| harlowja | nikhil_k hopefully answered your quetions on https://review.openstack.org/#/c/85211/ :) | 00:00 |
|---|---|---|
| *** TravT has joined #openstack-glance | 00:03 | |
| *** dkingshott has quit IRC | 00:04 | |
| *** annashen has joined #openstack-glance | 00:07 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 00:07 | |
| *** jasondotstar has joined #openstack-glance | 00:16 | |
| *** alex_xu_ has quit IRC | 00:23 | |
| *** flwang2 has left #openstack-glance | 00:30 | |
| *** flwang2 has joined #openstack-glance | 00:38 | |
| *** jasondotstar has quit IRC | 00:41 | |
| *** annashen has quit IRC | 00:53 | |
| *** david-lyle is now known as david-lyle_afk | 00:57 | |
| *** nellysmitt has joined #openstack-glance | 01:11 | |
| *** nellysmitt has quit IRC | 01:16 | |
| *** jasondotstar has joined #openstack-glance | 01:24 | |
| *** jamielennox is now known as jamielennox|away | 01:33 | |
| *** TravT has quit IRC | 01:34 | |
| flwang2 | zhiyan: ping | 01:36 |
| *** jasondotstar has quit IRC | 01:50 | |
| *** ericpeterson has joined #openstack-glance | 02:01 | |
| *** ericpeterson has quit IRC | 02:05 | |
| *** rwsu is now known as rwsu-afk | 02:11 | |
| *** haomaiwang has joined #openstack-glance | 02:20 | |
| *** ericpeterson has joined #openstack-glance | 02:22 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 02:26 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 02:27 | |
| *** diegows has quit IRC | 02:33 | |
| *** ericpeterson has quit IRC | 02:37 | |
| *** spzala has quit IRC | 03:01 | |
| *** nellysmitt has joined #openstack-glance | 03:12 | |
| zhiyan | flwang2: pong | 03:14 |
| *** nellysmitt has quit IRC | 03:16 | |
| *** harlowja is now known as harlowja_away | 03:28 | |
| *** annashen has joined #openstack-glance | 03:29 | |
| openstackgerrit | Sabari proposed openstack/glance_store: Check VMware session before uploading image https://review.openstack.org/141665 | 03:42 |
| *** mitz_ has quit IRC | 03:56 | |
| *** mitz has joined #openstack-glance | 03:56 | |
| *** TravT has joined #openstack-glance | 03:58 | |
| *** TravT has quit IRC | 04:02 | |
| *** annashen has quit IRC | 04:12 | |
| *** flwang2 has quit IRC | 04:13 | |
| *** annashen has joined #openstack-glance | 04:15 | |
| *** jyoti-ranjan has joined #openstack-glance | 04:23 | |
| *** julim has quit IRC | 04:27 | |
| *** annashen has quit IRC | 04:31 | |
| *** mgagne has joined #openstack-glance | 04:53 | |
| *** changbl has joined #openstack-glance | 04:57 | |
| *** annashen has joined #openstack-glance | 05:09 | |
| *** david-lyle_afk has quit IRC | 05:11 | |
| *** david-lyle_afk has joined #openstack-glance | 05:11 | |
| *** nellysmitt has joined #openstack-glance | 05:13 | |
| *** nellysmitt has quit IRC | 05:17 | |
| *** harlowja_away has quit IRC | 05:19 | |
| *** test2222 has joined #openstack-glance | 05:23 | |
| *** alex_xu has joined #openstack-glance | 05:24 | |
| *** test2222 has quit IRC | 05:27 | |
| *** sgotliv has joined #openstack-glance | 05:30 | |
| *** mtreinish has quit IRC | 05:40 | |
| *** TravT has joined #openstack-glance | 05:55 | |
| *** mtreinish has joined #openstack-glance | 05:56 | |
| *** annashen has quit IRC | 05:59 | |
| *** annashen has joined #openstack-glance | 06:01 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Imported Translations from Transifex https://review.openstack.org/146765 | 06:02 |
| *** nlevinki has joined #openstack-glance | 06:13 | |
| *** nlevinki has quit IRC | 06:14 | |
| *** nlevinki has joined #openstack-glance | 06:15 | |
| *** annashen has quit IRC | 06:17 | |
| *** sgotliv has quit IRC | 06:22 | |
| *** flwang has joined #openstack-glance | 06:29 | |
| *** sgotliv has joined #openstack-glance | 06:40 | |
| *** sgotliv has quit IRC | 06:52 | |
| openstackgerrit | Jun Hong Li proposed openstack/glance-specs: Add spec for https support in glance-replicator https://review.openstack.org/152809 | 07:04 |
| *** chlong has joined #openstack-glance | 07:06 | |
| openstackgerrit | Jun Hong Li proposed openstack/glance-specs: Add spec for https support in glance-replicator https://review.openstack.org/152809 | 07:08 |
| *** abhishekk has joined #openstack-glance | 07:09 | |
| *** nellysmitt has joined #openstack-glance | 07:13 | |
| *** groen692 has joined #openstack-glance | 07:14 | |
| *** nellysmitt has quit IRC | 07:18 | |
| *** chlong has quit IRC | 07:28 | |
| *** belmoreira has joined #openstack-glance | 07:36 | |
| *** nellysmitt has joined #openstack-glance | 07:42 | |
| *** tshefi has joined #openstack-glance | 08:14 | |
| *** markus_z has joined #openstack-glance | 08:23 | |
| *** TravT has quit IRC | 08:24 | |
| *** jistr has joined #openstack-glance | 08:29 | |
| *** sgotliv has joined #openstack-glance | 08:29 | |
| *** TravT has joined #openstack-glance | 08:39 | |
| *** TravT has quit IRC | 08:43 | |
| *** TravT has joined #openstack-glance | 08:52 | |
| *** TravT has quit IRC | 08:58 | |
| *** jyoti-ranjan has quit IRC | 09:04 | |
| *** pdb has joined #openstack-glance | 09:09 | |
| *** takedakn has joined #openstack-glance | 09:12 | |
| *** TravT has joined #openstack-glance | 09:15 | |
| *** mfedosin has quit IRC | 09:20 | |
| *** belmoreira has quit IRC | 09:21 | |
| *** TravT has quit IRC | 09:22 | |
| *** belmoreira has joined #openstack-glance | 09:22 | |
| *** takedakn has quit IRC | 09:29 | |
| nikhil_k | ativelkov: back yet? | 09:35 |
| *** boris-42 has joined #openstack-glance | 09:39 | |
| *** lakshmiS has joined #openstack-glance | 09:42 | |
| *** lakshmiS has quit IRC | 09:47 | |
| *** lakshmiS has joined #openstack-glance | 09:47 | |
| *** MattMan has joined #openstack-glance | 09:49 | |
| *** TravT has joined #openstack-glance | 09:56 | |
| *** TravT has quit IRC | 10:01 | |
| abhishekk | hi all, I am getting error pkg_resources.DistributionNotFound: glance==2015.1.dev87, while starting the glance api, registry service | 10:06 |
| abhishekk | does any one know how to resolve this issue? | 10:06 |
| nikhil_k | where abhishekk ? | 10:20 |
| abhishekk | nikhil_k: on screen | 10:21 |
| nikhil_k | devstack you mean? | 10:21 |
| abhishekk | I have installed using devstack | 10:21 |
| abhishekk | after installation services are running, but when I try to restart the service it gives ^^ error | 10:21 |
| nikhil_k | abhishekk: check your pkg info and see if matches | 10:29 |
