| *** TravT has joined #openstack-glance | 00:17 | |
| *** TravT has quit IRC | 00:21 | |
| *** nellysmitt has joined #openstack-glance | 01:08 | |
| *** nellysmitt has quit IRC | 01:12 | |
| *** buzztroll has joined #openstack-glance | 01:29 | |
| *** buzztroll has quit IRC | 01:32 | |
| *** takedakn has joined #openstack-glance | 01:47 | |
| *** takedakn has quit IRC | 01:55 | |
| *** takedakn has joined #openstack-glance | 01:55 | |
| *** changbl has quit IRC | 01:57 | |
| *** buzztroll has joined #openstack-glance | 01:58 | |
| *** buzztroll has quit IRC | 01:58 | |
| *** takedakn has quit IRC | 02:02 | |
| *** annashen has joined #openstack-glance | 02:06 | |
| *** TravT has joined #openstack-glance | 02:06 | |
| *** haomaiwang has joined #openstack-glance | 02:08 | |
| *** TravT has quit IRC | 02:11 | |
| *** TravT has joined #openstack-glance | 02:16 | |
| *** annashen has quit IRC | 02:20 | |
| *** takedakn has joined #openstack-glance | 02:29 | |
| *** annashen has joined #openstack-glance | 02:30 | |
| *** takedakn has quit IRC | 02:33 | |
| *** takedakn has joined #openstack-glance | 02:37 | |
| *** takedakn has quit IRC | 02:42 | |
| openstackgerrit | Merged openstack/glance: Fix copy-from when user_storage_quota is enabled https://review.openstack.org/143608 | 02:52 |
|---|---|---|
| *** changbl has joined #openstack-glance | 03:01 | |
| *** Longgeek has joined #openstack-glance | 03:04 | |
| *** Longgeek has quit IRC | 03:04 | |
| *** Longgeek has joined #openstack-glance | 03:04 | |
| *** nellysmitt has joined #openstack-glance | 03:08 | |
| *** nellysmitt has quit IRC | 03:13 | |
| *** flwang has quit IRC | 04:30 | |
| *** Longgeek has quit IRC | 04:30 | |
| *** Longgeek has joined #openstack-glance | 04:34 | |
| *** ozialien has joined #openstack-glance | 04:37 | |
| openstackgerrit | Merged openstack/python-glanceclient: Remove openstack.common.strutils https://review.openstack.org/145238 | 04:49 |
| *** ozialien has quit IRC | 04:53 | |
| *** ozialien has joined #openstack-glance | 05:02 | |
| *** annashen has quit IRC | 05:03 | |
| *** nellysmitt has joined #openstack-glance | 05:09 | |
| *** nellysmitt has quit IRC | 05:14 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 05:22 | |
| *** Longgeek has quit IRC | 05:27 | |
| *** buzztroll has joined #openstack-glance | 05:29 | |
| *** Longgeek has joined #openstack-glance | 05:32 | |
| *** buzztroll has quit IRC | 05:34 | |
| *** buzztroll has joined #openstack-glance | 05:50 | |
| *** buzztroll has quit IRC | 05:54 | |
| *** annashen has joined #openstack-glance | 05:55 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance: Imported Translations from Transifex https://review.openstack.org/146765 | 06:04 |
| *** buzztroll has joined #openstack-glance | 06:11 | |
| openstackgerrit | Zhi Yan Liu proposed openstack/glance_store: Rename oslo.concurrency to oslo_concurrency https://review.openstack.org/144631 | 06:15 |
| *** annashen has quit IRC | 06:20 | |
| *** annashen has joined #openstack-glance | 06:28 | |
| *** jyoti-ranjan has joined #openstack-glance | 06:31 | |
| sabari | zhiyan: Thanks for the reviews on https://review.openstack.org/#/c/141665/. | 06:40 |
| sabari | Ping me if you got some quick concerns. I will be around for a while :) | 06:40 |
| *** bfic has joined #openstack-glance | 06:41 | |
| *** groen692 has joined #openstack-glance | 06:44 | |
| *** takedakn has joined #openstack-glance | 06:45 | |
| *** sgotliv_ has joined #openstack-glance | 06:46 | |
| *** ozialien has quit IRC | 06:50 | |
| *** annashen has quit IRC | 06:57 | |
| zhiyan | sabari: Thanks! I will take a look asap! | 06:59 |
| *** takedakn has quit IRC | 07:01 | |
| *** nellysmitt has joined #openstack-glance | 07:10 | |
| *** ozialien has joined #openstack-glance | 07:11 | |
| *** nlevinki has joined #openstack-glance | 07:12 | |
| *** nellysmitt has quit IRC | 07:15 | |
| *** belmoreira has joined #openstack-glance | 07:21 | |
| *** Longgeek has quit IRC | 07:21 | |
| *** Longgeek has joined #openstack-glance | 07:30 | |
| *** Longgeek has quit IRC | 07:34 | |
| *** Longgeek has joined #openstack-glance | 07:49 | |
| *** sgotliv_ has quit IRC | 07:50 | |
| *** Longgeek has quit IRC | 07:52 | |
| *** jyoti-ranjan has quit IRC | 07:59 | |
| *** chlong has quit IRC | 08:00 | |
| *** sgotliv has joined #openstack-glance | 08:04 | |
| *** eglynn-pto has joined #openstack-glance | 08:10 | |
| openstackgerrit | Rohan Kanade proposed openstack/glance: Handle empty request body with chunked encoding https://review.openstack.org/78134 | 08:19 |
| *** markus_z has joined #openstack-glance | 08:22 | |
| *** sgotliv has quit IRC | 08:27 | |
| *** sgotliv has joined #openstack-glance | 08:28 | |
| *** buzztroll has quit IRC | 08:34 | |
| *** nellysmitt has joined #openstack-glance | 08:36 | |
| *** pkoniszewski has joined #openstack-glance | 08:43 | |
| *** eglynn-pto is now known as eglynn | 08:43 | |
| *** tshefi has joined #openstack-glance | 08:46 | |
| *** flwang has joined #openstack-glance | 08:49 | |
| *** rwsu has joined #openstack-glance | 08:49 | |
| *** rwsu is now known as rwsu-afk | 08:50 | |
| openstackgerrit | Julien Danjou proposed openstack/glance: Simplify context by using oslo.context https://review.openstack.org/143449 | 08:52 |
| *** takedakn has joined #openstack-glance | 08:56 | |
| *** annashen has joined #openstack-glance | 08:57 | |
| *** jistr has joined #openstack-glance | 08:58 | |
| *** nlevinki has quit IRC | 09:01 | |
| *** annashen has quit IRC | 09:02 | |
| *** jyoti-ranjan has joined #openstack-glance | 09:02 | |
| *** nlevinki has joined #openstack-glance | 09:10 | |
| *** eglynn has quit IRC | 09:14 | |
| *** pdb has joined #openstack-glance | 09:19 | |
| *** MattMan has joined #openstack-glance | 09:19 | |
| *** takedakn has quit IRC | 09:27 | |
| *** takedakn has joined #openstack-glance | 09:28 | |
| openstackgerrit | Merged openstack/python-glanceclient: Remove openstack.common.importutils https://review.openstack.org/145244 | 09:44 |
