| *** itlinux has joined #openstack-glance | 00:06 | |
| *** threestrands has joined #openstack-glance | 00:27 | |
| *** threestrands has quit IRC | 00:27 | |
| *** threestrands has joined #openstack-glance | 00:27 | |
| *** itlinux has quit IRC | 00:34 | |
| *** zhurong has joined #openstack-glance | 00:47 | |
| *** harlowja has joined #openstack-glance | 00:57 | |
| *** dalgaaf has quit IRC | 01:09 | |
| *** dalgaaf has joined #openstack-glance | 01:10 | |
| *** zhurong has quit IRC | 01:11 | |
| *** zhurong has joined #openstack-glance | 01:14 | |
| *** threestrands has quit IRC | 01:29 | |
| *** Nil_ has quit IRC | 01:38 | |
| *** r-daneel has quit IRC | 01:45 | |
| *** namnh has joined #openstack-glance | 02:37 | |
| *** ChanServ changes topic to "OpenStack Glance" | 03:10 | |
| *** ChanServ changes topic to "OpenStack Glance | priorities: http://lists.openstack.org/pipermail/openstack-dev/2018-February/127181.html" | 03:10 | |
| *** ChanServ changes topic to "OpenStack Glance | priorities: http://lists.openstack.org/pipermail/openstack-dev/2018-February/127181.html | logs at http://eavesdrop.openstack.org/irclogs/%23openstack-glance/" | 03:10 | |
| *** nicolasbock has quit IRC | 03:39 | |
| *** nicolasbock has joined #openstack-glance | 03:49 | |
| *** nicolasbock has quit IRC | 03:56 | |
| *** rosmaita has quit IRC | 04:02 | |
| *** udesale has joined #openstack-glance | 04:02 | |
| *** namnh has quit IRC | 04:13 | |
| *** abhishekk has joined #openstack-glance | 04:22 | |
| *** zhurong has quit IRC | 04:35 | |
| Shilpa | abhishekk: GM | 04:42 |
|---|---|---|
| abhishekk | Shilpa, hi, good morning | 04:47 |
| Shilpa | can we discuss LP1747869 now? | 04:47 |
| abhishekk | yes | 04:49 |
| *** links has joined #openstack-glance | 04:52 | |
| abhishekk | Shilpa, is it ok if we discuss it after an hour? | 04:53 |
| Shilpa | ok | 04:53 |
| *** links has quit IRC | 04:54 | |
| *** links has joined #openstack-glance | 05:03 | |
| *** links has quit IRC | 05:10 | |
| *** harlowja has quit IRC | 05:10 | |
| *** links has joined #openstack-glance | 05:12 | |
| openstackgerrit | baiwenteng proposed openstack/glance master: Fix typos in glance https://review.openstack.org/542597 | 05:23 |
| *** pdeore has joined #openstack-glance | 05:30 | |
| *** harlowja has joined #openstack-glance | 05:35 | |
| *** zhurong has joined #openstack-glance | 05:36 | |
| *** harlowja has quit IRC | 05:39 | |
| *** ratailor has joined #openstack-glance | 05:40 | |
| *** pgadige__ has joined #openstack-glance | 05:49 | |
| *** itlinux has joined #openstack-glance | 05:58 | |
| *** e0ne has joined #openstack-glance | 06:05 | |
| *** e0ne has quit IRC | 06:06 | |
| openstackgerrit | Xianjin Shao proposed openstack/glance master: Description about the parmeters 'op', 'path' and 'value' when update image property or metadata https://review.openstack.org/541508 | 06:26 |
| *** itlinux has quit IRC | 06:27 | |
| *** MattMan has quit IRC | 06:30 | |
| *** pbourke has quit IRC | 06:30 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/glance master: Imported Translations from Zanata https://review.openstack.org/542642 | 07:03 |
| *** namnh has joined #openstack-glance | 07:03 | |
| *** rcernin has quit IRC | 07:17 | |
| *** AlexeyAbashkin has joined #openstack-glance | 07:20 | |
| *** e0ne has joined #openstack-glance | 07:27 | |
| openstackgerrit | Xianjin Shao proposed openstack/glance master: Description about the parmeters 'op', 'path' and 'value' when update image property or metadata https://review.openstack.org/541508 | 07:28 |
| *** pcaruana has joined #openstack-glance | 07:35 | |
| *** ykarel has joined #openstack-glance | 07:42 | |
| ykarel | jokke_, hi | 07:43 |
