| *** randomhack has joined #openstack-barbican | 00:20 | |
| *** randomhack has quit IRC | 00:27 | |
| tdasilva | is Jeremy Liu in the channel? | 00:32 |
|---|---|---|
| *** liujiong has joined #openstack-barbican | 01:23 | |
| liujiong | tdasilva: hi | 01:30 |
| tdasilva | hello | 01:31 |
| liujiong | :) | 01:31 |
| tdasilva | liujiong: I submitted the bug as you asked, but when trying to fix the tests I started questioning if my patch is correct | 01:32 |
| tdasilva | liujiong: it does seem like the correct default version should be v1 and not just '1' | 01:32 |
| liujiong | tdasilva: but 'v1' works with openstack command, right? | 01:33 |
| tdasilva | but I also tried making this change instead: https://github.com/openstack/python-barbicanclient/blob/master/barbicanclient/osc_plugin.py#L18 | 01:33 |
| tdasilva | liujiong: master doesn't work | 01:34 |
| tdasilva | if I leave client.py as v1 and change osc_plugin.py to v1, then I get a different error | 01:34 |
| liujiong | tdasilva: I don't think we can change DEFAULT_API_VERSION of osc_plugin, that comes first | 01:35 |
| tdasilva | http://paste.openstack.org/show/615916/ | 01:35 |
| tdasilva | mmm..ok | 01:35 |
| tdasilva | so do you think changing client.py to '1' is correct? | 01:35 |
| liujiong | I suppose so | 01:36 |
| tdasilva | it seems like it will force me to make changes to functional tests that have been there for a while | 01:36 |
| liujiong | Yes, functionaltests need changed for the version change | 01:37 |
| liujiong | tdasilva: can you change functionaltest as well? we need your patch merged into Pike | 01:39 |
| tdasilva | liujiong: I'm not sure I know where to make the changes, if you would like, feel free to take over patch | 01:39 |
| tdasilva | i'm looking now, if you have pointers I can keep looking | 01:40 |
| tdasilva | first time I used barbican client was yesterday ;) | 01:40 |
| *** randomhack has joined #openstack-barbican | 01:41 | |
| liujiong | tdasilva: let me check | 01:43 |
| *** ssmith has quit IRC | 01:44 | |
| liujiong | tdasilva: you can refer to my refactor patch to see functionaltest changes | 01:45 |
| liujiong | tdasilva: functionaltests/client/base.py and functionaltests/client/test_client_connectivity.py | 01:46 |
| *** randomhack has quit IRC | 01:46 | |
| liujiong | tdasilva: changes should be in these two files ^ | 01:46 |
| tdasilva | looking | 01:46 |
| tdasilva | liujiong: so for example, i'm guessing you are saying one of the changes should be here? https://github.com/openstack/python-barbicanclient/blob/master/functionaltests/client/base.py#L48 | 01:52 |
| tdasilva | but versions is not required as it should take the default value | 01:52 |
| liujiong | tdasilvaļ¼Ahh, yes, i need see the logs of your first patchset | 01:54 |
| liujiong | tdasilva: your patchset 1 should do the right fix | 01:57 |
| tdasilva | liujiong: the only different to the patchset1 and patchset2 are the unit tests, no? | 01:58 |
| liujiong | tdasilva: can you fetch the patchset1 logs? | 02:02 |
| tdasilva | liujiong: yeah, which logs? py27 or functests? | 02:04 |
| liujiong | tdasilva: both | 02:07 |
| tdasilva | py27 http://logs.openstack.org/26/484926/1/check/gate-python-barbicanclient-python27-ubuntu-xenial/dcb5827/ | 02:08 |
| tdasilva | functest - http://logs.openstack.org/26/484926/1/check/gate-python-barbicanclient-devstack-dsvm-ubuntu-xenial/9eeebf5/ | 02:08 |
| liujiong | tdasilva: OK, let me check | 02:09 |
| tdasilva | liujiong: what I don't understand is that before your refactor, the default version was already 'v1': https://review.openstack.org/#/c/403604/27/barbicanclient/client.py@a33 | 02:09 |
| tdasilva | that's why I thought my change to make it just '1' was incorrect, no? | 02:09 |
| liujiong | tdasilva: yes, i see it. | 02:11 |
| liujiong | tdasilva: we add 'version' to the endpoint directly | 02:13 |
| liujiong | tdasilva: now, we need confirm why the refactor change breaks openstack command | 02:16 |
| tdasilva | liujiong: this change also works for me | 02:16 |