| *** eglynn has joined #openstack-glance | 09:51 | |
| *** takedakn has quit IRC | 10:02 | |
| *** eglynn has quit IRC | 10:02 | |
| *** takedakn has joined #openstack-glance | 10:21 | |
| openstackgerrit | Lakshmi N Sampath proposed openstack/glance: Catalog Index Service https://review.openstack.org/138269 | 10:22 |
| *** buzztroll has joined #openstack-glance | 10:23 | |
| *** takedakn has quit IRC | 10:24 | |
| *** buzztroll has quit IRC | 10:28 | |
| *** bfic has quit IRC | 10:30 | |
| *** bfic has joined #openstack-glance | 10:31 | |
| *** bfic has joined #openstack-glance | 10:32 | |
| *** bfic has quit IRC | 10:34 | |
| *** elig_mh has joined #openstack-glance | 10:41 | |
| *** bfic has joined #openstack-glance | 10:42 | |
| *** bfic has quit IRC | 10:58 | |
| *** annashen has joined #openstack-glance | 10:58 | |
| *** bfic has joined #openstack-glance | 10:59 | |
| *** annashen has quit IRC | 11:03 | |
| *** haomaiwang has quit IRC | 11:07 | |
| *** jyoti-ranjan has quit IRC | 11:22 | |
| openstackgerrit | Louis Taylor proposed openstack/python-glanceclient: Remove uuidutils from openstack-common https://review.openstack.org/152087 | 11:22 |
| *** tellesnobrega has joined #openstack-glance | 11:34 | |
| *** nlevinki has quit IRC | 11:39 | |
| *** buzztroll has joined #openstack-glance | 12:00 | |
| *** bfic has quit IRC | 12:01 | |
| *** buzztroll has quit IRC | 12:01 | |
| *** buzztroll has joined #openstack-glance | 12:02 | |
| *** diegows has joined #openstack-glance | 12:03 | |
| *** buzztroll has quit IRC | 12:03 | |
| *** buzztroll has joined #openstack-glance | 12:04 | |
| *** buzztroll has quit IRC | 12:05 | |
| *** buzztrol_ has joined #openstack-glance | 12:05 | |
| *** buzztrol_ has quit IRC | 12:07 | |
| *** buzztroll has joined #openstack-glance | 12:07 | |
| *** nlevinki has joined #openstack-glance | 12:08 | |
| *** buzztroll has quit IRC | 12:09 | |
| *** buzztroll has joined #openstack-glance | 12:09 | |
| *** buzztroll has quit IRC | 12:10 | |
| *** buzztroll has joined #openstack-glance | 12:11 | |
| *** buzztroll has quit IRC | 12:12 | |
| *** buzztroll has joined #openstack-glance | 12:13 | |
| *** buzztroll has quit IRC | 12:14 | |
| *** buzztroll has joined #openstack-glance | 12:14 | |
| *** buzztroll has quit IRC | 12:16 | |
| *** buzztroll has joined #openstack-glance | 12:16 | |
| openstackgerrit | Kamil Rykowski proposed openstack/glance: Notifications for metadefinition resources https://review.openstack.org/148546 | 12:17 |
| *** buzztroll has quit IRC | 12:18 | |
| *** buzztroll has joined #openstack-glance | 12:18 | |
| *** buzztroll has quit IRC | 12:19 | |
| *** buzztroll has joined #openstack-glance | 12:20 | |
| *** buzztroll has quit IRC | 12:21 | |
| *** buzztroll has joined #openstack-glance | 12:22 | |
| *** buzztroll has quit IRC | 12:23 | |
| *** buzztroll has joined #openstack-glance | 12:23 | |
| *** buzztrol_ has joined #openstack-glance | 12:25 | |
| *** buzztroll has quit IRC | 12:25 | |
| *** buzztroll has joined #openstack-glance | 12:27 | |
| *** buzztrol_ has quit IRC | 12:27 | |
| *** buzztroll has quit IRC | 12:28 | |
| *** buzztroll has joined #openstack-glance | 12:29 | |
| *** buzztroll has quit IRC | 12:30 | |
| *** buzztroll has joined #openstack-glance | 12:31 | |
| *** aix has joined #openstack-glance | 12:31 | |
| *** buzztrol_ has joined #openstack-glance | 12:32 | |
| *** buzztroll has quit IRC | 12:32 | |
| *** belmoreira has quit IRC | 12:33 | |
| *** buzztrol_ has quit IRC | 12:34 | |
| *** buzztroll has joined #openstack-glance | 12:34 | |
| *** buzztroll has quit IRC | 12:35 | |
| *** buzztroll has joined #openstack-glance | 12:36 | |
| *** buzztroll has quit IRC | 12:37 | |
| *** buzztroll has joined #openstack-glance | 12:38 | |
| *** nellysmitt has quit IRC | 12:38 | |
| *** nellysmi_ has joined #openstack-glance | 12:38 | |
| *** buzztroll has quit IRC | 12:39 | |
| *** nellysmi_ has quit IRC | 12:39 | |
| *** buzztroll has joined #openstack-glance | 12:40 | |
| *** delattec has joined #openstack-glance | 12:40 | |
| *** buzztrol_ has joined #openstack-glance | 12:41 | |
| *** buzztroll has quit IRC | 12:41 | |
| *** cdelatte has quit IRC | 12:42 | |
| *** buzztrol_ has quit IRC | 12:43 | |
| *** buzztroll has joined #openstack-glance | 12:43 | |
| *** buzztroll has quit IRC | 12:44 | |
| *** buzztroll has joined #openstack-glance | 12:45 | |
| *** buzztroll has quit IRC | 12:46 | |
| *** buzztroll has joined #openstack-glance | 12:47 | |
| *** buzztroll has quit IRC | 12:48 | |
| *** buzztroll has joined #openstack-glance | 12:48 | |
| *** buzztroll has quit IRC | 12:50 | |
| *** buzztroll has joined #openstack-glance | 12:50 | |
| *** buzztroll has quit IRC | 12:52 | |
| *** buzztroll has joined #openstack-glance | 12:52 | |
| *** buzztroll has quit IRC | 12:54 | |
| *** buzztroll has joined #openstack-glance | 12:54 | |
| *** buzztrol_ has joined #openstack-glance | 12:56 | |
| *** buzztroll has quit IRC | 12:56 | |
| *** buzztroll has joined #openstack-glance | 12:57 | |
| *** buzztrol_ has quit IRC | 12:57 | |
| *** nellysmitt has joined #openstack-glance | 12:58 | |
| *** buzztroll has quit IRC | 12:59 | |
| *** annashen has joined #openstack-glance | 12:59 | |
| *** buzztroll has joined #openstack-glance | 12:59 | |
| *** buzztrol_ has joined #openstack-glance | 13:01 | |
| *** buzztroll has quit IRC | 13:01 | |
| *** buzztrol_ has quit IRC | 13:02 | |
| *** buzztroll has joined #openstack-glance | 13:03 | |
| *** annashen has quit IRC | 13:04 | |
| *** buzztroll has quit IRC | 13:04 | |
| *** buzztroll has joined #openstack-glance | 13:05 | |
| *** buzztrol_ has joined #openstack-glance | 13:06 | |
| *** buzztroll has quit IRC | 13:06 | |
| *** buzztroll has joined #openstack-glance | 13:08 | |
| *** buzztrol_ has quit IRC | 13:08 | |
| *** buzztroll has quit IRC | 13:10 | |
| *** buzztroll has joined #openstack-glance | 13:10 | |
| *** pkoniszewski has quit IRC | 13:11 | |
