| *** masber has joined #openstack-keystone | 00:03 | |
| *** jessegler has quit IRC | 00:04 | |
| jamielennox | yea, i'd say deprecate it | 00:09 |
|---|---|---|
| jamielennox | however there was general approval for a file backed catalog previously | 00:09 |
| jamielennox | just one that was more version independant | 00:09 |
| jamielennox | specifically because it's really easy to drop with ansible or somethgni | 00:10 |
| eandersson | Would it be worth fixing the v3 support? or at least for older versions | 00:16 |
| *** gongysh has quit IRC | 00:17 | |
| eandersson | I have a patch, but no way of saying if my patch will make it worse for other situations | 00:22 |
| *** liujiong has joined #openstack-keystone | 00:24 | |
| *** aojea has joined #openstack-keystone | 00:32 | |
| *** aojea has quit IRC | 00:36 | |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/keystone master: [WIP] Fixing multi-region support in templated v3 catalog https://review.openstack.org/482364 | 00:44 |
| *** thorst has quit IRC | 00:45 | |
| *** edmondsw has joined #openstack-keystone | 00:49 | |
| *** edmondsw has quit IRC | 00:53 | |
| *** amyge has quit IRC | 01:05 | |
| eandersson | I am sure the above patch is far from fixing all the issues, but at the very least it behaves better with multi-regions now ^ | 01:05 |
| *** zzzeek_ has quit IRC | 01:09 | |
| *** zzzeek_ has joined #openstack-keystone | 01:11 | |
| *** thorst has joined #openstack-keystone | 01:28 | |
| *** zzzeek_ has quit IRC | 01:33 | |
| *** zzzeek_ has joined #openstack-keystone | 01:34 | |
| *** zzzeek_ has quit IRC | 01:36 | |
| *** Shunli has joined #openstack-keystone | 01:41 | |
| *** iurygregory has quit IRC | 01:43 | |
| *** iurygregory has joined #openstack-keystone | 01:46 | |
| *** zzzeek_ has joined #openstack-keystone | 01:47 | |
| openstackgerrit | wingwj proposed openstack/keystone master: [install] Clarify the paths of the rc files https://review.openstack.org/482373 | 01:55 |
| morgan | mordred: ++ | 02:01 |
| morgan | mordred: i can spin up a deprecation patch soon | 02:01 |
| morgan | i'll also look at a yaml alternative for folks who need it | 02:01 |
| *** zhurong has joined #openstack-keystone | 02:03 | |
| *** phalmos_ has quit IRC | 02:23 | |
| *** aojea has joined #openstack-keystone | 02:32 | |
| *** aojea has quit IRC | 02:37 | |
| *** edmondsw has joined #openstack-keystone | 02:37 | |
| *** aselius has quit IRC | 02:38 | |
| *** edmondsw has quit IRC | 02:42 | |
| *** ducttape_ has joined #openstack-keystone | 02:46 | |
| *** iurygregory has quit IRC | 02:49 | |
| *** iurygregory has joined #openstack-keystone | 02:53 | |
| *** namnh has joined #openstack-keystone | 03:07 | |
| *** ducttape_ has quit IRC | 03:20 | |
| *** gyee has quit IRC | 03:23 | |
| *** thorst has joined #openstack-keystone | 03:29 | |
| *** thorst has quit IRC | 03:34 | |
| *** links has joined #openstack-keystone | 03:41 | |
| *** jmlowe has joined #openstack-keystone | 03:58 | |
| *** dave-mccowan has quit IRC | 04:00 | |
| *** wasmum has quit IRC | 04:14 | |
| *** edmondsw has joined #openstack-keystone | 04:25 | |
| *** wasmum has joined #openstack-keystone | 04:28 | |
| *** edmondsw has quit IRC | 04:30 | |
| *** jrist has quit IRC | 04:33 | |
| *** d0ugal has joined #openstack-keystone | 05:22 | |
| *** thorst has joined #openstack-keystone | 05:30 | |
| *** aojea has joined #openstack-keystone | 05:33 | |
| *** thorst has quit IRC | 05:34 | |
| *** Shunli has quit IRC | 05:39 | |
| *** Shunli has joined #openstack-keystone | 05:40 | |
| *** rcernin has joined #openstack-keystone | 05:41 | |
| *** d0ugal has quit IRC | 05:52 | |
| *** nicolasbock has joined #openstack-keystone | 05:55 | |
| *** pcaruana has joined #openstack-keystone | 06:04 | |
| *** aojea has quit IRC | 06:07 | |
| *** aojea has joined #openstack-keystone | 06:07 | |
| *** aojea has quit IRC | 06:12 | |
| *** aojea has joined #openstack-keystone | 06:12 | |
| *** edmondsw has joined #openstack-keystone | 06:14 | |
| openstackgerrit | Merged openstack/keystone master: Move caching docs into admin-guide https://review.openstack.org/477678 | 06:15 |
| *** aojea has quit IRC | 06:16 | |
| *** edmondsw has quit IRC | 06:18 | |
| openstackgerrit | Merged openstack/keystone master: Fixing flushing tokens workflow https://review.openstack.org/480287 | 06:19 |
| *** basilAB has quit IRC | 06:23 | |
| *** vaishali has quit IRC | 06:24 | |
| *** jmlowe has quit IRC | 06:25 | |
| *** basilAB has joined #openstack-keystone | 06:28 | |
| *** vaishali has joined #openstack-keystone | 06:29 | |
| *** zhurong has quit IRC | 06:40 | |
| *** zhurong has joined #openstack-keystone | 06:44 | |
| *** belmoreira has joined #openstack-keystone | 06:46 | |
| *** namnh has quit IRC | 06:52 | |
| *** tobberydberg has joined #openstack-keystone | 06:53 | |
| *** tobberyd_ has joined #openstack-keystone | 06:55 | |
| *** tobberydberg has quit IRC | 06:58 | |
| *** namnh has joined #openstack-keystone | 07:01 | |
| *** tesseract has joined #openstack-keystone | 07:02 | |
| *** aojea has joined #openstack-keystone | 07:15 | |
| *** zhurong has quit IRC | 07:30 | |
| *** thorst has joined #openstack-keystone | 07:31 | |
| *** openstackgerrit has quit IRC | 07:33 | |
| *** thorst has quit IRC | 07:36 | |
| *** toddnni has joined #openstack-keystone | 07:40 | |
| *** toddnni has quit IRC | 07:40 | |
| *** namnh has quit IRC | 07:45 | |
| *** toddnni has joined #openstack-keystone | 07:45 | |
| *** openstackgerrit has joined #openstack-keystone | 07:46 | |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/keystone master: [WIP] Fixing multi-region support in templated v3 catalog https://review.openstack.org/482364 | 07:46 |
| *** markvoelker_ has quit IRC | 07:47 | |
| *** rajalokan has joined #openstack-keystone | 07:55 | |
| openstackgerrit | Samriddhi proposed openstack/keystone master: Expanded the best practices subsection in devdocs https://review.openstack.org/476541 | 07:58 |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-keystone | 08:00 | |
