| *** TravT has quit IRC | 00:04 | |
| *** jamielennox is now known as jamielennox|away | 00:25 | |
| *** ducttape_ has quit IRC | 00:27 | |
| *** jamielennox|away is now known as jamielennox | 00:31 | |
| *** zhurong has joined #openstack-glance | 00:32 | |
| *** dharinic has quit IRC | 00:32 | |
| *** dharinic has joined #openstack-glance | 00:51 | |
| *** ducttape_ has joined #openstack-glance | 01:10 | |
| *** lucasxu has joined #openstack-glance | 01:45 | |
| *** ducttape_ has quit IRC | 02:04 | |
| *** lucasxu has quit IRC | 02:10 | |
| *** lucasxu has joined #openstack-glance | 02:11 | |
| *** tshefi has joined #openstack-glance | 02:14 | |
| *** lucasxu has quit IRC | 02:19 | |
| *** ducttape_ has joined #openstack-glance | 02:53 | |
| *** namnh has joined #openstack-glance | 03:01 | |
| *** ducttape_ has quit IRC | 03:20 | |
| *** ducttape_ has joined #openstack-glance | 03:31 | |
| *** ducttape_ has quit IRC | 03:34 | |
| *** _ducttape_ has joined #openstack-glance | 03:39 | |
| *** knangia has quit IRC | 03:41 | |
| *** _ducttape_ has quit IRC | 03:48 | |
| *** ducttape_ has joined #openstack-glance | 03:51 | |
| *** tshefi has quit IRC | 04:03 | |
| *** ducttape_ has quit IRC | 04:05 | |
| *** zhurong has quit IRC | 04:09 | |
| *** ducttape_ has joined #openstack-glance | 04:19 | |
| *** ducttape_ has quit IRC | 04:27 | |
| *** ducttape_ has joined #openstack-glance | 04:28 | |
| *** zhurong has joined #openstack-glance | 04:30 | |
| *** ducttape_ has quit IRC | 04:33 | |
| *** pdeore has joined #openstack-glance | 04:41 | |
| *** pdeore has joined #openstack-glance | 04:47 | |
| *** adisky_ has joined #openstack-glance | 04:54 | |
| *** ratailor has joined #openstack-glance | 05:07 | |
| *** mine0901 has quit IRC | 05:07 | |
| *** udesale has joined #openstack-glance | 05:10 | |
| *** pdeore has quit IRC | 05:14 | |
| *** pdeore has joined #openstack-glance | 05:20 | |
| *** dharinic has quit IRC | 05:22 | |
| *** e0ne has joined #openstack-glance | 05:25 | |
| *** e0ne has quit IRC | 05:26 | |
| *** ducttape_ has joined #openstack-glance | 05:29 | |
| *** ducttape_ has quit IRC | 05:35 | |
| *** zhurong has quit IRC | 05:41 | |
| *** rcernin has joined #openstack-glance | 05:44 | |
| *** groen692 has joined #openstack-glance | 05:47 | |
| *** belmoreira has joined #openstack-glance | 05:55 | |
| *** jaosorior has joined #openstack-glance | 05:56 | |
| *** zhurong has joined #openstack-glance | 06:05 | |
| *** udesale__ has joined #openstack-glance | 06:29 | |
| *** udesale has quit IRC | 06:31 | |
| *** udesale has joined #openstack-glance | 06:33 | |
| *** udesale__ has quit IRC | 06:34 | |
| *** d0ugal has joined #openstack-glance | 06:52 | |
| *** d0ugal has quit IRC | 06:52 | |
| *** d0ugal has joined #openstack-glance | 06:52 | |
| *** d0ugal has quit IRC | 06:54 | |
| *** d0ugal has joined #openstack-glance | 06:58 | |
| *** d0ugal has quit IRC | 06:58 | |
| *** d0ugal has joined #openstack-glance | 06:58 | |
| *** ducttape_ has joined #openstack-glance | 07:00 | |
| *** ducttape_ has quit IRC | 07:05 | |
| *** tesseract has joined #openstack-glance | 07:08 | |
| *** ezoszed has joined #openstack-glance | 07:14 | |
| *** dharinic has joined #openstack-glance | 07:16 | |
| *** d0ugal has quit IRC | 07:22 | |
| *** dharinic has quit IRC | 07:23 | |
| *** d0ugal has joined #openstack-glance | 07:26 | |
| *** d0ugal has quit IRC | 07:26 | |
| *** d0ugal has joined #openstack-glance | 07:26 | |
| *** pcaruana has joined #openstack-glance | 07:27 | |
| *** udesale has quit IRC | 07:30 | |
| *** udesale has joined #openstack-glance | 07:30 | |
| *** ezoszed has quit IRC | 07:38 | |
| *** ezoszed has joined #openstack-glance | 07:40 | |