| *** buzztroll has quit IRC | 13:11 | |
| *** buzztroll has joined #openstack-glance | 13:12 | |
| *** buzztroll has quit IRC | 13:13 | |
| *** buzztroll has joined #openstack-glance | 13:14 | |
| *** takedakn has joined #openstack-glance | 13:14 | |
| *** buzztroll has quit IRC | 13:15 | |
| *** buzztroll has joined #openstack-glance | 13:15 | |
| *** buzztroll has quit IRC | 13:17 | |
| *** buzztroll has joined #openstack-glance | 13:17 | |
| *** belmoreira has joined #openstack-glance | 13:18 | |
| *** buzztroll has quit IRC | 13:19 | |
| *** buzztroll has joined #openstack-glance | 13:19 | |
| *** buzztroll has quit IRC | 13:23 | |
| *** mjturek has joined #openstack-glance | 13:32 | |
| openstackgerrit | Kamil Rykowski proposed openstack/glance: Notifications for metadefinition resources https://review.openstack.org/148546 | 13:33 |
| *** takedakn has quit IRC | 13:55 | |
| *** takedakn has joined #openstack-glance | 13:59 | |
| *** thangp has joined #openstack-glance | 14:00 | |
| *** takedakn has quit IRC | 14:08 | |
| *** esheffield has joined #openstack-glance | 14:17 | |
| *** belmoreira has quit IRC | 14:23 | |
| *** jaypipes has joined #openstack-glance | 14:30 | |
| *** zigo has quit IRC | 14:31 | |
| *** junhongl__ has quit IRC | 14:32 | |
| *** zigo has joined #openstack-glance | 14:35 | |
| *** gordc has joined #openstack-glance | 14:36 | |
| openstackgerrit | Merged openstack/python-glanceclient: Remove uuidutils from openstack-common https://review.openstack.org/152087 | 14:43 |
| *** jasondot_ has joined #openstack-glance | 14:45 | |
| *** junhongl__ has joined #openstack-glance | 14:46 | |
| *** peristeri has joined #openstack-glance | 14:48 | |
| *** openstackgerrit has quit IRC | 14:52 | |
| *** openstackgerrit has joined #openstack-glance | 14:52 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:55 | |
| *** flwang has quit IRC | 14:56 | |
| *** annashen has joined #openstack-glance | 15:00 | |
| *** annashen has quit IRC | 15:05 | |
| *** nlevinki has quit IRC | 15:08 | |
| *** nlevinki has joined #openstack-glance | 15:08 | |
| *** spzala has joined #openstack-glance | 15:11 | |
| *** TravT has quit IRC | 15:16 | |
| *** jyoti-ranjan has joined #openstack-glance | 15:36 | |
| openstackgerrit | Ian Cordasco proposed openstack/python-glanceclient: Ignore NoneType when encoding headers https://review.openstack.org/152159 | 15:39 |
| *** TravT has joined #openstack-glance | 15:40 | |
| *** TravT has quit IRC | 15:41 | |
| *** TravT has joined #openstack-glance | 15:41 | |
| *** cpallares has joined #openstack-glance | 15:47 | |
| *** ericpeterson has joined #openstack-glance | 15:56 | |
| *** vijendar has joined #openstack-glance | 15:57 | |
| nikhil_k | kragniz: ever heard back from stuart on this https://review.openstack.org/#/c/146437/ ? | 16:00 |
| kragniz | nikhil_k: he said he'd comment on the spec in a minute | 16:03 |
| kragniz | nikhil_k: he's still positive about changing the default | 16:03 |
| nikhil_k | ack | 16:04 |
| nikhil_k | kragniz: I still have hope for it to land on k2 | 16:05 |
| kragniz | okay! | 16:05 |
| kragniz | it's ready to go once we decide on a number :P | 16:05 |
| *** pkoniszewski has joined #openstack-glance | 16:05 | |
| nikhil_k | :) | 16:06 |
| *** zz_jgrimm is now known as jgrimm | 16:12 | |
| *** david-lyle_afk is now known as david-lyle | 16:19 | |
| *** ericpeterson has quit IRC | 16:20 | |
| *** elig_mh has quit IRC | 16:20 | |
| *** buzztroll has joined #openstack-glance | 16:25 | |
| *** nlevinki has quit IRC | 16:26 | |
| *** ericpeterson has joined #openstack-glance | 16:26 | |
| *** TravT_ has joined #openstack-glance | 16:27 | |
| *** TravT has quit IRC | 16:29 | |
| *** TravT_ has quit IRC | 16:39 | |
| *** pkoniszewski has quit IRC | 16:39 | |
| *** TravT has joined #openstack-glance | 16:39 | |
| *** rwsu-afk is now known as rwsu | 16:40 | |
| *** annegent_ has joined #openstack-glance | 16:49 | |
| *** groen692 has quit IRC | 16:50 | |
| openstackgerrit | Lakshmi N Sampath proposed openstack/glance-specs: Catalog Index Service https://review.openstack.org/138051 | 16:53 |
| ativelkov | folks, I still need one more +2 on https://review.openstack.org/#/c/148574/ - that's a critical we wanted to land in k2. Any volunteers? | 16:56 |
| kragniz | ativelkov: I +1ed | 16:56 |
| *** markus_z has quit IRC | 16:57 | |
| ativelkov | kragniz: thanks! | 16:57 |
| kragniz | ativelkov: I like the nice long commit message :) | 16:57 |
| *** tshefi has quit IRC | 17:00 | |
| *** annashen has joined #openstack-glance | 17:01 | |
| *** mjturek has quit IRC | 17:03 | |
| *** annegent_ has quit IRC | 17:04 | |
| *** annashen has quit IRC | 17:05 | |
| sigmavirus24 | ativelkov: technically we should -1 that because the commit message isn't following the guidelines but the change is too important | 17:11 |
| sigmavirus24 | nikhil_k: thoughts on https://review.openstack.org/#/c/148574/ | 17:12 |
| kragniz | sigmavirus24: which guidelines, other than the tense of the commit title? | 17:12 |
| sigmavirus24 | max line length of 72. | 17:12 |
| kragniz | sigmavirus24: and linewrapping to 72 | 17:12 |
| sigmavirus24 | That looks like it's longer | 17:12 |
| kragniz | ah, right | 17:12 |
| sigmavirus24 | just eyeballing it | 17:12 |
| kragniz | I thought the same, but whoever +As it can edit the message | 17:13 |
| sigmavirus24 | I could edit it right now but it would wipe out zhiyan's +2 | 17:13 |
| sigmavirus24 | also the other +1s | 17:13 |
| kragniz | right | 17:14 |
| sigmavirus24 | Still trying to help nikhil_k with his goal to get more reviewers to follow the guidelines more closely for review (which includes closely reviewing commit messages) | 17:16 |
| kragniz | yeah, it's wrapped to 80 char at the moment | 17:16 |
| sigmavirus24 | I was an early reviewer for that and missed it (or maybe I didn't, I don't remember) but it's priority is up there and at this point, I'm not sure it's worth it | 17:16 |
| nikhil_k | sigmavirus24: that's a good point | 17:17 |