| *** namnh has joined #openstack-keystone | 08:01 | |
| *** openstackgerrit has quit IRC | 08:03 | |
| *** gengchc has joined #openstack-keystone | 08:11 | |
| *** gengchc has quit IRC | 08:11 | |
| *** mvk has quit IRC | 08:26 | |
| *** zhurong has joined #openstack-keystone | 08:27 | |
| *** mvk has joined #openstack-keystone | 08:55 | |
| odyssey4me | lbragstad I think that the rolling-upgrade job can be vastly more simply configured: https://review.openstack.org/482474 | 09:18 |
| *** openstackgerrit has joined #openstack-keystone | 09:26 | |
| openstackgerrit | TommyLike proposed openstack/oslo.policy master: Fix parsing bug when config file is empty. https://review.openstack.org/482477 | 09:26 |
| *** zhurong has quit IRC | 09:27 | |
| *** Shunli has quit IRC | 09:29 | |
| openstackgerrit | TommyLike proposed openstack/oslo.policy master: Fix parsing bug when config file is empty https://review.openstack.org/482477 | 09:32 |
| *** thorst has joined #openstack-keystone | 09:32 | |
| *** mvk has quit IRC | 09:37 | |
| *** links has quit IRC | 09:37 | |
| *** thorst has quit IRC | 09:37 | |
| *** links has joined #openstack-keystone | 09:39 | |
| *** links has quit IRC | 09:44 | |
| *** markvoelker has joined #openstack-keystone | 09:48 | |
| *** links has joined #openstack-keystone | 09:49 | |
| *** edmondsw has joined #openstack-keystone | 09:50 | |
| *** mvk has joined #openstack-keystone | 09:51 | |
| *** d0ugal has joined #openstack-keystone | 09:53 | |
| *** edmondsw has quit IRC | 09:54 | |
| *** d0ugal has quit IRC | 09:59 | |
| *** namnh has quit IRC | 10:08 | |
| *** links has quit IRC | 10:13 | |
| *** liujiong has quit IRC | 10:20 | |
| *** ducttape_ has joined #openstack-keystone | 10:22 | |
| *** markvoelker has quit IRC | 10:22 | |
| *** stingaci has joined #openstack-keystone | 10:26 | |
| *** ducttape_ has quit IRC | 10:26 | |
| samueldmq | morning | 10:29 |
| *** thorst has joined #openstack-keystone | 10:50 | |
| *** stingaci has quit IRC | 10:54 | |
| *** dave-mccowan has joined #openstack-keystone | 11:09 | |
| *** markvoelker has joined #openstack-keystone | 11:20 | |
| knikolla | o/ | 11:24 |
| *** edmondsw has joined #openstack-keystone | 11:38 | |
| *** edmondsw has quit IRC | 11:42 | |
| *** links has joined #openstack-keystone | 11:46 | |
| *** markvoelker has quit IRC | 11:53 | |
| *** links has quit IRC | 11:59 | |
| *** raildo has joined #openstack-keystone | 12:16 | |
| *** markvoelker has joined #openstack-keystone | 12:19 | |
| *** rajalokan has quit IRC | 12:45 | |
| *** edmondsw has joined #openstack-keystone | 12:46 | |
| *** openstackgerrit has quit IRC | 12:47 | |
| *** spilla has joined #openstack-keystone | 12:59 | |
| *** rajalokan has joined #openstack-keystone | 13:13 | |
| lbragstad | o/ | 13:15 |
| bhagyashris | samueldmq, mordred, jamielennox: Hi, thanks for your opinion. Sorry I am not able to reply yesterday because i am working in IST time zone. Actually i am not fully clear about suggestion so can you please explain me little bit in detail? | 13:27 |
| bhagyashris | samueldmq, mordred, jamielennox: Regarding the topic log request-id at INFO level in keystoneauth. | 13:31 |
| *** openstackgerrit has joined #openstack-keystone | 13:35 | |
| openstackgerrit | Merged openstack/keystone master: Move upgrade documentation to admin-guide https://review.openstack.org/481381 | 13:35 |
| *** bknudson has joined #openstack-keystone | 13:42 | |
| hrybacki | lbragstad: question regarding backports -- is it generally acceptable to backport via the cherrypick -> <stable branch> in gerrit web UI or do folks do it manually and then submit via gerrit-review? | 13:48 |
| lbragstad | hrybacki: i typically do it manually | 13:48 |
| lbragstad | something like: | 13:49 |
| lbragstad | $ git checkout stable/ocata~0; git review -x <change-id> | 13:50 |
| hrybacki | ack. Thanks lbragstad. Adam's old work is now falling on my shoulders :) | 13:51 |
| lbragstad | :) | 13:51 |
| *** d0ugal has joined #openstack-keystone | 13:54 | |
| *** dklyle has quit IRC | 13:55 | |
| *** david-lyle has joined #openstack-keystone | 13:55 | |
| *** ducttape_ has joined #openstack-keystone | 13:57 | |
| *** jistr is now known as jistr|call | 13:58 | |
| *** d0ugal has quit IRC | 14:00 | |
| mordred | bhagyashris: no worries- the suggestion is to define a new named logger just for request-id logging | 14:01 |
| *** ducttape_ has quit IRC | 14:01 | |
| mordred | bhagyashris: so - currently logger is defined as: _logger = utils.get_logger(__name__) | 14:01 |
| mordred | bhagyashris: which defined a logger named "keystoneauth.session" | 14:01 |
| mordred | bhagyashris: an aditional one could be made like "_request_id_logger = utils.get_logger("keystoneauth.request-ids") | 14:02 |
| mordred | logger names can be completely arbitrary | 14:02 |
| mordred | bhagyashris: then at the call side, use the request_id logger intead of the normal logger | 14:02 |
| mordred | bhagyashris: this way a person consuming the library can say that they want to see messages for keystoneauth1.request-ids at DEBUG level but everything else at info level | 14:03 |
| mordred | bhagyashris: (also, it's keystoneauth1.session and keystoneauth1.request-ids fwiw) | 14:03 |
| mordred | bhagyashris: it will need to be plumbed through in the request, _http_log_request and _http_log_response methods so that a person can pass in their own request_ids logger .. and also so that a person who is today passing in a single logger to the logger option will have that logger used for both logger and request_ids_logger | 14:05 |
| mordred | (since that would be backwards compatible) | 14:05 |
| bhagyashris | mordred: ok. now I got point and border view. | 14:11 |
| bhagyashris | mordred: Thanks for whole explanation. :) | 14:12 |
| *** lwanderley has joined #openstack-keystone | 14:12 | |
| openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Use a specific logger for logging request ids https://review.openstack.org/482577 | 14:12 |
