| *** Guest7393 has quit IRC | 00:04 | |
| *** harshs has quit IRC | 00:04 | |
| *** mgarza has quit IRC | 00:11 | |
| *** mingdang1 has joined #openstack-glance | 00:11 | |
| *** chlong has joined #openstack-glance | 00:12 | |
| openstackgerrit | Wayne Okuma proposed openstack/glance: Glance metadef tables need unique constraints. https://review.openstack.org/195820 | 00:21 |
|---|---|---|
| *** tsekiyam_ has joined #openstack-glance | 00:22 | |
| *** dims has joined #openstack-glance | 00:23 | |
| *** jamielennox is now known as jamielennox|away | 00:24 | |
| *** mingdang_ has joined #openstack-glance | 00:25 | |
| *** tsekiyama has quit IRC | 00:26 | |
| *** tsekiyam_ has quit IRC | 00:27 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 00:27 | |
| *** mingdang1 has quit IRC | 00:29 | |
| *** mtanino has quit IRC | 00:30 | |
| *** jamielennox|away is now known as jamielennox | 00:35 | |
| *** jasondotstar has quit IRC | 00:43 | |
| *** ducttape_ has joined #openstack-glance | 00:58 | |
| *** Longgeek has joined #openstack-glance | 01:15 | |
| *** Longgeek has quit IRC | 01:20 | |
| *** spzala has joined #openstack-glance | 01:22 | |
| *** annegentle has joined #openstack-glance | 01:23 | |
| *** dims_ has joined #openstack-glance | 01:24 | |
| *** dims_ has quit IRC | 01:24 | |
| *** mingdang_ has quit IRC | 01:27 | |
| *** dims has quit IRC | 01:27 | |
| *** annegentle has quit IRC | 01:27 | |
| *** hogepodge has quit IRC | 01:29 | |
| *** wayne__ has quit IRC | 01:32 | |
| *** hogepodge has joined #openstack-glance | 01:41 | |
| openstackgerrit | Long Quan Sha proposed openstack/python-glanceclient: Fix the download error when the image locations are blank https://review.openstack.org/199367 | 01:42 |
| *** Ctina has joined #openstack-glance | 01:58 | |
| *** Ctina has quit IRC | 01:58 | |
| *** jskinner has joined #openstack-glance | 02:02 | |
| *** jskinner has quit IRC | 02:04 | |
| *** ericpeterson has joined #openstack-glance | 02:04 | |
| *** ducttape_ has quit IRC | 02:06 | |
| *** ericpeterson has quit IRC | 02:19 | |
| *** jasondotstar has joined #openstack-glance | 02:22 | |
| *** annegentle has joined #openstack-glance | 02:34 | |
| openstackgerrit | Long Quan Sha proposed openstack/glance: Fix downlaod error when the image status is not active https://review.openstack.org/199378 | 02:36 |
| *** bkopilov has quit IRC | 02:36 | |
| *** annegentle has quit IRC | 02:37 | |
| *** changbl has joined #openstack-glance | 02:48 | |
| openstackgerrit | Long Quan Sha proposed openstack/glance: Fix downlaod error when the image status is not active https://review.openstack.org/199378 | 02:54 |
| *** gberginc has joined #openstack-glance | 02:54 | |
| *** jasondotstar has quit IRC | 03:02 | |
| *** gberginc has quit IRC | 03:13 | |
| *** Longgeek has joined #openstack-glance | 03:16 | |
| *** Longgeek has quit IRC | 03:22 | |
| *** harshs has joined #openstack-glance | 03:32 | |
| *** annegentle has joined #openstack-glance | 03:37 | |
| *** annegentle has quit IRC | 03:41 | |
| *** boris-42 has quit IRC | 03:52 | |
| *** ayoung has quit IRC | 03:56 | |
| *** bkopilov has joined #openstack-glance | 03:57 | |
| *** arnaud___ has joined #openstack-glance | 03:58 | |
| *** annashen has joined #openstack-glance | 04:01 | |
| *** zxen has quit IRC | 04:05 | |
| *** zxen has joined #openstack-glance | 04:05 | |
| *** mtreinish has quit IRC | 04:07 | |
| *** gberginc has joined #openstack-glance | 04:09 | |
| *** r-daneel has joined #openstack-glance | 04:13 | |
| *** mtreinish has joined #openstack-glance | 04:16 | |
| *** david-ly_ has joined #openstack-glance | 04:19 | |
| *** sgotliv has joined #openstack-glance | 04:21 | |
| *** david-lyle has quit IRC | 04:22 | |
| *** annashen has quit IRC | 04:28 | |
| *** chlong has quit IRC | 04:35 | |
| *** arnaud___ has quit IRC | 04:37 | |
| *** chlong has joined #openstack-glance | 04:44 | |
| *** arnaud___ has joined #openstack-glance | 04:44 | |
| *** chlong has quit IRC | 04:51 | |
| *** changbl has quit IRC | 04:52 | |
| openstackgerrit | Long Quan Sha proposed openstack/glance: Fix downlaod error when the image status is not active https://review.openstack.org/199378 | 05:06 |
| *** dims has joined #openstack-glance | 05:11 | |
| openstackgerrit | Long Quan Sha proposed openstack/glance: Fix downlaod error when the image status is not active https://review.openstack.org/199378 | 05:14 |
| *** Longgeek has joined #openstack-glance | 05:16 | |
| openstackgerrit | Long Quan Sha proposed openstack/glance: Fix download error when the image status is not active https://review.openstack.org/199378 | 05:17 |
| *** arnaud___ has quit IRC | 05:20 | |
| *** chlong has joined #openstack-glance | 05:20 | |
| *** boris-42 has joined #openstack-glance | 05:21 | |
| *** ajayaa has quit IRC | 05:22 | |
| *** arnaud___ has joined #openstack-glance | 05:25 | |
| *** annegentle has joined #openstack-glance | 05:26 | |
| *** annegentle has quit IRC | 05:30 | |
| *** harshs has quit IRC | 05:34 | |
| *** dims has quit IRC | 05:40 | |
| *** ajayaa has joined #openstack-glance | 05:45 | |
| *** krykowski has joined #openstack-glance | 05:50 | |
| *** lakshmiS_ has joined #openstack-glance | 05:50 | |
| *** ig0r__ has joined #openstack-glance | 05:51 | |
| *** sgotliv has quit IRC | 05:52 | |
| *** sgotliv has joined #openstack-glance | 05:52 | |
