| *** itlinux has joined #openstack-glance | 00:16 | |
| *** edmondsw has quit IRC | 00:18 | |
| *** AlexeyAbashkin has joined #openstack-glance | 00:40 | |
| *** links has joined #openstack-glance | 00:40 | |
| *** AlexeyAbashkin has quit IRC | 00:44 | |
| *** btully has joined #openstack-glance | 01:16 | |
| *** itlinux has quit IRC | 01:19 | |
| *** btully has quit IRC | 01:21 | |
| *** gyee has quit IRC | 01:25 | |
| *** itlinux has joined #openstack-glance | 01:44 | |
| *** edmondsw has joined #openstack-glance | 01:46 | |
| *** rmcall has joined #openstack-glance | 01:46 | |
| *** edmondsw has quit IRC | 01:51 | |
| *** itlinux has quit IRC | 02:10 | |
| *** itlinux has joined #openstack-glance | 02:32 | |
| *** itlinux has quit IRC | 02:44 | |
| *** d0ugal has quit IRC | 02:49 | |
| *** d0ugal has joined #openstack-glance | 02:59 | |
| *** r-daneel has joined #openstack-glance | 03:23 | |
| *** harlowja has quit IRC | 03:30 | |
| *** edmondsw has joined #openstack-glance | 03:34 | |
| *** edmondsw has quit IRC | 03:39 | |
| *** links has quit IRC | 03:50 | |
| *** d0ugal has quit IRC | 03:55 | |
| *** pdeore has joined #openstack-glance | 03:58 | |
| *** udesale has joined #openstack-glance | 03:59 | |
| *** d0ugal has joined #openstack-glance | 03:59 | |
| *** links has joined #openstack-glance | 04:07 | |
| *** harlowja has joined #openstack-glance | 04:12 | |
| *** ykarel_ has joined #openstack-glance | 04:22 | |
| *** harlowja has quit IRC | 04:30 | |
| *** aniketh has joined #openstack-glance | 04:48 | |
| *** abhishekk has joined #openstack-glance | 04:51 | |
| *** bhagyashri_s is now known as bhagyashris | 04:52 | |
| *** dalgaaf has quit IRC | 04:59 | |
| *** dalgaaf has joined #openstack-glance | 05:01 | |
| *** Shilpa has joined #openstack-glance | 05:07 | |
| *** dillaman has joined #openstack-glance | 05:08 | |
| *** jdillaman has quit IRC | 05:09 | |
| *** harlowja has joined #openstack-glance | 05:14 | |
| Shilpa | abhishekk: Hi | 05:45 |
|---|---|---|
| abhishekk | Shilpa, hi, not well today | 05:55 |
| Shilpa | actually want to discuss 2 points, jokke's comment on 540736 and postgracesql support, will discuss the same tomorrow then | 05:58 |
| rosmaita | Shilpa yes, we will need another day before figuring out postgres support | 05:59 |
| rosmaita | i found out today that GET /v2/tasks has not worked with postgres since newton | 05:59 |
| rosmaita | so doesn't look like too many people have tried using it | 05:59 |
| rosmaita | abhishekk sorry you are not feeling well | 06:00 |
| abhishekk | rosmaita, does that mean its not use widely? | 06:00 |
| Shilpa | i have tried test migrations scripts and they are not getting executed on patch | 06:00 |
| abhishekk | rosmaita, no issues | 06:00 |
| rosmaita | abhishekk i guess postgres isn't being used by anyone who uses tasks, but not many people use tasks | 06:01 |
| rosmaita | although now we want them to ! | 06:01 |
| Shilpa | Matt Riedemann and Chris Friesen, posted comment on 433934 that facing issue since postgracesql not supported | 06:01 |
| abhishekk | rosmaita, agree | 06:01 |
| abhishekk | Shilpa, that is because they were not aware about the patch which restored that support | 06:02 |
| Shilpa | yes | 06:02 |
| abhishekk | rosmaita, again we need to think about uwsgi support as well, because even if we have marked 2.6 as current tasks are not executing | 06:03 |
| rosmaita | abhishekk they do execute if you set WSGI_MODE=mod_wsgi in local.conf for devstack | 06:04 |
| abhishekk | great, is it documented somewhere? | 06:04 |
| rosmaita | no | 06:04 |
| abhishekk | ok | 06:05 |
| *** pdeore_ has joined #openstack-glance | 06:05 | |
| rosmaita | we do need to publicize that | 06:05 |
| abhishekk | ok | 06:05 |
| rosmaita | that setting also makes keystone run in a weird way, if you need to restart keystone you have to restart apache2.service | 06:06 |
| abhishekk | Shilpa, could you please share your test scripts with me, which will save my time for creating new ones | 06:06 |
| abhishekk | yes, I know about that | 06:07 |
| Shilpa | yes, i have updated queens for test purpose, will share soon | 06:07 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance master: Imported Translations from Zanata https://review.openstack.org/544171 | 06:07 |
| *** pdeore has quit IRC | 06:07 | |
| abhishekk | Shilpa, great, thank you | 06:07 |
| rosmaita | Shilpa can you put them in a paste or something when they are ready, so we can all use them? | 06:07 |
| rosmaita | they don't have to be perfect, anything will be helpful | 06:08 |
| Shilpa | ok will do that, and share that link here | 06:08 |
| rosmaita | Shilpa excellent! | 06:08 |
| Shilpa | rosmaita: yes | 06:08 |