| mordred | bhagyashris: sure! there is a not-tested half-patch ^^ | 14:12 |
| mordred | bhagyashris: but might be a good basis to start from - feel free to take that over and modify it or whatnot - I figured tossing up a few lines of code might be the clearest :) | 14:13 |
| bhagyashris | mordred: yeah, sure I will try my best and will check and will also ask in case of any doubt. thanks. :) | 14:16 |
| *** ducttape_ has joined #openstack-keystone | 14:17 | |
| *** zzzeek_ has quit IRC | 14:18 | |
| *** zzzeek_ has joined #openstack-keystone | 14:21 | |
| *** d0ugal has joined #openstack-keystone | 14:25 | |
| *** spzala has joined #openstack-keystone | 14:27 | |
| *** lwanderley has quit IRC | 14:27 | |
| *** gyee has joined #openstack-keystone | 14:27 | |
| *** d0ugal has quit IRC | 14:31 | |
| *** rcernin has quit IRC | 14:38 | |
| *** rcernin has joined #openstack-keystone | 14:40 | |
| *** openstackgerrit has quit IRC | 14:48 | |
| *** openstackgerrit has joined #openstack-keystone | 14:50 | |
| openstackgerrit | Yaguang Tang proposed openstack/keystonemiddleware master: Remove PKI/PKIZ auth support and tests https://review.openstack.org/481895 | 14:50 |
| *** jmlowe has joined #openstack-keystone | 14:56 | |
| hrybacki | lbragstad: are we still accepting backports to Newton? The timeline for stable branches accepting backports reads a bit loose and fast | 14:59 |
| *** chandankumar has left #openstack-keystone | 15:00 | |
| *** phalmos has joined #openstack-keystone | 15:05 | |
| *** dklyle has joined #openstack-keystone | 15:11 | |
| *** david-lyle has quit IRC | 15:11 | |
| *** zzzeek_ has quit IRC | 15:12 | |
| *** dklyle is now known as david-lyle | 15:15 | |
| *** aselius_ has joined #openstack-keystone | 15:15 | |
| openstackgerrit | Omar Tleimat proposed openstack/keystone master: WIP: Add project tags https://review.openstack.org/470317 | 15:16 |
| *** rcernin has quit IRC | 15:18 | |
| *** jistr|call is now known as jistr | 15:22 | |
| *** rajalokan has quit IRC | 15:26 | |
| *** catintheroof has joined #openstack-keystone | 15:28 | |
| *** rderose has joined #openstack-keystone | 15:29 | |
| *** zzzeek_ has joined #openstack-keystone | 15:30 | |
| *** tobberydberg has joined #openstack-keystone | 15:35 | |
| *** tobberyd_ has quit IRC | 15:38 | |
| *** belmoreira has quit IRC | 15:38 | |
| raildo | lbragstad, backports for ocata and newton for the flushing tokens: https://review.openstack.org/#/c/482606/ https://review.openstack.org/#/c/482601/ | 15:39 |
| *** tobberydberg has quit IRC | 15:39 | |
| lbragstad | raildo: nice - thank you | 15:40 |
| lbragstad | raildo: looks like the original patch for stable/newton will have to be rebased once https://review.openstack.org/#/c/482606/1 merges? | 15:40 |
| raildo | lbragstad, hum, I'm not sure, since there are in different branches... | 15:48 |
| *** tobberydberg has joined #openstack-keystone | 15:56 | |
| *** jmlowe has quit IRC | 15:57 | |
| *** tobberydberg has quit IRC | 16:00 | |
| *** jmlowe has joined #openstack-keystone | 16:00 | |
| *** aojea has quit IRC | 16:03 | |
| *** sjain has joined #openstack-keystone | 16:10 | |
| *** edmondsw has quit IRC | 16:17 | |
| *** edmondsw_ has joined #openstack-keystone | 16:19 | |
| *** lwanderley has joined #openstack-keystone | 16:20 | |
| *** jmlowe has quit IRC | 16:22 | |
| *** phalmos has quit IRC | 16:23 | |
| *** edmondsw_ has quit IRC | 16:23 | |
| *** spzala has quit IRC | 16:37 | |
| *** spilla has quit IRC | 16:43 | |
| *** toddnni has quit IRC | 16:43 | |
| morgan | would someone be kind enough to re-spin this and fix comments:https://review.openstack.org/#/c/477566/4 (mordred) I'd like to get this landed so we can release KSA with lots of discovery goodness | 16:46 |
| openstackgerrit | Eric Fried proposed openstack/keystoneauth master: Miscellaneous cleanup in discover.py https://review.openstack.org/482271 | 16:47 |
| *** sjain_ has joined #openstack-keystone | 16:54 | |
| *** sjain has quit IRC | 16:57 | |
| *** spzala has joined #openstack-keystone | 17:02 | |
| *** spzala has quit IRC | 17:03 | |
| *** spzala has joined #openstack-keystone | 17:03 | |
| *** lwanderley has quit IRC | 17:07 | |
| *** lwanderley has joined #openstack-keystone | 17:07 | |
| *** lwanderley has quit IRC | 17:14 | |
| openstackgerrit | Nicolas Helgeson proposed openstack/keystone master: WIP: Add project tags https://review.openstack.org/470317 | 17:19 |
| *** lwanderley has joined #openstack-keystone | 17:22 | |
| *** feefifofum has joined #openstack-keystone | 17:23 | |
| feefifofum | hi | 17:24 |
| feefifofum | is there a set of default roles that are created for domains? | 17:24 |
| *** sjain_ has quit IRC | 17:31 | |
| *** edmondsw has joined #openstack-keystone | 17:35 | |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/keystone master: Fixing multi-region support in templated v3 catalog https://review.openstack.org/482364 | 17:37 |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/keystone master: Fixing multi-region support in templated v3 catalog https://review.openstack.org/482364 | 17:38 |
| lbragstad | feefifofum: unfortunately no | 17:39 |
| lbragstad | feefifofum: that's certainly something we'd like to work towards though | 17:40 |
| *** lwanderley has quit IRC | 17:41 | |
| *** leticiawanderley has joined #openstack-keystone | 17:41 | |
| *** leticiawanderley is now known as lwanderley | 17:41 | |
| feefifofum | lbragstad: darn.. ok | 17:42 |
| *** toddnni has joined #openstack-keystone | 17:44 | |
| *** aojea has joined #openstack-keystone | 17:44 | |
| *** zzzeek_ has quit IRC | 17:46 | |
| openstackgerrit | Merged openstack/keystoneauth master: Nix EndpointData.get_versioned_data(authenticated) https://review.openstack.org/482260 | 17:47 |
| openstackgerrit | Nicolas Helgeson proposed openstack/keystone master: WIP: Add project tags https://review.openstack.org/470317 | 17:48 |
| *** zzzeek_ has joined #openstack-keystone | 17:49 | |
| *** aojea has quit IRC | 17:49 | |