| *** ig0r_ has quit IRC | 05:52 | |
| *** zxen has quit IRC | 05:55 | |
| *** nlevinki has joined #openstack-glance | 05:55 | |
| *** zxen has joined #openstack-glance | 05:55 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/glance: Don't choose and set eventlet's hub type by Glance https://review.openstack.org/199402 | 05:57 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Imported Translations from Transifex https://review.openstack.org/197206 | 06:04 |
| *** nlevinki has quit IRC | 06:05 | |
| *** bkopilov has quit IRC | 06:19 | |
| *** tobe has joined #openstack-glance | 06:19 | |
| *** nlevinki has joined #openstack-glance | 06:20 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance_store: Imported Translations from Transifex https://review.openstack.org/194510 | 06:20 |
| *** mingdang1 has joined #openstack-glance | 06:21 | |
| *** abhishekk has joined #openstack-glance | 06:24 | |
| *** groen692 has joined #openstack-glance | 06:27 | |
| *** bkopilov has joined #openstack-glance | 06:32 | |
| *** mingdang1 has quit IRC | 06:45 | |
| *** markus_z has joined #openstack-glance | 06:46 | |
| *** markus_z has quit IRC | 06:58 | |
| *** arnaud___ has quit IRC | 07:02 | |
| *** arnaud___ has joined #openstack-glance | 07:06 | |
| *** arnaud___ has quit IRC | 07:06 | |
| *** annegentle has joined #openstack-glance | 07:14 | |
| *** annegentle has quit IRC | 07:19 | |
| *** e0ne has joined #openstack-glance | 07:25 | |
| *** ccrouch has quit IRC | 07:33 | |
| *** exploreshaifali has joined #openstack-glance | 07:38 | |
| *** dims has joined #openstack-glance | 07:41 | |
| *** dims_ has joined #openstack-glance | 07:42 | |
| *** jistr has joined #openstack-glance | 07:43 | |
| *** dims has quit IRC | 07:45 | |
| *** dims_ has quit IRC | 07:46 | |
| *** tobe has quit IRC | 07:46 | |
| *** e0ne is now known as e0ne_ | 07:49 | |
| *** e0ne_ is now known as e0ne | 07:51 | |
| *** e0ne is now known as e0ne_ | 07:57 | |
| *** e0ne_ is now known as e0ne | 07:59 | |
| *** e0ne has quit IRC | 08:00 | |
| *** tobe has joined #openstack-glance | 08:16 | |
| *** chlong has quit IRC | 08:21 | |
| *** pdb is now known as pbourke | 08:23 | |
| *** tobe has quit IRC | 08:25 | |
| *** tobe has joined #openstack-glance | 08:25 | |
| *** boris-42 has quit IRC | 08:32 | |
| *** belmoreira has joined #openstack-glance | 08:38 | |
| *** tobe has quit IRC | 08:40 | |
| *** dims has joined #openstack-glance | 08:43 | |
| *** dims_ has joined #openstack-glance | 08:44 | |
| *** dims__ has joined #openstack-glance | 08:45 | |
| *** dims___ has joined #openstack-glance | 08:46 | |
| openstackgerrit | Alexey Galkin proposed openstack/glance: Improve code readability in functional test for the WSGIServer https://review.openstack.org/198093 | 08:46 |
| *** dims has quit IRC | 08:47 | |
| *** dims_ has quit IRC | 08:49 | |
| *** dims__ has quit IRC | 08:49 | |
| *** dims___ has quit IRC | 08:50 | |
| *** GeetikaBatra has joined #openstack-glance | 09:02 | |
| *** e0ne has joined #openstack-glance | 09:03 | |
| *** annegentle has joined #openstack-glance | 09:03 | |
| *** annegentle has quit IRC | 09:08 | |
| *** r-daneel has quit IRC | 09:08 | |
| *** belmoreira has quit IRC | 09:12 | |
| *** markus_z has joined #openstack-glance | 09:16 | |
| *** tobe has joined #openstack-glance | 09:20 | |
| *** e0ne is now known as e0ne_ | 09:21 | |
| *** belmoreira has joined #openstack-glance | 09:24 | |
| *** e0ne_ is now known as e0ne | 09:26 | |
| *** mingdang1 has joined #openstack-glance | 09:26 | |
| *** mingdang_ has joined #openstack-glance | 09:29 | |
| *** mingdang1 has quit IRC | 09:32 | |
| *** haypo has joined #openstack-glance | 09:34 | |
| *** bdossant has joined #openstack-glance | 09:36 | |
| *** e0ne is now known as e0ne_ | 09:37 | |
| *** lakshmiS_ has quit IRC | 09:45 | |
| *** jasondotstar has joined #openstack-glance | 09:45 | |
| *** GeetikaBatra has quit IRC | 09:51 | |
| *** GeetikaBatra has joined #openstack-glance | 09:52 | |
| *** aix has joined #openstack-glance | 09:54 | |
| *** mingdang_ has quit IRC | 09:55 | |
| *** e0ne_ is now known as e0ne | 10:00 | |
| *** e0ne is now known as e0ne_ | 10:11 | |
| *** GeetikaBatra is now known as GB21 | 10:13 | |
| *** e0ne_ is now known as e0ne | 10:13 | |
| *** tobe has quit IRC | 10:29 | |
| *** GB21 has quit IRC | 10:36 | |
| *** e0ne is now known as e0ne_ | 10:40 | |
| *** e0ne_ is now known as e0ne | 10:42 | |
| *** annegentle has joined #openstack-glance | 10:52 | |
| *** mingdang1 has joined #openstack-glance | 10:56 | |
| *** annegentle has quit IRC | 10:56 | |
| *** vijendar has joined #openstack-glance | 10:59 | |
| *** mingdang1 has quit IRC | 11:00 | |
| *** GeetikaBatra has joined #openstack-glance | 11:12 | |
| *** GeetikaBatra has quit IRC | 11:13 | |
| *** GeetikaBatra has joined #openstack-glance | 11:16 | |
| *** GeetikaBatra is now known as GB21 | 11:25 | |
| *** mingdang1 has joined #openstack-glance | 11:29 | |
| *** krykowski_ has joined #openstack-glance | 11:32 | |
| *** krykowski has quit IRC | 11:34 | |
| *** e0ne is now known as e0ne_ | 11:34 | |
| *** e0ne_ is now known as e0ne | 11:35 | |
| *** e0ne has quit IRC | 11:38 | |
| *** shosagoudar has joined #openstack-glance | 11:39 | |
| *** Longgeek has quit IRC | 11:41 | |
| shosagoudar | hey guys, glance is failing to give result for "glance -d image-list" | 11:41 |