| nikhil_k | you may not have installed after changing (with appropriate permissions) | 10:29 |
| * nikhil_k afk | 10:30 | |
| abhishekk | nikhil_k: ok, thank you for the pointer | 10:30 |
| *** TravT has joined #openstack-glance | 10:32 | |
| *** TravT has quit IRC | 10:37 | |
| *** jasondotstar has joined #openstack-glance | 10:46 | |
| *** belmoreira has quit IRC | 10:48 | |
| *** TravT has joined #openstack-glance | 10:50 | |
| *** haomaiwang has quit IRC | 10:51 | |
| nikhil_k | flaper87: zhiyan: kragniz sigmavirus24_awa : anyone seeing metadef, tasks etc commands on the client cli? | 10:53 |
| *** Longgeek has joined #openstack-glance | 10:55 | |
| *** TravT has quit IRC | 10:55 | |
| *** jyoti-ranjan has joined #openstack-glance | 10:55 | |
| *** lakshmiS has quit IRC | 11:03 | |
| *** TravT has joined #openstack-glance | 11:27 | |
| *** tnurlygayanov has joined #openstack-glance | 11:27 | |
| *** lakshmiS has joined #openstack-glance | 11:30 | |
| *** TravT has quit IRC | 11:31 | |
| *** Longgeek has quit IRC | 11:33 | |
| openstackgerrit | Yusuke Ide proposed openstack/glance: Add detail description of image_cache_max_size https://review.openstack.org/152523 | 11:34 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts API https://review.openstack.org/136629 | 11:36 |
| *** lakshmiS_ has joined #openstack-glance | 11:43 | |
| *** lakshmiS has quit IRC | 11:46 | |
| *** TravT has joined #openstack-glance | 11:49 | |
| *** TravT has quit IRC | 11:53 | |
| *** diegows has joined #openstack-glance | 11:57 | |
| *** david-lyle_afk is now known as david-lyle | 12:01 | |
| *** abhishekk has quit IRC | 12:03 | |
| *** abhishekk has joined #openstack-glance | 12:03 | |
| *** TravT has joined #openstack-glance | 12:04 | |
| openstackgerrit | Lakshmi N Sampath proposed openstack/glance: Synchronising oslo-incubator service module and its dependencies to enable openstack common service launcher within glance. This will be used as the framework for event notifiers and listeners. https://review.openstack.org/152872 | 12:10 |
| *** TravT has quit IRC | 12:10 | |
| *** abhishekk has quit IRC | 12:11 | |
| openstackgerrit | Lakshmi N Sampath proposed openstack/glance: Synchronising oslo-incubator service module https://review.openstack.org/152872 | 12:13 |
| *** hemanth has joined #openstack-glance | 12:14 | |
| *** jyoti-ranjan has quit IRC | 12:15 | |
| *** rodrigod` has joined #openstack-glance | 12:18 | |
| *** jasondotstar has quit IRC | 12:20 | |
| *** hemanth_ has quit IRC | 12:20 | |
| *** zul has quit IRC | 12:20 | |
| *** rodrigods has quit IRC | 12:20 | |
| *** TravT has joined #openstack-glance | 12:21 | |
| *** rodrigod` is now known as rodrigods | 12:22 | |
| *** rodrigods has quit IRC | 12:22 | |
| *** rodrigods has joined #openstack-glance | 12:22 | |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts Repository - DB https://review.openstack.org/115998 | 12:27 |
| *** EmilienM|afk is now known as EmilienM | 12:27 | |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts plugin loader https://review.openstack.org/134300 | 12:28 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: A mixin for jsonpatch requests validation https://review.openstack.org/148588 | 12:28 |
| *** zul has joined #openstack-glance | 12:28 | |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts API https://review.openstack.org/136629 | 12:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts Domain https://review.openstack.org/132898 | 12:32 |
| *** david-lyle is now known as david-lyle_afk | 12:36 | |
| *** david-lyle_afk is now known as david-lyle | 12:36 | |
| *** david-lyle is now known as david-lyle_afk | 12:44 | |
| *** david-lyle_afk is now known as david-lyle | 12:45 | |
| *** takedakn has joined #openstack-glance | 12:47 | |
| openstackgerrit | Travis Tripp proposed openstack/glance: Softare Metadata Definitions https://review.openstack.org/148007 | 12:47 |
| *** jyoti-ranjan has joined #openstack-glance | 12:54 | |
| kragniz | nikhil_k: they all look fine to me | 12:57 |
| *** aix has joined #openstack-glance | 13:01 | |
| *** TravT has quit IRC | 13:12 | |
| *** TravT has joined #openstack-glance | 13:16 | |
| *** tshefi has quit IRC | 13:22 | |
| *** tshefi has joined #openstack-glance | 13:22 | |
| *** TravT has quit IRC | 13:26 | |
| *** mjturek has joined #openstack-glance | 13:28 | |
| *** lakshmiS_ has quit IRC | 13:49 | |
| *** lakshmiS_ has joined #openstack-glance | 13:49 | |
| *** TravT has joined #openstack-glance | 13:51 | |
| *** TravT has quit IRC | 13:58 | |
| *** krykowski has joined #openstack-glance | 13:59 | |
| *** takedakn has quit IRC | 14:07 | |
| *** jyoti-ranjan has quit IRC | 14:09 | |
| *** belmoreira has joined #openstack-glance | 14:11 | |
| *** julim has joined #openstack-glance | 14:19 | |
| *** julim has quit IRC | 14:21 | |
| kragniz | friendly reminder for you all to unleash the openstack lizards | 14:22 |
| *** julim has joined #openstack-glance | 14:24 | |
| nikhil_k | kragniz: is this what you see http://paste.openstack.org/show/166831/ ? | 14:25 |
| kragniz | nikhil_k: you need $ glance --os-image-api-version 2 help for them to appear | 14:27 |
| kragniz | nikhil_k: since they're part of the v2 api | 14:27 |
| *** TravT has joined #openstack-glance | 14:29 | |
| nikhil_k | v2 should be default no? | 14:32 |
| kragniz | not on the client | 14:33 |
| kragniz | we were talking about making that switch, but it hasn't happened yet | 14:33 |
| *** Longgeek has joined #openstack-glance | 14:33 | |
| nikhil_k | kragniz: got it, I was looking in the code in thw wrong place. sorry. and thanks! | 14:34 |
| kragniz | you're welcome! | 14:34 |
| *** Longgeek has quit IRC | 14:35 | |
| *** peristeri has joined #openstack-glance | 14:38 | |
| *** TravT has quit IRC | 14:42 | |
| *** andrearo1 has joined #openstack-glance | 14:43 | |
| *** andrearosa has quit IRC | 14:45 | |
| *** sgotliv has quit IRC | 14:47 | |
| nikhil_k | zhiyan: flaper87 : need your oslo experience a bit here https://review.openstack.org/#/c/85211/ | 14:51 |
| *** andrearo1 is now known as andrearosa | 14:52 | |
| nikhil_k | ativelkov: we've around 9 hours for fixing https://review.openstack.org/#/c/148574/ . I think we'd plan for this in early k3 | 14:52 |