| -openstackstatus- NOTICE: Jobs in gate are failing with POST_FAILURE. Infra roots are investigating | 07:42 | |
| *** ChanServ changes topic to "Jobs in gate are failing with POST_FAILURE. Infra roots are investigating" | 07:42 | |
| *** rcernin has quit IRC | 07:54 | |
| *** tesseract has quit IRC | 07:54 | |
| *** rcernin has joined #openstack-glance | 07:55 | |
| *** tesseract has joined #openstack-glance | 07:56 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zyun has joined #openstack-glance | 08:01 | |
| zyun | Hi all, May I ask a question here? | 08:02 |
|---|---|---|
| zyun | On vmware env, when I do capture/snapshot for an instance, could I specify the specified datastore to contain this template? | 08:04 |
| *** zzzeek has joined #openstack-glance | 08:04 | |
| zyun | Any one know this? Thanks in advance!!! | 08:08 |
| -openstackstatus- NOTICE: logs.openstack.org has corrupted disks, it's being repaired. Please avoid rechecking until this is fixed | 08:22 | |
| *** ChanServ changes topic to "logs.openstack.org has corrupted disks, it's being repaired. Please avoid rechecking until this is fixed" | 08:22 | |
| *** e0ne has joined #openstack-glance | 08:24 | |
| *** pbandark has joined #openstack-glance | 08:59 | |
| *** bjornar_ has joined #openstack-glance | 09:14 | |
| *** gcb has quit IRC | 09:21 | |
| *** zyun has quit IRC | 09:24 | |
| *** yatinkarel is now known as yatin_away | 09:26 | |
| *** udesale__ has joined #openstack-glance | 09:35 | |
| *** _dalgaaf has quit IRC | 09:37 | |
| *** udesale has quit IRC | 09:37 | |
| *** dalgaaf has joined #openstack-glance | 09:40 | |
| *** bjornar_ has quit IRC | 10:00 | |
| *** udesale__ has quit IRC | 10:03 | |
| *** bjornar_ has joined #openstack-glance | 10:03 | |
| *** e0ne has quit IRC | 10:13 | |
| *** e0ne has joined #openstack-glance | 10:16 | |
| *** ratailor has quit IRC | 10:18 | |
| *** zhurong has quit IRC | 10:18 | |
| *** namnh has quit IRC | 10:24 | |
| *** ratailor has joined #openstack-glance | 10:33 | |
| *** zhurong has joined #openstack-glance | 10:41 | |
| *** gcb has joined #openstack-glance | 10:51 | |
| *** ratailor has quit IRC | 11:00 | |
| *** ratailor has joined #openstack-glance | 11:01 | |
| openstackgerrit | Rajat Sharma proposed openstack/python-glanceclient master: Removing py34 from tox https://review.openstack.org/452153 | 11:19 |
| *** gcb has quit IRC | 11:20 | |
| *** pdeore has quit IRC | 11:26 | |
| *** knangia has joined #openstack-glance | 11:29 | |
| *** gcb has joined #openstack-glance | 11:33 | |
| *** zhurong has quit IRC | 11:59 | |
| *** ducttape_ has joined #openstack-glance | 12:17 | |
| *** ccesario has quit IRC | 12:26 | |
| *** ducttape_ has quit IRC | 12:30 | |
| *** catintheroof has joined #openstack-glance | 12:36 | |
| *** ccesario has joined #openstack-glance | 12:38 | |
| *** chlong has joined #openstack-glance | 12:40 | |
| *** pdeore has joined #openstack-glance | 13:00 | |
| *** e0ne has quit IRC | 13:06 | |
| *** ratailor has quit IRC | 13:19 | |
| *** ducttape_ has joined #openstack-glance | 13:30 | |
| *** ducttape_ has quit IRC | 13:35 | |
| *** ducttape_ has joined #openstack-glance | 13:38 | |
| *** e0ne has joined #openstack-glance | 13:53 | |
| *** e0ne has quit IRC | 13:53 | |
| *** amoralej is now known as amoralej|lunch | 13:54 | |
| *** Dinesh_Bhor has quit IRC | 13:55 | |
| *** e0ne has joined #openstack-glance | 13:56 | |
| *** e0ne has quit IRC | 13:56 | |
| *** lucasxu has joined #openstack-glance | 14:03 | |
| *** gcb has quit IRC | 14:05 | |
| *** ratailor has joined #openstack-glance | 14:05 | |
| *** gcb has joined #openstack-glance | 14:09 | |
| *** pdeore has quit IRC | 14:11 | |
| *** jamielennox is now known as jamielennox|away | 14:16 | |
| *** jroll has quit IRC | 14:32 | |