| *** ducttape_ has joined #openstack-glance | 11:41 | |
| *** Longgeek has joined #openstack-glance | 11:42 | |
| *** Longgeek has quit IRC | 11:42 | |
| *** Longgeek has joined #openstack-glance | 11:42 | |
| shosagoudar | from /var/log/glance/api.log, its saying "WARNING keystonemiddleware.auth_token [-] Authorization failed for token" | 11:42 |
| shosagoudar | what could be the issue/how to debug? | 11:43 |
| *** Longgeek has quit IRC | 11:43 | |
| *** GB21 has quit IRC | 11:46 | |
| *** GeetikaBatra has joined #openstack-glance | 11:46 | |
| *** dims has joined #openstack-glance | 11:47 | |
| *** zul has quit IRC | 11:48 | |
| *** dims_ has joined #openstack-glance | 11:48 | |
| *** dims__ has joined #openstack-glance | 11:49 | |
| *** GeetikaBatra_ has joined #openstack-glance | 11:50 | |
| *** exploreshaifali has quit IRC | 11:51 | |
| *** GeetikaBatra has quit IRC | 11:52 | |
| *** dims has quit IRC | 11:52 | |
| *** dims_ has quit IRC | 11:53 | |
| *** dims__ has quit IRC | 11:54 | |
| *** ducttape_ has quit IRC | 11:56 | |
| jokke_ | yelling out for help here as well. Any ideas why our kilo tests are failing in glance_store? http://logs.openstack.org/36/199136/1/check/gate-glance_store-python27/7d74387/console.html#_2015-07-07_14_57_14_258 | 11:57 |
| kragniz | jokke_: I'm going to blame pbr | 11:58 |
| kragniz | but that might be unfair | 11:59 |
| kragniz | shosagoudar: can you get a token from keystone using the creds you're using to talk to glance? | 12:00 |
| jokke_ | kragniz: _something_ seems to have requirement t oslo.config >=1.11.0 but I can't find what it is :( | 12:00 |
| *** krykowski has joined #openstack-glance | 12:04 | |
| *** ajayaa has quit IRC | 12:06 | |
| *** krykowski_ has quit IRC | 12:07 | |
| shosagoudar | kragniz : I am getting the following error when tried to issue the token using "openstack token issue" | 12:10 |
| shosagoudar | "openstack token issue --os-auth-type password --os-user-domain-name default --os-user-domain-id default --os-username glance | 12:10 |
| shosagoudar | ERROR: openstack Invalid username or password (Disable debug mode to suppress these details.) (HTTP 401) (Request-ID: req-e71b9dde-1fa0-4be4-b8a3-e4066e301dde)" | 12:10 |
| shosagoudar | but the same works for admin user, | 12:11 |
| shosagoudar | as its credentials are sources using "source admin.rc" | 12:11 |
| shosagoudar | is it the issue? | 12:11 |
| shosagoudar | the above command to get the token for glance user failed even without asking for password | 12:12 |
| kragniz | it sounds like a keystone issue, or the credentials are wrong or not set up properly | 12:13 |
| shosagoudar | but keystone is working fine for admin user | 12:15 |
| kragniz | you're probably using the wrong credentials, then | 12:16 |
| *** belmoreira has quit IRC | 12:17 | |
| *** krykowski has quit IRC | 12:19 | |
| *** gberginc has quit IRC | 12:19 | |
| shosagoudar | let me update the password then for glance | 12:20 |
| shosagoudar | kragniz: you are correct, openstack-keystone service has died, | 12:22 |
| *** belmoreira has joined #openstack-glance | 12:23 | |
| *** Longgeek has joined #openstack-glance | 12:25 | |
| *** jasondotstar has quit IRC | 12:25 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/glance: test with evenlet monkey_patch https://review.openstack.org/199539 | 12:25 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/glance: test with eventlet monkey_patch https://review.openstack.org/199539 | 12:28 |
| *** exploreshaifali has joined #openstack-glance | 12:28 | |
| *** GeetikaBatra has joined #openstack-glance | 12:29 | |
| *** annegentle has joined #openstack-glance | 12:30 | |
| *** zul has joined #openstack-glance | 12:30 | |
| *** chlong has joined #openstack-glance | 12:31 | |
| jokke_ | ok ... stable/kilo for _store and client should be on track if we get the changes on flight in and those dependency issues in _store gate fixed | 12:31 |
| *** GeetikaBatra_ has quit IRC | 12:32 | |
| *** GeetikaBatra has quit IRC | 12:39 | |
| *** mingdang1 has quit IRC | 12:43 | |
| *** abhishekk has quit IRC | 12:44 | |
| *** GeetikaBatra has joined #openstack-glance | 12:44 | |
| *** shosagoudar has quit IRC | 12:47 | |
| *** ninag has joined #openstack-glance | 12:49 | |
| *** dims has joined #openstack-glance | 12:50 | |
| *** dims has quit IRC | 12:54 | |
| *** shosagoudar has joined #openstack-glance | 12:55 | |
| GeetikaBatra | hello everyone, I am trying to run image scrubbers, apparently it was not working, so as I traced it, it says module' object has no attribute 'register_options' | 12:55 |
| GeetikaBatra | I searched a little bit, this is the case of mutual top-level imports | 12:56 |
| GeetikaBatra | can anyone please help me on this | 12:56 |
| openstackgerrit | Long Quan Sha proposed openstack/python-glanceclient: Fix the download error when the image locations are blank https://review.openstack.org/199549 | 12:57 |
| *** mfedosin has quit IRC | 12:59 | |
| *** ninag has quit IRC | 13:00 | |
| *** mfisch has quit IRC | 13:00 | |
| *** ninag has joined #openstack-glance | 13:00 | |
| *** mfedosin has joined #openstack-glance | 13:01 | |
| *** julim has joined #openstack-glance | 13:01 | |
| *** GeetikaBatra has quit IRC | 13:02 | |
| *** cdelatte has joined #openstack-glance | 13:03 | |
| *** jasondotstar has joined #openstack-glance | 13:05 | |
