zigo | dims: Am I right that the newer oslo.concurrency now sets lock_path to /tmp by default? | 00:04 |
---|---|---|
zigo | dims: I very much would prefer if it continued to set None, and produce error if no good settings is forced. | 00:05 |
zigo | dims: This is otherwise a very dangerous default. | 00:05 |
zigo | Your thoughts? | 00:05 |
*** edmondsw has quit IRC | 00:06 | |
openstackgerrit | Joshua Harlow proposed openstack/tooz: Add basic mysql table support (if provided a database) [POC] https://review.openstack.org/290178 | 00:09 |
openstackgerrit | Joshua Harlow proposed openstack/tooz: Add basic mysql table support (if provided a database) [POC] https://review.openstack.org/290178 | 00:27 |
openstackgerrit | Sachi King proposed openstack-dev/pbr: Do not silently drop markers that fail to evaluate https://review.openstack.org/289628 | 00:27 |
harlowja | zigo https://github.com/openstack/oslo.concurrency/commit/499d5aa36d352cf3b seemed to revert that | 00:27 |
*** salv-orlando has joined #openstack-oslo | 00:30 | |
*** suro-patz has quit IRC | 00:30 | |
*** browne has quit IRC | 00:33 | |
*** bapalm has quit IRC | 00:37 | |
*** salv-orlando has quit IRC | 00:40 | |
*** mriedem has joined #openstack-oslo | 00:44 | |
*** bapalm has joined #openstack-oslo | 00:54 | |
*** harlowja has quit IRC | 00:55 | |
*** mtanino has quit IRC | 01:02 | |
*** vilobhmm11 has quit IRC | 01:02 | |
*** vilobhmm11 has joined #openstack-oslo | 01:03 | |
*** vilobhmm11 has quit IRC | 01:04 | |
*** vilobhmm11 has joined #openstack-oslo | 01:04 | |
*** sdague has joined #openstack-oslo | 01:06 | |
*** browne has joined #openstack-oslo | 01:08 | |
*** gcb has joined #openstack-oslo | 01:10 | |
*** zz_dimtruck is now known as dimtruck | 01:25 | |
openstackgerrit | Merged openstack-dev/pbr: Use apt-cache generated packages to provide build deps https://review.openstack.org/289689 | 01:26 |
*** sdague has quit IRC | 01:38 | |
*** sdague has joined #openstack-oslo | 01:46 | |
*** mtanino has joined #openstack-oslo | 01:53 | |
*** _amrith_ is now known as amrith | 01:55 | |
*** salv-orlando has joined #openstack-oslo | 01:57 | |
dims | zigo : that change was never released. | 01:58 |
*** vilobhmm11 has quit IRC | 01:59 | |
*** vilobhmm11 has joined #openstack-oslo | 02:03 | |
*** vilobhmm11 has quit IRC | 02:03 | |
*** vilobhmm11 has joined #openstack-oslo | 02:03 | |
*** salv-orlando has quit IRC | 02:04 | |
*** sdague has quit IRC | 02:05 | |
*** yamamoto_ has joined #openstack-oslo | 02:06 | |
*** vilobhmm11 has quit IRC | 02:22 | |
*** john51 has quit IRC | 02:31 | |
lifeless | dims: around? | 02:32 |
lifeless | dims: could you +2A https://review.openstack.org/#/c/289628/ ? | 02:32 |
patchbot | lifeless: patch 289628 - openstack-dev/pbr - Do not silently drop markers that fail to evaluate | 02:32 |
*** john51 has joined #openstack-oslo | 02:33 | |
dims | lifeless : pong | 02:34 |
dims | lifeless : will review now | 02:35 |
lifeless | dims: thanks | 02:39 |
dims | lifeless : in the error message, the bit about "see the following error" seemed redundant. but that's ok | 02:40 |
lifeless | dims: its alittle awkward, but not really many better options given the structure we're dealing with | 02:42 |
dims | ack lifeless | 02:43 |
openstackgerrit | Merged openstack-dev/pbr: Clarifications around tags and version numbers https://review.openstack.org/289630 | 02:52 |
*** gcb has quit IRC | 02:52 | |
*** yamahata has quit IRC | 02:53 | |
*** mriedem has quit IRC | 02:54 | |
*** vilobhmm11 has joined #openstack-oslo | 02:55 | |
*** yamamoto_ has quit IRC | 02:56 | |
*** vilobhmm111 has joined #openstack-oslo | 02:57 | |
*** itisha has quit IRC | 02:59 | |
*** vilobhmm11 has quit IRC | 03:00 | |
*** gcb has joined #openstack-oslo | 03:13 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.config: doc:Log warning when can't get information from git https://review.openstack.org/290255 | 03:24 |
*** dims has quit IRC | 03:25 | |
*** dimtruck is now known as zz_dimtruck | 03:29 | |
*** zz_dimtruck is now known as dimtruck | 03:29 | |
openstackgerrit | zwei proposed openstack/oslo.service: Exception: message need '_' function https://review.openstack.org/290256 | 03:32 |
*** salv-orlando has joined #openstack-oslo | 03:32 | |
*** salv-orlando has quit IRC | 03:35 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.config: doc:Log warning when can't get information from git https://review.openstack.org/290255 | 03:37 |
*** suro-patz has joined #openstack-oslo | 03:38 | |
*** yamamoto_ has joined #openstack-oslo | 03:39 | |
*** dimtruck is now known as zz_dimtruck | 03:39 | |
openstackgerrit | Merged openstack-dev/pbr: Do not silently drop markers that fail to evaluate https://review.openstack.org/289628 | 03:41 |