| kragniz | nikhil_k: I'll help with looking into the taskflow patch | 14:54 |
| kragniz | nikhil_k: give me a few minutes for a clean devstack | 14:54 |
| nikhil_k | kragniz: appreciate it | 14:54 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:00 | |
| * sigmavirus24 is looking into taskflow issue | 15:03 | |
| sigmavirus24 | *too | 15:03 |
| *** TravT has joined #openstack-glance | 15:03 | |
| *** esheffield has joined #openstack-glance | 15:04 | |
| kragniz | sigmavirus24: I hope you've voted lizard | 15:04 |
| *** thangp has joined #openstack-glance | 15:04 | |
| sigmavirus24 | lol | 15:04 |
| * sigmavirus24 knows what you're talking about but hasn't read his ML threads yet | 15:04 | |
| * sigmavirus24 knows because he observed the TC meeting and earlier ML threads | 15:05 | |
| * kragniz thinks sigmavirus24 knows too much | 15:05 | |
| * sigmavirus24 is only functioning because he has already had a large quantity of coffee this morning | 15:06 | |
| * nikhil_k bbiab | 15:08 | |
| *** peristeri has quit IRC | 15:08 | |
| *** zz_jgrimm is now known as jgrimm | 15:09 | |
| *** peristeri has joined #openstack-glance | 15:10 | |
| *** EmilienM has quit IRC | 15:14 | |
| *** julim has quit IRC | 15:15 | |
| *** EmilienM has joined #openstack-glance | 15:16 | |
| *** cpallares has joined #openstack-glance | 15:16 | |
| *** julim has joined #openstack-glance | 15:18 | |
| *** peristeri has quit IRC | 15:23 | |
| nikhil_k | kragniz: I meant to be afk, did not yet. Any luck so far? | 15:23 |
| * nikhil_k afk | 15:25 | |
| kragniz | nikhil_k: a few minutes for a clean devstack was optimistic | 15:25 |
| sigmavirus24 | nikhil_k: the reason is that the option should be set to taskflow | 15:27 |
| sigmavirus24 | nikhil_k: we do https://github.com/openstack/glance/blob/90a9cc3cc8d7ba350db013a127a65b03cd6c0d2f/glance/domain/__init__.py#L464 to load the module | 15:27 |
| sigmavirus24 | so by using "taskflow_executor" it tries to import "glance.async.taskflow_executor_executor" | 15:28 |
| sigmavirus24 | the exception message is really bad though | 15:28 |
| sigmavirus24 | kragniz: ^ | 15:28 |
| kragniz | heh, that message sucks | 15:28 |
| sigmavirus24 | YEP | 15:29 |
| nikhil_k | sigmavirus24: I did not set as that | 15:30 |
| sigmavirus24 | Ah, was thinking of your second attempt to make it work | 15:31 |
| sigmavirus24 | So that's coming from https://github.com/openstack/oslo.utils/blob/master/oslo_utils/importutils.py#L27 and rpartition is working as expected as far as I can tell | 15:33 |
| *** krykowski has quit IRC | 15:34 | |
| *** nlevinki has quit IRC | 15:34 | |
| *** sgotliv has joined #openstack-glance | 15:35 | |
| *** EmilienM is now known as EmilienM|afk | 15:35 | |
| sigmavirus24 | hm | 15:37 |
| sigmavirus24 | nikhil_k: can you look in /usr/local/lib/python2.7/{dist,site}-packages/glance/async/ to see what's in there? | 15:44 |
| *** diegows has quit IRC | 15:44 | |
| *** TravT has quit IRC | 15:44 | |
| *** vijendar has joined #openstack-glance | 15:47 | |
| *** tshefi has quit IRC | 16:00 | |
| *** diegows has joined #openstack-glance | 16:05 | |
| *** sgotliv has quit IRC | 16:06 | |
| *** TravT has joined #openstack-glance | 16:09 | |
| nikhil_k | $ ls /usr/local/lib/python2.7/dist-packages/glance/async/taskflow_executor.py | 16:09 |
| nikhil_k | /usr/local/lib/python2.7/dist-packages/glance/async/taskflow_executor.py | 16:09 |
| nikhil_k | sigmavirus24: ^ | 16:09 |
| sigmavirus24 | hm, so it should be able to import it | 16:09 |
| sigmavirus24 | from teh commandline, if you do import oslo_utils; oslo_utils.import_class('glance.async.taskflow_executor.TaskExecutor') what happens? | 16:10 |
| zhiyan | Seems sigmavirus24 already got answer | 16:11 |
| nikhil_k | zhiyan: ?? | 16:11 |
| nikhil_k | zhiyan: welcome back! | 16:11 |
| nikhil_k | zhiyan: how are you doing now? | 16:11 |
| sigmavirus24 | zhiyan: not quite | 16:11 |
| *** TravT has quit IRC | 16:11 | |
| zhiyan | So what about to change taskfow_executor to just taskfow? | 16:12 |
| *** TravT has joined #openstack-glance | 16:12 | |
| zhiyan | Hi nikhil_k! I feel I'm better now. | 16:13 |
| nikhil_k | zhiyan: ah glad to hear! | 16:13 |
| nikhil_k | sigmavirus24: zhiyan : AttributeError: 'module' object has no attribute 'import_object' | 16:13 |
| sigmavirus24 | zhiyan: nikhil_k wasn't doing that | 16:13 |
| nikhil_k | and | 16:13 |
| nikhil_k | AttributeError: 'module' object has no attribute 'import_class' | 16:13 |
| sigmavirus24 | nikhil_k: sorry, import oslo_utils.importutils and oslo_utils.importutils.import_class(...) | 16:14 |
| nikhil_k | sigmavirus24: zhiyan : I tried 2 values | 16:14 |
| sigmavirus24 | so import object is what I think we should be using here anyway, but that's besides the point =P | 16:14 |
| zhiyan | So I was thinking 23:27:45 <sigmavirus24> nikhil_k: we do https://github.com/openstack/glance/blob/90a9cc3cc8d7ba350db013a127a65b03cd6c0d2f/glance/domain/__init__.py#L464 to load the module is the point no? | 16:14 |
| nikhil_k | taskflow and glance.async.taskflow | 16:14 |
| sigmavirus24 | nikhil_k: the difference is oslo_utils versus oslo_utils**.importutils** | 16:14 |
| nikhil_k | sigmavirus24: right, flaper87 was against it as it was sup to be deprecated | 16:14 |
| sigmavirus24 | nikhil_k: ah okay | 16:14 |
| * sigmavirus24 didn't have that context | 16:14 | |
| zhiyan | Humm I'm taking with you on my ip6 hard to do some try | 16:15 |
| *** EmilienM|afk is now known as EmilienM | 16:18 | |
| nikhil_k | >>> oslo_utils.importutils.import_class('glance.async.taskflow_executor.TaskExecutor') | 16:23 |
| nikhil_k | <class 'glance.async.taskflow_executor.TaskExecutor'> | 16:23 |
| nikhil_k | (in a meeting to availability is limited) | 16:23 |
| nikhil_k | s/to/so/ | 16:23 |
| sigmavirus24 | nikhil_k: is this a cloud server that I could ssh into? | 16:27 |
| nikhil_k | sigmavirus24: umm, you seem like a hacker kind of a person :P | 16:27 |
| * nikhil_k means that in a good way | 16:27 | |
| sigmavirus24 | So you want me to write a gui in VB to track your IP address back to the cloudserver? | 16:28 |
| sigmavirus24 | Done. | 16:28 |
| nikhil_k | VB :P | 16:29 |
| sigmavirus24 | Reference: http://youtu.be/hkDD03yeLnU | 16:29 |
| sigmavirus24 | I keep forgetting openstack uses rfc3986 so everytime I see it installed I giggle | 16:34 |