| *** peristeri has joined #openstack-glance | 13:05 | |
| *** GeetikaBatra has joined #openstack-glance | 13:13 | |
| *** shosagoudar has quit IRC | 13:16 | |
| mclaren | jokke_: I'm motivated to look into that | 13:17 |
| *** changbl has joined #openstack-glance | 13:19 | |
| *** ayoung has joined #openstack-glance | 13:20 | |
| *** dims has joined #openstack-glance | 13:22 | |
| *** GeetikaBatra_ has joined #openstack-glance | 13:25 | |
| *** GeetikaBatra has quit IRC | 13:27 | |
| jokke_ | mclaren: please let me know if you find the cause ... I'm trying to dig into that as well | 13:32 |
| *** shosagoudar has joined #openstack-glance | 13:32 | |
| mclaren | sure | 13:33 |
| *** ducttape_ has joined #openstack-glance | 13:34 | |
| mclaren | jokke_: wondering if we need to explicitly set/bound python-keystone client in requirements.txt | 13:37 |
| mclaren | jokke_: yeah, I think that fixes it. | 13:38 |
| *** annegentle has quit IRC | 13:45 | |
| *** annegentle has joined #openstack-glance | 13:45 | |
| mclaren | it looks very similar to https://bugs.launchpad.net/designate/+bug/1463988 | 13:48 |
| openstack | Launchpad bug 1463988 in keystonemiddleware kilo "requirements.txt out of sync for stable/kilo" [Low,Confirmed] | 13:48 |
| *** vijendar has quit IRC | 13:51 | |
| *** vijendar has joined #openstack-glance | 13:51 | |
| jokke_ | we need to get Doug unblock this https://review.openstack.org/#/c/173977/ | 13:51 |
| jokke_ | mclaren: lets see if we get that sync to pytho-cinderclient in and if that fixes our issue as I suspect | 13:53 |
| *** annegentle has quit IRC | 14:02 | |
| *** bkopilov has quit IRC | 14:03 | |
| *** mtanino has joined #openstack-glance | 14:04 | |
| *** edmondsw has joined #openstack-glance | 14:05 | |
| *** GeetikaBatra_ has quit IRC | 14:09 | |
| *** shosagoudar has quit IRC | 14:17 | |
| *** mfisch has joined #openstack-glance | 14:19 | |
| *** mfisch has quit IRC | 14:20 | |
| *** mfisch has joined #openstack-glance | 14:20 | |
| *** pbourke has quit IRC | 14:20 | |
| *** pbourke has joined #openstack-glance | 14:20 | |
| *** vijendar has quit IRC | 14:22 | |
| *** vijendar has joined #openstack-glance | 14:22 | |
| *** vijendar1 has joined #openstack-glance | 14:26 | |
| *** annegentle has joined #openstack-glance | 14:27 | |
| *** vijendar has quit IRC | 14:27 | |
| *** kebray has joined #openstack-glance | 14:33 | |
| *** flwang has quit IRC | 14:40 | |
| *** nlevinki has quit IRC | 14:44 | |
| *** ajayaa has joined #openstack-glance | 14:45 | |
| *** GeetikaBatra has joined #openstack-glance | 14:47 | |
| *** bpoulos has joined #openstack-glance | 14:47 | |
| openstackgerrit | Mike Fedosin proposed openstack/glance_store: Fix expired token issue while uploading chunks to swift https://review.openstack.org/199049 | 14:48 |
| openstackgerrit | Mike Fedosin proposed openstack/glance_store: Fix expired token issue while uploading chunks to swift https://review.openstack.org/199049 | 14:52 |
| *** belmoreira has quit IRC | 14:53 | |
| *** GeetikaBatra_ has joined #openstack-glance | 14:58 | |
| *** GeetikaBatra_ is now known as GB21 | 14:58 | |
| *** GeetikaBatra has quit IRC | 14:58 | |
| mfedosin | sigmavirus24_awa, jokke_ , mclaren , hi, folks! I really want to hear your opinions about https://review.openstack.org/#/c/199049/3 it fixes a very nasty bug with token expiration during upload that we encountered many times. | 14:58 |
| *** ninag has quit IRC | 15:00 | |
| mfedosin | ativelkov, you're welcome too :) ^ | 15:00 |
| *** ninag has joined #openstack-glance | 15:01 | |
| nikhil_k | hemanthm: ^ | 15:02 |
| * hemanthm looking | 15:03 | |
| *** dims has quit IRC | 15:04 | |
| *** dims has joined #openstack-glance | 15:05 | |
| mfedosin | nikhil_k, hemanthm \o/ review it too, please | 15:05 |
| openstackgerrit | Long Quan Sha proposed openstack/python-glanceclient: Fix the download error when the image locations are blank https://review.openstack.org/199549 | 15:06 |
| jokke_ | mfedosin: elegant | 15:07 |
| *** ninag has quit IRC | 15:07 | |
| *** shosagoudar has joined #openstack-glance | 15:08 | |
| jokke_ | mfedosin: we could use that with swift+config as well, no? | 15:08 |
| *** dims has quit IRC | 15:10 | |
| hemanthm | mfedosin: we've been bitten by this a lot. I had a quick look, it looks ok to me. However, I don't necessarily like to include re-aunthenticating logic here because it really belongs in the client. | 15:10 |
| *** bkopilov has joined #openstack-glance | 15:10 | |
| hemanthm | mfedosin: swiftclient does re-authenticate though. But, it fails miserably to resume the upload after re-authenticating when its working with a stream | 15:10 |
| mfedosin | jokke_, I think we can | 15:10 |
| hemanthm | it works perfectly with a file though because it can seek and tell | 15:11 |
| hemanthm | let me point you to a patch that's trying to solve the same problem but in a different way | 15:11 |
| mfedosin | hemanthm, yes please | 15:11 |
| jokke_ | hemanthm: that's what I was thinking as well | 15:11 |
| hemanthm | mfedosin, jokke_: https://review.openstack.org/#/c/120891/ | 15:12 |
| hemanthm | that patch needs some love. I can start the work if there is interest | 15:13 |
| *** mclaren has quit IRC | 15:14 | |
| *** bkopilov has quit IRC | 15:15 | |
| *** primechuck has joined #openstack-glance | 15:15 | |