| rosmaita | abhishekk i am having trouble getting erno's original web-download stuff (without the whitelist/blacklist) working | 06:09 |
| abhishekk | rosmaita, what issue you are facing | 06:09 |
| Shilpa | Abhishek: i want you to go through jokee's comment on validation patch i.e. 540736, and discuss, so that i will work accordingly on it | 06:10 |
| rosmaita | he expects to find import_req in the kwargs, but it's not there ... the kwargs have task_id and task_repo, so I can get the import_req but first i have to get the task from the task repo | 06:10 |
| rosmaita | which is fine | 06:11 |
| rosmaita | but it's weird | 06:11 |
| rosmaita | (not sure that makes sense) | 06:11 |
| abhishekk | Shilpa, I will but right now I am on the way to see doctor and using mobile | 06:12 |
| rosmaita | was seeing a lot of 'NoneType' object has no attribute '__getitem__' in the logs | 06:12 |
| rosmaita | abhishekk hope you feel better | 06:12 |
| abhishekk | once I get back I will have a look | 06:12 |
| Shilpa | abhishek: ok, no issues, we can have that discussion later | 06:12 |
| rosmaita | it's late here, i am going to sign off | 06:13 |
| Shilpa | Take care | 06:13 |
| abhishekk | rosmaita, improving | 06:13 |
| Shilpa | rosmaita: GN | 06:13 |
| rosmaita | abhishekk i'll put up a DNM patch so you can look if you have time | 06:13 |
| abhishekk | rosmaita, could you share your curl commands as well | 06:13 |
| *** pdeore_ has quit IRC | 06:13 | |
| rosmaita | Shilpa have a good day | 06:13 |
| rosmaita | abhishekk sure | 06:14 |
| abhishekk | rosmaita, ack, good night | 06:14 |
| rosmaita | abhishekk will email info to you | 06:14 |
| abhishekk | rosmaita, thank you | 06:14 |
| *** pdeore has joined #openstack-glance | 06:15 | |
| Shilpa | rosmaita/abhishek: added test scripts to verify migrations is working or not in case of PostgreSQL support at http://paste.openstack.org/show/671578/ | 06:19 |
| rosmaita | ack | 06:19 |
| *** masber has joined #openstack-glance | 06:22 | |
| abhishekk | Shilpa, this migration script will not execute as we are not calling migrate command explicitly | 06:22 |
| Shilpa | yes,as we discussed yesterday, going forward we are using only alembic_migrations | 06:24 |
| *** harlowja has quit IRC | 06:24 | |
| abhishekk | the code which you have added in your migration script (has_migrations()) is called from migrate command only, I guess for offline migrations this will never be called | 06:24 |
| Shilpa | yes, so for offline migrations, we need to consider this | 06:25 |
| Shilpa | if installing fresh, migration step one can skip since no data at all, but during upgrading offline, one can't skip migration | 06:26 |
| abhishekk | Shilpa, if you have time please check how migrations were working using sync command (not EMC) that will help us to understand better what changes we require now | 06:27 |
| Shilpa | there were pike01/ocata01 scripts at alembic_migration which sync command uses (not EMC) | 06:28 |
| Shilpa | these are consolidated sccripts of what EMC scripts will have | 06:28 |
| abhishekk | Shilpa, So on similar note is it possible for you to create one new migration script for testing purpose? | 06:31 |
| *** ykarel_ is now known as ykarel | 06:31 | |
| Shilpa | but that is not possible since alembic won't accept files with same revision | 06:35 |
| abhishekk | you can create new revision, right? | 06:41 |
| *** btully has joined #openstack-glance | 06:42 | |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: DNM testing image import https://review.openstack.org/544184 | 06:43 |
| *** btully has quit IRC | 06:46 | |
| *** rosmaita has quit IRC | 06:51 | |
| *** belmoreira has joined #openstack-glance | 07:01 | |
| *** edmondsw has joined #openstack-glance | 07:11 | |
| *** edmondsw has quit IRC | 07:15 | |
| Shilpa | abhishekk: for you, only for testing purpose, will create and share with you | 07:16 |
| *** AlexeyAbashkin has joined #openstack-glance | 07:16 | |
| *** alexchadin has joined #openstack-glance | 07:17 | |
| *** AlexeyAbashkin has quit IRC | 07:21 | |
| *** AlexeyAbashkin has joined #openstack-glance | 07:26 | |
| *** aniketh has quit IRC | 07:28 | |
| *** ykarel_ has joined #openstack-glance | 07:47 | |
| *** ykarel has quit IRC | 07:48 | |
| *** ykarel__ has joined #openstack-glance | 07:52 | |
| *** ykarel has joined #openstack-glance | 07:54 | |
| *** ykarel_ has quit IRC | 07:55 | |
| *** pcaruana has joined #openstack-glance | 07:56 | |
| *** bhagyashri_s has joined #openstack-glance | 07:56 | |
| *** ShilpaSD has joined #openstack-glance | 07:56 | |
| *** ykarel__ has quit IRC | 07:57 | |
| *** pdeore has quit IRC | 07:57 | |
| *** pooja-jadhav has joined #openstack-glance | 07:58 | |
| *** nsingh has joined #openstack-glance | 07:58 | |
| *** niraj_singh has quit IRC | 07:59 | |
| *** Shilpa has quit IRC | 07:59 | |