| tdasilva | http://paste.openstack.org/show/615951/ | 02:16 |
| liujiong | tdasilva: ahh, that's confusing | 02:17 |
| tdasilva | well..not really...before refactor you didn't use the version to instantiate that Client class | 02:18 |
| tdasilva | and I think you supported either 1 or v1 | 02:18 |
| tdasilva | but now you created _SUPPORTED_API_VERSION_MAP and you are using that to instantiate the client class | 02:18 |
| liujiong | tdasilva: oh, we set different values for the default version from barbicanclient side and osc_plugin side | 02:21 |
| tdasilva | right | 02:21 |
| liujiong | tdasilva: so, we need support both of them | 02:21 |
| tdasilva | exactly | 02:22 |
| liujiong | tdasilva: just make that change | 02:22 |
| tdasilva | ok | 02:22 |
| liujiong | tdasilva: what do you think that we change this to 'v1': https://github.com/openstack/python-barbicanclient/blob/master/barbicanclient/osc_plugin.py#L18 | 02:23 |
| tdasilva | that doesnt work: you get this error: http://paste.openstack.org/show/615916/ | 02:24 |
| openstackgerrit | Thiago da Silva proposed openstack/python-barbicanclient master: fix default version https://review.openstack.org/484926 | 02:27 |
| liujiong | tdasilva: OK, to fix this, we need the 'version' args be 'v1': https://github.com/openstack/python-barbicanclient/blob/master/barbicanclient/client.py#L38 | 02:30 |
| liujiong | tdasilva: the rest is just how we handle the version map to support osc_plugin | 02:31 |
| tdasilva | liujiong: maybe we could also just revert this change: https://review.openstack.org/#/c/403604/27/barbicanclient/osc_plugin.py | 02:35 |
| tdasilva | that way the osc_plugin will just use the default version defined in client.py | 02:36 |
| tdasilva | as it used to do before refactor | 02:36 |
| tdasilva | liujiong: current patchset 3 passed all tests, let me know what you think | 02:46 |
| tdasilva | I need to drop off now | 02:46 |
| liujiong | tdasilva: I prefer we revert the change in osc_plugin.py | 02:50 |
| openstackgerrit | Thiago da Silva proposed openstack/python-barbicanclient master: fix default version https://review.openstack.org/484926 | 02:53 |
| liujiong | tdasilva: thank you for making the fix! | 02:57 |
| openstackgerrit | Kiran Totad proposed openstack/barbican master: Stop using deprecated 'message' attribute in Exception https://review.openstack.org/485420 | 03:30 |
| *** chlong_ has joined #openstack-barbican | 04:20 | |
| *** dave-mccowan has quit IRC | 04:23 | |
| *** liujiong has quit IRC | 04:33 | |
| openstackgerrit | Kiran Totad proposed openstack/barbican master: Stop using deprecated 'message' attribute in Exception https://review.openstack.org/485420 | 04:36 |
| *** randomhack has joined #openstack-barbican | 05:23 | |
| *** chlong_ has quit IRC | 05:26 | |
| *** randomhack has quit IRC | 05:28 | |
| *** liujiong has joined #openstack-barbican | 06:00 | |
| *** andreas_s has joined #openstack-barbican | 06:15 | |
| *** mathiasb has quit IRC | 08:41 | |
| *** mathiasb has joined #openstack-barbican | 08:42 | |
| *** salmankhan has joined #openstack-barbican | 08:59 | |
| *** randomhack has joined #openstack-barbican | 09:24 | |
| *** pbourke has joined #openstack-barbican | 09:25 | |
| pbourke | Hi, was wondering can anyone tell me, how important is it to use a byte array when storing opaque secret types | 09:26 |
| pbourke | as both byte arrays and strings seem to be accepted | 09:26 |
| *** randomhack has quit IRC | 09:28 | |
| *** salmankhan has quit IRC | 09:32 | |
| *** salmankhan has joined #openstack-barbican | 09:32 | |
| *** salmankhan has quit IRC | 10:02 | |
| *** liujiong has quit IRC | 10:13 | |
| *** salmankhan has joined #openstack-barbican | 10:24 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/castellan master: Use Stevedore for better extensions https://review.openstack.org/484015 | 11:16 |
| *** dave-mccowan has joined #openstack-barbican | 11:21 | |