*** yamahata has joined #openstack-oslo | 03:47 | |
*** vilobhmm111 has quit IRC | 03:51 | |
*** jamielennox is now known as jamielennox|away | 03:54 | |
*** suro-patz has quit IRC | 03:58 | |
*** sputnik13 has quit IRC | 03:59 | |
*** links has joined #openstack-oslo | 04:03 | |
*** amrith is now known as _amrith_ | 04:10 | |
*** vilobhmm11 has joined #openstack-oslo | 04:12 | |
*** zz_dimtruck is now known as dimtruck | 04:33 | |
*** yamahata has quit IRC | 04:37 | |
*** yamahata has joined #openstack-oslo | 04:38 | |
*** dimtruck is now known as zz_dimtruck | 04:54 | |
*** EinstCrazy has joined #openstack-oslo | 04:57 | |
*** eliqiao has quit IRC | 05:04 | |
*** eliqiao has joined #openstack-oslo | 05:04 | |
*** regXboi has joined #openstack-oslo | 05:05 | |
*** regXboi has quit IRC | 05:06 | |
openstackgerrit | zwei proposed openstack/oslo.service: Exception: message need '_' function https://review.openstack.org/290256 | 05:07 |
*** salv-orlando has joined #openstack-oslo | 05:07 | |
openstackgerrit | Merged openstack-dev/pbr: fix some variable names https://review.openstack.org/285249 | 05:08 |
*** salv-orlando has quit IRC | 05:13 | |
*** claudiub has quit IRC | 05:15 | |
*** EinstCrazy has quit IRC | 05:28 | |
*** harlowja_at_home has joined #openstack-oslo | 05:29 | |
*** vilobhmm11 has quit IRC | 05:31 | |
*** vilobhmm11 has joined #openstack-oslo | 05:33 | |
*** gcb has quit IRC | 05:41 | |
*** nkrinner has joined #openstack-oslo | 05:54 | |
*** gcb has joined #openstack-oslo | 05:59 | |
openstackgerrit | Joshua Harlow proposed openstack/oslo.serialization: Make msgpack registries copyable (and add __contains__) https://review.openstack.org/289611 | 06:08 |
*** rcernin has joined #openstack-oslo | 06:09 | |
*** gcb has quit IRC | 06:10 | |
*** bnemec has quit IRC | 06:11 | |
*** amotoki has joined #openstack-oslo | 06:25 | |
*** salv-orlando has joined #openstack-oslo | 06:31 | |
openstackgerrit | Dmitriy Ukhlov proposed openstack/oslo.messaging: test https://review.openstack.org/285239 | 06:32 |
*** salv-orlando has quit IRC | 06:36 | |
*** fhermeni has joined #openstack-oslo | 06:44 | |
*** harlowja_at_home has quit IRC | 07:03 | |
*** suro-patz has joined #openstack-oslo | 07:07 | |
*** claudiub has joined #openstack-oslo | 07:07 | |
*** jeckersb is now known as jeckersb_gone | 07:20 | |
*** sputnik13_ has joined #openstack-oslo | 07:23 | |
*** suro-patz has quit IRC | 07:24 | |
*** nkrinner has quit IRC | 07:27 | |
*** nkrinner has joined #openstack-oslo | 07:31 | |
*** sigmavirus24_awa has quit IRC | 07:36 | |
*** odyssey4me has quit IRC | 07:36 | |
*** cloudnull has quit IRC | 07:38 | |
*** odyssey4me has joined #openstack-oslo | 07:41 | |
*** sputnik13_ has quit IRC | 07:55 | |
*** itisha has joined #openstack-oslo | 08:07 | |
*** d0ugal has joined #openstack-oslo | 08:09 | |
*** dguitarbite has joined #openstack-oslo | 08:10 | |
*** e0ne has joined #openstack-oslo | 08:10 | |
*** binarin has joined #openstack-oslo | 08:11 | |
*** browne has quit IRC | 08:11 | |
*** e0ne has quit IRC | 08:19 | |
*** vilobhmm11 has quit IRC | 08:25 | |
*** jschwarz has joined #openstack-oslo | 08:28 | |
*** fhermeni has quit IRC | 08:35 | |
*** ndipanov has quit IRC | 08:35 | |
*** pblaho has quit IRC | 08:41 | |
*** shardy has joined #openstack-oslo | 08:47 | |
*** openstackgerrit has quit IRC | 08:47 | |
*** openstackgerrit has joined #openstack-oslo | 08:47 | |
*** ihrachys has joined #openstack-oslo | 08:48 | |
jschwarz | hello - I have a question on tooz :) | 08:48 |
jschwarz | I'm looking at the documentations and it specifies one needs to send a heartbeat every once in a while | 08:48 |
jschwarz | say I have a server which is using multiple threads, each one is an oslo.messaging RPC server | 08:49 |
jschwarz | Is there any easy way to tell oslo.messaging to also send heartbeats (or run some function of mine every once in a while)? | 08:49 |
*** ndipanov has joined #openstack-oslo | 08:55 | |
*** links has quit IRC | 08:55 | |
*** amotoki has quit IRC | 09:10 | |
*** bapalm has quit IRC | 09:14 | |
*** lucas-dinner is now known as lucasagomes | 09:16 | |
*** salv-orlando has joined #openstack-oslo | 09:21 | |
*** yamahata has quit IRC | 09:23 | |
*** amotoki has joined #openstack-oslo | 09:25 | |
*** cloudnull has joined #openstack-oslo | 09:27 | |
*** yassine has joined #openstack-oslo | 09:27 | |
*** mhickey_ has joined #openstack-oslo | 09:27 | |
*** sigmavirus24_awa has joined #openstack-oslo | 09:27 | |
*** bapalm has joined #openstack-oslo | 09:27 | |
*** salv-orlando has quit IRC | 09:29 | |
*** amotoki has quit IRC | 09:34 | |
*** _amrith_ is now known as amrith | 09:41 | |
*** vks has joined #openstack-oslo | 09:43 | |
*** amotoki has joined #openstack-oslo | 09:43 | |
*** amotoki has quit IRC | 09:51 | |
*** e0ne has joined #openstack-oslo | 09:51 | |
vks | hi, i have written a sample neutron agent which subscribe ROUTER AFTER_CREATE event. But I am not getting the call in my callback method. | 09:52 |
vks | 2016-03-09 01:36:08.220 7075 DEBUG neutron.callbacks.manager [-] Subscribe: <function add_router at 0x7fdbbc482ed8> router after_create subscribe /opt/stack/neutron/neutron/callbacks/manager.py:41 | 09:54 |