| nikhil_k | ativelkov: we can mark it as a blocker for k2 | 17:17 |
| nikhil_k | that way you don't have to worry about missing it | 17:17 |
| nikhil_k | if the author edits it using the GUI, will it flush the +2? | 17:18 |
| *** EmilienM is now known as EmilienM|afk | 17:18 | |
| kragniz | nikhil_k: yes, since the git hash changes | 17:19 |
| sigmavirus24 | I think so but I'm not 100% sure | 17:19 |
| kragniz | unless gerrit has changed and does something magic now | 17:19 |
| nikhil_k | hmm, swear I've seem some cases when jenkins remembers | 17:19 |
| kragniz | nikhil_k: you may be thinking about rebases? | 17:19 |
| ativelkov | It does. I fixed some typo in some commit and it wiped the +1's | 17:19 |
| nikhil_k | gah | 17:19 |
| nikhil_k | ok, that makes sense. then it's prolly rebases only as kragniz is saying .. | 17:20 |
| ativelkov | Sorry, I always forget about "max charcount in commit messages" | 17:20 |
| nikhil_k | ativelkov: you might want to wait a tiny bit as I will review it in next 30 or so mins | 17:20 |
| ativelkov | nikhil_k: thanks | 17:20 |
| kragniz | ativelkov: set the linewrap to 72 for commit messages in your editor | 17:21 |
| kragniz | ativelkov: you'll never have to remember again! | 17:21 |
| ativelkov | I edit commit messages in vim ) | 17:21 |
| sigmavirus24 | We should just add a job to check that for us | 17:21 |
| nikhil_k | I'm looking to test this functionally today https://review.openstack.org/#/c/85211/22 | 17:21 |
| sigmavirus24 | ativelkov: that's easy to do in vim | 17:21 |
| nikhil_k | more volunteers are welcome@ | 17:21 |
| nikhil_k | sigmavirus24: was there one before, do you know? | 17:22 |
| sigmavirus24 | ativelkov: add "autocmd FileType gitcommit setlocal spell tw=72" to your vimrc | 17:22 |
| sigmavirus24 | nikhil_k: I think some other projects have it and hacking used to have a check for commit style | 17:22 |
| sigmavirus24 | I think hacking removed it though because it was causing issues or something | 17:22 |
| kragniz | nikhil_k: I'll have a test later | 17:22 |
| sigmavirus24 | ativelkov: you don't need spell for spellchecking unless you want it, but the rest is correct | 17:23 |
| kragniz | sigmavirus24: people were annoyed at being nitted so much about commit messages iirc | 17:23 |
| sigmavirus24 | kragniz: well too bad | 17:23 |
| nikhil_k | sigmavirus24: hmm, that makes sense. Hacking did the trick ! | 17:23 |
| sigmavirus24 | it's in the review guidelines | 17:23 |
| nikhil_k | yeah, pointless to put in guideline and not in hacking | 17:23 |
| sigmavirus24 | I don't care how annoying it may seem, it helps everyone when you're looking for what broke the project | 17:23 |
| nikhil_k | may be it was the urls etc | 17:23 |
| kragniz | I think we may be able to enable it again | 17:24 |
| sigmavirus24 | Admittedly, no one says "I totally broke how the engine powering this will work under these conditions" in their message, but they damn well should | 17:24 |
| sigmavirus24 | =P | 17:24 |
| nikhil_k | having a short url might annoy some people | 17:24 |
| kragniz | http://lists.openstack.org/pipermail/openstack-dev/2014-December/052519.html | 17:28 |
| *** sgotliv has quit IRC | 17:28 | |
| kragniz | sigmavirus24: nikhil_k ^ | 17:29 |
| sigmavirus24 | kragniz: yeah, that's what I was thinking of | 17:32 |
| sigmavirus24 | nikhil_k: well shortening services are subject to availability, so using one (even if openstack provided) could provide more harmful than helpful | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Declarative definitions of type-specific Artifact properties https://review.openstack.org/119174 | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: A mixin for jsonpatch requests validation https://review.openstack.org/148588 | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts API https://review.openstack.org/136629 | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: SemVer utility to store SemVer-compatible object versions in DB https://review.openstack.org/107676 | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts Repository - DB https://review.openstack.org/115998 | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts Domain https://review.openstack.org/132898 | 17:32 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts plugin loader https://review.openstack.org/134300 | 17:32 |
| nikhil_k | and not everyone likes google ;) | 17:37 |
| nikhil_k | sigmavirus24: I see your point :) | 17:37 |
| *** jyoti-ranjan has quit IRC | 17:37 | |
| sigmavirus24 | the counter argument, of course, is that links rot no matter what | 17:37 |
| sigmavirus24 | so | 17:37 |
| sigmavirus24 | we need archive.openstack.org to run an internet archive like service =P | 17:38 |
| nikhil_k | let's ask TravT if his team is willing to write another elastic search | 17:39 |
| nikhil_k | brb | 17:39 |
| sigmavirus24 | lol | 17:39 |
| *** jistr has quit IRC | 17:39 | |
| sigmavirus24 | TravT: you're rewriting elastic search in Python? Cool! =P | 17:39 |
| *** pkoniszewski has joined #openstack-glance | 17:39 | |
| kragniz | nikhil_k: sounds like a plan! | 17:40 |
| sigmavirus24 | should we bolt this onto glance too or... ? | 17:42 |
| kragniz | sigmavirus24: it can go in artifacts | 17:42 |
| *** pdb has quit IRC | 17:52 | |
| *** TravT has quit IRC | 17:54 | |
| *** TravT has joined #openstack-glance | 17:55 | |
| *** EmilienM|afk is now known as EmilienM | 18:00 | |
| *** annegent_ has joined #openstack-glance | 18:04 | |
| openstackgerrit | Jesse J. Cook proposed openstack/glance: Replace snet config with endpoint config https://review.openstack.org/150144 | 18:09 |
| *** harlowja has joined #openstack-glance | 18:11 | |