| openstackgerrit | Kiran Totad proposed openstack/barbican master: Stop using deprecated 'message' attribute in Exception https://review.openstack.org/485420 | 11:40 |
| *** raildo has joined #openstack-barbican | 11:56 | |
| *** chlong_ has joined #openstack-barbican | 12:16 | |
| *** randomhack has joined #openstack-barbican | 12:23 | |
| *** randomhack has quit IRC | 12:28 | |
| *** catintheroof has joined #openstack-barbican | 12:36 | |
| *** pbourke has quit IRC | 12:58 | |
| *** pbourke has joined #openstack-barbican | 13:00 | |
| *** alee_ has quit IRC | 13:22 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/castellan master: Use Stevedore for better extensions https://review.openstack.org/484015 | 13:36 |
| openstackgerrit | Kaitlin Farr proposed openstack/barbican-tempest-plugin master: Add ephemeral disk encryption scenario test https://review.openstack.org/455459 | 13:38 |
| *** chlong_ has quit IRC | 13:40 | |
| *** ssmith has joined #openstack-barbican | 13:44 | |
| *** cpusmith has joined #openstack-barbican | 13:45 | |
| *** ssmith has quit IRC | 13:49 | |
| *** cpusmith_ has joined #openstack-barbican | 14:00 | |
| *** cpusmith has quit IRC | 14:04 | |
| *** alee has joined #openstack-barbican | 14:11 | |
| *** cpusmith_ has quit IRC | 14:15 | |
| *** ssmith has joined #openstack-barbican | 14:15 | |
| *** l4bytt has joined #openstack-barbican | 14:17 | |
| *** cpusmith has joined #openstack-barbican | 14:21 | |
| *** ssmith has quit IRC | 14:25 | |
| l4bytt | Hi there. Is the github documentation, especially for barbican, up to date? I am currenty following this guide and the first command fails -> https://github.com/cloudkeep/barbican/wiki/Barbican-Quick-Start-Guide#generate-a-secret | 14:41 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/castellan master: Use Stevedore for better extensions https://review.openstack.org/484015 | 14:49 |
| *** l4bytt has quit IRC | 14:52 | |
| openstackgerrit | Merged openstack/barbican-tempest-plugin master: Add ephemeral disk encryption scenario test https://review.openstack.org/455459 | 14:57 |
| *** chlong_ has joined #openstack-barbican | 14:58 | |
| *** andreas_s has quit IRC | 15:39 | |
| *** chlong_ has quit IRC | 16:19 | |
| *** aspiers has joined #openstack-barbican | 16:44 | |
| aspiers | hi all, we've just noticed an issue where barbican won't stop cleanly with more than one asynchronous_worker configured - is that a known issue? | 16:45 |
| aspiers | 2017-07-20 16:26:31.023 17964 INFO barbican.queue.server [-] Halting the TaskServer | 16:47 |
| aspiers | 2017-07-20 16:26:31.031 17950 INFO oslo_service.service [-] Caught SIGTERM, stopping children | 16:47 |
| aspiers | 2017-07-20 16:26:31.032 17950 INFO barbican.queue.server [-] Halting the TaskServer | 16:47 |
| aspiers | 2017-07-20 16:27:01.033 17950 WARNING oslo_messaging.server [-] Possible hang: stop is waiting for start to complete | 16:47 |
| aspiers | if I change asynchronous_workers from 2 to 1 then the problem vanishes | 16:47 |
| aspiers | this is true whether queue.enable is True or False | 16:48 |
| aspiers | I can report a bug if that makes sense | 16:55 |
| aspiers | ahah, I found the same bug in sahara, and porting the fix works :) | 17:19 |
| *** kfarr has joined #openstack-barbican | 17:23 | |
| kfarr | aspiers I don't think that's a known issue, please open a bug report | 17:24 |
| aspiers | kfarr: already doing so :) I'll submit to gerrit too | 17:24 |
| kfarr | l4bytt that documentation you're referencing is pretty old | 17:26 |
| kfarr | thanks aspiers ! | 17:26 |
| kfarr | pbourke the important thing is if the object you get when you retrieve it is the same as the original object | 17:28 |
| kfarr | and when you say byte array, do you mean bytestring? | 17:29 |
| *** chlong_ has joined #openstack-barbican | 17:33 | |
| openstackgerrit | Adam Spiers proposed openstack/barbican master: Start task RPC service before waiting https://review.openstack.org/485755 | 17:44 |
| aspiers | kfarr: there you go ... but really I have not much clue how this stuff is supposed to work, so it definitely needs careful sanity checking ;-) | 17:45 |