| *** spilla_ has joined #openstack-keystone | 17:53 | |
| *** sjain has joined #openstack-keystone | 17:57 | |
| lbragstad | reminder that the keystone meeting is about to start in #openstack-meeting | 17:58 |
| *** sjain_ has joined #openstack-keystone | 18:00 | |
| *** sjain_ has quit IRC | 18:00 | |
| *** tobberydberg has joined #openstack-keystone | 18:04 | |
| *** ducttape_ has quit IRC | 18:12 | |
| *** ducttape_ has joined #openstack-keystone | 18:13 | |
| *** spzala has quit IRC | 18:14 | |
| *** spzala has joined #openstack-keystone | 18:16 | |
| *** ducttape_ has quit IRC | 18:17 | |
| *** sjain__ has joined #openstack-keystone | 18:18 | |
| *** sjain has quit IRC | 18:20 | |
| *** spzala has quit IRC | 18:21 | |
| *** edmondsw_ has joined #openstack-keystone | 18:23 | |
| *** ducttape_ has joined #openstack-keystone | 18:23 | |
| *** spzala has joined #openstack-keystone | 18:24 | |
| *** tesseract has quit IRC | 18:25 | |
| *** edmondsw has quit IRC | 18:25 | |
| *** spzala_ has joined #openstack-keystone | 18:26 | |
| *** spzala has quit IRC | 18:27 | |
| *** ducttape_ has quit IRC | 18:27 | |
| *** ducttape_ has joined #openstack-keystone | 18:32 | |
| *** switch_aesch has joined #openstack-keystone | 18:36 | |
| *** switch_aesch has quit IRC | 18:37 | |
| *** aojea has joined #openstack-keystone | 18:45 | |
| *** dtroyer has quit IRC | 18:47 | |
| *** dtroyer has joined #openstack-keystone | 18:49 | |
| *** nicolasbock has quit IRC | 18:50 | |
| gagehugo | brb, then can do office hours | 18:57 |
| lbragstad | #startmeeting office-hours | 19:00 |
| openstack | Meeting started Tue Jul 11 19:00:06 2017 UTC and is due to finish in 60 minutes. The chair is lbragstad. Information about MeetBot at http://wiki.debian.org/MeetBot. | 19:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 19:00 |
| openstack | The meeting name has been set to 'office_hours' | 19:00 |
| morgan | o/ | 19:00 |
| lbragstad | o/ | 19:01 |
| morgan | ok everyone, office hours! turn on the music, order pizza, party time :P | 19:01 |
| *** tobberydberg has quit IRC | 19:02 | |
| *** tobberydberg has joined #openstack-keystone | 19:02 | |
| *** aojea has quit IRC | 19:04 | |
| *** tobberydberg has quit IRC | 19:06 | |
| *** tobberydberg has joined #openstack-keystone | 19:06 | |
| *** spzala_ has quit IRC | 19:06 | |
| *** sjain__ has quit IRC | 19:07 | |
| *** tobberydberg has quit IRC | 19:10 | |
| *** mvk has quit IRC | 19:11 | |
| openstackgerrit | Matthew Edmonds proposed openstack/keystone master: fix assert_admin https://review.openstack.org/482359 | 19:13 |
| openstackgerrit | Eric Fried proposed openstack/keystoneauth master: Make Discover.version_data accept null max_version https://review.openstack.org/482250 | 19:14 |
| lbragstad | morgan: i agree! | 19:15 |
| eandersson | lbragstad, https://bugs.launchpad.net/keystone/+bug/1703666 | 19:18 |
| openstack | Launchpad bug 1703666 in OpenStack Identity (keystone) "Templated catalog does not handle multi-regions properly" [Undecided,New] | 19:18 |
| eandersson | I failed a bit at the markup :D | 19:18 |
| *** tobberydberg has joined #openstack-keystone | 19:18 | |
| eandersson | Wish you could preview before you posted a bug | 19:19 |
| morgan | go ahead and edit it :) | 19:19 |
| lbragstad | eandersson: you should be able to edit the description | 19:19 |
| eandersson | Oh lol | 19:20 |
| eandersson | Is there markup for code? | 19:21 |
| lbragstad | eandersson: unfortunately no | 19:21 |
| *** rajalokan has joined #openstack-keystone | 19:23 | |
| *** rajalokan has quit IRC | 19:23 | |
| *** ducttape_ has quit IRC | 19:23 | |
| *** ducttape_ has joined #openstack-keystone | 19:23 | |
| gagehugo | o/ | 19:25 |
| gagehugo | I was looking at https://bugs.launchpad.net/keystone/+bug/1702211 yesterday, it's an odd bug | 19:25 |
| openstack | Launchpad bug 1702211 in OpenStack Identity (keystone) "test_password_history_not_enforced_in_admin_reset failed in tempest test" [Undecided,Confirmed] | 19:25 |
| openstackgerrit | Merged openstack/keystoneauth master: Fix _run_discovery caching https://review.openstack.org/481754 | 19:27 |
| cmurphy | catching up on meeting logs | 19:29 |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/keystone master: Fixing multi-region support in templated v3 catalog https://review.openstack.org/482364 | 19:29 |
| cmurphy | samueldmq: lbragstad the sample dsta script was useful for having data to work with without having to set up a devstac | 19:29 |
| cmurphy | but i don't feel strongly about getting rid of it | 19:30 |
| cmurphy | i always devstack now | 19:30 |
| lbragstad | gagehugo: interesting - was that a timing issue? | 19:31 |
| lbragstad | cmurphy: yeah - i'm in the same boat for the most part :-/ | 19:31 |
| *** aojea has joined #openstack-keystone | 19:31 | |
| gagehugo | lbragstad, no idea. either that or maybe some weird race condition | 19:32 |
| gagehugo | from the frequency log that mriedem posted it looks like it started failing on 07/01 | 19:32 |
| openstackgerrit | Lance Bragstad proposed openstack/keystone master: WIP: Implement global role assignments https://review.openstack.org/481781 | 19:34 |
| *** aojea has quit IRC | 19:39 | |
| *** tobberydberg has quit IRC | 19:44 | |
| *** tobberydberg has joined #openstack-keystone | 19:44 | |
| *** tobberydberg has quit IRC | 19:46 | |
| *** tobberydberg has joined #openstack-keystone | 19:46 | |
| openstackgerrit | Matthew Edmonds proposed openstack/keystone master: don't validate trust in policy https://review.openstack.org/482190 | 19:47 |
| *** lwanderley has quit IRC | 19:48 | |
| lbragstad | here's a patch that closes a bug https://review.openstack.org/#/c/470425/16 | 19:51 |
| * cmurphy is home and ready to officehours | 19:54 | |
| lbragstad | \o/ | 19:54 |
| cmurphy | lbragstad: did you see my comments on that one? | 19:54 |
| cmurphy | it doesn't fix the bug | 19:54 |
| lbragstad | cmurphy: https://review.openstack.org/#/c/470425/16 ? | 19:54 |
| cmurphy | lbragstad: ya | 19:55 |
| lbragstad | cmurphy: checking | 19:55 |