*** amotoki has joined #openstack-oslo | 09:55 | |
*** amotoki has quit IRC | 09:56 | |
*** e0ne has quit IRC | 10:01 | |
*** claudiub|2 has joined #openstack-oslo | 10:03 | |
*** e0ne has joined #openstack-oslo | 10:04 | |
*** claudiub has quit IRC | 10:06 | |
*** amotoki has joined #openstack-oslo | 10:06 | |
*** claudiub|2 has quit IRC | 10:32 | |
openstackgerrit | javeme proposed openstack/oslo.messaging: Missing version parameter in can_send_version() https://review.openstack.org/290385 | 10:35 |
openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Reduce number of connections https://review.openstack.org/287094 | 10:39 |
*** yamamoto_ has quit IRC | 10:42 | |
openstackgerrit | Bernhard M. Wiedemann proposed openstack/stevedore: dont claim copyright for future years https://review.openstack.org/290388 | 10:44 |
*** salv-orlando has joined #openstack-oslo | 10:51 | |
*** cdent has joined #openstack-oslo | 10:54 | |
*** salv-orlando has quit IRC | 10:56 | |
*** sdague has joined #openstack-oslo | 11:06 | |
*** dims has joined #openstack-oslo | 11:07 | |
*** takedakn has joined #openstack-oslo | 11:19 | |
*** yamamoto_ has joined #openstack-oslo | 11:33 | |
*** links has joined #openstack-oslo | 11:48 | |
*** amrith is now known as _amrith_ | 11:51 | |
lxsli | dhellmann: fyi added a mutable config session idea to the newton pad | 11:51 |
*** takedakn has quit IRC | 11:53 | |
*** takedakn has joined #openstack-oslo | 11:53 | |
*** salv-orlando has joined #openstack-oslo | 12:06 | |
openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: [zmq] Fix ceilometer https://review.openstack.org/289292 | 12:09 |
*** amotoki has quit IRC | 12:10 | |
openstackgerrit | Merged openstack-dev/pbr: Correct typo - s/enabeld/enabled/ https://review.openstack.org/289623 | 12:12 |
*** EinstCrazy has joined #openstack-oslo | 12:15 | |
*** yamamoto_ has quit IRC | 12:16 | |
*** sputnik13_ has joined #openstack-oslo | 12:16 | |
*** vks has quit IRC | 12:18 | |
*** lucasagomes is now known as lucas-hungry | 12:21 | |
dims | lxsli : nice thanks | 12:22 |
*** amotoki has joined #openstack-oslo | 12:24 | |
*** dguitarbite has quit IRC | 12:27 | |
*** nkrinner has quit IRC | 12:27 | |
*** pkholkin_ has quit IRC | 12:27 | |
*** pkholkin__ has joined #openstack-oslo | 12:27 | |
*** bapalm has quit IRC | 12:27 | |
*** takedakn has quit IRC | 12:27 | |
*** bapalm has joined #openstack-oslo | 12:27 | |
*** amotoki has quit IRC | 12:29 | |
*** sputnik13_ has quit IRC | 12:30 | |
*** nkrinner has joined #openstack-oslo | 12:40 | |
*** dguitarbite has joined #openstack-oslo | 12:41 | |
*** yamamoto has joined #openstack-oslo | 12:45 | |
openstackgerrit | Feng Xi Yan proposed openstack/oslo.vmware: Decode special chars in vSphere objects' propset https://review.openstack.org/187457 | 12:46 |
*** yamamoto has quit IRC | 12:47 | |
*** yamamoto has joined #openstack-oslo | 12:48 | |
*** thumpba has joined #openstack-oslo | 12:50 | |
*** yamamoto has quit IRC | 12:53 | |
openstackgerrit | Ilya Shakhat proposed openstack/oslo.messaging: [WIP] Simulator: calculate message latency statistics https://review.openstack.org/288383 | 13:02 |
*** dims has quit IRC | 13:05 | |
*** dims_ has joined #openstack-oslo | 13:05 | |
*** gchung has joined #openstack-oslo | 13:09 | |
*** edmondsw has joined #openstack-oslo | 13:10 | |
*** dims_ has quit IRC | 13:10 | |
*** dims has joined #openstack-oslo | 13:14 | |
*** pblaho has joined #openstack-oslo | 13:15 | |
*** yamamoto has joined #openstack-oslo | 13:17 | |
*** yamamoto has quit IRC | 13:17 | |
*** yamamoto has joined #openstack-oslo | 13:19 | |
*** _amrith_ is now known as amrith | 13:20 | |
openstackgerrit | Julien Danjou proposed openstack/oslo.middleware: cors: prevent WebOb setting a default Content-Type https://review.openstack.org/290482 | 13:21 |
*** lucas-hungry is now known as lucasagomes | 13:28 | |
*** jeckersb_gone is now known as jeckersb | 13:32 | |
*** links has quit IRC | 13:35 | |
*** vks has joined #openstack-oslo | 13:37 | |
*** jpena is now known as jpena|lunch | 13:39 | |
*** rakhmerov has quit IRC | 13:41 | |
*** ativelkov has quit IRC | 13:41 | |
*** kzaitsev_ws has quit IRC | 13:41 | |
*** binarin has quit IRC | 13:41 | |
*** rlrossit has joined #openstack-oslo | 13:48 | |
*** amotoki has joined #openstack-oslo | 13:54 | |
*** salv-orlando has quit IRC | 13:55 | |
*** yamamoto has quit IRC | 13:56 | |
*** yamamoto has joined #openstack-oslo | 13:57 | |
*** dtardivel has joined #openstack-oslo | 14:03 | |
*** kgiusti has joined #openstack-oslo | 14:05 | |
*** salv-orlando has joined #openstack-oslo | 14:07 | |
*** pradk has quit IRC | 14:08 | |
*** rakhmerov has joined #openstack-oslo | 14:12 | |
*** salv-orlando has quit IRC | 14:14 | |
*** ativelkov has joined #openstack-oslo | 14:15 | |
openstackgerrit | Julien Danjou proposed openstack/oslo.middleware: cors: prevent WebOb setting a default Content-Type https://review.openstack.org/290482 | 14:19 |
*** mriedem has joined #openstack-oslo | 14:19 | |
*** mtanino has quit IRC | 14:19 | |
*** nikhil_k has joined #openstack-oslo | 14:24 | |
openstackgerrit | Merged openstack/devstack-plugin-amqp1: Enable qpid-dispatch-router as alternative amqp backend https://review.openstack.org/289481 | 14:28 |