| *** jroll has joined #openstack-glance | 14:34 | |
| *** amoralej|lunch is now known as amoralej | 14:36 | |
| *** adisky_ has quit IRC | 14:39 | |
| *** dillaman has joined #openstack-glance | 14:41 | |
| *** ducttape_ has quit IRC | 14:44 | |
| *** ducttape_ has joined #openstack-glance | 14:44 | |
| *** ducttape_ has quit IRC | 14:45 | |
| *** ducttape_ has joined #openstack-glance | 14:45 | |
| *** ducttape_ has quit IRC | 14:46 | |
| *** ducttape_ has joined #openstack-glance | 14:46 | |
| *** ducttape_ has quit IRC | 14:47 | |
| *** bjornar_ has quit IRC | 14:47 | |
| *** ratailor has quit IRC | 14:49 | |
| hemanthm | rosmaita: https://blueprints.launchpad.net/glance/+spec/alembic-migrations | 14:52 |
| hemanthm | milestone target for that is set as pike-1 | 14:52 |
| hemanthm | should I change it to glance-ocata-final or stable? | 14:53 |
| *** ducttape_ has joined #openstack-glance | 14:58 | |
| *** ezoszed has quit IRC | 15:06 | |
| *** rcernin has quit IRC | 15:08 | |
| *** belmoreira has quit IRC | 15:10 | |
| *** e0ne has joined #openstack-glance | 15:18 | |
| *** ducttape_ has quit IRC | 15:19 | |
| *** ryandiet has joined #openstack-glance | 15:21 | |
| *** hongbin has joined #openstack-glance | 15:23 | |
| *** pdeore has joined #openstack-glance | 15:27 | |
| *** jaosorior has quit IRC | 15:28 | |
| *** lucasxu has quit IRC | 15:31 | |
| *** pbandark has quit IRC | 15:37 | |
| *** e0ne has quit IRC | 15:39 | |
| *** pcaruana has quit IRC | 15:46 | |
| *** lucasxu has joined #openstack-glance | 15:55 | |
| *** ducttape_ has joined #openstack-glance | 16:05 | |
| *** ducttape_ has quit IRC | 16:05 | |
| *** gcb has quit IRC | 16:07 | |
| *** gcb has joined #openstack-glance | 16:09 | |
| *** ducttape_ has joined #openstack-glance | 16:12 | |
| *** bjornar_ has joined #openstack-glance | 16:13 | |
| *** groen692 has quit IRC | 16:13 | |
| *** dillaman has quit IRC | 16:14 | |
| *** lucasxu has quit IRC | 16:29 | |
| *** pdeore has quit IRC | 16:29 | |
| *** ducttape_ has quit IRC | 16:36 | |
| *** dharinic has joined #openstack-glance | 16:38 | |
| *** ducttape_ has joined #openstack-glance | 16:39 | |
| *** ducttape_ has quit IRC | 16:50 | |
| *** ducttape_ has joined #openstack-glance | 16:53 | |
| *** lucasxu has joined #openstack-glance | 16:54 | |
| *** gyee has joined #openstack-glance | 16:59 | |
| *** ducttape_ has quit IRC | 17:03 | |
| openstackgerrit | Srinivas Sakhamuri proposed openstack/glance_store master: Make the image uploads atomic https://review.openstack.org/242084 | 17:14 |
| *** MVenesio has joined #openstack-glance | 17:18 | |
| *** TravT has joined #openstack-glance | 17:22 | |
| *** pdeore has joined #openstack-glance | 17:42 | |
| *** groen692 has joined #openstack-glance | 17:46 | |
| *** e0ne has joined #openstack-glance | 17:50 | |
| *** ryandiet has quit IRC | 18:06 | |
| *** lucasxu has quit IRC | 18:12 | |
| stevelle | for https://review.openstack.org/#/c/441501/ -- I'm inclined to suggest we intentionally throw a 500 when multiple segments are requested in a Range | 18:18 |
| stevelle | that is arguably "better" than incorrectly sending a 216 when we don't return what was asked for, and we know it | 18:18 |
| *** tesseract has quit IRC | 18:21 | |
| *** pdeore has quit IRC | 18:30 | |
| *** e0ne has quit IRC | 18:40 | |
| *** e0ne has joined #openstack-glance | 18:41 | |
| hemanthm | stevelle: it is definitely a server-side issue, so I'm opening up to 500 suggestion | 18:41 |
| hemanthm | makes sense too | 18:42 |
| hemanthm | stevelle: also, just posted a comment on 400 vs 416 discussion | 18:42 |
| stevelle | I don't like the idea of giving the wrong answer and telling someone there was nothing wrong | 18:42 |
| hemanthm | +1 | 18:42 |
