| *** markvoelker has quit IRC | 00:17 | |
| *** markvoelker has joined #openstack-glance | 00:17 | |
| *** markvoelker has quit IRC | 00:21 | |
| *** gyee has quit IRC | 00:29 | |
| *** mhen has quit IRC | 01:05 | |
| *** mhen has joined #openstack-glance | 01:07 | |
| *** markvoelker has joined #openstack-glance | 02:18 | |
| *** markvoelker has quit IRC | 02:51 | |
| *** markvoelker has joined #openstack-glance | 03:48 | |
| *** markvoelker has quit IRC | 04:21 | |
| *** abhishekk has joined #openstack-glance | 04:28 | |
| openstackgerrit | Abhishek Kekane proposed openstack/glance-specs master: Spec Lite: Rethinking our filesystem access https://review.openstack.org/605584 | 04:31 |
|---|---|---|
| *** markvoelker has joined #openstack-glance | 05:18 | |
| *** udesale has joined #openstack-glance | 05:41 | |
| *** ratailor has joined #openstack-glance | 05:43 | |
| *** markvoelker has quit IRC | 05:52 | |
| *** mosulica has joined #openstack-glance | 05:54 | |
| *** dims has quit IRC | 06:24 | |
| *** dims has joined #openstack-glance | 06:26 | |
| *** dims has quit IRC | 06:34 | |
| *** dims has joined #openstack-glance | 06:35 | |
| *** dpawlik has joined #openstack-glance | 06:39 | |
| *** mosulica has quit IRC | 06:44 | |
| *** pcaruana has joined #openstack-glance | 06:51 | |
| *** nguyenhai93 has joined #openstack-glance | 07:10 | |
| *** sdin has joined #openstack-glance | 07:10 | |
| *** MattMan_1 has quit IRC | 07:18 | |
| *** MattMan_1 has joined #openstack-glance | 07:19 | |
| *** nguyenhai_ has quit IRC | 07:19 | |
| *** sdinescu has quit IRC | 07:19 | |
| *** odyssey4me has quit IRC | 07:19 | |
| *** odyssey4me has joined #openstack-glance | 07:32 | |
| *** rcernin has quit IRC | 07:55 | |
| *** odyssey4me has quit IRC | 08:13 | |
| *** markvoelker has joined #openstack-glance | 08:18 | |
| *** odyssey4me has joined #openstack-glance | 08:27 | |
| *** odyssey4me has quit IRC | 08:51 | |
| *** markvoelker has quit IRC | 08:51 | |
| *** odyssey4me has joined #openstack-glance | 09:05 | |
| *** priteau has joined #openstack-glance | 09:13 | |
| *** sdin has quit IRC | 09:52 | |
| *** sdin has joined #openstack-glance | 09:52 | |
| *** sdin has quit IRC | 09:52 | |
| *** sdinescu has joined #openstack-glance | 09:53 | |
| *** kukacz has quit IRC | 10:12 | |
| *** kukacz has joined #openstack-glance | 10:13 | |
| *** abhishekk has quit IRC | 10:21 | |
| *** udesale has quit IRC | 10:26 | |
| *** mvkr has quit IRC | 10:48 | |
| *** markvoelker has joined #openstack-glance | 10:49 | |
| *** mvkr has joined #openstack-glance | 11:02 | |
| *** mosulica has joined #openstack-glance | 11:12 | |
| *** mosulica has quit IRC | 11:17 | |
| *** mosulica has joined #openstack-glance | 11:20 | |
| *** markvoelker has quit IRC | 11:22 | |
| *** ratailor has quit IRC | 11:37 | |
| *** udesale has joined #openstack-glance | 12:03 | |
| *** rosmaita has joined #openstack-glance | 12:38 | |
| *** rosmaita has quit IRC | 12:45 | |
| *** rosmaita has joined #openstack-glance | 12:46 | |
| openstackgerrit | Merged openstack/glance master: Add missing tests for 2.6 statuses https://review.openstack.org/595299 | 12:51 |
| *** udesale has quit IRC | 13:05 | |
| *** mvkr has quit IRC | 13:23 | |
| *** mvkr has joined #openstack-glance | 13:50 | |
| *** dpawlik has quit IRC | 14:27 | |
| *** dpawlik has joined #openstack-glance | 15:19 | |
| *** aniketh has joined #openstack-glance | 15:23 | |
| *** dpawlik has quit IRC | 15:24 | |
| *** pcaruana has quit IRC | 15:33 | |
| *** mosulica has quit IRC | 15:38 | |
| *** Luzi has joined #openstack-glance | 15:47 | |
| *** gyee has joined #openstack-glance | 16:06 | |
| *** dims has quit IRC | 16:28 | |
| *** dims_ has joined #openstack-glance | 16:35 | |
| *** Luzi has quit IRC | 17:02 | |
| *** mvkr has quit IRC | 17:39 | |
| imacdonn | rosmaita: around? | 18:16 |
| rosmaita | imacdonn: yep | 18:17 |