| bknudson | why is it only the token header gets trimmed? seems like all headers should get the same treatment | 19:56 |
| bknudson | also, you'd expect the web server would handle fixing up the request. | 19:57 |
| cmurphy | i have no idea | 19:58 |
| *** aojea has joined #openstack-keystone | 19:58 | |
| cmurphy | i could be totally wrong, maybe edmondsw_ could test it to see if it actually solves his problem | 19:59 |
| edmondsw_ | cmurphy I'll try to do that | 20:00 |
| *** lwanderley has joined #openstack-keystone | 20:00 | |
| edmondsw_ | I've got 4 policy-related bug fixes out for review if anyone wants me to give pointers to them | 20:01 |
| lbragstad | bknudson: yeah - that's a good point | 20:03 |
| lbragstad | edmondsw_: i just started reviewing https://review.openstack.org/#/c/482142/ | 20:03 |
| edmondsw_ | lbragstad tx. The really bad one is https://review.openstack.org/482359 | 20:05 |
| *** edmondsw_ is now known as edmondsw | 20:07 | |
| *** lwanderley has quit IRC | 20:08 | |
| openstackgerrit | Matthew Edmonds proposed openstack/keystone master: remove default rule https://review.openstack.org/482164 | 20:11 |
| morgan | lbragstad: i'm going to propose a deprecation of the template catalog and (hopefully) a YAML-based one to replace that is actually tested... | 20:11 |
| morgan | lbragstad: that is based upon convos yesterday | 20:11 |
| lbragstad | morgan: it sounds like eandersson is using the templated catalog | 20:12 |
| morgan | right, hence the yaml replacement | 20:13 |
| lbragstad | morgan: so what exactly would we be deprecating? | 20:14 |
| morgan | the current templated one | 20:14 |
| morgan | there would be a new one, that is named something else | 20:14 |
| lbragstad | what format is that in? | 20:14 |
| lbragstad | er - does it have a format? | 20:14 |
| morgan | the current one is basically write out a json doc and we replace some stuff in it | 20:14 |
| morgan | it's not formatted really. | 20:14 |
| lbragstad | ah | 20:14 |
| morgan | it is terrible | 20:15 |
| lbragstad | that makes sense | 20:15 |
| bknudson | I think we still need to support replacement since swift puts the project in the URL? | 20:16 |
| morgan | that is the plan | 20:16 |
| lbragstad | morgan: since eandersson is currently relying on it - i want to make sure that coordination happens | 20:16 |
| morgan | just going to make the input enforced. | 20:16 |
| morgan | expect regions, etc all in a yaml format that would be similar to the current in-db model | 20:17 |
| *** rderose has quit IRC | 20:17 | |
| lbragstad | morgan: does eandersson on board with the deprecation of the existing templated catalog? | 20:20 |
| morgan | dunno | 20:20 |
| lbragstad | s/does/is/ | 20:20 |
| morgan | he was part of the convo esterday | 20:20 |
| morgan | it will be much better with something that is actually using the same mechanisms as the DB to render | 20:21 |
| morgan | the current templated catalog is ... frightening. | 20:21 |
| lbragstad | there is also https://github.com/openstack/keystone/blob/0731dab01a5d2da9650b67ebe8b91e825795c0ba/keystone/catalog/backends/templated.py#L244-L297 | 20:22 |
| morgan | those will mostly be the same | 20:22 |
| morgan | this is a FS based catalog | 20:22 |
| morgan | no writes allowed, CMS is there to do that job | 20:22 |
| lbragstad | that makes sense | 20:23 |
| morgan | i mean... we *could* support writes... but lets not do that silly thing | 20:23 |
| eandersson | yes - for sure we should deprecate it | 20:29 |
| eandersson | and if needed replace it with a better alternative | 20:30 |
| *** feefifofum has quit IRC | 20:30 | |
| eandersson | lbragstad, can't we just remove those overrides, as they are already implemented in the base class? | 20:30 |
| eandersson | e.g. https://github.com/openstack/keystone/blob/0731dab01a5d2da9650b67ebe8b91e825795c0ba/keystone/catalog/backends/base.py#L378 | 20:31 |
| lbragstad | eandersson: i don't see a problem with that | 20:32 |
| lbragstad | we take that approach elsewhere in keystone | 20:32 |
| morgan | you can't remove with them being abstract | 20:32 |
| morgan | you must redefine abstract methods on the subclass | 20:32 |
| edmondsw | lbragstad cmurphy was right, https://review.openstack.org/#/c/470425/16 doesn't fix the bug | 20:32 |
| morgan | edmondsw: that code looked suspect | 20:32 |
| openstackgerrit | Eric Fried proposed openstack/keystoneauth master: Update docs and add a release note https://review.openstack.org/477566 | 20:33 |
| eandersson | lbragstad, Yea I agree, but not all of them are implemented at the moment | 20:33 |
| eandersson | (sorry defined, obviously not implemented) | 20:33 |
| morgan | basically... don't send the /r /n etc in headers... you will be sad | 20:33 |
| cmurphy | ya...found a few sources saying just don't do that | 20:33 |
| morgan | cmurphy: ++ | 20:33 |
| edmondsw | morgan yeah, but when a customer does that... | 20:33 |
| morgan | edmondsw: you point them at the docs in curl saying "yeah don't" | 20:34 |
| lbragstad | edmondsw: removed my +1 accordingly | 20:34 |
| edmondsw | morgan it was odd that neutron worked fine and keystone didn't | 20:34 |
| morgan | edmondsw: apache makes a difference | 20:34 |
| gagehugo | is it a bug then? | 20:34 |
| morgan | unlikely | 20:34 |
| *** thorst has quit IRC | 20:35 | |
| edmondsw | morgan :) sure but it was a really hard to pin down issue... wasn't obvious they had the /r in the header unless you're an awk god | 20:35 |
| morgan | heh | 20:35 |
| *** ducttape_ has quit IRC | 20:36 | |
| edmondsw | morgan I don't think it's an apache thing, actually... the neutron curl command was targeted at an apache reverse proxy | 20:37 |
| morgan | ah | 20:37 |
| lbragstad | eandersson: are you using the templated catalog backend? | 20:37 |
| lbragstad | s/are/aren't/ | 20:37 |
| cmurphy | edmondsw: morgan my test reproduced it with just uwsgi | 20:37 |
| morgan | ah | 20:38 |
| morgan | *shrug* | 20:38 |