krotscheck | Anyone around to help me brainstorm a bug? | 14:33 |
krotscheck | I _think_ the set_defaults which the whole cors config discussion ended up asking for is actually a red herring. | 14:33 |
krotscheck | Assuming a paste deply environment.... | 14:34 |
krotscheck | The middleware chain is initialized before the app. | 14:34 |
krotscheck | So, our ConfigurableMiddleware goes and grabs the config file, and whatever defaults have been set at that point. | 14:35 |
krotscheck | In the case of the CORS middleware, it reads the values and stores them independently of oslo_config's config instance. | 14:35 |
krotscheck | (For easy access) | 14:35 |
krotscheck | So, the set_default values are simply not honored in a paste.ini environment. | 14:36 |
krotscheck | Do I have that right? | 14:39 |
*** jpena|lunch is now known as jpena | 14:47 | |
ryansb | Is there a more ideal way to copy a config object than using copy.copy? | 14:49 |
ryansb | I've seen some crashes caused by the copy module getting in a loop of calling _get and __getattr__ on a ConfigOpts object | 14:50 |
*** mriedem is now known as mriedem_meeting | 14:52 | |
*** bnemec has joined #openstack-oslo | 14:53 | |
*** zz_dimtruck is now known as dimtruck | 15:03 | |
*** xek_ is now known as xek | 15:05 | |
*** mtanino has joined #openstack-oslo | 15:06 | |
dims | rbradfor : dhellmann : fyi, here's a single log with deprecation messages from a dvsm/tempest - http://logs.openstack.org/01/289901/3/check/gate-tempest-dsvm-full/71b213c/logs/deprecations.txt.gz | 15:07 |
*** pblaho has quit IRC | 15:08 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 15:09 | |
*** pradk has joined #openstack-oslo | 15:09 | |
rbradfor | dims, is that frequency you added to prefix the line? | 15:26 |
openstackgerrit | Michael Krotscheck proposed openstack/oslo.middleware: CORS Middleware now honors upstream Vary header https://review.openstack.org/290583 | 15:28 |
*** suro-patz has joined #openstack-oslo | 15:28 | |
*** suro-patz has quit IRC | 15:32 | |
dims | rbradfor : yes | 15:36 |
krotscheck | dims, rbradfor, gordc: I'm throwing up some bug fixes to oslo_middleware.cors while I wait for the other patches to land. I'm not certain they're mitaka-required, but I want to get them out of my brain. Feel free to WF-2 them. | 15:36 |
dims | krotscheck : that's ok, dhellmann and i will get stable branches built from last set of releases so even if your changes land we are good | 15:37 |
krotscheck | dims: coolio | 15:38 |
rbradfor | dims, so how do you test devstack-gate script changes manually? | 15:38 |
dims | rbradfor : i just test snippets by hand | 15:39 |
rbradfor | dims, I was trying to work out how to actually run devstack-vm-gate-wrap.sh locally. | 15:39 |
*** mriedem_meeting is now known as mriedem | 15:40 | |
*** gchung is now known as gordc | 15:42 | |
*** browne has joined #openstack-oslo | 15:45 | |
dims | rbradfor : seen this? https://github.com/openstack-infra/devstack-gate#simulating-devstack-gate-tests | 15:46 |
*** nkrinner has quit IRC | 15:46 | |
rbradfor | dims, thanks | 15:46 |
*** EinstCrazy has quit IRC | 15:56 | |
*** regXboi has joined #openstack-oslo | 15:59 | |
*** suro-patz has joined #openstack-oslo | 16:00 | |
openstackgerrit | Dmitriy Ukhlov proposed openstack/oslo.messaging: test https://review.openstack.org/285239 | 16:00 |
*** zzzeek has quit IRC | 16:03 | |
*** zzzeek has joined #openstack-oslo | 16:04 | |
*** gcb has joined #openstack-oslo | 16:05 | |
*** yamahata has joined #openstack-oslo | 16:10 | |
*** gcb has quit IRC | 16:17 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.config: Disallow config option name as same as attribute of ConfigOpts https://review.openstack.org/289752 | 16:18 |
*** itisha has quit IRC | 16:19 | |
openstackgerrit | Ken Giusti proposed openstack/oslo.messaging: Fix the driver shutdown/failover logic https://review.openstack.org/290631 | 16:20 |
*** suro-patz has quit IRC | 16:27 | |
*** suro-patz has joined #openstack-oslo | 16:28 | |
*** gcb_ has joined #openstack-oslo | 16:30 | |
*** dims has quit IRC | 16:33 | |
*** harlowja_at_home has joined #openstack-oslo | 16:34 | |
vks | krotscheck, hi i need some info on oslo_messaging | 16:39 |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.config: doc:Log warning when can't get information from git https://review.openstack.org/290255 | 16:40 |
*** gcb_ has quit IRC | 16:41 | |
*** suro-patz has quit IRC | 16:41 | |
*** gcb_ has joined #openstack-oslo | 16:43 | |
*** dims has joined #openstack-oslo | 16:44 | |
krotscheck | vks: Oslo messaging isn't my specialty. rbradfor? | 16:49 |
*** e0ne has quit IRC | 16:50 | |
vks | krotscheck, thnx | 16:50 |
vks | rbradfor, hi | 16:50 |
rbradfor | vks, krotscheck sorry, I do nothing in oslo.messaging either. | 16:50 |
*** yamahata has quit IRC | 16:50 | |
krotscheck | vks: You can check the project's core list in gerrit for names to ping. | 16:50 |