| stevelle | hemanthm: yeah, just saw the 400 vs416. My thinking was that it is a problem in the Range header either way, and it being malformed. so 416 | 18:42 |
| hemanthm | 500 with a message sounds good to me | 18:43 |
| *** e0ne has quit IRC | 18:43 | |
| stevelle | will think a bit on your distinction | 18:43 |
| hemanthm | I usually see anything malformed in the request as 400 | 18:45 |
| hemanthm | The request could not be understood by the server due to malformed syntax. The client SHOULD NOT repeat the request without modifications. | 18:45 |
| hemanthm | ^ from https://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html | 18:45 |
| *** e0ne has joined #openstack-glance | 18:45 | |
| stevelle | "... or that the set of ranges requested has been rejected due to invalid ranges or..." | 18:46 |
| stevelle | 2616 is obsoleted btw | 18:47 |
| hemanthm | ah | 18:48 |
| stevelle | which makes that really tough to interpret, because it isn't wholely replaced :( | 18:48 |
| stevelle | https://tools.ietf.org/html/rfc7233#section-4.4 | 18:48 |
| hemanthm | considering the fact that a server can ignore the Range header and if it does the request wouldn't be malformed, maybe 416 makes more sense? | 18:49 |
| stevelle | that's where it gets really sticky | 18:50 |
| stevelle | "many implementations will simply respond with the entire selected representation in a 200 (OK) response." | 18:50 |
| hemanthm | so, there's room for taking a side :) | 18:51 |
| stevelle | "clients cannot depend on receiving a 416 (Range Not Satisfiable) response even when it is most appropriate." | 18:51 |
| stevelle | there is absolutely room, on both questions... the 500 vs 216 for multi-range, as well as this response | 18:51 |
| stevelle | I wonder if 416 is stronger here because it at least tells you exactly what header you got wrong | 18:52 |
| hemanthm | we can 400 with a specific error message but yeah 416 is definitely stronger | 18:52 |
| hemanthm | *can do | 18:52 |
| *** amoralej is now known as amoralej|off | 18:54 | |
| *** lucasxu has joined #openstack-glance | 19:00 | |
| *** lucasxu has quit IRC | 19:02 | |
| *** e0ne has quit IRC | 19:02 | |
| hemanthm | stevelle: https://review.openstack.org/#/c/441501/8/glance/tests/functional/v2/test_images.py | 19:03 |
| hemanthm | we should probably do add more unit tests like you said, but you think asserting the data returned there is not needed | 19:03 |
| hemanthm | without asserting the data returned, I feel like we are only testing the functionality partially | 19:04 |
| rosmaita | hemanthm: yeah, i think alembic-migrations should be changed to ocata-final | 19:04 |
| hemanthm | rosmaita: ack | 19:04 |
| rosmaita | not sure what ocata-stable-1 would be for, i guess to backport a big change? | 19:05 |
| stevelle | hemanthm: to be clear, I think asserting on the data in the unit test makes more sense than in the functional test | 19:05 |
| hemanthm | guess so, rosmaita | 19:05 |
| hemanthm | changed it to ocata-final | 19:05 |
| stevelle | intra-function coverage and edge cases are best in unit tests, imo | 19:05 |
| stevelle | I think of functional testing as light coverage with the intent to ensure things are glued together correct. maybe that is where the difference is? | 19:09 |
| *** MVenesio has quit IRC | 19:10 | |
| hemanthm | stevelle: Yes, we shouldn't be testing all possible combinations of input in functional tests. A set of typical positive or negative test cases should be enough. | 19:12 |
| *** gcb has quit IRC | 19:12 | |
| hemanthm | I'm saying that even for those one or two typical test cases, we should probably test for everything. Headers as well as data. | 19:12 |
| *** gcb has joined #openstack-glance | 19:13 | |
| hemanthm | I'm open to leaving it as-is as long as we have coverge | 19:13 |