| openstackgerrit | Jesse J. Cook proposed openstack/glance_store: Replace snet config with endpoint config https://review.openstack.org/139726 | 18:11 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Declarative definitions of type-specific Artifact properties https://review.openstack.org/119174 | 18:13 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: A mixin for jsonpatch requests validation https://review.openstack.org/148588 | 18:13 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts API https://review.openstack.org/136629 | 18:13 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts Repository - DB https://review.openstack.org/115998 | 18:13 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts Domain https://review.openstack.org/132898 | 18:13 |
| openstackgerrit | Inessa Vasilevskaya proposed openstack/glance: Artifacts plugin loader https://review.openstack.org/134300 | 18:13 |
| *** flwang has joined #openstack-glance | 18:15 | |
| *** jasondot_ has quit IRC | 18:20 | |
| *** flwang has quit IRC | 18:25 | |
| *** flwang has joined #openstack-glance | 18:30 | |
| *** TravT has quit IRC | 18:30 | |
| *** cpallares has quit IRC | 18:31 | |
| *** annashen has joined #openstack-glance | 18:34 | |
| *** eglynn has joined #openstack-glance | 18:41 | |
| *** eglynn has quit IRC | 18:45 | |
| *** TravT has joined #openstack-glance | 18:47 | |
| *** TravT has quit IRC | 18:54 | |
| *** buzztroll has quit IRC | 18:55 | |
| *** aix has quit IRC | 18:59 | |
| *** jasondot_ has joined #openstack-glance | 19:02 | |
| *** echevemaster has joined #openstack-glance | 19:11 | |
| *** flwang has quit IRC | 19:12 | |
| kragniz | nikhil_k: stuart posted comments a little while ago, if you hadn't already seen them | 19:15 |
| nikhil_k | kragniz: just about to | 19:15 |
| *** EmilienM is now known as EmilienM|afk | 19:25 | |
| nikhil_k | sigmavirus24: https://review.openstack.org/#/c/146437/ , if you've a few moments | 19:28 |
| sigmavirus24 | nikhil_k: yes? | 19:29 |
| nikhil_k | sigmavirus24: seems like Stuart is of the opinion that this default incr value is delyaed already | 19:30 |
| nikhil_k | based on your input about os-ansible like projects facing issues, it bring us into a trixy situation | 19:31 |
| nikhil_k | Usualy, I'd be more careful about increasing defaults like this however, he seems to have a good point | 19:32 |
| sigmavirus24 | So by default, I think we uses whatever glance has a default in this particular case | 19:32 |
| sigmavirus24 | Regardless, I do think there's value in configuring this for os-ansible-deployment regardless of the defaults in glance | 19:33 |
| sigmavirus24 | Unless the defaults are really really good ;) | 19:33 |
| *** nellysmitt has quit IRC | 19:34 | |
| nikhil_k | heh | 19:34 |
| sigmavirus24 | (part of the reason for all my work on the glance policy stuff) | 19:35 |
| nikhil_k | ok, that does add more weight | 19:35 |
| nikhil_k | (to your counter argument) | 19:35 |
| nikhil_k | was about to say geh instead of gah | 19:37 |
| nikhil_k | then realized that it may not be the best place to do so | 19:37 |
| nikhil_k | (place/moment/situation etc) | 19:37 |
| sigmavirus24 | nikhil_k: I think you're confusing me for brian? | 19:37 |
| nikhil_k | lol | 19:37 |
| nikhil_k | ok, switching the topic | 19:38 |
| sigmavirus24 | 'ian' in 'brian' == True | 19:38 |
| nikhil_k | :D | 19:38 |
| sigmavirus24 | 'ian' != 'brian' ;) | 19:38 |
| nikhil_k | sigmavirus24: oh no, I did see both of your comments | 19:38 |
| kragniz | sigmavirus24: so what you're saying is you ate brian | 19:39 |
| nikhil_k | and have pinged him separately | 19:39 |
| sigmavirus24 | no no | 19:39 |
| sigmavirus24 | brian at me | 19:39 |
| nikhil_k | sigmavirus24: https://review.openstack.org/#/c/148574/4/glance/common/utils.py | 19:39 |
| *** eglynn has joined #openstack-glance | 19:39 | |
| sigmavirus24 | Oh. Hm. nikhil_k I'm also easy swayed by both sides because I'm far too empathetic | 19:39 |
| nikhil_k | pep error on the docstring | 19:39 |
| sigmavirus24 | why did hacking not catch that? | 19:40 |
| sigmavirus24 | =( | 19:40 |
| nikhil_k | egzaktli | 19:40 |
| sigmavirus24 | huh? | 19:40 |
| nikhil_k | exactly | 19:40 |
| nikhil_k | gunna test some tests on that patch, seems important | 19:41 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 19:43 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 19:43 | |
| nikhil_k | am getting a bunch of pep8 errors on that review, completely unrelated | 19:46 |
| sigmavirus24 | nikhil_k: what hacking version are you running? | 19:47 |
| nikhil_k | sigmavirus24: rebuilding the venv now | 19:48 |
| nikhil_k | they are in the wsme utils | 19:48 |
| nikhil_k | about old style objectts and such | 19:48 |
| sigmavirus24 | I remember ye olde classes of yore | 19:49 |
| kragniz | I didn't think that check was merged in hacking yet? | 19:49 |
| *** EmilienM|afk is now known as EmilienM | 19:50 | |
| kragniz | ignore me, I was thinking of the static method one | 19:51 |
| nikhil_k | I think it merge quite a while ago | 19:51 |
| nikhil_k | jd had some patches for the same | 19:51 |
| kragniz | yeah, I'm sure | 19:51 |
| nikhil_k | wonder why I'm seeing this now | 19:52 |
| *** belmoreira has joined #openstack-glance | 19:52 | |
| kragniz | it wasn't the one I was thinking of | 19:52 |
| kragniz | it's odd the gate didn't catch it | 19:52 |
| nikhil_k | anyone use xiki much? | 19:52 |
| nikhil_k | yeah | 19:52 |
| kragniz | let me run tests in my env | 19:52 |
| kragniz | no, what's xiki? | 19:52 |
| nikhil_k | https://github.com/trogdoro/xiki | 19:52 |
| nikhil_k | seems pretty cool | 19:53 |
| nikhil_k | gunna get some lunch while the tests run | 19:54 |
| kragniz | looks cool | 19:54 |
| kragniz | nikhil_k: okay, have a good lunch | 19:55 |
| *** TravT has joined #openstack-glance | 19:56 | |
| sigmavirus24 | why does xiki's site remind me of fish | 19:58 |
| kragniz | nikhil_k: tox -e pep8 is showing up clean in my env on ativelkov's patch | 19:59 |
| sigmavirus24 | kragniz: the tox.ini ignores may be insightful | 19:59 |
| kragniz | sigmavirus24: right | 20:01 |
| sigmavirus24 | would be funny if nikhil_k had somehow upgraded pep8 to 1.5.7 without angering pbr | 20:02 |
| kragniz | heh | 20:02 |
| *** jasondot_ has quit IRC | 20:02 | |
| * kragniz needs to go home | 20:03 | |