| imacdonn | so I had this crazy idea... | 18:17 |
| * rosmaita is listening | 18:17 | |
| imacdonn | still exploring it ... what if we had the concept of _immutable_properties ... which you could "add", but not "replace" | 18:17 |
| openstackgerrit | Victor Coutellier proposed openstack/glance master: Fix cleaning of web-download image import https://review.openstack.org/607680 | 18:18 |
| imacdonn | seemed like the code change would be really trivial ... problem is, the logic to not allow "add" on something that's already set is weird | 18:18 |
| imacdonn | https://github.com/openstack/glance/blob/master/glance/api/v2/images.py#L302-L306 | 18:19 |
| imacdonn | seems it only does the check if schema version is 4 | 18:19 |
| *** imacdonn has quit IRC | 18:21 | |
| *** imacdonn has joined #openstack-glance | 18:21 | |
| imacdonn | grr, stupid proxy server | 18:22 |
| openstackgerrit | Victor Coutellier proposed openstack/glance master: Fix cleaning of web-download image import https://review.openstack.org/607680 | 18:22 |
| *** pcaruana has joined #openstack-glance | 18:24 | |
| *** itlinux has joined #openstack-glance | 18:24 | |
| rosmaita | imacdonn: iirc, that's because in version 4 you can't do an 'add' for an existing property; in the later version, add acts like replace if the value exists | 18:25 |
| imacdonn | rosmaita: bummer ... if it worked the "old" way, it'd be quite elegant | 18:25 |
| imacdonn | I had this in mind: http://paste.openstack.org/show/731426/ | 18:26 |
| rosmaita | well, we should probably look into it and make sure what's implemented fits the standard | 18:26 |
| openstackgerrit | Victor Coutellier proposed openstack/glance master: Fix cleaning of web-download image import https://review.openstack.org/607680 | 18:28 |
| rosmaita | imacdonn: your idea for immutable properties is a good one, but i thought that part of the reason to keep them in the readonly_properties list was so that we wouldn't have to write a bunch of new tests to make sure they're not changed anywhere | 18:31 |
| imacdonn | rosmaita: seems to me that "readonly" is defined in multiple different places | 18:31 |
| rosmaita | probably, and i wonder whether the lists are consistent with each other | 18:32 |
| imacdonn | rosmaita: the only place I had to change to make PATCH work is https://github.com/openstack/glance/blob/master/glance/api/v2/images.py#L463-L466 - but that's NOT the schema (right?) | 18:32 |
| imacdonn | rosmaita: schema at https://github.com/openstack/glance/blob/master/glance/api/v2/images.py#L1052-L1056 - at least | 18:33 |
| imacdonn | rosmaita: or is that even the schema? :/ | 18:34 |
| *** mvkr has joined #openstack-glance | 18:34 | |
| imacdonn | I guess it is - it's used in get_schema() at line 1184 | 18:34 |
| rosmaita | right | 18:35 |
| imacdonn | so apparently the PATCH mechanism doesn't care about the schema..... ? | 18:37 |
| rosmaita | i think _validate_change() calls schema.validate() | 18:40 |
| rosmaita | i think part of the problem is that this code was written before the json-patch standard had stabilized, so a lot is managed by hand instead of by library | 18:42 |
| imacdonn | hmm | 18:43 |
| imacdonn | it does call schema.validate(), but that doesn't seem to catch the read-only attribute of checksum | 18:47 |
| imacdonn | https://stackoverflow.com/questions/48153961/how-to-validate-input-and-output-on-a-single-json-schema-edge-cases-with-readon | 18:49 |
| imacdonn | says "readOnly never causes JSON Schema's validation process to fail. It's not an assertion, just an annotation that applications can use to take action as they please. So your server application can handle that however it wants- raise an error if someone attempts to change it, or just silently ignore the new value." | 18:49 |
| rosmaita | that's pretty nice | 18:49 |
| rosmaita | that's exactly what we want | 18:49 |