| *** rwsu-afk is now known as rwsu | 16:40 | |
| *** diegows has quit IRC | 16:41 | |
| *** groen692 has quit IRC | 16:46 | |
| *** belmoreira has quit IRC | 16:52 | |
| *** TravT has quit IRC | 16:54 | |
| *** diegows has joined #openstack-glance | 16:57 | |
| *** annegent_ has joined #openstack-glance | 16:58 | |
| zhiyan | seems i can create taskflow executor now successfully now.. i set "task_executor = taskflow" in glance-api.conf and do a replace on glance/async/taskflow_executor.py L109 , replace 'llistener.DynamicLogListener(engine, logger=LOG)' to 'llistener.DynamicLoggingListener(engine, log=LOG)" (base on | 16:59 |
| zhiyan | https://github.com/openstack/taskflow/blob/master/taskflow/listeners/logging.py#L114 ) | 16:59 |
| zhiyan | nikhil_k: sigmavirus24 ^ | 16:59 |
| *** esheffield has quit IRC | 17:00 | |
| sigmavirus24 | zhiyan: why would that cause the import error? | 17:00 |
| zhiyan | i just used 'taskflow' for task_executor option in config, then DynamicLogListener not found raised up. | 17:02 |
| sigmavirus24 | We're not seeing that | 17:03 |
| *** ericpeterson has joined #openstack-glance | 17:04 | |
| zhiyan | sigmavirus24: any exception still there in you env ? | 17:06 |
| sigmavirus24 | yes. I'm digging in | 17:07 |
| zhiyan | in my local env, the task import task can be create correctly, and state of it turned/worked well currently. | 17:09 |
| zhiyan | ..with taskflow executor | 17:10 |
| sigmavirus24 | so here's the weird thing, if I debug into the running api server, the import fails, outside of that it passes | 17:12 |
| zhiyan | weird | 17:14 |
| sigmavirus24 | I think I might see the problem | 17:15 |
| sigmavirus24 | also that taskflow stuff is just fine in the local copy zhiyan | 17:15 |
| zhiyan | sigmavirus24: which is fine? | 17:15 |
| sigmavirus24 | the taskflow logging information | 17:15 |
| zhiyan | you mean my comment about change DynamicLogListener to DynamicLoggingListener? | 17:16 |
| zhiyan | or the logging out about taskflow...actually i found a wield logging in taskflow side and glance-api.. | 17:17 |
| *** esheffield has joined #openstack-glance | 17:17 | |
| zhiyan | .. which it's a 'failed' import image task, but taskflow says task (its concept) just successful .. | 17:18 |
| sigmavirus24 | I know what's' going on | 17:18 |
| *** TravT has joined #openstack-glance | 17:19 | |
| sigmavirus24 | the version of glance being run is coming out of /opt/stack/glance/glance | 17:19 |
| sigmavirus24 | now the question is to find out why | 17:20 |
| zhiyan | need to sleep. can talk later if needed..also can direct to add comments on gerrit. at least, ify, it works for me locally. | 17:20 |
| sigmavirus24 | zhiyan: yeah this is an installation problem =P | 17:21 |
| zhiyan | btw, i just tested in my devstack env as well..not in /opt/stack however. | 17:21 |
| sigmavirus24 | yep | 17:21 |
| zhiyan | i think so sigmavirus24 | 17:21 |
| * sigmavirus24 scolds nikhil_k | 17:21 | |
| sigmavirus24 | now to fixit | 17:21 |
| *** TravT_ has joined #openstack-glance | 17:22 | |
| *** TravT has quit IRC | 17:22 | |
| zhiyan | sigmavirus24: and we def need to change DynamicLogListener to DynamicLoggingListener to make it work | 17:22 |
| sigmavirus24 | I believe you. I would guess we're not seeing that yet because of the installation issue. | 17:23 |
| zhiyan | otherwise two exceptions raise up: AttributeError: 'module' object has no attribute 'DynamicLogListener' and TypeError: __init__() got an unexpected keyword argument 'logger' | 17:24 |
| zhiyan | sigmavirus24: yep, just share my info here | 17:25 |
| sigmavirus24 | thanks zhiyan | 17:25 |
| *** jistr has quit IRC | 17:25 | |
| zhiyan | sigmavirus24: np. i'm just interested in why i didn't met that installation issue.. | 17:26 |
| sigmavirus24 | perhaps zhiyan you ran devstack from a clean machine? | 17:26 |
| zhiyan | sigmavirus24: i think it's not really :) it's really 'old'. | 17:27 |
| *** annashen has joined #openstack-glance | 17:27 | |
| zhiyan | sigmavirus24: so you verified patch on a clean devstack dev successfully ? | 17:28 |
| sigmavirus24 | No I was being super lazy | 17:28 |
| sigmavirus24 | I'll devstack now | 17:29 |
| zhiyan | sigmavirus24: ok. i need to get some sleep, ttly, or direct add comments in gerrit if needed. thanks for your efforts. | 17:29 |
| sigmavirus24 | yep | 17:29 |
| sigmavirus24 | have a good night | 17:30 |
| zhiyan | have a good one too | 17:30 |
| *** pdb has quit IRC | 17:40 | |
| *** annegent_ has quit IRC | 17:53 | |
| *** groen692 has joined #openstack-glance | 17:58 | |
| *** aix has quit IRC | 18:05 | |
| *** harlowja has joined #openstack-glance | 18:11 | |
| *** jasondotstar has joined #openstack-glance | 18:11 | |
| *** jasondotstar has quit IRC | 18:14 | |
| *** groen692 has quit IRC | 18:25 | |
| *** ericpete_ has joined #openstack-glance | 18:34 | |
| *** ericpeterson has quit IRC | 18:37 | |
| *** lakshmiS_ has quit IRC | 18:38 | |
| *** ericpeterson has joined #openstack-glance | 18:42 | |
| *** ericpete_ has quit IRC | 18:46 | |
| *** annegent_ has joined #openstack-glance | 18:54 | |
| *** annegent_ has quit IRC | 18:59 | |
| *** annegent_ has joined #openstack-glance | 19:00 | |
| *** sgotliv has joined #openstack-glance | 19:15 | |
| *** markus_z has quit IRC | 19:17 | |
| *** flwang has quit IRC | 19:21 | |
| nikhil_k | sigmavirus24: back | 19:23 |
| nikhil_k | sigmavirus24: so, I installed it from /home/nikhil/glance | 19:23 |
| nikhil_k | and that's an issue, why? | 19:23 |
| sigmavirus24 | nikhil_k: if you look at the logs, things are being used from /opt/stack/glance/* | 19:32 |
| sigmavirus24 | So the question may be better asked, why is it using that? | 19:32 |
| sigmavirus24 | The first suspect (as is always the case is pbr) =P | 19:32 |
| nikhil_k | sigmavirus24: gotcha, well I'm detached from the screen now | 19:34 |
| sigmavirus24 | I think -x will allow us to both be attached | 19:34 |
| *** ericpeterson has quit IRC | 19:34 | |
| nikhil_k | not bad, screens sizes match :) | 19:34 |
| nikhil_k | to some extent | 19:34 |
| sigmavirus24 | need me to make mine bigger? | 19:35 |
| sigmavirus24 | nikhil_k: and now we're at the problem that zhiyan mentioned earlier | 19:37 |
| *** annashen has quit IRC | 19:37 | |
| nikhil_k | sigmavirus24: nah, it looks good. thanks | 19:37 |