| *** flwang1 has quit IRC | 20:05 | |
| *** annashen has quit IRC | 20:06 | |
| *** spzala_ has joined #openstack-glance | 20:07 | |
| sigmavirus24 | go home kragniz | 20:08 |
| *** TravT has quit IRC | 20:08 | |
| sigmavirus24 | kragniz: are you still there? | 20:08 |
| *** spzala has quit IRC | 20:08 | |
| *** spzala_ is now known as spzala | 20:08 | |
| *** TravT has joined #openstack-glance | 20:09 | |
| *** flwang has joined #openstack-glance | 20:09 | |
| *** TravT has quit IRC | 20:09 | |
| sigmavirus24 | kragniz-reply | awk '/y(es)?/i { print "Go home kragniz" }' | 20:09 |
| *** pkoniszewski has quit IRC | 20:09 | |
| *** flwang1 has joined #openstack-glance | 20:18 | |
| *** sgotliv has joined #openstack-glance | 20:18 | |
| *** sgotliv has quit IRC | 20:19 | |
| *** mjturek has joined #openstack-glance | 20:22 | |
| *** cpallares has joined #openstack-glance | 20:30 | |
| *** annashen has joined #openstack-glance | 20:33 | |
| *** jasondot_ has joined #openstack-glance | 20:33 | |
| openstackgerrit | Steve Lewis proposed openstack/glance: Rewrite SSL tests https://review.openstack.org/148400 | 20:39 |
| *** annashen has quit IRC | 20:40 | |
| *** annashen has joined #openstack-glance | 20:45 | |
| *** jasondot_ has quit IRC | 20:45 | |
| *** nellysmitt has joined #openstack-glance | 20:47 | |
| openstackgerrit | Steve Lewis proposed openstack/glance: Rewrite SSL tests https://review.openstack.org/148400 | 20:47 |
| *** nellysmitt has quit IRC | 20:54 | |
| *** diegows has quit IRC | 20:54 | |
| nikhil_k | sigmavirus24: while the other things are fixed | 21:00 |
| nikhil_k | what do you think about https://review.openstack.org/#/c/148574/4/glance/tests/unit/common/test_utils.py | 21:00 |
| nikhil_k | Copyright 2015? | 21:00 |
| nikhil_k | and I'm wondering why he removed the ws before Licensed | 21:00 |
| nikhil_k | rest of the things look good | 21:01 |
| nikhil_k | ativelkov: still around by any chance? | 21:01 |
| nikhil_k | I know it's 12am your time :) | 21:01 |
| sigmavirus24 | nikhil_k: left comments | 21:02 |
| *** thumpba has joined #openstack-glance | 21:09 | |
| thumpba | i am trying to upload an image over 5gb. i got this error "413 Request Entity Too Large" | 21:10 |
| nikhil_k | thumpba: check your storage quota | 21:11 |
| nikhil_k | or glance config for size limit | 21:11 |
| nikhil_k | sigmavirus24: https://www.python.org/dev/peps/pep-0257/#multi-line-docstrings is not enforcing blank line on the first one, and did our hacking change? | 21:12 |
| sigmavirus24 | nikhil_k: hacking doesn't use pep257 because docstrings aren't that important | 21:13 |
| thumpba | nikhil_k: in glance-api.conf "#image_size_cap=1099511627776" | 21:13 |
| nikhil_k | heh, I remeber being pestered on this | 21:13 |
| sigmavirus24 | also pep257 has a few false positives too | 21:13 |
| thumpba | nikhil_k: so just uncomment | 21:13 |
| nikhil_k | thumpba: sure, be careful though | 21:14 |
| nikhil_k | if your node can't handle that much data, it may go down | 21:14 |
| thumpba | nikhil_k: im migrating a instance from another openstack env. the snapshot is raw format 5.3GB | 21:14 |
| *** krtaylor has joined #openstack-glance | 21:16 | |
| *** diegows has joined #openstack-glance | 21:17 | |
| *** changbl has quit IRC | 21:18 | |
| nikhil_k | I getting a bit worried on the size of buffer being too large here | 21:23 |
| ativelkov | nikhil_k: I am here :) | 21:23 |
| nikhil_k | ativelkov: hey! | 21:23 |
| ativelkov | I am still in Bay Area, have a flight back to Russia in an hour | 21:24 |
| nikhil_k | ativelkov: oh ok, no issues | 21:25 |
| nikhil_k | ativelkov: will you be available tomorrow? | 21:25 |
| ativelkov | Yes, hopefully :) | 21:25 |
| nikhil_k | ativelkov: I'm trying to investigate a potential memory dos using this patch | 21:26 |
| *** changbl has joined #openstack-glance | 21:26 | |
| nikhil_k | ativelkov: ok, in that case - I will propose k2 to be closed on wed sometime | 21:26 |
| *** belmoreira has quit IRC | 21:26 | |
| ativelkov | I still have about 40 minutes before the boarding starts. Let's take a look | 21:26 |
| ativelkov | So, you are concerned by the buffer being allocated? | 21:27 |
| nikhil_k | ativelkov: yes, what can be the maximum size of the buffer? | 21:28 |
| ativelkov | It is set equal to the value returned by the underlying iterator | 21:29 |
| ativelkov | so, it actually does not take any extra memory | 21:29 |
| ativelkov | it just remembers the last chunk read from the external source | 21:29 |
| ativelkov | The sie of that chunk is usually defined by READ_CHUNK_SIZE of appropriate backend | 21:29 |
| ativelkov | size* | 21:29 |
| ativelkov | In case of HTTP store it is 16 Mb which is a bit too much, I believe | 21:30 |
| *** chipmanc has joined #openstack-glance | 21:30 | |
| ativelkov | But that is a separate issue, as this patch does not take a single bit extra - it just makes sure that if the consumer has requested less then that size, then the remaining is not lost | 21:31 |
| ativelkov | sigmavirus24: are you around? | 21:33 |
| sigmavirus24 | ativelkov: o/ | 21:33 |
| ativelkov | About your comment to that patch. | 21:34 |
| ativelkov | The test case in test_cooperative_reader_preserves_size_chunk_less_then_read was intended to demonstrate that the sequence of data in the whole content is correct regardless of the chunk sizes | 21:35 |
| ativelkov | so I made it to read the whole data into memory and then verified | 21:35 |
| sigmavirus24 | ativelkov: those three tests are all the same code with different parameters | 21:36 |
| sigmavirus24 | They can be written as one function that each of those three calls with different parameters | 21:36 |
| ativelkov | ahh, got it | 21:36 |
| ativelkov | I thought you want them to do chunk-by-chunk assertion | 21:36 |
| ativelkov | Yes, sure, will do | 21:36 |
| sigmavirus24 | sorry I wasn't clear | 21:38 |
| nikhil_k | sorry, I think I may be misreading it here | 21:39 |