| ykarel | jokke_, please check my comment in your merged patch: https://review.openstack.org/#/c/541740 | 07:44 |
| *** AlexeyAbashkin has quit IRC | 07:44 | |
| *** alexchadin has joined #openstack-glance | 07:48 | |
| *** AlexeyAbashkin has joined #openstack-glance | 07:54 | |
| Shilpa | abhishekk: is it ok to discuss LP1747869 now? | 07:55 |
| openstackgerrit | yatin proposed openstack/glance master: Fix wrong usage of extend in list_image_import_opts https://review.openstack.org/542699 | 07:58 |
| *** MattMan has joined #openstack-glance | 08:01 | |
| openstackgerrit | yatin proposed openstack/glance master: Fix wrong usage of extend in list_image_import_opts https://review.openstack.org/542699 | 08:02 |
| *** dtantsur|afk is now known as dtantsur | 08:08 | |
| *** mosulica has joined #openstack-glance | 08:09 | |
| openstackgerrit | Xianjin Shao proposed openstack/glance master: Description about the parmeters 'op', 'path' and 'value' when update image property or metadata https://review.openstack.org/541508 | 08:25 |
| *** tesseract has joined #openstack-glance | 08:26 | |
| *** namnh has quit IRC | 08:27 | |
| *** d0ugal has quit IRC | 08:33 | |
| *** d0ugal has joined #openstack-glance | 08:47 | |
| openstackgerrit | Shilpa Devharakar proposed openstack/glance master: Add validation to check if E-M-C is already in sync https://review.openstack.org/540736 | 08:54 |
| *** pbourke has joined #openstack-glance | 08:58 | |
| *** alexchadin has quit IRC | 09:07 | |
| *** alexchadin has joined #openstack-glance | 09:08 | |
| *** ykarel is now known as ykarel|lunch | 09:08 | |
| *** zhurong has quit IRC | 09:16 | |
| openstackgerrit | Xianjin Shao proposed openstack/glance master: Description about the parmeters 'op', 'path' and 'value' when update image property or metadata https://review.openstack.org/541508 | 09:19 |
| abhishekk | jokke_, you around | 09:24 |
| jokke_ | yeah | 09:24 |
| abhishekk | jokke_, please have a look at https://review.openstack.org/#/c/542699/2, its important we need to backport it as well | 09:24 |
| openstackgerrit | yatin proposed openstack/glance master: Fix wrong usage of extend in list_image_import_opts https://review.openstack.org/542699 | 09:26 |
| jokke_ | yep ... it seems like our config refresher did not have that patch it ... | 09:32 |
| abhishekk | right | 09:32 |
| jokke_ | just waiting for the gate | 09:34 |
| jokke_ | ykarel|lunch: thanks, nice catch | 09:35 |
| *** edmondsw has joined #openstack-glance | 09:40 | |
| abhishekk | jokke_, ok | 09:41 |
| abhishekk | jokke_, Pushing one patch for postgresql fix, I will set workflow to -1 initially so that you and rosmaita can decide whether its good approach or not | 09:42 |
| jokke_ | abhishekk: thanks, I like it already | 09:44 |
| abhishekk | :D | 09:44 |
| *** edmondsw has quit IRC | 09:45 | |
| abhishekk | Shilpa, hi I am sorry,I haven't saw your message | 09:49 |
| Shilpa | ok, actually i am also working on solution to resovle postgresql fix, and before finalizing, need to discuss the same with you | 09:49 |
| abhishekk | Shilpa, wait for sometime, I am proposing a patch soon to fix it | 09:50 |
| abhishekk | till then you can ask me your doubts | 09:50 |
| Shilpa | is glance supports any other engine for offline upgrade other than ,ysql/sqlite/postgres | 09:52 |
| Shilpa | mysql* | 09:52 |
| Shilpa | or we need to consider issue for postgres only? | 09:53 |
| *** AlexeyAbashkin has quit IRC | 09:56 | |
| abhishekk | no consider postgresql only | 09:56 |
| *** AlexeyAbashkin has joined #openstack-glance | 09:57 | |
| Shilpa | ok | 09:59 |
| Shilpa | as of now i have verified configuring postgres and verifying running EMC scripts on it, and they are executing with no errors | 09:59 |
| Shilpa | i have also verified original defect, upgrading from ocata to pike on postgres and yes that also success | 10:00 |