| *** annashen has joined #openstack-glance | 19:38 | |
| nikhil_k | nice | 19:39 |
| nikhil_k | it worked, so devstack problem and not the code | 19:39 |
| nikhil_k | this will make it to k2 hopefully ! | 19:39 |
| nikhil_k | ok, I'm gunna push up a PS | 19:39 |
| nikhil_k | sigmavirus24: or do you want to? | 19:39 |
| sigmavirus24 | nikhil_k: after you | 19:39 |
| sigmavirus24 | I have lots on my plate you and I need to fix jeepyb because I doubt anyone's going to fix it for us | 19:40 |
| * sigmavirus24 logged off your cloudserver | 19:40 | |
| nikhil_k | sigmavirus24: appreciate your help on this. Makes a ton of difference esp if this lands in k2 | 19:41 |
| *** ericpeterson has joined #openstack-glance | 19:41 | |
| sigmavirus24 | nikhil_k: happy to help | 19:41 |
| nikhil_k | and with the productive meeting I'd about k3 | 19:41 |
| *** diegows has quit IRC | 19:41 | |
| *** harlowja is now known as harlowja_away | 19:50 | |
| *** EmilienM is now known as EmilienM|afk | 19:53 | |
| *** peristeri has joined #openstack-glance | 19:53 | |
| *** ericpete_ has joined #openstack-glance | 19:58 | |
| *** annegent_ has quit IRC | 19:58 | |
| *** diegows has joined #openstack-glance | 19:59 | |
| *** foobloo has joined #openstack-glance | 20:01 | |
| foobloo | does policy.json restrict public images too? | 20:01 |
| *** ericpeterson has quit IRC | 20:01 | |
| *** annashen has quit IRC | 20:02 | |
| foobloo | i tried both is_admin and role:admin for image(s) get, but can still list them as a non-privileged user | 20:02 |
| openstackgerrit | nikhil komawar proposed openstack/glance: Add Support for TaskFlow Executor https://review.openstack.org/85211 | 20:02 |
| foobloo | same for delete, et al | 20:02 |
| *** annegent_ has joined #openstack-glance | 20:04 | |
| nikhil_k | flaper87: flwang1 zhiyan ativelkov : https://review.openstack.org/#/c/85211/ | 20:06 |
| nikhil_k | and https://review.openstack.org/#/c/148007/ | 20:10 |
| sigmavirus24 | foobloo: what version of glance | 20:17 |
| sigmavirus24 | nikhil_k: btw, any word about the drivers group on launchpad? | 20:17 |
| nikhil_k | nothing yet, let me try again | 20:17 |
| flwang1 | nikhil_k: yes, sir | 20:21 |
| nikhil_k | hello sir | 20:21 |
| nikhil_k | flwang1: need your help with a couple of reviews | 20:22 |
| nikhil_k | https://review.openstack.org/#/c/85211/ and https://review.openstack.org/#/c/148007/ | 20:22 |
| nikhil_k | they'd be pretty straightforward | 20:22 |
| nikhil_k | I'm happy to chat if you need more info | 20:22 |
| *** flwang has joined #openstack-glance | 20:23 | |
| *** TravT_ has quit IRC | 20:23 | |
| nikhil_k | sigmavirus24: what do you think about postponing https://review.openstack.org/#/c/148574/ ? (if you are busy, we can talk later) | 20:23 |
| *** TravT has joined #openstack-glance | 20:23 | |
| kragniz | sigmavirus24: nikhil_k: sorry, got pulled away for other things | 20:23 |
| nikhil_k | kragniz: no issues, we got it resolved | 20:23 |
| kragniz | all is happy with the taskflow patch now? | 20:23 |
| kragniz | awesome | 20:24 |
| nikhil_k | it was some weird installation problem | 20:24 |
| kragniz | okay - just devstack being itself? | 20:24 |
| sigmavirus24 | nikhil_k: No I think that's devstack's fault | 20:24 |
| sigmavirus24 | kragniz: yes | 20:24 |
| sigmavirus24 | there seems to be a link dropped to /opt/stack/glance so that seems to override dist-packages and I suspect now that it might be a PYTHONPATH thing | 20:25 |
| * sigmavirus24 didn't think of this before | 20:25 | |
| sigmavirus24 | nikhil_k: I have no strong opinions about postponing the Cooperative Reader stuff | 20:25 |
| sigmavirus24 | I think it makes sense to postpone it even though it is a high priority bug | 20:25 |
| sigmavirus24 | It just isn't ready to merge and putting off fixing it up doesn't make a whole lot of sense to me. Plus we threw the idea around to see if any members of the OSSG wanted to give us opinions about the security impact it might have | 20:26 |
| nikhil_k | sigmavirus24: ah ok, I was just curious about that. Did you hear whispers floating in other places? | 20:27 |
| flwang1 | nikhil_k: I'm happy to review the taskflow patch, but i'm not really familiar with the metadata def code | 20:27 |
| nikhil_k | in any case, thanks for the reassurance. We'd postpone! | 20:27 |
| sigmavirus24 | Ah, I haven't pinged anyone specifically to look at it nikhil_k | 20:27 |
| nikhil_k | flwang1: do ask TravT or me if you need more info on the metadef patch. It's not a big change | 20:28 |
| nikhil_k | sigmavirus24: ahk | 20:28 |
| * sigmavirus24 puts it on the meeting notes for the next OSSG meeting | 20:28 | |
| flwang1 | nikhil_k: yep, seems only some db, service definitions | 20:28 |
| flwang1 | ok, I will ask if i see some questions | 20:28 |
| nikhil_k | yeah | 20:28 |
| nikhil_k | cool, ty | 20:29 |
| TravT | nikhil_k: flwang1: yep, let me know if you have questions | 20:29 |
| flwang1 | TravT: cool, cheers | 20:29 |
| *** harlowja_away is now known as harlowja | 20:33 | |
| *** IlyaE has joined #openstack-glance | 20:34 | |
| flwang1 | TravT: ping | 20:35 |
| flwang1 | TravT: in https://review.openstack.org/#/c/148007/7/glance/tests/etc/test-metadefs.json i saw the 'protected' attribute value is false, but in all the json files you're adding, it's true | 20:36 |
| flwang1 | so i'm wondering if this case can be covered by test | 20:37 |
| TravT | pong | 20:38 |
| TravT | let me look | 20:38 |
| TravT | ok, so that is the difference between where they are located | 20:39 |
| TravT | that particular file is in test data. | 20:39 |
| TravT | the other ones are deployed in etc/metadefs | 20:39 |
| TravT | and get loaded by default | 20:40 |
| flwang | yep, I know | 20:40 |
| TravT | in devstack / etc | 20:40 |
| *** IlyaE has quit IRC | 20:40 | |
| *** IlyaE has joined #openstack-glance | 20:40 | |
| TravT | those particular test metadefs aren't used by an automated test yet | 20:40 |
| flwang | TravT: so does the test-metadefs is used for metadata def testing? | 20:40 |
| TravT | and I could drop them | 20:40 |
| TravT | but, I created them for testing out some things in the Horizon UI | 20:40 |
| TravT | and decided to contribute them here | 20:41 |
| flwang | so which file will call this test-metadef.json? | 20:41 |