| nikhil_k | if your allocate result = bytearray() and do result.extend(chunk) | 21:41 |
| nikhil_k | it is storing extra data in memory right? | 21:41 |
| nikhil_k | and like you are saying, the max size would be chunk size marked by external source | 21:42 |
| ativelkov | Ah, this on. Got it | 21:42 |
| ativelkov | Yes, this would be equal to the read size requested by consumer | 21:44 |
| nikhil_k | This is the reason why I was proposing to have common constants or constants specific to stores in a place which can be monitored, collaborated upon, contained, say what you may | 21:44 |
| ativelkov | rounded up by the chunk_size | 21:44 |
| nikhil_k | This _may_ be an issue if your glance node is of small memory and the external source driver keeps a high chunk constants | 21:45 |
| nikhil_k | and someone with this internal knowledge requests a bit too many of copy-from operations | 21:46 |
| ativelkov | well, we may limit the buffer size then | 21:46 |
| nikhil_k | yeah, exactly | 21:46 |
| nikhil_k | how complicated will that become? | 21:47 |
| nikhil_k | (do you know?) | 21:47 |
| ativelkov | Thinking | 21:47 |
| ativelkov | The issue is not the read chunk size | 21:48 |
| ativelkov | We should just limit the requested read length | 21:49 |
| ativelkov | Actually, read chunk size is the issue we may not avoid at all | 21:50 |
| ativelkov | I mean, if it is set to 16 Gb and the host just has 8 Gb of RAM, we are in trouble anyway | 21:51 |
| ativelkov | because even a single read operation will lead to OOM | 21:52 |
| ativelkov | But what we can do is to avoid very large read values | 21:53 |
| nikhil_k | hmm | 21:53 |
| ativelkov | We actually have two options if the consumer requests too large read size: we may either raise an exception OR return less the requested | 21:55 |
| *** chipmanc has quit IRC | 21:56 | |
| nikhil_k | ativelkov: so, in the v1 code it is feeding off the LimitingReader | 21:56 |
| ativelkov | the latter is a little worse as some consumers may interpret that as the end of stream which is not true | 21:56 |
| nikhil_k | ativelkov: right | 21:57 |
| ativelkov | Yes, the LimitingReader raises an exception | 21:57 |
| ativelkov | Ok, we may use the same approach here | 21:57 |
| nikhil_k | yeah | 21:57 |
| ativelkov | nikhil_k: do you suggest to use the same configuration parameter as the limit value? Or use something lower? | 21:59 |
| nikhil_k | ativelkov: heh, thinking | 22:00 |
| ativelkov | Actually this parameter should not be too high: this is the size of the buffer, not the size of the whole image | 22:00 |
| *** flwang has quit IRC | 22:01 | |
| ativelkov | My built-in random number generator suggests 128 Mb as a very pessimistic estimate of the worst possible read size chunk which may be meanigfull | 22:02 |
| nikhil_k | ativelkov: yeah, so in the BufferedReader we use something | 22:02 |
| nikhil_k | lemme see what that is | 22:02 |
| *** spzala has quit IRC | 22:02 | |
| sigmavirus24 | similarly, a user with 8 GB and a read asking for 5GB could also have problems if we're duplicating that 5GB anywhere, we should be trying to use memoryviews if possible to avoid copies | 22:03 |
| sigmavirus24 | also for consideration, simultaneously issuing two reads that summed up exceed the system capabilities | 22:04 |
| nikhil_k | ok, so this is not using any limits too! https://review.openstack.org/#/c/120866/5 | 22:04 |
| sigmavirus24 | (two or more reads ofc) | 22:04 |
| nikhil_k | sigmavirus24: yeah, although accessing the memory looks a bit icky as is ... | 22:06 |
| ativelkov | Well, we should bot be asking for 5Gb at all | 22:06 |
| ativelkov | For what we have now, the read size is determined by the httplib of whatever is used by the store client | 22:06 |
| ativelkov | In swift it is 8K | 22:07 |
| nikhil_k | ( ativelkov: the issue is that Cooperative reader is not optional) | 22:07 |
| nikhil_k | yeah, just that some random driver adds a mechanism to read a whole chunk of data | 22:08 |
| nikhil_k | s/whole/huge/ | 22:08 |
| *** spzala has joined #openstack-glance | 22:09 | |
| nikhil_k | anyway, we may be over-projecting it here | 22:09 |
| sigmavirus24 | ativelkov: so uploads in httplib are by default 8192 bytes | 22:09 |
| sigmavirus24 | no way to change it | 22:09 |
| sigmavirus24 | yeah | 22:09 |
| sigmavirus24 | if you want, add securityimpact to the commit message adn send openstack-security a message explaining the concerns | 22:09 |
| ativelkov | sigmavirus24: yes. | 22:09 |
| nikhil_k | in essence, we need to provide a way for the reader to be smart about not reading gigantic data | 22:09 |
| sigmavirus24 | nikhil_k: yeah, and maybe the openstack-security has a good idea about how to do this ;) | 22:10 |
| nikhil_k | sigmavirus24: surely! | 22:10 |
| sigmavirus24 | swift may also have experience in this regard | 22:10 |
| nikhil_k | good point | 22:10 |
| nikhil_k | although, they rate limit pretty smart is what I hear | 22:11 |
| ativelkov | Folks, the boarding to my flight starts in about a minute. So, I will address sigmavirus24 's questions to tests, add some reasonable upper limit to read length and submit a new patch | 22:11 |
| nikhil_k | ativelkov: sure, that sounds like a good plan | 22:11 |
| ativelkov | will return back online in about 16 hours | 22:11 |
| nikhil_k | please drop me a message when you are back | 22:11 |
| nikhil_k | I will try to hold off k2 | 22:12 |
| nikhil_k | ativelkov: safe travels ;) | 22:12 |
| nikhil_k | I may need to -2 this unfortunately | 22:15 |
| nikhil_k | and we don't have IRC logs for today! | 22:16 |
| nikhil_k | fun fun fun.. | 22:16 |
| *** peristeri has quit IRC | 22:17 | |
| sigmavirus24 | nikhil_k: why don't we have logs? | 22:22 |
| nikhil_k | sigmavirus24: prolly, the cron job did not upload for today | 22:26 |
| nikhil_k | it may be every 24hrs | 22:26 |
| sigmavirus24 | weird. I thought they updated live | 22:26 |
| * sigmavirus24 wonders if other channels are having the same issues | 22:27 | |