vks | rbradfor, :) | 16:50 |
rbradfor | I suggest you propose your question, people will read it, but not everybody is looking at IRC right now. | 16:50 |
rbradfor | between different timezones, meetings and other work, it may take a bit for the right person to respond. | 16:51 |
*** browne has quit IRC | 16:53 | |
openstackgerrit | Doug Hellmann proposed openstack/oslo.config: Update reno for stable/mitaka https://review.openstack.org/290658 | 16:53 |
openstackgerrit | Doug Hellmann proposed openstack/oslo.db: Update reno for stable/mitaka https://review.openstack.org/290662 | 16:54 |
openstackgerrit | Doug Hellmann proposed openstack/oslo.log: Update reno for stable/mitaka https://review.openstack.org/290665 | 16:54 |
*** EinstCrazy has joined #openstack-oslo | 16:56 | |
*** gcb_ has quit IRC | 16:59 | |
*** EinstCrazy has quit IRC | 17:03 | |
krotscheck | Anyone know if there's a way for me to ask a ConfigOpts instance whether it has a value, or whether the returned value is the default? | 17:03 |
* krotscheck is digging through configOpts right now, but wouldn't mind pointers. | 17:04 | |
bknudson | I hope dhellmann has a script for updating .gitreviews | 17:06 |
toabctl | why are there 2 different versions mentioned for pbr in i.e. oslo.db (setup.py has >=1.8, but requirements.txt has >= 1.6) | 17:06 |
toabctl | ? | 17:06 |
*** kzaitsev_ws has joined #openstack-oslo | 17:07 | |
*** rcernin has quit IRC | 17:09 | |
vks | hi i need some info on oslo_messaging. I am writing a neutron agent, which should act on router create/delete/update event. It look like oslo_messaging can be used to subscribe for the rpc event | 17:10 |
vks | can it be used ? | 17:10 |
*** yamahata has joined #openstack-oslo | 17:10 | |
*** browne has joined #openstack-oslo | 17:18 | |
*** jschwarz has quit IRC | 17:25 | |
*** dimtruck is now known as zz_dimtruck | 17:30 | |
*** zz_dimtruck is now known as dimtruck | 17:30 | |
krotscheck | What is oslo's versiond eprecation policy? i.e. how long do I have to support things? | 17:35 |
*** mhickey_ has quit IRC | 17:46 | |
*** vilobhmm11 has joined #openstack-oslo | 17:47 | |
*** vilobhmm111 has joined #openstack-oslo | 17:48 | |
*** thumpba has quit IRC | 17:50 | |
*** thumpba has joined #openstack-oslo | 17:51 | |
*** vilobhmm11 has quit IRC | 17:51 | |
openstackgerrit | Dmitriy Ukhlov proposed openstack/oslo.messaging: test https://review.openstack.org/285239 | 17:52 |
*** amotoki has quit IRC | 17:57 | |
*** vks has quit IRC | 17:59 | |
*** shardy has quit IRC | 18:03 | |
*** sputnik13 has joined #openstack-oslo | 18:07 | |
*** vilobhmm111 has quit IRC | 18:08 | |
*** ihrachys has quit IRC | 18:10 | |
*** vks has joined #openstack-oslo | 18:13 | |
*** jpena is now known as jpena|zzz | 18:14 | |
*** lucasagomes is now known as lucas-dinner | 18:18 | |
openstackgerrit | Merged openstack/osprofiler: It's unnecessary set deprecate group for option 'enabled' https://review.openstack.org/285898 | 18:25 |
*** e0ne has joined #openstack-oslo | 18:29 | |
*** takedakn has joined #openstack-oslo | 18:29 | |
*** takedakn has quit IRC | 18:30 | |
*** mriedem has quit IRC | 18:36 | |
*** mriedem has joined #openstack-oslo | 18:36 | |
*** dims has quit IRC | 18:44 | |
*** yamamoto has quit IRC | 18:45 | |
*** dims has joined #openstack-oslo | 18:46 | |
*** vks has quit IRC | 18:55 | |
*** rlrossit has quit IRC | 18:58 | |
*** rlrossit has joined #openstack-oslo | 19:00 | |
*** cdent has quit IRC | 19:01 | |
*** vilobhmm11 has joined #openstack-oslo | 19:06 | |
*** boris-42 has quit IRC | 19:14 | |
*** yamamoto has joined #openstack-oslo | 19:15 | |
*** SergeyLukjanov has quit IRC | 19:19 | |
*** SergeyLukjanov has joined #openstack-oslo | 19:21 | |
*** yamamoto has quit IRC | 19:21 | |
*** gordc has quit IRC | 19:22 | |
openstackgerrit | Merged openstack/oslo.db: Update reno for stable/mitaka https://review.openstack.org/290662 | 19:25 |
*** yassine has quit IRC | 19:40 | |
dhellmann | bknudson : yes, I definitely do. | 19:40 |
dhellmann | krotscheck : why do you want to know if an option is set to something other than the default? | 19:41 |
openstackgerrit | Merged openstack/oslo.config: Update reno for stable/mitaka https://review.openstack.org/290658 | 19:41 |
bknudson | dhellmann: dims: looks like we're ready to lift all the -2s? | 19:42 |
dhellmann | bknudson : yeah, I think so, but I'd wait for dims just in case | 19:43 |
bknudson | I've got my own script to try out | 19:43 |
dims | bknudson : go for it! | 19:44 |
dhellmann | dims : for the joint oslo presentation, do you want to prepare a single set of slides or treat it like lightning talks and change laptops for each speaker? | 19:44 |
*** yamahata has quit IRC | 19:44 | |
*** sdague has quit IRC | 19:45 | |
*** sigmavirus24 is now known as sigmavirus24_awa | 19:48 | |
dims | dhellmann : lets do lightning talks format for now to have some freedom though the downside is that we may lose time (unless we load up all presos into the same laptop). | 19:48 |