| imacdonn | yeah, if it's stable, it is | 18:49 |
| *** gyee has quit IRC | 18:50 | |
| *** pcaruana has quit IRC | 18:50 | |
| imacdonn | http://json-schema.org/latest/json-schema-validation.html#rfc.section.10.3 | 18:51 |
| rosmaita | i was just looking at thta | 18:52 |
| rosmaita | *that | 18:52 |
| rosmaita | "If "readOnly" has a value of boolean true, it indicates that the value of the instance is managed exclusively by the owning authority, and attempts by an application to modify the value of this property are expected to be ignored or rejected by that owning authority." | 18:52 |
| rosmaita | i guess that doesn't strictly rule out allowing the value to be set? | 18:53 |
| imacdonn | I guess it's up to the "owning authority" to decide | 18:53 |
| imacdonn | I wonder if a lawyer wrote this doc ;) | 18:53 |
| rosmaita | :) | 18:54 |
| rosmaita | section 3.3 seems to indicate that the annotations (like "readOnly") are informative, but not binding | 18:55 |
| imacdonn | from 10. : "These general-purpose annotation keywords provide commonly used information for documentation and user interface display purposes. They are not intended to form a comprehensive set of features. Rather, additional vocabularies can be defined for more complex annotation-based applications." | 18:56 |
| imacdonn | I wonder if we need to define something like "Immutable" | 18:56 |
| rosmaita | either that or "controlled" | 18:59 |
| rosmaita | but it does look like we can keep them as readOnly and still allow them to be set in specific circumstances, which is really what we want | 19:00 |
| imacdonn | that feels a little awkward to me .... they're either readonly or not .... seems wrong to make them "readonly ... except if the full moon falls on the third Wednesday of October" | 19:02 |
| rosmaita | i don't know, they're *mostly* readOnly -- i mean if you create an image the normal way, you don't get an option | 19:03 |
| imacdonn | the other part of this that I don't full get ..... the assertion has been made that setting checksum (etc.) should only be allowed if the image status is "queued" | 19:04 |
| imacdonn | why would it be bad it we allowed them to be set if (and only if) they are not already set ? | 19:04 |
| imacdonn | This is why I'm thinking of Immutable, as a generic mechanism for things like this, that might be reusable in the future | 19:05 |
| rosmaita | that's a good question. I think the idea is that once the image goes 'active', all these immutable properties should be frozen | 19:08 |
| rosmaita | but i don't have a really good argument against your proposal off the top of my head | 19:08 |
| imacdonn | does the schema get exposed via API somewhere? (so a client could use it to form requests) | 19:10 |
| rosmaita | yes, there are GET v2/schema/* calls | 19:11 |
| rosmaita | https://developer.openstack.org/api-ref/image/v2/index.html#image-schemas | 19:12 |
| imacdonn | so I'd be a bit concerned about perverting the meaning of standard attrs like readOnly ... in case a future client implementation "takes the schema seriously" | 19:12 |
| rosmaita | yes, that's probably why 'id' isn't readOnly in the schema | 19:13 |
| *** spartakos has joined #openstack-glance | 19:25 | |
| *** spartakos has quit IRC | 20:17 | |
| *** gyee has joined #openstack-glance | 20:24 | |
| *** spartakos has joined #openstack-glance | 20:55 | |
| *** priteau has quit IRC | 21:01 | |
| *** spartakos has quit IRC | 21:08 | |
| *** spartakos has joined #openstack-glance | 21:26 | |
| *** rosmaita has left #openstack-glance | 21:35 | |
| *** aniketh has quit IRC | 21:43 | |
| *** itlinux has quit IRC | 22:09 | |
| *** spartakos has quit IRC | 22:17 | |
| *** rcernin has joined #openstack-glance | 22:25 | |
| *** spartakos has joined #openstack-glance | 23:02 | |
| *** itlinux has joined #openstack-glance | 23:48 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!