| edmondsw | cmurphy yeah, I'm not sure why the author of this test isn't just reproducing the problem and then testing their fix against it until it actually fixes it | 20:38 |
| edmondsw | s/test/fix/ | 20:38 |
| eandersson | I am yes | 20:38 |
| cmurphy | edmondsw: yeah :/ | 20:38 |
| lbragstad | eandersson: i'm just wondering why you wouldn't be opposed to deprecating and removing it then | 20:39 |
| lbragstad | unless i'm missing something obvious | 20:39 |
| eandersson | I'll rather move over to sql | 20:40 |
| eandersson | and we are not going to pike anytime soon | 20:40 |
| eandersson | So we would have to stick with the semi-functional templated implementation for too long | 20:40 |
| gagehugo | morgan cmurphy edmondsw: I'll ping kaerie about it | 20:41 |
| edmondsw | gagehugo tx | 20:41 |
| eandersson | Moving to a yaml based variant would be an alternative, but we couldn't really go there unless we backported that to Mitaka, or maybe Newton/Ocata. | 20:41 |
| gagehugo | I can mess around with it too after I take a look at this random failing tempest test | 20:42 |
| morgan | eandersson: right, but you might be able to backport the code yourself. | 20:42 |
| morgan | for your install | 20:42 |
| morgan | until you move to pike | 20:42 |
| eandersson | Yep - that is for sure an alternative | 20:43 |
| *** ducttape_ has joined #openstack-keystone | 20:46 | |
| eandersson | It just makes more sense for us to go move to the production ready alternative. | 20:46 |
| lbragstad | eandersson: morgan so - let's recap the options for the templated catalog backend | 20:47 |
| lbragstad | 1.) formally deprecate it for removal | 20:47 |
| lbragstad | 2.) support it in a well known format (like yaml) | 20:47 |
| morgan | yep | 20:49 |
| *** zzzeek_ has quit IRC | 20:49 | |
| lbragstad | is that it? | 20:49 |
| lbragstad | i guess there would be a 3rd option | 20:50 |
| lbragstad | 3.) perform option 1, deprecating all existing templated catalog stuff and start fresh with option 2 introducing a new backend for YAML officially | 20:51 |
| lbragstad | so - maybe option 3 is actually option 2 just spelled out | 20:51 |
| eandersson | Nr 3 is what I would recommend as well | 20:52 |
| *** spilla_ has quit IRC | 20:56 | |
| *** tobberydberg has quit IRC | 20:57 | |
| morgan | option 3 was what i was planning | 20:57 |
| lbragstad | i certainly wouldn't be opposed to #3 if someone is willing to do the work | 20:57 |
| *** tobberydberg has joined #openstack-keystone | 20:57 | |
| morgan | it isn't a ton of work | 20:58 |
| morgan | i have to grab my laptop, ... anyway | 20:58 |
| lbragstad | it sounds like we're in agreement that deprecation of the existing templated catalog is in order | 20:59 |
| *** tobberydberg has quit IRC | 21:01 | |
| eandersson | having fun writing tests for the multi-region patch | 21:04 |
| *** catintheroof has quit IRC | 21:04 | |
| *** jmlowe has joined #openstack-keystone | 21:05 | |
| eandersson | not sure how to handle attributes | 21:05 |
| eandersson | > v3_catalog[service_type][attr] = value | 21:06 |
| eandersson | guessing the endpoint should have the id? | 21:07 |
| lbragstad | it should be consistent with what the sql implementation does with the exception of write operations | 21:08 |
| lbragstad | edmondsw: i'm missing the bit here at line 202 - https://review.openstack.org/#/c/482359/2/keystone/common/authorization.py | 21:08 |
| lbragstad | edmondsw: the patch ends up doing the same thing as before, right? | 21:09 |
| lbragstad | the action ends up being `identity:<operation>` right? | 21:09 |
| edmondsw | lbragstad if you call check_protection, yes... but if you call assert_admin, no | 21:10 |
| edmondsw | lbragstad the places that use check_protection expect that to be added... the places that call assert_admin don't | 21:10 |
| *** zzzeek_ has joined #openstack-keystone | 21:11 | |
| lbragstad | oh - line 130 | 21:11 |
| edmondsw | yep, 129 | 21:11 |
| edmondsw | oh, right, 130 on the new file | 21:11 |
| lbragstad | edmondsw: how'd you stumble across this? | 21:15 |
| lbragstad | edmondsw: testing a custom policy? | 21:15 |
| edmondsw | lbragstad digging into why tests failed for https://review.openstack.org/#/c/482164/ | 21:16 |
| *** raildo has quit IRC | 21:16 | |
| edmondsw | which I had proposed after digging into and proposing https://review.openstack.org/482142 | 21:17 |
| edmondsw | which was a result of reviewing our customized policy changes for pike and noticing that didn't look right | 21:17 |
| edmondsw | https://review.openstack.org/482190 also came out of that as well | 21:18 |
| edmondsw | lbragstad so it was quite a chain of events | 21:18 |
| edmondsw | I'm quite happy to be fixing a 4 year old defect with that last one... | 21:18 |
| lbragstad | so - correct me if i'm wrong | 21:19 |
| lbragstad | but https://review.openstack.org/#/c/482359/2 will be tested by default once https://review.openstack.org/#/c/482164/2 merges? | 21:20 |
| edmondsw | lbragstad yes... once there's no default rule, that can't hide issues like this | 21:22 |
| lbragstad | because identity:admin_required didn't exist and was getting caught by the default - which ended up having the same result | 21:22 |
| edmondsw | lbragstad exactly | 21:23 |
| lbragstad | hmmm tricky | 21:23 |
| *** pcaruana has quit IRC | 21:23 | |
| edmondsw | yep | 21:23 |
| edmondsw | I think we may also need to merge https://review.openstack.org/482142 before the default rule is removed | 21:23 |
| edmondsw | s/may // | 21:23 |
| edmondsw | but I can't base https://review.openstack.org/482164 on multiple changes | 21:24 |
| lbragstad | edmondsw: yeah - that one looks good | 21:24 |
| eandersson | So there is a second bug with templated - the endpoint id is used as the id for the service | 21:24 |
| eandersson | which is an expected bug | 21:24 |
| lbragstad | edmondsw: since a policy is changing, a release note would make sense i think, would you agree? | 21:24 |
| edmondsw | lbragstad I'm adding a rel note | 21:24 |
| edmondsw | lbragstad do you think it should be a security or fixes note? | 21:25 |