| hemanthm | *coverage | 19:13 |
| *** p_arch has joined #openstack-glance | 19:18 | |
| stevelle | fair point there too, testing the data in one case makes fine sense | 19:20 |
| p_arch | Hello all - I have a problem that I'm hoping someone can help with. I have an instance on Newton that I snapshot'd and downloaded from glance. I then created an image on Ocata using it. When I try to boot the instance, I get a boot failure with "out of memory" in grub. I get a boot menu but anything I select ends up in black screen. Is this a known issue or am I missing a step? | 19:21 |
| p_arch | I can successfully import the image back on Newton and boot it | 19:21 |
| *** ducttape_ has joined #openstack-glance | 19:22 | |
| stevelle | p_arch: using the same flavor? | 19:22 |
| stevelle | typically that could happen if trying to use a flavor with less memory | 19:22 |
| p_arch | @stevelle - I recreated the same flavor params | 19:22 |
| p_arch | I'll double check | 19:22 |
| nikhil | rosmaita: https://review.openstack.org/#/c/447479/ | 19:23 |
| nikhil | some suggestions | 19:23 |
| stevelle | nikhil: good suggestions | 19:23 |
| p_arch | @stevelle - thanks, I found the problem, and it's embarassing :( MB vs GB grr! | 19:27 |
| *** groen692 has quit IRC | 19:27 | |
| rosmaita | nikhil: around? | 19:29 |
| *** ducttape_ has quit IRC | 19:31 | |
| * hemanthm wonders if nikhil can chat in his dematerialized state | 19:31 | |
| rosmaita | apparently not | 19:31 |
| hemanthm | stevelle: for multi-range requests, what do you think about returning the entire image with 200 | 19:41 |
| stevelle | p_arch: glad I could bobble-head :D | 19:41 |
| hemanthm | essentially just ignoring the header | 19:41 |
| stevelle | hemanthm: +1 | 19:42 |
| hemanthm | which is valid according to the spec | 19:42 |
| stevelle | good thinking, should "just work" more accurately | 19:42 |
| stevelle | probably a best-case failure | 19:42 |
| hemanthm | yeah | 19:43 |
| hemanthm | dharinic ^, what do you think? | 19:43 |
| dharinic | hemanthm: I am working on doing exactly what we should be doing with the multi-range request | 19:44 |
| hemanthm | supporting them, you mean dharinic ? | 19:44 |
| dharinic | If not, i think for now, we should refrain from doing anything at all. (Not do what we arent supposed to do) | 19:44 |
| dharinic | Yes hemanthm | 19:44 |
| dharinic | This is kinda tricky, sticky | 19:45 |
| dharinic | working on the logic in image_data.py | 19:45 |
| hemanthm | dharinic: it's an overkill, IMO | 19:46 |
| dharinic | I will either support it properly or i will make sure we throw a warning or something | 19:46 |
| hemanthm | users can make multiple single range requests, it's not efficient I get that | 19:46 |
| dharinic | adding logic to support that is breaking my existing logic | 19:46 |
| hemanthm | but without a strong use-case for it, I don't think we should attempt it | 19:46 |
| hemanthm | besides that http://httpd.apache.org/security/CVE-2011-3192.txt | 19:46 |
| dharinic | Hmmm i understand your point. I just felt we should do stuff in its entirety. either way, my existing logic in image_data.py needs some modifications. | 19:47 |
| hemanthm | well, the spec doesn't say that both single and multiple range requests must be supported | 19:48 |
| dharinic | :D | 19:49 |
| hemanthm | in fact, you can ignore all range requests completely | 19:49 |
| dharinic | all range requests? | 19:49 |
| hemanthm | as in, you can ignore the range header and just return the entire content | 19:49 |
| dharinic | in the case of a multi range? | 19:50 |
| dharinic | Sure. And we can just throw a warning. | 19:50 |
| stevelle | +1 for "without a strong use-case..." | 19:50 |