| jokke_ | hemanthm: Honestly, what Mike is proposing is a bugfix, that dependency patch of the review you linked is still new feature IMO and I think we should get spec out for that if someone wants to revive it | 15:17 |
| jokke_ | I'm personally not huge fan of the buffered approach as that brings us the thread pool issues back | 15:18 |
| *** boris-42 has joined #openstack-glance | 15:21 | |
| mfedosin | hemanthm, yeah, very cool! but i'm interested in one thing. Swift doesn't know that it's a stream, because it accepts separated objects each time. so if there was a Unauthorized error, we don't start chunk upload and nothing changes for swift and the stream and the number of uploaded bytes. | 15:22 |
| mfedosin | I'm going to ask Alexey Galkin to test it very well today and tomorrow | 15:22 |
| mfedosin | with different configurations and in HA mode | 15:22 |
| *** dims has joined #openstack-glance | 15:23 | |
| mfedosin | I will tell you the results when it's done :) | 15:23 |
| hemanthm | mfedosin: cool, thanks! | 15:23 |
| hemanthm | jokke_: +1, I'm not necessarily saying 'no' to this approach. Even with bufferedReader, it's not that everyone will use it. It's configurable. But, just saying that I'm not necessarily a fan of having re-authenticating logic in there, which belongs in the client to me. Every case is different anyway | 15:25 |
| *** tsekiyama has joined #openstack-glance | 15:27 | |
| *** bkopilov has joined #openstack-glance | 15:27 | |
| jokke_ | hemanthm: I hear ya! I think we need to accept that if you use stream as input you will loose some benefits of static file | 15:28 |
| jokke_ | hemanthm: that's just reality, but we should not have that as an issue as the failure comes at the point when we try to upload next chunk, no? So our data point should be always on the correct place? | 15:30 |
| *** spzala has quit IRC | 15:30 | |
| jokke_ | hemanthm: so do you have an idea where the swiftclient actually loses the sync on that? | 15:31 |
| hemanthm | jokke_: +1 | 15:31 |
| hemanthm | jokke_: I tries to seek on the stream, which is not available. Let me link you to the code | 15:32 |
| jokke_ | aha | 15:32 |
| jokke_ | ok, that makes sense | 15:32 |
| *** vijendar1 has quit IRC | 15:33 | |
| *** vijendar has joined #openstack-glance | 15:33 | |
| jokke_ | mfedosin: don't we loose the chunk we're uploading on your approach as well? | 15:35 |
| jokke_ | mfedosin: we do reauthenticate, but we do not resend the chunk after that, we continue to the next chunk | 15:35 |
| hemanthm | jokke_: +1, good point | 15:36 |
| mfedosin | jokke_, I thought we resent it, but I need to check it again :) thanks | 15:37 |
| jokke_ | mfedosin: flagged it in the review | 15:38 |
| mfedosin | technically we can't start upload if it was a error... my test checks it btw - that there were 7 uploaded chunks | 15:39 |
| *** e0ne has joined #openstack-glance | 15:40 | |
| *** zul has quit IRC | 15:41 | |
| *** david-ly_ is now known as david-lyle | 15:42 | |
| *** exploreshaifali has quit IRC | 15:42 | |
| jokke_ | mfedosin: hmm-m ... that's tricky ... that is expecting that the swiftclient does not read anything from the input on that failure scenario | 15:43 |
| jokke_ | mfedosin: have you pulled the image down and verified checksum that it actually matches with the original, as I'm pretty sure that would corrupt the data. | 15:45 |
| *** annegentle has quit IRC | 15:45 | |
| *** annegentle has joined #openstack-glance | 15:46 | |
| *** groen692 has quit IRC | 15:47 | |
| mfedosin | jokke_, I'm not sure with swiftclient at all :) I'm currently talking with Alexey, he's going to start testing it right now. | 15:47 |
| *** jasondotstar has quit IRC | 15:48 | |
| jokke_ | now I don't know swiftclient well enough so it might depend when the token expires. If they do it same way we do that they create the object and then upload it it's fine as long as the token expires between the two objects but we're again screwed if it expires between creating the object and uploading it as that would read stuff from the reader and corrupt again | 15:48 |
| jokke_ | damn, this is actually really nasty scenario to solve without buffering of some kind | 15:49 |
| *** spzala__ has joined #openstack-glance | 15:55 | |
| jokke_ | I'm more and more leaning towards the trusts and renewing the trust-token between every chunk | 15:56 |
| *** jistr has quit IRC | 15:56 | |
| *** jasondotstar has joined #openstack-glance | 15:58 | |
| mfedosin | jokke_, this patch is the last desperate attempt to avoid the use of trusts :D | 15:59 |
| *** spzala__ is now known as spzala | 15:59 | |
| jokke_ | mfedosin: I'm more and more thinking we shouldn't avoid them ... specially as this change does not even address the issue for normal multitenant case | 16:00 |
| mfedosin | are there some alternatives? | 16:01 |
| mfedosin | I think no | 16:01 |
| mfedosin | so, yes, you're right, we should start moving in that way | 16:02 |
| *** kebray has quit IRC | 16:04 | |
| *** zul has joined #openstack-glance | 16:08 | |
| *** dontalton has joined #openstack-glance | 16:10 | |
| *** bdossant has quit IRC | 16:13 | |
| *** arnaud___ has joined #openstack-glance | 16:14 | |
| *** markus_z has quit IRC | 16:15 | |
| *** dims has quit IRC | 16:15 | |
| *** dims has joined #openstack-glance | 16:16 | |
| *** flwang has joined #openstack-glance | 16:18 | |