| lbragstad | well - the default is still in place and that's true for all previous releases, right? | 21:25 |
| openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Add paragraph clarifying major and micro versions https://review.openstack.org/482710 | 21:26 |
| lbragstad | edmondsw: can you walk me through the case where it *should* be considered a security issue? | 21:26 |
| *** aojea has quit IRC | 21:26 | |
| edmondsw | lbragston see breton's comments in the bug | 21:27 |
| edmondsw | lbragstad ^ | 21:27 |
| edmondsw | not sure what my fingers were doing there :) | 21:27 |
| lbragstad | oh - so it would be considered a security issue if the deployer relaxed the default | 21:28 |
| lbragstad | but didn't realize they were also relaxing get_identity_providers | 21:29 |
| lbragstad | hmm | 21:29 |
| *** aojea has joined #openstack-keystone | 21:33 | |
| *** aojea has quit IRC | 21:33 | |
| edmondsw | lbragstad yeah... I'm inclined to put it in the security section of the rel notes, but not backport anything | 21:34 |
| edmondsw | lbragstad I should probably also add a rel note to the change removing the default rule | 21:35 |
| lbragstad | so - policy in code has only been effective for pike | 21:35 |
| *** aojea has joined #openstack-keystone | 21:35 | |
| edmondsw | lbragstad yes, but the typo for identity:get_identity_providers goes back to the default policy.json file we shipped in past releases | 21:35 |
| lbragstad | so - if someone wanted to "backport" the fix it would consist of correcting their policy operation | 21:35 |
| *** thorst has joined #openstack-keystone | 21:35 | |
| morgan | lbragstad: deprecation patch about to be posted (catalog) | 21:36 |
| edmondsw | lbragstad right, the backport would be to update the default policy.json file and/or to change the code to look for the typo version... neither of which I like | 21:36 |
| morgan | i should have a yaml-loading catalog in a couple hours as well. | 21:36 |
| lbragstad | morgan: awesome - thank you | 21:36 |
| lbragstad | cc eandersson ^ | 21:36 |
| edmondsw | taht's really an or, not an and/or | 21:36 |
| lbragstad | edmondsw: so the options are | 21:36 |
| lbragstad | 1.) backport the default that corrects the type | 21:37 |
| lbragstad | typo* | 21:37 |
| lbragstad | 2.) correct the code to look for get_identity_providers (which breaks conventions) | 21:37 |
| lbragstad | i agree in that option 2 seems like the wrong approach | 21:37 |
| lbragstad | but what's wrong with option 1? | 21:37 |
| eandersson | You'll probably have that done before I get working unit tests working for this lol | 21:38 |
| edmondsw | lbragstad I don't know that #1 really helps anyone | 21:38 |
| edmondsw | lbragstad if you've already customized policy, you're not going to take a new default policy.json file and apply it | 21:38 |
| lbragstad | is there a negative side-effect outside of that? | 21:38 |
| edmondsw | if you're not customizing policy, then things are already working ok | 21:38 |
| lbragstad | what about consuming a new release note proposed to stable/ocata and stable/newton? | 21:39 |
| edmondsw | lbragstad probably not... | 21:39 |
| edmondsw | lbragstad we could certainly propose patches to the stable releases that a) updates the policy.json to fix the typo and b) adds a rel note warning about the issue | 21:40 |
| lbragstad | edmondsw: that would at least flag things to deployers that read release notes | 21:40 |
| openstackgerrit | Morgan Fainberg proposed openstack/keystone master: Deprecate the templated catalog https://review.openstack.org/482714 | 21:40 |
| edmondsw | lbragstad that read release notes long after the release :) | 21:40 |
| lbragstad | how they apply that change is obviously up to them - but at least they'd know | 21:40 |
| morgan | lbragstad: ^ very simple patch. | 21:40 |
| lbragstad | edmondsw: true - it's more or less only for following procedure ;) | 21:41 |
| openstackgerrit | Matthew Edmonds proposed openstack/keystone master: fix identity:get_identity_providers typo https://review.openstack.org/482142 | 21:41 |
| edmondsw | lbragstad I'm fine with it | 21:41 |
| *** ducttape_ has quit IRC | 21:41 | |
| lbragstad | edmondsw: ok - i'll update the bug report then | 21:41 |
| edmondsw | lbragstad ^ that adds a rel note | 21:41 |
| *** thorst has quit IRC | 21:42 | |
| eandersson | Writing tests for this is going to require a lot more fixes than what I did :p | 21:46 |
| eandersson | since this is being back-ported from the v2 catalog, IDs are not at all implemented properly | 21:47 |
| openstackgerrit | Matthew Edmonds proposed openstack/keystone master: remove default rule https://review.openstack.org/482164 | 21:48 |
| lbragstad | edmondsw: looks good - one minor nit inline | 21:49 |
| lbragstad | s/inline/in the release note/ | 21:50 |
| eandersson | do we even care about ids for this? they are not in the example config https://github.com/openstack/keystone/blob/master/etc/default_catalog.templates | 21:51 |
| lbragstad | eandersson: i would imagine that would get rewritten to yaml - based on morgan's work | 21:51 |
| edmondsw | lbragstad fixed | 21:52 |
| eandersson | sure - but do we want to backport the yaml work as well? | 21:52 |
| openstackgerrit | Matthew Edmonds proposed openstack/keystone master: fix identity:get_identity_providers typo https://review.openstack.org/482142 | 21:52 |
| eandersson | or do we want to fix the current implementation pre-pike? | 21:52 |
| openstackgerrit | Merged openstack/keystoneauth master: Make Discover.version_data accept null max_version https://review.openstack.org/482250 | 21:52 |
| lbragstad | eandersson: we won't be able to fix stable/ocata without landing something in pike first | 21:57 |
| *** jmlowe has quit IRC | 21:57 | |
| lbragstad | but i doubt we'd backport fixes for the templated catalog to ocata anyway | 21:57 |
| eandersson | Yep - so do we want to just skip my patch and go straight to yaml? | 21:57 |
| *** ducttape_ has joined #openstack-keystone | 21:58 | |