| flwang | in glance | 20:41 |
| TravT | none of them. | 20:41 |
| TravT | which is why I could drop it if you want | 20:41 |
| TravT | but I have found it useful | 20:42 |
| TravT | for testing from command line | 20:42 |
| TravT | manually | 20:42 |
| nikhil_k | It would be hard to test that in Glance, no? | 20:42 |
| flwang | TravT: may i know how? | 20:42 |
| TravT | let me type it up in a terminal and copy over | 20:42 |
| TravT | just a sec | 20:42 |
| flwang | TravT: cool, cheers | 20:43 |
| *** annashen has joined #openstack-glance | 20:43 | |
| TravT | just FYI, i've got some guys here talking to me, so trying to keep my brain from splitting in half | 20:44 |
| TravT | flwang: here's the first way | 20:49 |
| TravT | glance --os-image-api-version 2 md-namespace-import --file /opt/stack/glance/glance/tests/etc/test-metadefs.json | 20:49 |
| flwang | ok | 20:53 |
| flwang | so it's most like a sample file to help you verify something, right? | 20:53 |
| TravT | yeah | 20:53 |
| TravT | maybe if anything it should be under glance/tests/etc/metadefs though | 20:53 |
| TravT | you can also do it from glance-manage db_load_metadefs | 20:54 |
| TravT | but my VM is not responding right now | 20:55 |
| flwang | TravT: if that's the case, personally i prefer to place it in doc and embedded in a file about how to test | 20:55 |
| flwang | but it's not a real blocker | 20:55 |
| TravT | i think I put it to sleep 1 to many times | 20:55 |
| foobloo | sigmavirus24, sorry, got pulled away. I'm using glance-2014.1.3 | 20:55 |
| flwang | nikhil_k: thoughts? | 20:55 |
| TravT | I could put it in doc if needed. | 20:56 |
| *** nellysmitt has quit IRC | 20:56 | |
| nikhil_k | flwang: I'm okay with moving (if you will be around a bit) :) | 20:56 |
| nikhil_k | and thanks for that extra check | 20:56 |
| nikhil_k | TravT: let's DOIT | 20:56 |
| TravT | HTTP 1.2 DOIT | 20:56 |
| flwang | nikhil_k: I will be here for next 8 hours :D | 20:56 |
| TravT | 200 (OK) | 20:57 |
| nikhil_k | flwang: ah ok, cool | 20:57 |
| TravT | so, you want me to put in doc in this patch? | 20:57 |
| TravT | or just drop it out of this patch | 20:57 |
| nikhil_k | ninja (OK) | 20:57 |
| TravT | I can drop it in this patch instantly | 20:57 |
| TravT | if in doc, are you wanting it in an RST file? | 20:57 |
| flwang | TravT: how about let's drop it for now | 20:57 |
| TravT | I have another doc change that nikhil wants anyway | 20:57 |
| flwang | and you can post another patch when you're free ;) | 20:57 |
| nikhil_k | either works | 20:58 |
| openstackgerrit | Travis Tripp proposed openstack/glance: Softare Metadata Definitions https://review.openstack.org/148007 | 20:58 |
| TravT | ^ | 20:58 |
| nikhil_k | (dropping now and adding later or changing now) | 20:58 |
| TravT | ask an ye shall receive | 20:59 |
| nikhil_k | we can wait for jenkins | 20:59 |
| sigmavirus24 | TravT: what other things can we ask for? =P | 20:59 |
| nikhil_k | gold | 21:00 |
| TravT | Sounds like I was asking for it | 21:00 |
| TravT | lol | 21:00 |
| nikhil_k | ;) | 21:00 |
| *** eglynn has quit IRC | 21:00 | |
| *** IlyaE has quit IRC | 21:02 | |
| *** IlyaE has joined #openstack-glance | 21:02 | |
| *** IlyaE has quit IRC | 21:06 | |
| *** zul has quit IRC | 21:08 | |
| *** zul has joined #openstack-glance | 21:08 | |
| *** IlyaE has joined #openstack-glance | 21:08 | |
| sigmavirus24 | nikhil_k: regarding your feedback on https://review.openstack.org/#/c/148387/, do we really want the very verbose details every time? | 21:09 |
| sigmavirus24 | It's seems like that may occasionally have more details than the user cares about | 21:09 |
| *** foobloo_ has joined #openstack-glance | 21:09 | |
| *** foobloo_ is now known as foobloo2 | 21:09 | |
| *** IlyaE has quit IRC | 21:10 | |
| *** annegent_ has quit IRC | 21:10 | |
| nikhil_k | sigmavirus24: it seems to give the details on xml | 21:11 |
| foobloo2 | sigmavirus24, sorry got kicked. the version i mentioned is what rh osp5 installs, 2014.1.3 | 21:11 |
| nikhil_k | why not json? | 21:11 |
| nikhil_k | err, html not xml | 21:11 |
| nikhil_k | trying to do multiple things at once, does not help | 21:11 |
| *** foobloo has quit IRC | 21:12 | |
| sigmavirus24 | The HTML is only returned if we're using webob's default messages which typically return very little detail in the first place though | 21:12 |
| sigmavirus24 | (Assuming they return detail at all ;) | 21:12 |
| openstackgerrit | Ian Cordasco proposed openstack/glance: Fix for CooperativeReader to properly process read length https://review.openstack.org/148574 | 21:17 |
| nikhil_k | it was a nit anyway | 21:18 |
| nikhil_k | I dont want another session of bikeshedding | 21:19 |
| nikhil_k | see things like these people should just fix | 21:20 |
| nikhil_k | anyways, +2 there | 21:20 |
| kragniz | nikhil_k: it's still better than it was | 21:21 |
| kragniz | even if it's a little inconsistant | 21:21 |
| nikhil_k | yep | 21:22 |
| *** IlyaE has joined #openstack-glance | 21:22 | |
| *** TravT has quit IRC | 21:26 | |
| *** IlyaE has quit IRC | 21:30 | |
| *** IlyaE has joined #openstack-glance | 21:34 | |
| *** IlyaE has quit IRC | 21:36 | |
| *** TravT has joined #openstack-glance | 21:36 | |
| sigmavirus24 | to be fair, I really think there needs to be an associated bug in glance for that glanceclient issue | 21:37 |
| sigmavirus24 | I don't think we should ever be returning HTML from glance-api | 21:38 |
| *** TravT has quit IRC | 21:41 | |
| *** TravT has joined #openstack-glance | 21:41 | |
| *** sgotliv has quit IRC | 21:41 | |
| *** TravT has quit IRC | 21:42 | |
| *** julim has quit IRC | 21:45 | |
| mgagne | nikhil_k: do you have any update on this subject? https://etherpad.openstack.org/p/havana-image-cloning-to-other-regions | 21:45 |
| *** EmilienM|afk is now known as EmilienM | 21:45 | |
| nikhil_k | mgagne: So, the only thing I can mention to you is that it is planned and being prioritized. However, it being a complex issue has minimal chance of getting in Kilo. The blockers are glance_store refactor to support user tokens. | 21:47 |
| *** foobloo2 is now known as foobloo | 21:47 | |
| nikhil_k | And the preceeding BPs | 21:47 |
| *** d34dh0r53 is now known as blarnath | 21:48 | |
| nikhil_k | we can chat if you need more info or have devs interested in contributing | 21:48 |