| *** bhagyashris has quit IRC | 07:59 | |
| *** pooja_jadhav has quit IRC | 07:59 | |
| *** MattMan has quit IRC | 08:01 | |
| *** MattMan has joined #openstack-glance | 08:02 | |
| *** masber has quit IRC | 08:05 | |
| *** pdeore has joined #openstack-glance | 08:08 | |
| openstackgerrit | Abhishek Kekane proposed openstack/glance master: Following changes are required for web-download (without wlist/blist) https://review.openstack.org/544184 | 08:11 |
| openstackgerrit | Abhishek Kekane proposed openstack/glance master: Do NOT MERGE: Following changes are required for web-download (without wlist/blist) https://review.openstack.org/544184 | 08:15 |
| *** e0ne has joined #openstack-glance | 08:22 | |
| *** tesseract has joined #openstack-glance | 08:28 | |
| *** ShilpaD has joined #openstack-glance | 08:28 | |
| *** bhagyashris_ has joined #openstack-glance | 08:28 | |
| *** bhagyashris_ has quit IRC | 08:28 | |
| *** pooja_jadhav has joined #openstack-glance | 08:30 | |
| *** niraj_singh has joined #openstack-glance | 08:31 | |
| *** pooja-jadhav has quit IRC | 08:31 | |
| *** ShilpaSD has quit IRC | 08:31 | |
| *** bhagyashris has joined #openstack-glance | 08:31 | |
| *** nsingh has quit IRC | 08:32 | |
| *** bhagyashri_s has quit IRC | 08:32 | |
| *** rcernin has quit IRC | 08:33 | |
| abhishekk | jokke_, hi | 08:50 |
| *** links has quit IRC | 08:53 | |
| *** aniketh has joined #openstack-glance | 09:00 | |
| *** links has joined #openstack-glance | 09:06 | |
| *** ykarel is now known as ykarel|lunch | 09:19 | |
| openstackgerrit | Abhishek Kekane proposed openstack/glance master: Do Not MERGE: Following changes are required for web-download (without wlist/blist) https://review.openstack.org/544184 | 09:44 |
| *** alexchadin has quit IRC | 09:56 | |
| *** alexchadin has joined #openstack-glance | 09:57 | |
| *** ykarel|lunch is now known as ykarel | 10:00 | |
| *** pcaruana has quit IRC | 10:05 | |
| *** pcaruana has joined #openstack-glance | 10:20 | |
| *** ykarel_ has joined #openstack-glance | 10:26 | |
| *** ykarel has quit IRC | 10:29 | |
| *** ykarel_ is now known as ykarel | 10:37 | |
| *** kgz has joined #openstack-glance | 10:41 | |
| *** abhishekk has quit IRC | 10:44 | |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/glance master: Fix some api tests when behind http proxy https://review.openstack.org/541390 | 10:44 |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/glance master: Fix some api tests when behind http proxy https://review.openstack.org/541390 | 10:45 |
| *** edmondsw has joined #openstack-glance | 10:47 | |
| *** arcolife has joined #openstack-glance | 10:51 | |
| *** ykarel_ has joined #openstack-glance | 10:52 | |
| *** edmondsw has quit IRC | 10:52 | |
| *** ykarel has quit IRC | 10:54 | |
| *** pbourke has joined #openstack-glance | 11:01 | |
| *** AlexeyAbashkin has quit IRC | 11:05 | |
| *** pdeore has quit IRC | 11:10 | |
| *** arcolife has quit IRC | 11:13 | |
| *** pbourke has quit IRC | 11:29 | |
| *** pcaruana has quit IRC | 11:30 | |
| *** pbourke has joined #openstack-glance | 11:30 | |
| *** udesale_ has joined #openstack-glance | 11:36 | |
| *** alexchadin has quit IRC | 11:38 | |
| *** udesale has quit IRC | 11:38 | |
| *** AlexeyAbashkin has joined #openstack-glance | 11:38 | |
| *** udesale has joined #openstack-glance | 11:41 | |
| *** udesale_ has quit IRC | 11:41 | |
| *** pcaruana has joined #openstack-glance | 11:42 | |
| *** alexchadin has joined #openstack-glance | 11:47 | |
| *** udesale has quit IRC | 11:47 | |
| *** achuprin has joined #openstack-glance | 12:12 | |
| *** dtantsur|afk is now known as dtantsur | 12:21 | |
| *** edmondsw has joined #openstack-glance | 12:36 | |
| *** edmondsw has quit IRC | 12:40 | |
| achuprin | Hello guys | 12:42 |
| achuprin | what is the right way to disable glance cache? I didn't found how to do that via glance-api/cache.conf. | 12:42 |
| *** mdbooth has joined #openstack-glance | 12:46 | |
| achuprin | will it be enough to remove cachemanagement from [pipeline:glance-api-keystone+cachemanagement] | 12:50 |
| *** edmondsw has joined #openstack-glance | 12:50 | |
| mdbooth | Afternoon, folks. I'm just starting to think about hurdles for implementing Nova local storage using Cinder, and I'm considering snapshots. Snapshots data is typically uploaded to glance. I'm wondering if it's possible to specify that a glance image's data is contained within a specific cinder volume, and therefore not have to upload it? | 12:50 |
| *** udesale has joined #openstack-glance | 12:51 | |
| *** r-daneel has quit IRC | 12:54 | |
| *** takedakn has joined #openstack-glance | 12:56 | |
| *** mvk_ has quit IRC | 12:58 | |
| *** takedakn has quit IRC | 13:01 | |
| *** ykarel__ has joined #openstack-glance | 13:03 | |
