined #openstack-oslo | 23:44 | |
*** ekarlso has joined #openstack-oslo | 23:47 | |
*** ekarlso has quit IRC | 23:48 | |
*** ekarlso has joined #openstack-oslo | 23:48 | |
*** harlowja_at_home has joined #openstack-oslo | 23:55 | |
*** browne has quit IRC | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!