| mgagne | nikhil_k: thanks. we are facing the same challenges here and was looking for existing tools instead of reinventing one of my own =) | 21:48 |
| *** belmoreira has joined #openstack-glance | 21:48 | |
| *** TravT has joined #openstack-glance | 21:49 | |
| nikhil_k | haha ok. Yeah, across DC will be a real challenge. We've discussed it lot and I will be happy to point you to someone who is eager to get this in. | 21:49 |
| mgagne | nikhil_k: I found https://github.com/hep-gc/glint which looks to be promising but it feels (IMO) like killing a fly with a bazooka. And the bazooka requires heavy assembly and uses non-standard rockets. | 21:52 |
| *** TravT has quit IRC | 21:55 | |
| *** thangp has quit IRC | 21:57 | |
| *** IlyaE has joined #openstack-glance | 22:00 | |
| nikhil_k | mgagne: hmm yeah. I've heard of it. BTW, I'm planning to chat with the person who impl squid proxy stuff for CERN later this week | 22:01 |
| mgagne | nikhil_k: cool, his presentation was great. | 22:02 |
| nikhil_k | mgagne: Glint has a lot of potential however, it won't scale for all the use caes | 22:03 |
| nikhil_k | the way I remember, it deals with a tiny image and if that's what you want to do replicate across DC then it's a good one | 22:04 |
| *** annegent_ has joined #openstack-glance | 22:04 | |
| *** ericpeterson has joined #openstack-glance | 22:04 | |
| *** ericpete_ has quit IRC | 22:04 | |
| mgagne | nikhil_k: are you referring to glint? | 22:04 |
| nikhil_k | however, people might want to close multiple gigs of data and that too of a proprietary image! And there we need some challenging engg | 22:05 |
| nikhil_k | mgagne: yes, glint | 22:05 |
| *** blarnath is now known as d34dh0r53 | 22:05 | |
| *** annegent_ has quit IRC | 22:05 | |
| nikhil_k | at least that's what the person told us. I haven't looked through the docs. | 22:05 |
| nikhil_k | (hope this glint is from UBC group) | 22:06 |
| mgagne | nikhil_k: IIC, there is a "master" glance where "golden" images are stored and a mapping of openstack installations is done in glint. You add/remove them to the list of destinations. The "trick" is the image is created and points to the actual image URL, the image itself is not "copied" to the installation. | 22:07 |
| mgagne | nikhil_k: but each provider has its own use cases. We version our images, cache images in Cinder for faster cloning, etc. and this tool would have to support it in some way. | 22:07 |
| mgagne | nikhil_k: because everyone is special =) | 22:08 |
| *** IlyaE has quit IRC | 22:08 | |
| *** TravT has joined #openstack-glance | 22:13 | |
| *** TravT has quit IRC | 22:14 | |
| *** TravT_ has joined #openstack-glance | 22:14 | |
| *** TravT_ has quit IRC | 22:15 | |
| nikhil_k | mgagne: ah gotcha. cache images in cinder huh, nice | 22:27 |
| nikhil_k | sigmavirus24: (just curious) Don't they get a auto email? And thanks for sending the same. | 22:28 |
| sigmavirus24 | nikhil_k: they do, but I think most people ignore those unless it piques their interest. So I replied to it to beg them for input =P | 22:29 |
| nikhil_k | sigmavirus24: haha ok. I think we can have another (secret) backdoor. | 22:31 |
| nikhil_k | Let me try that tomorrow ;) | 22:31 |
| sigmavirus24 | Super secret backdoors are the best hideouts | 22:32 |
| kragniz | does glance-replicator need to be bundled with glance? | 22:34 |
| kragniz | might it be more natural for it to live with glanceclient? | 22:34 |
| kragniz | sigmavirus24: ^ | 22:34 |
| sigmavirus24 | kragniz: Idk | 22:35 |
| kragniz | idk either | 22:35 |
| mgagne | kragniz: glance-replicator is more related to an app than a client library | 22:35 |
| * sigmavirus24 thought so | 22:36 | |
| sigmavirus24 | =P | 22:36 |
| mgagne | kragniz: I can only see the same story repeating itself just like keystonemiddleware: moved to keystoneclient and than move to its own repository. | 22:36 |
| * sigmavirus24 didn't really have any opinions | 22:36 | |
| nikhil_k | yeah, it does store related stuff | 22:36 |
| kragniz | mgagne: this is true | 22:37 |
| kragniz | mgagne: it would be much easier if it used glanceclient, though | 22:37 |
| sigmavirus24 | nikhil_k: mgagne context is https://review.openstack.org/#/c/149467/ | 22:37 |
| kragniz | thanks sigmavirus24 | 22:37 |
| mgagne | sigmavirus24: thanks for context. even without context, I still think it would be a bad idea to move it to glanceclient =) | 22:38 |
| sigmavirus24 | It would seem that using glanceclient from glance-replicator would be convenient, of course, that makes glance dependent glanceclient which seems odd | 22:38 |
| ndonegan | glance-replicator needs a lot of love. | 22:43 |
| ndonegan | Waiting for "legal" to confirm here that I'm covered by Corporate CLA thingy, and I'll be be pushing up a few small fixes. | 22:45 |
| kragniz | ndonegan: cool! | 22:46 |
| kragniz | ndonegan: what company? | 22:47 |
| ndonegan | Workday. | 22:47 |
| ndonegan | Using glane-replicator in anger between DCs, and found a whole load of small little things, some probably due to our abuse of meta-data. | 22:48 |
| *** mjturek has quit IRC | 22:48 | |
| mgagne | ndonegan: tell me more about your abuse :D | 22:55 |
| *** nellysmitt has joined #openstack-glance | 22:57 | |
| ndonegan | mgagne: The custom properties are being used to track what QA etc. images have gone through along with other data used for deciding where an image is allowed to be deployed. | 22:57 |
| mgagne | ndonegan: this is relevant to my interests =) | 22:57 |
| ndonegan | Where we've run into fun is with ha-proxy's max header size. | 22:58 |
| ndonegan | So QA did some magic to restric the max amount of properties associated with the images. | 22:59 |
| mgagne | oh yea, that one | 22:59 |
| ndonegan | This works grand locally now. However then I hit it again with glance-replicator :) | 22:59 |
| mgagne | ndonegan: I think we tweaked tune.maxrewrite to allow more room for the header values | 23:00 |
| ndonegan | The only change I've made so far is to catch and handle the socket error thrown by ha-proxy's behaviour. | 23:00 |
| *** esheffield has quit IRC | 23:00 | |
| *** jaypipes has quit IRC | 23:00 | |
| ndonegan | mgagne: Looked at that, however the HA-Proxy seem to thing the sky will fall down if that's done! | 23:00 |
| mgagne | ndonegan: running 1.5 ? | 23:00 |