| sigmavirus24 | nikhil_k: http://eavesdrop.openstack.org/irclogs/%23openstack-glance/%23openstack-glance.2015-02-02.log live updating | 22:27 |
| sigmavirus24 | (So long as you refresh) | 22:27 |
| nikhil_k | darn, I'm on the meetings page :P | 22:28 |
| nikhil_k | url fail | 22:28 |
| sigmavirus24 | It's okay nikhil_k. Happens to all of us =P | 22:28 |
| nikhil_k | lol | 22:29 |
| *** annashen has quit IRC | 22:30 | |
| *** ericpete_ has joined #openstack-glance | 22:32 | |
| *** ericpeterson has quit IRC | 22:34 | |
| nikhil_k | sigmavirus24: in the worst case, say 401 - this change https://review.openstack.org/#/c/146437/ would retry 25 times (if using swiftclient)! | 22:39 |
| sigmavirus24 | brb, getting off vpn | 22:40 |
| nikhil_k | (the information missing is swiftclient retries the entire request 5 times already) | 22:40 |
| sigmavirus24 | so the reality being that swiftclient is probably right in retrying the whole request each time | 22:41 |
| sigmavirus24 | although | 22:41 |
| sigmavirus24 | hm | 22:41 |
| sigmavirus24 | So this is the thing | 22:41 |
| nikhil_k | (may be not 401 then, though in some random scenaario that I don't know of) | 22:41 |
| sigmavirus24 | httplib prevents requests from properly supporting the "Expects" header being sent by a client | 22:41 |
| sigmavirus24 | That would give us the opportunity to not upload the entire thing first | 22:41 |
| *** annegent_ has quit IRC | 22:42 | |
| sigmavirus24 | 401 being authentication possible but failed iirc, that would return instead of a 100-continue | 22:42 |
| sigmavirus24 | so that would make this a lot better | 22:42 |
| sigmavirus24 | so that's a fundamental limitation of swiftclient and python's httplib as a whole | 22:42 |
| sigmavirus24 | (Arent' you glad I know all of this so you don't have to? =P) | 22:43 |
| stevelle | sigmavirus24: do you also do windows? | 22:43 |
| * nikhil_k ponders into wilderness | 22:43 | |
| sigmavirus24 | stevelle: I've been branching out into a brick & mortar bitcoin store for artisinally crafted bitcoin. I'm calling it Mordor | 22:44 |
| stevelle | sigmavirus24: I have just the logo for your new store... | 22:44 |
| sigmavirus24 | I swear. 3 years (ish) maintaining requests and I know way too much about things I never thought I'd ever care about | 22:44 |
| *** annashen has joined #openstack-glance | 22:45 | |
| nikhil_k | sigmavirus24: sorry, I did not get your point in specific. Also, the 401 might not be the case for 25 retries, something like a 500 could be. | 22:49 |
| sigmavirus24 | nikhil_k: yeah, I'm thinking we want retries in a much more specific way | 22:50 |
| nikhil_k | Although, 100-continue would be applicable in case of the auth retry right? and you're saying swiftclient does not work well with it? | 22:50 |
| sigmavirus24 | this is like 80% of the way there | 22:50 |
| sigmavirus24 | so 401 is an authentication failure | 22:50 |
| sigmavirus24 | if you do a post with Expects: 100-continue | 22:50 |
| sigmavirus24 | then technically the client should wait for a "response" (it isn't really a full http response but semantics) of 100 Continue (status line) | 22:51 |
| sigmavirus24 | then the client would upload the server | 22:51 |
| *** flwang has joined #openstack-glance | 22:51 | |
| nikhil_k | (yes, we are a bit too familiar with this part of swiftclient) | 22:51 |
| sigmavirus24 | since that's comes only if the headers a complete, it could issue a 401 immediately and the body would never be uploaded | 22:51 |
| sigmavirus24 | so really | 22:51 |
| sigmavirus24 | those 25 retries wouldn't be so bad if we weren't uploading the body | 22:52 |
| sigmavirus24 | which is something we can't avoid thanks to httplib | 22:52 |
| nikhil_k | you're right, although we're retrying to the LB 25 times | 22:52 |
| nikhil_k | and that _can_ global rate limit you | 22:52 |
| nikhil_k | basically, you don't want glance user to be bloked | 22:53 |
| sigmavirus24 | Yeah | 22:53 |
| sigmavirus24 | yep | 22:53 |
| nikhil_k | :(( | 22:53 |
| nikhil_k | ok, let's block this for now | 22:53 |
| *** nellysmitt has joined #openstack-glance | 22:55 | |
| *** ericpete_ has quit IRC | 22:55 | |
| *** ericpeterson has joined #openstack-glance | 22:55 | |
| sigmavirus24 | nikhil_k: am I doing the math right on that, 25 retries with exponential retry back off means that the total time would be over 4600 hours? | 22:55 |
| *** ericpete_ has joined #openstack-glance | 22:55 | |
| sigmavirus24 | *total time used by sleep | 22:55 |
| *** alex_xu_ has quit IRC | 22:58 | |
| *** ericpeterson has quit IRC | 22:59 | |
| *** nellysmitt has quit IRC | 23:00 | |
| *** alex_xu has joined #openstack-glance | 23:01 | |
| *** annashen has quit IRC | 23:03 | |
| nikhil_k | sigmavirus24: yeah, though that's just defaults | 23:04 |
| nikhil_k | mumhh, yeah may be we can let it slip dunno. Depends on how operators see this | 23:04 |
| sigmavirus24 | openstack-operators would be a good place to discuss this | 23:05 |
| nikhil_k | do you mind adding your excellent suggestion? | 23:05 |
| nikhil_k | don't want to take away your IP rights here :P | 23:05 |
| nikhil_k | (was sincere in the first comment and kidding in the second) | 23:06 |
| nikhil_k | (IRC can be hard) | 23:06 |
| *** vijendar has quit IRC | 23:07 | |
| sigmavirus24 | heh | 23:08 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:11 | |
| *** chlong has joined #openstack-glance | 23:12 | |
| *** TravT has joined #openstack-glance | 23:15 | |
| *** TravT has quit IRC | 23:24 | |
| *** TravT has joined #openstack-glance | 23:25 | |
| *** gordc has quit IRC | 23:25 | |
| *** jwang__ has quit IRC | 23:28 | |
| *** eglynn has quit IRC | 23:34 | |
| *** mjturek has quit IRC | 23:35 | |
| *** annegent_ has joined #openstack-glance | 23:42 | |
| *** TravT has quit IRC | 23:43 | |
| *** TravT has joined #openstack-glance | 23:44 | |
| *** annegent_ has quit IRC | 23:49 | |
| *** thangp has quit IRC | 23:53 | |
| *** annegent_ has joined #openstack-glance | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!