| hemanthm | why would anyone want a partial image? I'd guess to resume an image download. | 19:50 |
| dharinic | that we do not support multi-range...so we gave u the full image | 19:50 |
| hemanthm | single range request should do for resuming image download | 19:51 |
| hemanthm | unless you are a torrent client, multi-range requests are pointless I think | 19:51 |
| hemanthm | dharinic: just return the entire image with 200 if multi-range is given | 19:52 |
| dharinic | okay, that saves us some extra lines of confused code | 19:53 |
| hemanthm | or like stevelle suggested, 500 with a specific message | 19:53 |
| dharinic | Everybody agree with hemanthm's suggestion? | 19:53 |
| dharinic | I dont know if we have the rights to throw an error here. | 19:54 |
| dharinic | Dont want ppl filing bugs that Glance besides from not supporting rfc in its entirety is also throwing 500. | 19:55 |
| hemanthm | 500 is a server-side error, you have all the rights :) | 19:56 |
| -openstackstatus- NOTICE: lists.openstack.org will be offline from 20:00 to 23:00 UTC for planned upgrade maintenance | 19:57 | |
| dharinic | Hmmm. Okay. I think I will give it some thought and do either of those 2 suggestions. | 19:59 |
| rosmaita | can we throw 501? | 19:59 |
| hemanthm | 501 is usually talking about the request method | 20:00 |
| rosmaita | i thought that was 405 | 20:00 |
| rosmaita | "The server does not support the functionality required to fulfill the request." | 20:01 |
| dharinic | hemanthm: About "Range" header being provided as empty, what was your opinion? 416 or 400? | 20:02 |
| rosmaita | says you can use it if you don't recoginze the request method, but doesn't say that's the only time to use it | 20:02 |
| hemanthm | 405 is more like a method not supported on a specific resource | 20:02 |
| stevelle | we will get bug reports on the 500. I think full image is stronger atm | 20:03 |
| dharinic | thats what I thought too. | 20:03 |
| stevelle | and I think paired with that we should document clearly in the api-ref that we only support single-segment, and that multi-segment will result in a full image / 200 | 20:04 |
| dharinic | +1 stevelle | 20:04 |
| rosmaita | nobody reads the docs, how about a 501 for multi-segment request? | 20:04 |
| stevelle | doesn't matter if they read the docs, we do the best thing | 20:06 |
| rosmaita | yeah, but how is giving them what they didn't ask for better than throwing an error? | 20:07 |
| stevelle | based on the RFC the full image w/ 200 makes a lot of sense ( https://tools.ietf.org/html/rfc7233#section-4.4 ) | 20:07 |
| stevelle | I can see 500 / 501 maybe | 20:07 |
| stevelle | that was my thinking before hemanthm suggested the full image idea | 20:07 |
| rosmaita | i like the 501, it's specific and indicates do not retry | 20:09 |
| *** arighi has quit IRC | 20:10 | |
| rosmaita | the problem is that a 200 with full image would be OK if we didn't support Range at all, but since we do support Range, we need to distinguish between supported requests and unsupported requests | 20:11 |
| *** ducttape_ has joined #openstack-glance | 20:11 | |
| stevelle | I think both would work for different reasons. | 20:11 |
| hemanthm | rosmaita: users won't be fooled if we return the entire image because of the response code | 20:12 |
| rosmaita | yeah, it's a judgement call, too bad you can't request a partial server from nova, otherwise we could just do what they do | 20:12 |
| hemanthm | lol | 20:12 |
| rosmaita | hemanthm: ah, right, you mean 206 vs 200 | 20:12 |
| hemanthm | yes | 20:12 |
| nikhil | stevelle: thanks | 20:12 |
| stevelle | asking in #openstack-sdks for more brains to help decide 200 vs 501 | 20:13 |
| nikhil | rosmaita: hemanthm have had a really long week and even longer day.. | 20:13 |