| lbragstad | eandersson: yeah - that would seem reasonable | 21:58 |
| eandersson | It's just weird to have a feature so broken :D | 21:58 |
| lbragstad | eandersson: yeah - the templated stuff is a mess | 21:58 |
| eandersson | Could we maybe update our documentation for newton/ocata? | 21:58 |
| lbragstad | eandersson: saying what exactly? | 21:59 |
| eandersson | Don't try to use with multiple regions! | 21:59 |
| openstackgerrit | Gage Hugo proposed openstack/keystone master: WIP - Trims whitespace from request headers https://review.openstack.org/470425 | 21:59 |
| eandersson | or not intended for production :D | 21:59 |
| eandersson | basically just slap a warning lable on it | 21:59 |
| *** jmlowe has joined #openstack-keystone | 22:01 | |
| openstackgerrit | Eric Fried proposed openstack/keystoneauth master: Update docs and add a release note https://review.openstack.org/477566 | 22:03 |
| openstackgerrit | Merged openstack/keystoneauth master: Expand some discover.py docstrings https://review.openstack.org/482207 | 22:03 |
| lbragstad | eandersson: we document a few of the warts with it already https://docs.openstack.org/keystone/latest/configuration.html#service-catalog | 22:04 |
| lbragstad | #endmeeting | 22:04 |
| openstack | Meeting ended Tue Jul 11 22:04:40 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 22:04 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/office_hours/2017/office_hours.2017-07-11-19.00.html | 22:04 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/office_hours/2017/office_hours.2017-07-11-19.00.txt | 22:04 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/office_hours/2017/office_hours.2017-07-11-19.00.log.html | 22:04 |
| openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Add paragraph clarifying major and micro versions https://review.openstack.org/482710 | 22:05 |
| lbragstad | thanks for coming to office hours - i'll be sending a recap tonight | 22:08 |
| lbragstad | stepping away for a bit | 22:08 |
| *** jmlowe has quit IRC | 22:09 | |
| *** aojea has quit IRC | 22:12 | |
| *** dave-mccowan has quit IRC | 22:14 | |
| openstackgerrit | Colleen Murphy proposed openstack/keystoneauth master: Add paragraph clarifying major and micro versions https://review.openstack.org/482710 | 22:16 |
| *** markvoelker has quit IRC | 22:16 | |
| *** markvoelker has joined #openstack-keystone | 22:17 | |
| openstackgerrit | Merged openstack/keystoneauth master: normalize_version_number([1]) => (1, 0) and docs https://review.openstack.org/481309 | 22:20 |
| *** edmondsw has quit IRC | 22:26 | |
| *** bknudson has quit IRC | 22:28 | |
| morgan | lbragstad, eandersson: see this as an example yaml config | 22:33 |
| morgan | https://www.irccloud.com/pastebin/oqFD3CG0/ | 22:33 |
| morgan | mordred: ^ cc | 22:33 |
| morgan | the in-memory data structure looks like: | 22:34 |
| morgan | https://www.irccloud.com/pastebin/eu1cyeK2/ | 22:34 |
| lbragstad | aha - nice | 22:34 |
| morgan | hmm. something is wonky there | 22:37 |
| morgan | anyway, something like that | 22:37 |
| mordred | morgan: neat! | 22:40 |
| morgan | the yaml is a bit off, working it out | 22:40 |
| mordred | morgan, cmurphy: btw- efried found another thing that we need to fix beore release | 22:40 |
| morgan | but it's close | 22:40 |
| mordred | with ksa | 22:40 |
| * mordred is working on patch now | 22:41 | |
| mordred | it's almost like reviewing the docs pointed out places where there was a problem :) | 22:41 |
| morgan | oh hah. re-serializing it did the back-refs weirdly | 22:42 |
| morgan | that explains it | 22:42 |
| *** brad[] has quit IRC | 22:46 | |
| *** brad[] has joined #openstack-keystone | 22:49 | |
| morgan | mordred: ok here we go | 22:50 |
| morgan | https://www.irccloud.com/pastebin/CHcq8KLI/ | 22:50 |
| morgan | That is the expected input yaml | 22:50 |
| morgan | or so | 22:50 |
| morgan | not sure if i like that or if I would prefer to not do the back-ref stuff | 22:51 |
| mordred | morgan: wow - fun | 22:52 |
| mordred | morgan: I mean- the back-refs are 0_o - otoh - they totally work | 22:52 |
| morgan | yeah | 22:53 |
| morgan | if we support the backrefs a non-backref version would be really explicit | 22:53 |
| morgan | if we just verify region in "regions" | 22:53 |
| morgan | for endpoints, its simpler to read | 22:53 |
| morgan | thoughts? | 22:55 |
| *** ducttape_ has quit IRC | 22:55 | |
| * mordred staring | 22:55 | |
| morgan | if we do something without backrefs (meaning no full data struct) it would be like: | 22:56 |
| mordred | morgan: I think the backrefs are easy enough to copy-pasta for folks - I think most folks dont' actually understand yaml backreferences - but it's easy enough to read for the endpoint section | 22:57 |
| morgan | https://www.irccloud.com/pastebin/C5EpG4Tz/ | 22:57 |
| morgan | ^ that is no back-refs. | 22:57 |
| mordred | morgan: that's also pretty readable - assuming region is validated | 22:57 |
| morgan | and a version of the original one without backrefs, (aka expanded) | 22:57 |
| morgan | https://www.irccloud.com/pastebin/P7oLSZoi/ | 22:58 |
| morgan | if we support backrefs, this last one is the non-backref expanded version | 22:58 |
| morgan | we could just validate region/service | 22:58 |
| morgan | in the no-backref-supported version | 22:58 |
| * morgan leans towards not supporting backrefs | 22:59 | |
| mordred | yah. I think it's less for people to mess up | 23:00 |
| morgan | ok. i'll go with that but i'm going to supply examples that are more json-y | 23:00 |
| openstackgerrit | Gage Hugo proposed openstack/keystone master: Trim invalid characters from token authentication https://review.openstack.org/470425 | 23:31 |
| openstackgerrit | Gage Hugo proposed openstack/keystone master: Trim invalid characters from token authentication https://review.openstack.org/470425 | 23:33 |
| *** d0ugal has joined #openstack-keystone | 23:36 | |
| *** thorst has joined #openstack-keystone | 23:47 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!