dhellmann | dims : ok, sounds good | 19:49 |
*** salv-orlando has joined #openstack-oslo | 19:51 | |
*** salv-orl_ has joined #openstack-oslo | 19:53 | |
*** sdague has joined #openstack-oslo | 19:54 | |
*** salv-orlando has quit IRC | 19:56 | |
krotscheck | dhellmann: I'm trying to properly implement set_defaults for the cors middleware, and it's turning out to be tricky; the middleware creates an internal copy of the existing configopts to use for subsequent configuration blocks, which are deep-copied so that the defaults can be overridden. | 19:58 |
*** rlrossit has quit IRC | 19:58 | |
krotscheck | dhellmann: The second issue there is that middleware is occasionally initialized before application code, meaning that set_defaults won't actually do anything. | 19:59 |
dhellmann | krotscheck : set_defaults() always only changes the default value, and allows a user-specified value to override that regardless of the order of calls | 19:59 |
krotscheck | dhellmann: I'm talking about this line: http://git.openstack.org/cgit/openstack/oslo.middleware/tree/oslo_middleware/cors.py#n143 | 19:59 |
dhellmann | you want to set the defaults of a subgroup option to be the default of the master option? | 20:00 |
krotscheck | dhellmann: Hence my question about deprecating as well. The whole prefix config option feature may be something we just deprecate, now that allowed_origins is a list_opt | 20:00 |
krotscheck | dhellmann: I want to set the value of the subgroup to what the master is, IFF the subgroup does not have a value set. | 20:01 |
krotscheck | That way I can use one single GroupAttr instance instead of two. | 20:01 |
*** rlrossi__ has joined #openstack-oslo | 20:02 | |
krotscheck | Or, well, n | 20:02 |
dhellmann | krotscheck : right, so just use set_default(). It changes the default, but not the actual value of the option. If the user sets the option value in their config file, it overrides the default. | 20:02 |
krotscheck | dhellmann: Maybe I'm not quite clear. | 20:02 |
ryansb | Hey, can someone help me out with copying a config object? I'm having trouble in python 3.4 using copy.copy | 20:02 |
*** gordc has joined #openstack-oslo | 20:02 | |
krotscheck | dhellmann: Right now, the value chain in a subgroup option is subgroup_value -> master_value, not subgroup_value -> master_default. | 20:03 |
krotscheck | So I can change it to just use one configopt, but that'd be a breaking change. | 20:03 |
krotscheck | Sorry | 20:03 |
krotscheck | Basically skip the clone altogether. | 20:03 |
* krotscheck is trying to brainstorm a path forward. | 20:04 | |
dhellmann | do you want subgroup_default to be master_default? | 20:04 |
dhellmann | krotscheck : I'm afraid I'm still not clear on exactly what you're trying to do, and I would like to because I think I can help you avoid writing something complicated | 20:06 |
dhellmann | there's already support for various levels of default and override setting | 20:06 |
bknudson | dims: dhellmann: ok, all the -2s are lifted. | 20:07 |
dhellmann | bknudson : thanks! | 20:07 |
krotscheck | dhellmann: I want to rip a bunch of things out of cors middleware that, quite frankly, aren't necessary. | 20:07 |
dhellmann | ok | 20:07 |
dhellmann | that seems a worthy goal | 20:07 |
krotscheck | dhellmann: But most of those are breaking changes, so I have to figure out a way to fix bugs in the meantime. | 20:07 |
dhellmann | makes sense | 20:07 |
krotscheck | The current order of execution bug I'm trying to figure out is this: | 20:08 |
krotscheck | Paste deploy initializes cors Middleware. Middleware parses config file for [cors] and [cors.*] (this second one creates a deep clone of CORS_OPTS). Then the application is initialized, and set_defaults is called. CORS_OPTS defaults are updated, the subgroup opts are not. | 20:09 |
krotscheck | Second bug: | 20:09 |
openstackgerrit | Joshua Harlow proposed openstack/tooz: Add basic mysql table support (if provided a database) [POC] https://review.openstack.org/290178 | 20:09 |
dhellmann | can you change the middleware so that it registers the options before it needs to use them, instead of when it is loaded? | 20:10 |
krotscheck | dhellmann: Yes, that's one approach I need to take anyway. | 20:10 |
krotscheck | 2nd bug: | 20:11 |
dhellmann | if you register the options inside the set_defaults function exposed by the middleware *and* at runtime before they are checked, that should solve your ordering problem | 20:11 |
krotscheck | The current behavior is that if something is set in [cors], but not in [cors.*], then [cors.*] has its value set to the value that is set in [cors], via set_defaults. | 20:12 |
krotscheck | Basically, there's config inheritance between [cors] and [cors.*] | 20:13 |
dhellmann | ok | 20:13 |
krotscheck | RIght now that's done via set_defaults. I'd like to actually set the value itself. | 20:13 |
dhellmann | why? | 20:13 |
dhellmann | if the user has set a value, why do you want to override it? | 20:14 |