| *** dims has quit IRC | 16:21 | |
| *** arnaud___ has quit IRC | 16:22 | |
| mfedosin | kragniz, can you look? https://review.openstack.org/#/c/199061/ we have an issue with bad import and many people are struggling from this bug: https://review.openstack.org/#/c/199549/ for example | 16:24 |
| *** jasondotstar has quit IRC | 16:25 | |
| *** GeetikaBatra has joined #openstack-glance | 16:27 | |
| *** changbl has quit IRC | 16:29 | |
| *** jaypipes has quit IRC | 16:29 | |
| *** kebray has joined #openstack-glance | 16:29 | |
| *** GB21 has quit IRC | 16:30 | |
| *** ajayaa has quit IRC | 16:33 | |
| *** dims has joined #openstack-glance | 16:40 | |
| *** dontalton is now known as bitblt | 16:41 | |
| *** bitblt has quit IRC | 16:41 | |
| *** bitblt has joined #openstack-glance | 16:41 | |
| *** bitblt has quit IRC | 16:42 | |
| *** primechuck has quit IRC | 16:44 | |
| *** arnaud___ has joined #openstack-glance | 16:45 | |
| openstackgerrit | Merged openstack/glance: Improve code readability in functional test for the WSGIServer https://review.openstack.org/198093 | 16:46 |
| *** harshs has joined #openstack-glance | 16:47 | |
| *** arnaud___ has quit IRC | 16:53 | |
| *** ayoung has quit IRC | 16:59 | |
| *** ByPasS has joined #openstack-glance | 17:01 | |
| jokke_ | mfedosin: still around? | 17:04 |
| ByPasS | Hi guys, having a question about snapshot quotas. In my setup I'm using ceph as a backend. Pool being user from creating a new VM is VMs. When I do a nova image-create (snapshop) the snapshot ends up in the image pool which is being used by glance and can list snapshots from glance if I look in the properties of the image ( image type snapshot ). What are the actual quotas being used for the amount of snapshots and/or images that can be created | 17:05 |
| ByPasS | by a tenant ? | 17:05 |
| mfedosin | jokke_, yep, I'm here | 17:06 |
| jokke_ | ByPasS: honestly I didn't know we have snapshot quotas | 17:06 |
| jokke_ | mfedosin: I'm trying to understand that piece of code | 17:06 |
| jokke_ | ref the upload to swift | 17:06 |
| haypo | hey, i'm still looking for reviews on my glance_store patches: https://review.openstack.org/#/c/184601/ & https://review.openstack.org/#/c/184603/ | 17:07 |
| mfedosin | jokke_, what line #? | 17:07 |
| ByPasS | there is a quota section but it doesnt seem to be related directly on the actual amount or not clear to me (image_member_quota / image_property_quota / image_tag_quota / image_location_quota / user_storage_quota) | 17:08 |
| jokke_ | mfedosin: currently I'm amazed around 546 | 17:08 |
| jokke_ | mfedosin: so how the ChunkReader knows where to read from? | 17:08 |
| jokke_ | mfedosin: so if I understand correctly it just waks through the file descriptor | 17:09 |
| jokke_ | walks | 17:09 |
| nikhil_k | ByPasS: there are not quotas on that | 17:09 |
| nikhil_k | no* | 17:09 |
| jokke_ | mfedosin: and even we create new one every round of that while loop, it just continues where the marker is in that fd | 17:10 |
| jokke_ | mfedosin: that would be on L955ish + | 17:13 |
| mfedosin | jokke_, yes. to change the number of read bytes swift must execute 'read', but it doesn't happen, so the position must be the same, I think | 17:14 |
| jokke_ | mfedosin: it must read something for that request ... ref my comment failure happening on upload vs. object creation | 17:15 |
| jokke_ | mfedosin: so if it fails on upload swift has read x amount of data to send with the request, we move forward on fd and never seek back, we retry and that chunk of data (rather than swift large object chunk) is lost | 17:16 |
| jokke_ | == corrupted image | 17:16 |
| mfedosin | okay, I got you. It's easy to fix anyway. | 17:17 |
| jokke_ | is it? | 17:17 |
| jokke_ | isn't the fd stream rather that static file? | 17:17 |
| mfedosin | the obvious thing is to check token expiration before each chunk upload | 17:18 |
| *** mingdang1 has joined #openstack-glance | 17:18 | |
| mfedosin | and reauth if it happened | 17:19 |
| jokke_ | mfedosin: yes, and make sure that the lifetime of that token is enough to be 100% sure we can send that request within the time | 17:19 |
| mfedosin | so no upload - no problems | 17:19 |
| *** primechuck has joined #openstack-glance | 17:19 | |
| jokke_ | ++ I like your way of thinking ;) | 17:19 |
| jokke_ | just wanted to make sure I'm not getting crazy with this corrupted image scenario of mine ;) | 17:20 |
| hemanthm | mfedosin: is that going to add one extra auth call for every chunk? | 17:20 |
| jokke_ | hemanthm: yup | 17:20 |
| hemanthm | how about caching the expiration timestamp instead? | 17:21 |
| * jokke_ hopes that we actually get the expiration time from the keystone | 17:21 | |
| jokke_ | not just True|False if it's valid | 17:21 |
| mfedosin | hemanthm, it's not reauth on each chunk, just a small check. it doesn't have big overhead | 17:21 |
| hemanthm | mfedosin: must be a head call right? | 17:22 |
| mfedosin | hemanthm, yes, why not? | 17:22 |
| *** mingdang1 has quit IRC | 17:22 | |
| hemanthm | it may result in too many requests | 17:23 |
| mfedosin | btw, do we have an access to token exp time param from glance_store? | 17:23 |
| hemanthm | mfedosin: chunk size in configurable | 17:23 |
| hemanthm | so if you choose a smalle chunk size, it will result in too many auth requests | 17:23 |