| stevelle | don't expect to get an answer though | 20:13 |
| hemanthm | how about api-wg? | 20:13 |
| nikhil | sorry if I had to step out for lunch right after I did the review | 20:13 |
| hemanthm | nikhil: no fun, get rest | 20:13 |
| rosmaita | nikhil: np | 20:13 |
| rosmaita | what hemanthm said | 20:13 |
| stevelle | hemanthm: that -sdks is where they hang out | 20:13 |
| hemanthm | ah, ok | 20:14 |
| stevelle | there was only 2 of us at the meeting this week, many seem busy | 20:14 |
| stevelle | and it's getting late on a friday so... | 20:14 |
| nikhil | rosmaita: what was the ping about? | 20:14 |
| rosmaita | your -1 on the cpl doc | 20:15 |
| nikhil | yes? | 20:15 |
| rosmaita | i was going to say i will make the change real quick but then you have to +2 it | 20:15 |
| rosmaita | but i confused it with hemanthm 's patch for the EMC stuff | 20:15 |
| rosmaita | so there's no rush | 20:16 |
| hemanthm | I can post an update | 20:16 |
| rosmaita | hemanthm: why don't you add nikhil's bullets (use the gerrit edit interface) and then nikhil & I can +2 right now | 20:17 |
| rosmaita | i'll +A when the logs.o.o is back | 20:17 |
| nikhil | I may have extra couple suggestions | 20:17 |
| rosmaita | ok, then let's not do anything | 20:17 |
| hemanthm | we can iterate on this too | 20:17 |
| rosmaita | the doc is only usefut to hemanthm ATM | 20:17 |
| rosmaita | and he knows what it says | 20:17 |
| hemanthm | there are no immediate consumers of this documentation other than me | 20:17 |
| rosmaita | ++ | 20:18 |
| nikhil | rosmaita: so I should stop reviewing ? | 20:18 |
| stevelle | I read that as "let's not rush merging" | 20:18 |
| rosmaita | nikhil: only until after you get some sleep | 20:18 |
| rosmaita | what i'm saying is, there is no rush to get changes in on that patch | 20:18 |
| nikhil | stevelle: ah | 20:19 |
| nikhil | rosmaita: I guess I can add comments and we can post another patch? | 20:19 |
| nikhil | flavio +A it | 20:19 |
| rosmaita | nikhil: well, that patch isn't going anywhere until logs.o.o gets fixed | 20:19 |
| nikhil | so only way to block the merge is to -2 | 20:19 |
| nikhil | gotcha | 20:19 |
| nikhil | that clears it for me | 20:19 |
| nikhil | thanks | 20:19 |
| *** dharinic has quit IRC | 20:19 | |
| rosmaita | cool | 20:20 |
| *** dharinic has joined #openstack-glance | 20:24 | |
| rosmaita | nikhil: did nova begin using v2 in newton or ocata? | 20:27 |
| rosmaita | newton, i think | 20:27 |
| nikhil | rosmaita: Newton | 20:27 |
| rosmaita | ty | 20:27 |
| nikhil | hemanthm: rosmaita stevelle : added a few more https://review.openstack.org/#/c/447479/ (and those are it from me for now) | 20:38 |
| rosmaita | ok | 20:39 |
| rosmaita | hemanthm: ed leafe thinks 400 + informative error message is the way to go | 21:17 |
| *** mvk has quit IRC | 21:18 | |
| hemanthm | 400? | 21:18 |
| stevelle | dharinic: ^ | 21:18 |
| stevelle | 416 is a bit specific about the nature of the invalid request | 21:18 |
| stevelle | and that leads away from the actual cause | 21:19 |
| rosmaita | yeah, as long as we have documented somewhere that we don't accept multiple ranges | 21:19 |
| hemanthm | 400 suggests that it's a client-side error | 21:19 |
| rosmaita | which it is | 21:19 |
| rosmaita | we could always return a 440 | 21:19 |
| rosmaita | then they would definitely have to look it up | 21:20 |
| stevelle | wat | 21:20 |
| rosmaita | didn't realize that mapped to an in-use code | 21:21 |
| hemanthm | 444? | 21:21 |
| dharinic | Hmmm, i thought 501 was a good idea | 21:21 |
| stevelle | " If multi-range is documented as not supported, and a client requests multi-range, that's client error" | 21:21 |
| hemanthm | was there any arguments against 200? | 21:21 |