| *** r-daneel has joined #openstack-glance | 13:04 | |
| *** ykarel_ has quit IRC | 13:05 | |
| *** e0ne has quit IRC | 13:10 | |
| *** e0ne has joined #openstack-glance | 13:13 | |
| *** rosmaita has joined #openstack-glance | 13:14 | |
| rosmaita | jokke_: good morning | 13:22 |
| *** mvk_ has joined #openstack-glance | 13:24 | |
| achuprin | nevermind found how to do that via [paste_deploy] | 13:24 |
| *** achuprin has quit IRC | 13:37 | |
| *** rmcall has quit IRC | 13:43 | |
| *** konan has joined #openstack-glance | 13:52 | |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: Cleanup basic import tasks https://review.openstack.org/544184 | 13:56 |
| *** btully has joined #openstack-glance | 13:57 | |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: Cleanup basic import tasks https://review.openstack.org/544184 | 13:58 |
| konan | rosmaita: hi abhishek here | 14:00 |
| rosmaita | konan nice work on that patch! | 14:01 |
| rosmaita | just about to email, everything seems to work | 14:01 |
| rosmaita | https://review.openstack.org/#/c/544184/ | 14:01 |
| konan | Just a small change, IMO we should set image status uploading before adding data to store? | 14:01 |
| rosmaita | hope you are feeling better, by the way | 14:01 |
| *** btully has quit IRC | 14:02 | |
| konan | Yes improving hopefully will recover in a day or two? | 14:02 |
| rosmaita | konan not sure ... is it possible that the add could fail and raise an exception? | 14:02 |
| rosmaita | i think it's the general problem we have with image status changes | 14:03 |
| konan | rosmaita: might be, let's see what jokke_ has to say | 14:03 |
| rosmaita | my thoughts exactly | 14:03 |
| rosmaita | i will put a comment on the patch so he doesn't miss it | 14:04 |
| konan | Now will my +2 matter here? | 14:04 |
| konan | rosmaita: ok | 14:04 |
| rosmaita | maybe we each +2 each others work on the patch, and leave it to jokke_ to +A ? | 14:05 |
| *** alexchadin has quit IRC | 14:08 | |
| *** alexchadin has joined #openstack-glance | 14:08 | |
| *** alexchadin has quit IRC | 14:08 | |
| *** alexchadin has joined #openstack-glance | 14:09 | |
| *** alexchadin has quit IRC | 14:09 | |
| *** alexchadin has joined #openstack-glance | 14:10 | |
| *** alexchadin has quit IRC | 14:10 | |
| *** alexchadin has joined #openstack-glance | 14:11 | |
| *** r-daneel has quit IRC | 14:11 | |
| *** alexchadin has quit IRC | 14:11 | |
| *** alexchadin has joined #openstack-glance | 14:11 | |
| rosmaita | konan: get some sleep! | 14:12 |
| *** alexchadin has quit IRC | 14:12 | |
| *** alexchadin has joined #openstack-glance | 14:12 | |
| *** alexchadin has quit IRC | 14:12 | |
| konan | rosmaita: sounds good | 14:14 |
| *** alexchadin has joined #openstack-glance | 14:22 | |
| *** rmcall has joined #openstack-glance | 14:24 | |
| *** ykarel__ is now known as ykarel | 14:26 | |
| *** alexchadin has quit IRC | 14:26 | |
| *** udesale has quit IRC | 14:29 | |
| *** r-daneel has joined #openstack-glance | 14:29 | |
| *** udesale has joined #openstack-glance | 14:29 | |
| *** links has quit IRC | 14:33 | |
| *** udesale has quit IRC | 14:36 | |
| *** bhagyashris has quit IRC | 14:50 | |
| *** ShilpaD has quit IRC | 14:50 | |
| *** nsingh has joined #openstack-glance | 14:50 | |
| *** nsingh has quit IRC | 14:51 | |
| *** ShilpaD has joined #openstack-glance | 14:52 | |
| *** bhagyashris has joined #openstack-glance | 14:52 | |
| *** pooja-jadhav has joined #openstack-glance | 14:53 | |
| jokke_ | me say what? | 14:53 |
| rosmaita | jokke_ hello -- need you to look at https://review.openstack.org/#/c/544184/ | 14:54 |
| *** pooja_jadhav has quit IRC | 14:55 | |
| *** niraj_singh has quit IRC | 14:55 | |
| *** alexchadin has joined #openstack-glance | 15:00 | |
| *** alexchadin has quit IRC | 15:03 | |
| jokke_ | hmm-m | 15:03 |
| -openstackstatus- NOTICE: Due to a race in stable/queens branch creation and some job removals, Zuul has reported syntax errors for the past hour; if you saw a syntax error reported for "Job tripleo-ci-centos-7-ovb-containers-oooq not defined" you can safely recheck now | 15:04 | |
| jokke_ | rosmaita: bit of oversight from my end, it seems .... so the status changes | 15:09 |
| jokke_ | I did not check how we did it, I assumed the status would have been importing already at the point when the web-dowload plugin is executed | 15:10 |
| jokke_ | it looks like we do that change at the very late stage of the whole taskflow | 15:10 |
| jokke_ | which is very muh nae good | 15:11 |
| rosmaita | i think you are right though, should probably be 'importing' in the web-download | 15:12 |
| rosmaita | that means we need to check the allowed image state transitions | 15:13 |
| *** r-daneel has quit IRC | 15:14 | |