| mfedosin | I mean can we know its value? | 17:23 |
| hemanthm | multiply that by the number of simultaneous images that maybe happening | 17:23 |
| mfedosin | by default it's 200mb | 17:24 |
| mfedosin | what the reason to make it lower? | 17:24 |
| jokke_ | hemanthm: has a point ... we might hit to the ratelimiting on big deployments ... this is good point to improve documentation and advice those big environment deployers to make sure their chunk size is large enough to avoid that | 17:25 |
| mfedosin | jokke_, ++ | 17:26 |
| hemanthm | mfedosin: there isn't one specific reason why some operators may choose to go with smaller chunks | 17:27 |
| jokke_ | I think default large object size is something around 5GB so that 200MB default is bit ridiculous anyways | 17:27 |
| jokke_ | hemanthm: but this is good reason for them to go bigger ;) | 17:27 |
| hemanthm | for instance: parallel downloads | 17:27 |
| *** bkopilov has quit IRC | 17:28 | |
| hemanthm | jokke_: yes but on the other hand, if you have a bigger chunk, you are probably wasting the bandwidth on errors | 17:28 |
| *** e0ne has quit IRC | 17:28 | |
| hemanthm | retrying a 1GB chunk vs 200mb or whatever | 17:29 |
| hemanthm | just trade-offs really | 17:29 |
| jokke_ | hemanthm: well at the moment we're wasting the bandwidth of everything uploaded thus far. I think we cannot go worse | 17:29 |
| hemanthm | jokke_: yes, yes, I'm just saying there are reasons why one could go either way | 17:29 |
| mfedosin | I talked with keystone guys, they told that token cached in their middleware | 17:30 |
| jokke_ | hemanthm: thus the advice "if you are (likely) hitting the ratelimits, increase the chunk size or live with it" Problem solved/tradeoff made | 17:30 |
| hemanthm | jokke_: its not really a trade-off in my view, it appers to be an upper cap | 17:31 |
| jokke_ | mfedosin: so is there some way we can query the token metadata from the middleware? | 17:31 |
| mfedosin | no, but there are two options: 1. normally (in case of big chunk size) we send a head request to keystone and check token expiration | 17:32 |
| *** spzala_ has joined #openstack-glance | 17:33 | |
| mfedosin | 2. in case of small chunks token cached in keystone middleware for 5 mins | 17:33 |
| jokke_ | mfedosin: will that query return the expiration time or will it return True|False depending if the token is still valid? | 17:33 |
| mfedosin | so we can easily checks it's status without any request to keystone | 17:34 |
| mfedosin | true|false | 17:34 |
| jokke_ | so that is absolutely pointless then | 17:34 |
| jokke_ | as it will not remove our problem anywhere | 17:34 |
| *** spzala has quit IRC | 17:35 | |
| jokke_ | as we will still do the new request without knowing that we have time to send the request while the token is still valid | 17:35 |
| *** spzala_ is now known as spzala | 17:35 | |
| mfedosin | jokke_, agree, there is that possibility | 17:35 |
| *** bkopilov has joined #openstack-glance | 17:36 | |
| jokke_ | but if we move to trusts, we know exactly how long the trust is valid, we can extend it by ourselves and all these issues are just gone | 17:36 |
| jokke_ | + it works on all of our tenancy modes not just single or swift+config | 17:37 |
| mfedosin | jokke_, can not argue :) but anyway I will think about other possible solutions... | 17:39 |
| *** dims has quit IRC | 17:39 | |
| *** dims has joined #openstack-glance | 17:39 | |
| mfedosin | jokke_, hemanth excuse me, I have to go. see you tomorrow! thanks for support, folks | 17:40 |
| jokke_ | mfedosin: sure, good night! o/~ | 17:40 |
| hemanthm | mfedosin: later, have a good one | 17:41 |
| *** vijendar1 has joined #openstack-glance | 17:41 | |
| *** vijendar has quit IRC | 17:43 | |
| *** vijendar1 has quit IRC | 17:43 | |
| *** vijendar has joined #openstack-glance | 17:43 | |
| *** exploreshaifali has joined #openstack-glance | 17:48 | |
| *** delattec has joined #openstack-glance | 17:52 | |
| *** annashen has joined #openstack-glance | 17:53 | |
| *** arnaud___ has joined #openstack-glance | 17:54 | |
| *** bkopilov has quit IRC | 17:54 | |
| *** cdelatte has quit IRC | 17:56 | |
| *** Longgeek has quit IRC | 17:57 | |
| *** Longgeek has joined #openstack-glance | 17:57 | |
| *** Longgeek has quit IRC | 17:57 | |
| *** aix has quit IRC | 17:57 | |
| *** Longgeek has joined #openstack-glance | 17:59 | |
| *** arnaud___ has quit IRC | 18:00 | |
| *** ducttape_ has quit IRC | 18:00 | |
| *** spzala_ has joined #openstack-glance | 18:06 | |
| *** kebray has quit IRC | 18:06 | |
| *** spzala has quit IRC | 18:08 | |
| *** spzala_ is now known as spzala | 18:08 | |
| *** bkopilov has joined #openstack-glance | 18:09 | |
| *** ayoung has joined #openstack-glance | 18:11 | |
| *** ducttape_ has joined #openstack-glance | 18:13 | |
| *** GeetikaBatra is now known as GB21 | 18:17 | |
| *** ajayaa has joined #openstack-glance | 18:20 | |
| *** boris-42 has quit IRC | 18:22 | |
| openstackgerrit | Chris St. Pierre proposed openstack/glance: Purge dead file-backed scrubber queue code https://review.openstack.org/161051 | 18:23 |
| *** kebray has joined #openstack-glance | 18:25 | |
| *** exploreshaifali has quit IRC | 18:36 | |
| *** jamielennox is now known as jamielennox|away | 18:47 | |
| *** TravT has quit IRC | 18:48 | |
| *** TravT has joined #openstack-glance | 18:48 | |