| hemanthm | yep, I see that argument | 21:22 |
| rosmaita | dharinic: the argument against 501 is that it might be interpreted as "we don't support download at all" | 21:22 |
| rosmaita | (but i liked 501, too) | 21:22 |
| rosmaita | hemanthm: yes, we are returning something different than they asked for | 21:22 |
| hemanthm | which is okay according to the spec | 21:23 |
| hemanthm | anyway, I think I like 400 | 21:23 |
| rosmaita | yes, it is rfc-OK, but not nice | 21:23 |
| rosmaita | (the 200, i mean) | 21:23 |
| dharinic | Hmm. 501 with the right message that multi range functionality alone is not supported wont do? | 21:23 |
| hemanthm | yeah | 21:23 |
| rosmaita | dharinic: ed said that it's not really a server-side issue if we have documented that we don't support multi-range requests | 21:24 |
| dharinic | stevelle: Where did you see "If multi-range is documented as not supported, and a client requests multi-range, that's client error" | 21:24 |
| dharinic | hmm okay rosmaita | 21:24 |
| rosmaita | i think he's quoting ed | 21:24 |
| dharinic | Ahhh i see. | 21:25 |
| dharinic | So we go with document it properly + 400 ? | 21:25 |
| rosmaita | yes, and if possible, in the error message something about only single-range requests are supported | 21:25 |
| rosmaita | and please RTFM | 21:26 |
| stevelle | F stands for Fabulous! | 21:26 |
| rosmaita | (in the error message) | 21:26 |
| hemanthm | there's certainly room for interpretation here, but I think 400 makes the most sense | 21:26 |
| rosmaita | hemanthm: i agree with both of your points | 21:26 |
| dharinic | webob.byterange does not seem to have a way to identify the presence of multiple ranges. I guess we need to have some string introspection and throw a 400 | 21:26 |
| hemanthm | yep, webob doesn't support it | 21:27 |
| dharinic | RTFM? | 21:27 |
| hemanthm | dharinic: you can just looking for , | 21:27 |
| dharinic | I was looking for a more robust way hemanthm. but i think "," is the thing to do now | 21:27 |
| stevelle | RTFM -> "please see the Glance api-ref documentation for additional information." | 21:27 |
| rosmaita | stevelle: you are too nice | 21:28 |
| rosmaita | and you forgot the word 'fabulous' | 21:28 |
| hemanthm | :) | 21:28 |
| dharinic | thats spreading wrong information ^ | 21:28 |
| dharinic | :P | 21:28 |
| hemanthm | dharinic: multi-range headers needs to contains ',' anyway | 21:28 |
| stevelle | :sadpanda: | 21:28 |
| dharinic | hemanthm: agreed | 21:28 |
| hemanthm | *contain | 21:28 |
| dharinic | New PS coming up in just a bit. | 21:28 |
| rosmaita | cool | 21:29 |
| -openstackstatus- NOTICE: The upgrade maintenance for lists.openstack.org has been completed and it is back online. | 21:50 | |
| *** catintheroof has quit IRC | 22:36 | |
| *** ducttape_ has quit IRC | 22:42 | |
| *** ducttape_ has joined #openstack-glance | 22:43 | |
| *** ducttape_ has quit IRC | 22:43 | |
| *** ducttape_ has joined #openstack-glance | 22:47 | |
| *** ducttape_ has quit IRC | 22:51 | |
| openstackgerrit | Dharini Chandrasekar proposed openstack/glance master: Accept Range requests and set appropriate response https://review.openstack.org/441501 | 22:53 |
| dharinic | rosmaita, stevelle, hemanthm ^ | 22:54 |
| dharinic | Have to add content verification in the unit tests. Apart from that if I missed anything, do let me know. | 22:54 |
| *** bjornar_ has quit IRC | 23:25 | |
| *** ducttape_ has joined #openstack-glance | 23:26 | |
| *** ducttape_ has quit IRC | 23:28 | |
| *** ducttape_ has joined #openstack-glance | 23:28 | |
| *** ducttape_ has quit IRC | 23:29 | |
| *** bjornar_ has joined #openstack-glance | 23:29 | |
| *** ducttape_ has joined #openstack-glance | 23:30 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!