| rosmaita | i think this is ok: https://github.com/openstack/glance/blob/master/glance/domain/__init__.py#L104 | 15:15 |
| *** AlexeyAbashkin has quit IRC | 15:17 | |
| *** r-daneel has joined #openstack-glance | 15:22 | |
| jokke_ | rosmaita: couple of comments in the review | 15:29 |
| rosmaita | thanks | 15:30 |
| *** alexchadin has joined #openstack-glance | 15:30 | |
| rosmaita | jokke_ replied ... take a look and then let's talk | 15:35 |
| jokke_ | soo ... by the looks of it we do the status transition way too late | 15:37 |
| jokke_ | as that happens in the ImportToStore class' execute | 15:38 |
| jokke_ | I think we need to change that | 15:38 |
| rosmaita | let;s make that a separate bug | 15:39 |
| rosmaita | do it now | 15:39 |
| rosmaita | but keep it separate | 15:39 |
| rosmaita | quick question about https://review.openstack.org/#/c/544184/6/glance/async/flows/api_image_import.py line 374 | 15:39 |
| *** links has joined #openstack-glance | 15:40 | |
| rosmaita | the "http" hard coded there | 15:41 |
| rosmaita | if someone whitelists "ftp", we could be looking at somethihg else there for web-download | 15:41 |
| rosmaita | couldn't we? | 15:41 |
| *** AlexeyAbashkin has joined #openstack-glance | 15:42 | |
| *** itlinux has joined #openstack-glance | 15:50 | |
| konan | Just a comment, if I don't add image status uploading then while ImportToStore task execution it raises conflict error as status transition from queued to import is not allowed | 15:52 |
| jokke_ | ok, yes, good point ... do we actually want to allow anything else than http(s) transports? | 15:52 |
| jokke_ | urllib supports all kind of things, but do we want to open that can of worms? | 15:53 |
| jokke_ | konan: then we need to fix the allowed transitions | 15:53 |
| jokke_ | I thought I did that when we introduced the poc in the first place 'though | 15:54 |
| rosmaita | konan: should be allowed: https://github.com/openstack/glance/blob/master/glance/domain/__init__.py#L104 | 15:55 |
| konan | Could you please confirm, I am away now to see the doctor (once back I can able to check) | 15:57 |
| *** ykarel is now known as ykarel|away | 15:57 | |
| rosmaita | sure, i will try it | 15:58 |
| rosmaita | do we want to make it 'importing' for now, and then jokke_ can put up a patch to move where status changes happen? | 15:58 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: Make the Image status transition early https://review.openstack.org/544511 | 15:59 |
| rosmaita | jokke_ : don't know what to do about the schemes | 16:00 |
| jokke_ | ^^ | 16:00 |
| *** alexchadin has quit IRC | 16:02 | |
| rosmaita | now i'm getting confused about the order to merge these patches ... should we roll 544511 in with 544184? | 16:03 |
| jokke_ | so after 511 184 does not need to make that status transition at all anymore and once it's not there, the merge order does not matter | 16:04 |
| *** ykarel|away has quit IRC | 16:05 | |
| jokke_ | you wanted to have it as separate bug/patch so I made it separate | 16:05 |
| rosmaita | jokke_ i am thinking we remove the http(s) restriction and let the black/whitelist take care of it | 16:10 |
| rosmaita | we could even make the default ['http','https'] for the scheme whitelist | 16:11 |
| rosmaita | ok, so i am going to push a new version of 544184 without the status change, and then pull in 544511 and test locally | 16:12 |
| *** pcaruana has quit IRC | 16:13 | |
| jokke_ | rosmaita: sounds good, the http was there just to make sure that the uri even remotely looked like we could process it | 16:17 |
| rosmaita | ok, cool | 16:18 |
| jokke_ | rosmaita: that 377-379 ... I need to look how that parameter is supposed to be passed to taskflow | 16:21 |
| rosmaita | yeah, i think maybe you don't have to pass it explicitly at line 383, i think when you declare it available in the flow, it's made available to the following flows by taskflow? | 16:22 |
| rosmaita | i remember having trouble with that at the end of the last cycle, too, though | 16:23 |
| jokke_ | https://github.com/openstack/glance/blob/master/glance/async/flows/_internal_plugins/web_download.py#L34 this should set it for the rest of the taskflow | 16:23 |
| jokke_ | so we just might need to introduce that variable with none value somewhere for python to handle it until taskflow populates it | 16:24 |
| jokke_ | but if that your change works for now leave it | 16:24 |
| rosmaita | yeah, i am thinking file a bug to clean it up later | 16:24 |
| jokke_ | I put figuring that out to my todo list | 16:24 |
| rosmaita | yeah | 16:25 |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: Cleanup basic import tasks https://review.openstack.org/544184 | 16:27 |
| *** belmoreira has quit IRC | 16:28 | |