| aspiers | it fixed the issue for me on 3.0.1 at least | 17:46 |
| aspiers | but I have no idea why the issue would appear with 2 workers and not with 1 | 17:46 |
| kfarr | aspiers unfortunately the folks who did most of the work for the workers functionality aren't working on OpenStack anymore :( | 17:46 |
| kfarr | aspiers so your guess is as good as mine | 17:46 |
| aspiers | hah | 17:46 |
| kfarr | but will be happy to look over it | 17:47 |
| aspiers | thanks! | 17:47 |
| kfarr | glad you were able to find a solution | 17:47 |
| aspiers | I find it particularly weird that changing the queue size has an effect regardless of whether queue.enabled is True or False | 17:47 |
| * aspiers shrugs | 17:47 | |
| *** diazjf has joined #openstack-barbican | 17:48 | |
| *** diazjf has quit IRC | 17:50 | |
| openstackgerrit | Merged openstack/castellan master: Replace LOG.warn with LOG.warning https://review.openstack.org/485073 | 17:57 |
| openstackgerrit | Merged openstack/python-barbicanclient master: fix default version https://review.openstack.org/484926 | 17:57 |
| openstackgerrit | Merged openstack/python-barbicanclient master: Update doc references https://review.openstack.org/484238 | 17:57 |
| openstackgerrit | Merged openstack/python-barbicanclient master: Switch from oslosphinx to openstackdocstheme https://review.openstack.org/479123 | 17:57 |
| *** salmankhan has quit IRC | 18:25 | |
| alee | dave-mccowan, when are the barbican sessions being held at the ptg? | 19:29 |
| alee | kfarr, are you ptg-ing? | 19:30 |
| kfarr | alee yes! | 19:31 |
| kfarr | not sure what the schedule is though | 19:31 |
| kfarr | I'll be there m-f | 19:31 |
| alee | kfarr, cool - I'm not confirmed yet - but there is a good chance I will be too | 19:31 |
| kfarr | alee awesome :) | 19:32 |
| alee | :) | 19:32 |
| *** cpusmith has quit IRC | 19:39 | |
| *** cpusmith_ has joined #openstack-barbican | 19:39 | |
| *** cpusmith has joined #openstack-barbican | 19:40 | |
| *** cpusmith_ has quit IRC | 19:44 | |
| *** openstack has joined #openstack-barbican | 19:58 | |
| *** raildo has quit IRC | 20:07 | |
| *** openstackgerrit has quit IRC | 20:17 | |
| dave-mccowan | alee kfarr i'll be there Monday through Thursday | 20:18 |
| dave-mccowan | security group has a room monday and tuesday, barbican has a room wednesday and thursday | 20:19 |
| dave-mccowan | here's the current schedule https://docs.google.com/spreadsheets/d/1xmOdT6uZ5XqViActr5sBOaz_mEgjKSCY7NEWcAEcT-A/pubhtml?gid=397241312&single=true | 20:20 |
| dave-mccowan | alee kfarr i started an etherpad: https://etherpad.openstack.org/p/barbican-ptg-queens | 20:29 |
| *** cpusmith has quit IRC | 20:47 | |
| alee | dave-mccowan, cool - thaks | 21:04 |
| *** catintheroof has quit IRC | 21:10 | |
| *** chlong_ has quit IRC | 21:16 | |
| *** randomhack has joined #openstack-barbican | 21:23 | |
| *** randomhack has quit IRC | 21:27 | |
| *** alee has quit IRC | 21:31 | |
| *** dave-mccowan has quit IRC | 21:31 | |
| *** openstackgerrit has joined #openstack-barbican | 21:35 | |
| openstackgerrit | Merged openstack/castellan master: Improve docs around configuring Castellan https://review.openstack.org/483461 | 21:35 |
| *** ssmith has joined #openstack-barbican | 21:41 | |
| *** dave-mccowan has joined #openstack-barbican | 22:06 | |
| openstackgerrit | Merged openstack/barbican master: Add PBR wsgi_scripts entrypoint for barbican api https://review.openstack.org/476995 | 22:06 |
| openstackgerrit | Merged openstack/barbican master: Update internal doc reference according to doc migration result https://review.openstack.org/484240 | 22:07 |
| *** kfarr has quit IRC | 22:10 | |
| *** dave-mccowan has quit IRC | 22:24 | |
| *** alee has joined #openstack-barbican | 22:52 | |
| openstackgerrit | Merged openstack/barbican master: Set entry point for dogtag config correctly https://review.openstack.org/483853 | 23:17 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!