| ndonegan | I'm sort of half tempted to look at LVS rather than ha-proxy for the duty. | 23:01 |
| ndonegan | mgagne: Yup. | 23:01 |
| *** nellysmitt has quit IRC | 23:01 | |
| mgagne | ndonegan: yea, we used to use LVS quite extensively before. can't say anything wrong about it. but haproxy is the new trend so. =) | 23:02 |
| ndonegan | mgagne: I'm not all that trendy :) | 23:02 |
| mgagne | ndonegan: getting the job done is more trendy =) | 23:02 |
| ndonegan | Actually, related to https://review.openstack.org/#/c/149467/ from a security standpoint is token expiry. | 23:02 |
| *** IlyaE has joined #openstack-glance | 23:02 | |
| ndonegan | Security don't seem to like long living tokens. | 23:03 |
| sigmavirus24 | metadata really doesnt' belong in the headers | 23:03 |
| ndonegan | And glance-replicator tends to be fairly long running, at least during the initial dumes or loads. | 23:03 |
| * sigmavirus24 starts considering a spec to fix tht | 23:03 | |
| ndonegan | sigmavirus24: A PUT/POST option for the metadata json would be nice. | 23:04 |
| sigmavirus24 | ndonegan: can you start an etherpad with this and email it to me? | 23:04 |
| sigmavirus24 | or just ping me in here with it | 23:04 |
| ndonegan | sigmavirus24: Can do. | 23:04 |
| sigmavirus24 | I feel like glance-replicator needs some love but there will be glance improvements necessary | 23:04 |
| ndonegan | sigmavirus24: I have a whole load of Jira's open internall which I'm slogging though on an internal fork. | 23:05 |
| ndonegan | Would really prefer it upstream. | 23:05 |
| *** IlyaE has quit IRC | 23:05 | |
| sigmavirus24 | heh | 23:06 |
| sigmavirus24 | We'll see what we can get upstream | 23:06 |
| sigmavirus24 | I think metadata as headers is such a bad idea given the typical limit on header values | 23:06 |
| ndonegan | THe other fun one is an equivilent of "rsync --delete". That one I'm a bit more scared off, but there's a PoC working internally which hasn't done anything stupid yet. | 23:06 |
| sigmavirus24 | FWIW, this is why we're arguing with swift folks over on the API-WG because metadata in headers is such a bad idea and they don't want to give that up | 23:06 |
| sigmavirus24 | ndonegan: /yet/ ;) | 23:07 |
| ndonegan | sigmavirus24: I'm with you on the meta-data as headers. | 23:07 |
| sigmavirus24 | It's currently being proposed as a guideline | 23:07 |
| *** echevemaster has joined #openstack-glance | 23:08 | |
| ndonegan | Oh, the other thing implemented internally is the adding the --property-filter from glance client to glance-replicator. | 23:08 |
| ndonegan | No point having images which haven't passed QA heading out to PROD. | 23:09 |
| sigmavirus24 | ndonegan: you're adding fuel to kragniz's fire =P | 23:09 |
| ndonegan | sigmavirus24: Good :)D | 23:09 |
| sigmavirus24 | So we're going to vendor glanceclient =P | 23:09 |
| * sigmavirus24 writes new spec | 23:09 | |
| sigmavirus24 | We'll hide it in glance/openstack/common/ so the packagers don't notice it =P | 23:10 |
| *** IlyaE has joined #openstack-glance | 23:11 | |
| mgagne | ndonegan: is glance-replicator running in a while(True) loop or is it triggered from a workflow/pipeline in the QA process? | 23:11 |
| ndonegan | Currently croned to run fairly regularly. | 23:12 |
| ndonegan | The images are small enough, and the network fast enough, that for the most part token rotation is not too much of an issue. | 23:12 |
| kragniz | sigmavirus24: I'm warming to the idea of a glance-tools repo | 23:13 |
| kragniz | I doubt anyone else will, though :P | 23:13 |
| ndonegan | It's only on full dumps or initial population of new glance installs that it "while true" is pulled out of the tool box. | 23:13 |
| mgagne | trying to fuel ideas around image replication here. I have to deploy images for yesterday :-/ | 23:13 |
| ndonegan | Oh, another really fun one... | 23:14 |
| mgagne | for now, dumping a bucket of water onto the fuel will be enough but soon I'll need a bigger bucket | 23:14 |
| mgagne | fire* | 23:14 |
| sigmavirus24 | kragniz: github.com/sigmavirus24/glance-toolbelt | 23:15 |
| ndonegan | Haven't checked this past Havanna yet, but when a snapshot of an image is created, the kernel_id and some other properties are set to NULL in MySQL. | 23:15 |
| kragniz | ~github.com/sigmavirus24/glance-toolbelt | 23:15 |
| kragniz | 23:15:30 < ndonegan> Haven't checked this past Havanna yet, but when a snapshot of an image is created, the kernel_id and some other properties are set to NULL in MySQL. | 23:15 |
| ndonegan | glance-replicator dumps these out as None in the json. | 23:15 |
| kragniz | woops | 23:15 |
| sigmavirus24 | ndonegan: weird | 23:16 |
| ndonegan | On load, that None is populated as a string containing 'None' in Mysql. | 23:16 |
| sigmavirus24 | oh wait, you haven't checked it after havanna | 23:16 |
| ndonegan | Causes ALL sorts of fun issues. | 23:16 |
| sigmavirus24 | ndonegan: ouch | 23:16 |
| ndonegan | sigmavirus24: Yet | 23:16 |
| sigmavirus24 | yeah I get that | 23:17 |
| sigmavirus24 | that's a painful bug though, regardless of series | 23:17 |
| ndonegan | Meh, the fix was fairly handy. | 23:18 |
| *** g4rg4m3|_ has quit IRC | 23:18 | |
| *** jasondotstar has joined #openstack-glance | 23:19 | |
| kragniz | ndonegan: you're using HP public cloud, right? | 23:19 |
| *** IlyaE has quit IRC | 23:19 | |
| *** belmoreira has quit IRC | 23:20 | |
| ndonegan | No idea, possiblly some team is. I've only been working on an internal one. | 23:20 |
| kragniz | ah, okay | 23:20 |
| *** annashen has quit IRC | 23:22 | |
| *** IlyaE has joined #openstack-glance | 23:23 | |
| *** peristeri has quit IRC | 23:31 | |
| *** vijendar has quit IRC | 23:33 | |
| *** cpallares has quit IRC | 23:39 | |
| openstackgerrit | Merged openstack/python-glanceclient: Add validation to --property-filter in v1 shell https://review.openstack.org/130264 | 23:40 |
| openstackgerrit | Merged openstack/python-glanceclient: Disable progress bar if image is piped into client https://review.openstack.org/142176 | 23:41 |
| sigmavirus24 | nikhil_k: reminder: API-WG meeting in 15 minutes ish | 23:46 |
| *** flwang has quit IRC | 23:48 | |
| *** annegent_ has joined #openstack-glance | 23:55 | |
| *** ericpeterson has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!