| *** jamielennox|away is now known as jamielennox | 18:54 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 18:56 | |
| *** arnaud___ has joined #openstack-glance | 18:56 | |
| *** arnaud___ has quit IRC | 19:01 | |
| shosagoudar | kragniz: This entry @ /etc/glance/glance-api.conf => [keystone_authtoken] helped to fix the issue | 19:02 |
| shosagoudar | auth_host = controller | 19:02 |
| shosagoudar | auth_port = 35357 | 19:02 |
| shosagoudar | auth_protocol = http | 19:02 |
| shosagoudar | admin_tenant_name = service | 19:02 |
| shosagoudar | admin_user = XXXXXX | 19:02 |
| shosagoudar | admin_password = XXXXXX | 19:02 |
| shosagoudar | auth_uri = __http://controller:5000 | 19:02 |
| *** MentalRay has joined #openstack-glance | 19:08 | |
| MentalRay | Hello. Is anyone has a juno or kilo setup running on 10gbe and has tested the snapshot speed? | 19:08 |
| *** e0ne has joined #openstack-glance | 19:09 | |
| *** bitblt has joined #openstack-glance | 19:13 | |
| *** ajayaa has quit IRC | 19:22 | |
| *** r-daneel has joined #openstack-glance | 19:23 | |
| *** kebray has quit IRC | 19:28 | |
| nikhil_k | GB21: ^ | 19:33 |
| *** zul has quit IRC | 19:34 | |
| *** shosagoudar has quit IRC | 19:34 | |
| *** kebray has joined #openstack-glance | 19:36 | |
| *** changbl has joined #openstack-glance | 19:37 | |
| *** e0ne has quit IRC | 19:47 | |
| *** ajayaa has joined #openstack-glance | 19:47 | |
| *** arnaud___ has joined #openstack-glance | 19:57 | |
| *** bpoulos has left #openstack-glance | 20:00 | |
| *** arnaud___ has quit IRC | 20:02 | |
| *** exploreshaifali has joined #openstack-glance | 20:05 | |
| *** vilobhmm has joined #openstack-glance | 20:08 | |
| *** openstackgerrit has quit IRC | 20:10 | |
| *** openstackgerrit has joined #openstack-glance | 20:10 | |
| *** e0ne has joined #openstack-glance | 20:11 | |
| *** e0ne has quit IRC | 20:12 | |
| *** wayne__ has joined #openstack-glance | 20:17 | |
| *** belmoreira has joined #openstack-glance | 20:18 | |
| *** GB21 has quit IRC | 20:19 | |
| *** kebray has quit IRC | 20:20 | |
| *** kebray has joined #openstack-glance | 20:23 | |
| *** sgotliv has quit IRC | 20:26 | |
| *** shosagoudar has joined #openstack-glance | 20:30 | |
| *** shosagoudar has quit IRC | 20:35 | |
| *** jasondotstar has joined #openstack-glance | 20:40 | |
| *** wayne__ has quit IRC | 20:42 | |
| *** kebray has quit IRC | 20:43 | |
| *** kebray has joined #openstack-glance | 20:48 | |
| *** spzala has quit IRC | 20:50 | |
| *** annegentle has quit IRC | 20:55 | |
| *** jasondotstar has quit IRC | 21:08 | |
| *** ducttape_ has quit IRC | 21:15 | |
| *** annegentle has joined #openstack-glance | 21:25 | |
| *** ducttape_ has joined #openstack-glance | 21:26 | |
| *** e0ne has joined #openstack-glance | 21:29 | |
| *** fifieldt has quit IRC | 21:30 | |
| *** TravT has quit IRC | 21:33 | |
| *** belmoreira has quit IRC | 21:34 | |
| *** ajayaa has quit IRC | 21:36 | |
| *** e0ne has quit IRC | 21:37 | |
| *** TravT has joined #openstack-glance | 21:39 | |
| *** TravT has quit IRC | 21:42 | |
| *** fifieldt has joined #openstack-glance | 21:43 | |
| *** arnaud___ has joined #openstack-glance | 21:45 | |
| *** jasondotstar has joined #openstack-glance | 21:48 | |
| *** annegentle has quit IRC | 21:49 | |
| *** annegentle has joined #openstack-glance | 21:49 | |
| *** arnaud___ has quit IRC | 21:50 | |
| *** jasondotstar has quit IRC | 22:00 | |
| openstackgerrit | Ian Cordasco proposed openstack/python-glanceclient: Remove custom SSL compression handling https://review.openstack.org/189344 | 22:02 |
| *** mingdang1 has joined #openstack-glance | 22:04 | |
| *** boris-42 has joined #openstack-glance | 22:04 | |
| *** spzala has joined #openstack-glance | 22:04 | |
| *** ducttape_ has quit IRC | 22:07 | |
| *** TravT has joined #openstack-glance | 22:07 | |
| jamielennox | kragniz: around? | 22:08 |
| *** peristeri has quit IRC | 22:10 | |
| *** tpeoples has quit IRC | 22:11 | |
| *** kebray has quit IRC | 22:16 | |
| *** MentalRay has quit IRC | 22:18 | |
| *** ducttape_ has joined #openstack-glance | 22:18 | |
| *** annegentle has quit IRC | 22:21 | |
| *** mingdang1 has quit IRC | 22:22 | |
| *** Guest69012 has joined #openstack-glance | 22:23 | |
| *** ericpeterson has joined #openstack-glance | 22:24 | |
| *** ducttape_ has quit IRC | 22:24 | |
| *** ducttape_ has joined #openstack-glance | 22:25 | |
| *** ericpeterson has quit IRC | 22:25 | |
| *** shosagoudar has joined #openstack-glance | 22:31 | |
| *** Guest69012 has quit IRC | 22:32 | |
| *** shosagoudar has quit IRC | 22:35 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:35 | |
| *** primechuck has quit IRC | 22:47 | |
| *** wayne__ has joined #openstack-glance | 22:47 | |
| *** edmondsw has quit IRC | 22:50 | |
| *** boris-42 has quit IRC | 22:56 | |
| *** boris-42 has joined #openstack-glance | 22:57 | |
| *** ducttape_ has quit IRC | 23:09 | |
| *** annashen has quit IRC | 23:12 | |
| *** shalq has quit IRC | 23:17 | |
| *** shalq has joined #openstack-glance | 23:17 | |
| *** ByPasS has quit IRC | 23:18 | |
| *** exploreshaifali has quit IRC | 23:21 | |
| *** zul has joined #openstack-glance | 23:29 | |
| *** arnaud___ has joined #openstack-glance | 23:33 | |
| *** arnaud___ has quit IRC | 23:38 | |
| *** TravT has quit IRC | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!