| konan | So does it work after removing state transition uploading? | 16:35 |
| rosmaita | jokke_ konan : just did a successful test of both glance-direct and web-download | 16:35 |
| rosmaita | konan yes | 16:35 |
| konan | rosmaita: great | 16:36 |
| rosmaita | jokke_ konan test was done in glance with both 544511 and 544184 applied | 16:37 |
| jokke_ | Guys, we're getting there! :) | 16:38 |
| *** alexchadin has joined #openstack-glance | 16:40 | |
| rosmaita | yes! | 16:40 |
| rosmaita | ok, getting back to testing the uri filtering | 16:40 |
| *** itlinux has quit IRC | 16:41 | |
| konan | jokke_: minor change in commit message :( | 16:42 |
| konan | Should we ignore it to save time? | 16:42 |
| *** e0ne has quit IRC | 16:43 | |
| jokke_ | oh yeah, typo there ... what ever you think is important | 16:43 |
| rosmaita | i can go either way | 16:44 |
| *** alexchadin has quit IRC | 16:44 | |
| rosmaita | i think it's fine, it's clear what you mean | 16:44 |
| konan | I guess, let's save the time | 16:44 |
| rosmaita | cool | 16:44 |
| konan | Will approve as soon as Jenkins passes | 16:44 |
| rosmaita | excellent! | 16:44 |
| jokke_ | had it been factual error, I'd be changing it already ;) | 16:45 |
| jokke_ | I do hate this keyboard | 16:45 |
| konan | Tomorrow, I will try to put some time for migration for postgres | 16:45 |
| rosmaita | me too! | 16:45 |
| rosmaita | ok, how does this sound ... when all these merge in a few hours, i will put up cherry picks to stable/queens | 16:46 |
| rosmaita | any objection to me ninja-approving them? | 16:46 |
| konan | No | 16:46 |
| konan | The only headache is preparing appropriate migration script :( | 16:47 |
| rosmaita | once they've all merged to stable/queens, i'll put up a patch for RC-2 | 16:47 |
| rosmaita | and then we can do the migration stuff for RC-3 | 16:47 |
| rosmaita | (and anything else that comes up) | 16:47 |
| konan | No objection rosmaita if my vote matters | 16:47 |
| jokke_ | rosmaita: ping me ... I will likely be awake still unless the gate takes half a day | 16:47 |
| rosmaita | ok, i will ping first | 16:48 |
| jokke_ | rosmaita: I do not mind, but looks better if we don't need to ninja | 16:48 |
| rosmaita | i will have some docs changes, but they can be for RC-3 | 16:48 |
| rosmaita | jokke_ agree, better if we don't have to ninja | 16:48 |
| konan | So when will be rc-3? | 16:48 |
| rosmaita | maybe tuesday? | 16:48 |
| rosmaita | i think the final RC has to be ready by thursday | 16:49 |
| jokke_ | rosmaita: as per process if you backport I can +2A anyways and vice versa | 16:49 |
| konan | Any way we can discuss that tomorrow | 16:49 |
| konan | In the meeting | 16:49 |
| rosmaita | yes | 16:49 |
| rosmaita | konan try to get some sleep! hope you feel better | 16:49 |
| jokke_ | konan: what rosmaita just said! | 16:50 |
| konan | rosmaita, jokke_ yes just waiting for Jenkins results to put +2A | 16:50 |
| jokke_ | konan: you know, none of us are kids anymore ... can't keep doing 22hr days and think you'll be fine long term :P | 16:50 |
| konan | :P | 16:51 |
| *** itlinux has joined #openstack-glance | 16:52 | |
| konan | jokke_: between, have you opted time for team photo? | 16:52 |
| jokke_ | konan: if you +2 what needs to be +2'd we can cross +A with rosmaita once zuul is done ... you don't need to wait hours ... if you are ok with that | 16:52 |
| jokke_ | yes, I think I did | 16:52 |
| konan | jokke_: I never thought of that ;) | 16:53 |
| jokke_ | 14:30 at Tue | 16:53 |
| konan | Great | 16:53 |
| konan | Signing of now, just drop me a mail if something needs to be taken care during my day time, have a good time ahead | 16:54 |
| jokke_ | konan: take care of yourself buddy | 16:54 |
| konan | jokke_: yup, thank you | 16:55 |
| *** konan has quit IRC | 16:55 | |
| *** rmcall has quit IRC | 16:58 | |
| *** rosmaita has quit IRC | 17:01 | |
| *** rmcall has joined #openstack-glance | 17:03 | |
| *** _alastor__ has joined #openstack-glance | 17:06 | |
| *** AlexeyAbashkin has quit IRC | 17:07 | |
| *** rmcall has quit IRC | 17:08 | |
| *** dtantsur is now known as dtantsur|afk | 17:09 | |
| *** gyee has joined #openstack-glance | 17:11 | |
| *** rosmaita has joined #openstack-glance | 17:13 | |
| *** AlexeyAbashkin has joined #openstack-glance | 17:15 | |
| *** alexchadin has joined #openstack-glance | 17:16 | |
| *** pcaruana has joined #openstack-glance | 17:17 | |
| *** AlexeyAbashkin has quit IRC | 17:20 | |
| *** links has quit IRC | 17:21 | |
| *** alexchadin has quit IRC | 17:50 | |
| rosmaita | jokke_ https://review.openstack.org/#/c/544184/ has passed tests if you want to +A it | 17:50 |
| rosmaita | https://review.openstack.org/#/c/544105/ also | 17:51 |
| *** rmcall has joined #openstack-glance | 17:56 | |
| jokke_ | rosmaita: just did | 17:57 |