krotscheck | dhellmann: Hrm. Similar order of operations case as above - I think there'll be aproblem that if I set the default to what is currently set in CORS_OPTS (via _get) and then the default on CORS_OPTS updates, then the defaults are now out of sync. | 20:16 |
dhellmann | if the sub-options aren't registered until they are used, won't that be after the application has changed the main option defaults? | 20:18 |
dhellmann | do you want to sync the defaults, or do you want to make the value of the sub-option match the parent? | 20:18 |
dhellmann | I'm confused about which sets of values you're trying to sync | 20:18 |
dhellmann | maybe if you used an actual example it would help make it clear | 20:18 |
krotscheck | Well, the end goal is to remove subotps altogether, because allowed_origin is now a ListOpt. | 20:19 |
krotscheck | And, well, I've yet to talk to someone who cares about having such fine-tune control over CORS> | 20:20 |
krotscheck | Mostly because, as an attack vector, just using an API directly is a better idea. | 20:20 |
dhellmann | I guess you could do a few things | 20:21 |
dhellmann | you could check if the value of the subopt is different from the parent and emit a warning, then either use set_override to change the subopt or just go with what it's set to | 20:22 |
dhellmann | starting with a warning puts you on the path to deprecation | 20:22 |
dhellmann | if you want the subopt to default to the value of the parent opt, unless overridden by the user, you can use set_default() instead of set_override() | 20:22 |
bknudson | btw - is there a good python library for gerrit? | 20:23 |
bknudson | gerrit rest api | 20:23 |
dhellmann | bknudson : the infra team might know | 20:23 |
bknudson | not that it was hard to write one | 20:23 |
dhellmann | ryansb : what kind of object are you trying to copy? | 20:24 |
*** ihrachys has joined #openstack-oslo | 20:25 | |
ryansb | dhellmann: cfg.ConfigOpts | 20:25 |
dhellmann | ryansb : why are you cloning that? what's the use case? | 20:26 |
dhellmann | I don't think we expect anyone to do that | 20:26 |
ryansb | We're using it to make dynamic configs to support multiple zaqar storage backends | 20:26 |
ryansb | specifically https://github.com/openstack/zaqar/blob/stable/liberty/zaqar/storage/utils.py#L53 | 20:26 |
dhellmann | why not just make a new instance? what's the benefit of cloning the existing thing? | 20:27 |
ryansb | We do if there isn't an existing one | 20:28 |
dhellmann | so if you have an instance you clone it and then change the clone? why not just make a new instance and initialize it? | 20:29 |
dhellmann | or, better, register the backend options in a custom group where they can all live together in the same instance | 20:30 |
dhellmann | IIUC, that's how cinder does it | 20:30 |
ryansb | The issue I'm seeing is a stackoverflow exception between the _get and the __getattr__ methods on ConfigOpts | 20:31 |
dhellmann | during the copy? | 20:31 |
ryansb | yes, during the copy | 20:32 |
dhellmann | what attribute is it looking up? we probably don't have the copy hooks in place for that class because we didn't expect anyone to do this | 20:32 |
ryansb | it looks like _get calls "self.__cache" which tries to resolve using getattr | 20:32 |
dhellmann | yeah | 20:32 |
ryansb | but then the __getattr__ calls _get | 20:32 |
ryansb | and so on until explosion | 20:32 |
dhellmann | my recommendation is that you stop trying to make a copy of the object, and we figure out a better way for you to have what you're trying to do | 20:33 |
ryansb | oslo.config==3.9.0 for what it's worth | 20:33 |
dhellmann | but for that, I need more details about what exactly you *are* trying to accomplish by having a copy | 20:33 |
ryansb | Yeah, sure, so as I understand it (I didn't write this bit, I happened across the exception on the .copy) | 20:34 |
*** fhermeni has joined #openstack-oslo | 20:36 | |
ryansb | we use it in the part of zaqar that maps queues to the right storage pool | 20:37 |
bknudson | apparently gertty's got a batch review feature that I could have used | 20:37 |
fhermeni | hi there | 20:37 |
dhellmann | bknudson : oh? I'm going to have to give that a look | 20:37 |
ryansb | So we share options across different pools | 20:37 |
dhellmann | bknudson : is there a user guide or something? | 20:37 |
bknudson | dhellmann: that's the thing, apparently it's poorly documented. | 20:38 |
dhellmann | ryansb : so there is a separate configopts instance for each pool? | 20:38 |
dhellmann | hi, fhermeni | 20:38 |
ryansb | dhellmann: yeah | 20:40 |
dhellmann | bknudson : you don't have a script I could use to recheck a bunch of patches all at once, do you? | 20:40 |
bknudson | dhellmann: I could modify this script pretty easily to post a recheck review. | 20:41 |
dhellmann | ryansb : ok. I still recommend instantiating independent ConfigOpts objects and populating them with the right options and config files. | 20:41 |
bknudson | let me post the tool | 20:41 |
ryansb | and the copy takes the options from the pool and populates these options | 20:41 |
dhellmann | ryansb : *OR* not using separate objects and registering options in pool-specific groups, but that's probably a bigger change | 20:41 |
ryansb | Yeah, it likely would be | 20:41 |
ryansb | Is there a way to, say, serialize the settings of a config option and use them to instantiate new ones | 20:42 |
openstackgerrit | Merged openstack/oslo.log: Update reno for stable/mitaka https://review.openstack.org/290665 | 20:43 |
ryansb | or is the way just to instantiate the objects normally based on the __init__ params? | 20:43 |
*** salv-orl_ has quit IRC | 20:43 | |
dhellmann | we have no facilities right now for assisting with cloning a ConfigOpts instance using any API because that is not a thing we have expected anyone to be doing | 20:43 |
dhellmann | it seems that whoever wrote this didn't really understand how to use oslo.config, and didn't ask for advice, so they've designed you into a bit of a corner | 20:44 |
ryansb | heh, that's possible | 20:44 |
ryansb | anyways, thank you for the tips. Hopefully it's not too tight a corner | 20:44 |
dhellmann | it appears that dynamic_conf() is trying to make a new ConfigOpt, ensure that some options are registered, and then ensure that some of those options have specific values? | 20:45 |
ryansb | yes | 20:45 |
ryansb | that'd be correct | 20:45 |
dhellmann | ok, then I would just do that. call ConfigOpts(), call register_opts(), call __call__() to initialize it from the config files, and then call set_override() | 20:46 |
ryansb | That makes sense. So I'll just figure out what that base object has set, and then I can recreate it without the copying | 20:47 |
ryansb | thanks again :) | 20:47 |
dhellmann | there are some public attributes of the object that you could use to pass to __call__() on the new one, but you should not reuse the same Opt instances between objects if you're calling set_override() and set_default() | 20:48 |
bknudson | dhellmann: this is the tool -- https://github.com/brantlk/update_reviews/blob/master/update_reviews/__init__.py not much to it | 20:49 |
bknudson | if you think it'd be useful to have it in an openstack repo I can propose it there, too | 20:50 |
dhellmann | I need something like that but I need it to let me specify the query and comment as args | 20:50 |
dhellmann | but having this as a basis would be a start, we could make it more flexible | 20:51 |
bknudson | easy enhancements. | 20:51 |
dhellmann | if you want to put it in release-tools maybe? | 20:51 |
bknudson | I'll take a stab at putting it in release-tools. | 20:51 |
dhellmann | excellent, thanks | 20:51 |
*** ihrachys has quit IRC | 20:51 | |
*** e0ne has quit IRC | 20:53 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 20:58 | |
*** dims has quit IRC | 21:05 | |
bknudson | dhellmann: posted the code to release-tools : https://review.openstack.org/#/c/290853/ | 21:08 |
patchbot | bknudson: patch 290853 - openstack-infra/release-tools - Add update_reviews tool | 21:08 |
*** thumpba has quit IRC | 21:13 | |
*** thumpba has joined #openstack-oslo | 21:14 | |
*** thumpba has quit IRC | 21:18 | |
*** sdague_ has joined #openstack-oslo | 21:23 | |
dhellmann | bknudson : ack, I'll take a look when I'm fresh tomorrow | 21:33 |
openstackgerrit | Ilya Shakhat proposed openstack/oslo.messaging: [WIP] Simulator: calculate message latency statistics https://review.openstack.org/288383 | 21:33 |
*** dims has joined #openstack-oslo | 21:37 | |
*** yamahata has joined #openstack-oslo | 21:47 | |
*** dtardivel has quit IRC | 21:48 | |
*** dims has quit IRC | 21:56 | |
*** ndipanov has quit IRC | 21:57 | |
*** kgiusti has left #openstack-oslo | 21:57 | |
*** dims has joined #openstack-oslo | 21:58 | |
*** dimtruck is now known as zz_dimtruck | 22:07 | |
openstackgerrit | Jason Dunsmore proposed openstack/oslo.service: Add ability to cancel Threads and ThreadGroups https://review.openstack.org/290880 | 22:14 |
*** gordc has quit IRC | 22:16 | |
*** sdague_ has quit IRC | 22:17 | |
*** sdague has quit IRC | 22:17 | |
*** david-lyle has quit IRC | 22:19 | |
*** david-lyle has joined #openstack-oslo | 22:20 | |
*** amrith is now known as _amrith_ | 22:24 | |
openstackgerrit | Miguel Angel Ajo proposed openstack/oslo.versionedobjects: Add BaseEnumField valid_values introspection https://review.openstack.org/287106 | 22:27 |
openstackgerrit | Eric Brown proposed openstack/oslo.context: Use project instead of tenant https://review.openstack.org/289588 | 22:28 |
*** dims has quit IRC | 22:29 | |
openstackgerrit | Miguel Angel Ajo proposed openstack/oslo.versionedobjects: Add BaseEnumField valid_values introspection https://review.openstack.org/287106 | 22:30 |
*** mriedem has quit IRC | 22:31 | |
*** fhermeni has quit IRC | 22:32 | |
*** yamahata has quit IRC | 22:35 | |
*** zz_dimtruck is now known as dimtruck | 22:38 | |
*** regXboi has quit IRC | 22:43 | |
*** dims has joined #openstack-oslo | 22:44 | |
*** rlrossi__ has quit IRC | 22:46 | |
*** browne has quit IRC | 22:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!