| jokke_ | ros also 542956 passed tests couple of days ago :P | 17:58 |
| jokke_ | rosmaita: so about 544105 ... does that change affect current deployments ... I mean is there any issues if there is mixed values in the db? | 18:00 |
| rosmaita | there won't be mixed values in the db | 18:01 |
| rosmaita | mysql will always have 0 or 1 | 18:01 |
| rosmaita | postgres always t or f | 18:01 |
| jokke_ | aha! ... ok cool that makes lots of sense then | 18:02 |
| rosmaita | yeah, i actually tested to make sure | 18:02 |
| jokke_ | sqlalchemy just does the translation on the fly when T/F is used but not when 1/0 is used I assume | 18:02 |
| rosmaita | yes, that seems to be what was happening | 18:03 |
| jokke_ | gating | 18:03 |
| rosmaita | about your uri patch, i thnk the opt help text will need revision, do you want to hit it on a followup patch? | 18:03 |
| rosmaita | also, there's a weird situation with the default port for a scheme | 18:03 |
| rosmaita | you can blacklist 80, but a request to http://whatever.net/path/here will go through | 18:04 |
| *** hoonetorg has quit IRC | 18:05 | |
| rosmaita | but http://whatever.net:80/path/here will fail | 18:05 |
| rosmaita | i don't know whether that's something to worry about or not | 18:06 |
| jokke_ | ohh darn, true ... so we might want to document that beahvior bit better at least | 18:06 |
| jokke_ | the problem really is that urllib.parse does not return the default ports | 18:07 |
| rosmaita | right | 18:07 |
| rosmaita | i will put up a doc patch, i was writing down how this is working | 18:08 |
| jokke_ | do you want to put hell of a lot messy logic to that filter function to work around that or should we just document it really clearly that if one wants to block http calls to port 80 they really need to block http scheme not the port | 18:08 |
| *** mvk_ has quit IRC | 18:08 | |
| rosmaita | i'm not sure, you may want to allow http on 8080 but disallow it on 80 | 18:10 |
| rosmaita | but i agree the logic is going to be ugly | 18:11 |
| jokke_ | how about we document it for now, being the limitation of the implementation and we do not clutter that specific function | 18:11 |
| rosmaita | i'm cool with that | 18:11 |
| rosmaita | we do need tests for the util function, though ... i started to spot test, but there is way too much to look at | 18:12 |
| jokke_ | if there comes need for something like blocking the default ports and allowing to some other port, we write separate function that pulls us the default ports of all available schemes and crosschecks | 18:12 |
| rosmaita | yeah, that would be a good solution | 18:12 |
| jokke_ | I think that filter function is bit difficult to understand as it is already now, it was just honestly the cleanest way I could come up with to implement it | 18:12 |
| rosmaita | actually the function was easier for me to understand than the help text! | 18:13 |
| jokke_ | they do complement eachother ... I'm not great explaining split logic like that | 18:14 |
| rosmaita | i will put up that doc patch right away | 18:14 |
| rosmaita | the help texts are not easy to write | 18:15 |
| jokke_ | it took me quite a while to get those 3 layer nested conditions right :P | 18:16 |
| *** hoonetorg has joined #openstack-glance | 18:22 | |
| *** alexchadin has joined #openstack-glance | 18:32 | |
| *** openstackgerrit has quit IRC | 18:33 | |
| *** openstackgerrit has joined #openstack-glance | 18:34 | |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: Update admin docs for web-download import method https://review.openstack.org/544596 | 18:34 |
| rosmaita | jokke_ ^^ | 18:34 |
| *** alexchadin has quit IRC | 18:36 | |
| jokke_ | all the c.s pprobably should have continue to +1 as well for clarity | 18:39 |
| jokke_ | not sure 'though | 18:40 |
| jokke_ | or the b. should have "Otherwise, skip c" rather than "Otherwise, continue to +1" | 18:41 |
| rosmaita | i was wondering about that | 18:43 |
| rosmaita | yeah, probably better to be completely explicit | 18:44 |
| rosmaita | i'll get that after lunch | 18:44 |
| rosmaita | also noticed that i have to get the "experimental" stuff out of there, will do in a different patch | 18:44 |
| jokke_ | sure ... making comment of one other error in the PS so you have note in there | 18:44 |
| rosmaita | ok, cool, ty | 18:45 |
| *** e0ne has joined #openstack-glance | 18:52 | |
| *** tesseract has quit IRC | 18:54 | |
| *** alexchadin has joined #openstack-glance | 19:07 | |
| *** harlowja has joined #openstack-glance | 19:21 | |
| *** Jimbagwell has joined #openstack-glance | 19:34 | |
| *** alexchadin has quit IRC | 19:39 | |
| *** alexchadin has joined #openstack-glance | 20:21 | |
| *** alexchadin has quit IRC | 20:25 | |
| rosmaita | jokke_ if you are around, i think i found a problem with https://review.openstack.org/#/c/542956/7 (should be a quick fix, though) | 20:51 |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: Update admin docs for web-download import method https://review.openstack.org/544596 | 20:59 |
| *** aniketh has quit IRC | 21:10 | |
| *** btully has joined #openstack-glance | 21:12 | |
| *** konan has joined #openstack-glance | 21:13 | |
| *** btully has quit IRC | 21:17 | |
| *** pcaruana has quit IRC | 21:28 | |
| openstackgerrit | Merged openstack/glance master: Cleanup basic import tasks https://review.openstack.org/544184 | 21:34 |
| openstackgerrit | Merged openstack/glance master: Use bool instead of int for boolean filter value https://review.openstack.org/544105 | 21:34 |
| *** mvk_ has joined #openstack-glance | 21:38 | |
| *** mdbooth has quit IRC | 21:39 | |
| *** rcernin has joined #openstack-glance | 22:00 | |
| *** rmcall has quit IRC | 22:05 | |
| jokke_ | now I am | 22:05 |
| jokke_ | rosmaita: still around? | 22:09 |
| rosmaita | jokke_ yep | 22:11 |
| rosmaita | bad luck on 544511, gate seems kind of slow even though there's not a lot of stuff in it | 22:12 |
| jokke_ | as said it got back to normal | 22:14 |
| jokke_ | so he port option, yes good catch ... I was actually testing that feature when I was still splitting the urlparse netloc output before I realized that it provides hostname and port ... didn't realize that port was actually integer returned not string | 22:16 |
| jokke_ | I mean testing the function | 22:16 |
| rosmaita | right, i wasn't sure but i checked, it's <type 'int'> | 22:17 |
| rosmaita | came up when i was writing the doc, was wondering about the string | 22:17 |
| jokke_ | yes it is and the behavious ir exactly like you said ,,, it won't match on the list | 22:17 |
| jokke_ | is | 22:17 |
| rosmaita | i think the logic looks good on that, i may write some tests later because there's too much to spot check | 22:19 |
| jokke_ | just reading about the hostaddresss | 22:22 |
| rosmaita | pretty sure dharini wrote that one | 22:22 |
| rosmaita | took a long time to merge into oslo.config iirc | 22:22 |
| jokke_ | I quess it will be ok to use that instead of string as we block using file:// already by demanding that there is host | 22:25 |
| rosmaita | that's a good point, though | 22:27 |
| rosmaita | couldn't you fake it if you had a file like file://openstack.org/whatever.file ? | 22:28 |
| rosmaita | oh you need file:/// | 22:28 |
| jokke_ | yup and file:/// gives always empty netloc no hostname nor port | 22:29 |
| rosmaita | cool, glad it's covered | 22:30 |
| *** threestrands has joined #openstack-glance | 22:31 | |
| *** threestrands has quit IRC | 22:31 | |
| *** threestrands has joined #openstack-glance | 22:31 | |
| jokke_ | ok, so there is one case I can see the hostaddress not working due to it's too strict value check, but I'm not sure how likely/valid usecase that is | 22:31 |
| rosmaita | just out of curiosity, what is it? | 22:33 |
| jokke_ | nah can't be bothered as that will just fuck it up between ipv6 and host:port anyways | 22:34 |
| rosmaita | if it comes up, we can always file a bug with oslo.config :) | 22:35 |
| rosmaita | i can see that the enhancement request for this will be regex matching | 22:36 |
| jokke_ | so say the deployment is using numeric hostnames and they rely on default search in dns like mycloud.local ... one could have valid request to something like http://25735/my.img which will resolve for urllib through dns 25735.mycloud.local = 172.25.7.35 . | 22:36 |
| jokke_ | but thta such a corner case compared to the benefits the hostaddress would give us | 22:36 |
| jokke_ | and I do need to push new patch anyways | 22:37 |
| rosmaita | well, we can think about it for RC-3 | 22:37 |
| rosmaita | or Rocky | 22:37 |
| rosmaita | i will carry over abhishek's +2 to the new PS and +2A it | 22:38 |
| rosmaita | you can hit the cherry-picks to stable/queens when you start work in the morning | 22:39 |
| jokke_ | was there anything else than those 4 list item types? | 22:39 |
| rosmaita | no, that was it, i haven't had time to try to reword the config help text | 22:40 |
| jokke_ | feel free to propose follow up patch for those :P | 22:40 |
| rosmaita | will do | 22:40 |
| rosmaita | gotta go cook dinner, be back in a few hours ... happy valentine's day | 22:41 |
| jokke_ | U2 | 22:41 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: URI filtering for web-download https://review.openstack.org/542956 | 22:49 |
| *** itlinux has quit IRC | 22:53 | |
| *** edmondsw has quit IRC | 23:12 | |
| *** Jimbagwell has quit IRC | 23:18 | |
| *** konan has quit IRC | 23:23 | |
| *** e0ne has quit IRC | 23:36 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!