| abhishekk | Shilpa, that is because in EMC script we don't have any constraint which will not work on postgresql | 10:00 |
| abhishekk | If you add that constraint then EMC will fail for postgresql | 10:00 |
| openstackgerrit | Abhishek Kekane proposed openstack/glance master: Offline migration support for postgresql https://review.openstack.org/542756 | 10:06 |
| *** bhagyashri_s has joined #openstack-glance | 10:06 | |
| *** nsingh has joined #openstack-glance | 10:06 | |
| *** pooja-jadhav has joined #openstack-glance | 10:06 | |
| *** pooja_jadhav has quit IRC | 10:07 | |
| *** Shilpa has quit IRC | 10:07 | |
| *** Shilpa has joined #openstack-glance | 10:07 | |
| *** bhagyashris has quit IRC | 10:08 | |
| *** niraj_singh has quit IRC | 10:08 | |
| abhishekk | Shilpa, ^^^ | 10:08 |
| Shilpa | sorry Abhishekk, got disconnected | 10:08 |
| abhishekk | jokke_, ^^^ | 10:08 |
| abhishekk | Shilpa, no issues, I have just pushed a patch, please have a look | 10:08 |
| Shilpa | ok, sure | 10:09 |
| *** ShilpaSD has joined #openstack-glance | 10:09 | |
| *** jcfro- has joined #openstack-glance | 10:11 | |
| jcfro- | (_) | | | 10:11 |
| jcfro- | (_) | | | 10:11 |
| jcfro- | (_) | | | 10:11 |
| *** ykarel|lunch is now known as ykarel | 10:11 | |
| jcfro- | _ _ __ ___ ___ _ _ _ __ ___ _ __ _ __ ___| |_ ___ ___ _ __ __ _ | 10:11 |
| jcfro- | _ _ __ ___ ___ _ _ _ __ ___ _ __ _ __ ___| |_ ___ ___ _ __ __ _ | 10:11 |
| jcfro- | _ _ __ ___ ___ _ _ _ __ ___ _ __ _ __ ___| |_ ___ ___ _ __ __ _ | 10:11 |
| jcfro- | | | '__/ __| / __| | | | '_ \ / _ \ '__| '_ \ / _ \ __/ __| / _ \| '__/ _` | | 10:11 |
| jcfro- | | | '__/ __| / __| | | | '_ \ / _ \ '__| '_ \ / _ \ __/ __| / _ \| '__/ _` | | 10:11 |
| jcfro- | | | '__/ __| / __| | | | '_ \ / _ \ '__| '_ \ / _ \ __/ __| / _ \| '__/ _` | | 10:11 |
| jcfro- | | | | | (__ _\__ \ |_| | |_) | __/ | | | | | __/ |_\__ \| (_) | | | (_| | | 10:11 |
| jcfro- | | | | | (__ _\__ \ |_| | |_) | __/ | | | | | __/ |_\__ \| (_) | | | (_| | | 10:11 |
| jcfro- | | | | | (__ _\__ \ |_| | |_) | __/ | | | | | __/ |_\__ \| (_) | | | (_| | | 10:11 |
| jcfro- | |_|_| \___(_)___/\__,_| .__/ \___|_| |_| |_|\___|\__|___(_)___/|_| \__, | | 10:11 |
| jcfro- | |_|_| \___(_)___/\__,_| .__/ \___|_| |_| |_|\___|\__|___(_)___/|_| \__, | | 10:11 |
| jcfro- | |_|_| \___(_)___/\__,_| .__/ \___|_| |_| |_|\___|\__|___(_)___/|_| \__, | | 10:11 |
| jcfro- | | | __/ | | 10:12 |
| jcfro- | | | __/ | | 10:12 |
| jcfro- | | | __/ | | 10:12 |
| jcfro- | |_| |___/ | 10:12 |
| jcfro- | |_| |___/ | 10:12 |
| jcfro- | |_| |___/ | 10:12 |
| jcfro- | ShilpaSD Shilpa pooja-jadhav nsingh bhagyashri_s AlexeyAbashkin alexchadin pbourke d0ugal tesseract mosulica MattMan ykarel|lunch pcaruana e0ne pgadige__ ratailor pdeore links abhishekk udesale dalgaaf mvk_ gaurangt lbragstad DuncanT cargonza openstackgerrit esberglu tbarron sapd mnaser _alastor__ hoonetorg masber pdr_ idlemind jdillaman pliu wxy NostawRm mtreinish jungleboyj jdurgin kairat zzzeek keekz twouters clayton amea | 10:12 |
| *** jcfro- has quit IRC | 10:12 | |
| ShilpaSD | abhishekk: Hi | 10:12 |
| abhishekk | ShilpaSD, hi | 10:13 |
| ShilpaSD | saw your changes | 10:13 |
| ShilpaSD | constraint issue will be there for postgras, right | 10:14 |
| abhishekk | ShilpaSD, yes in case of online migrations | 10:15 |
| ShilpaSD | but online migrations always support MySQL, right | 10:15 |
| abhishekk | yes | 10:16 |
| ShilpaSD | but during oflline migrations also issue of constraint will be there | 10:16 |
| ShilpaSD | for postgreSQL | 10:17 |
| *** pgadige__ has left #openstack-glance | 10:17 | |
| abhishekk | Shilpa, yes but during offline migration operator can resolve those issues and it will not break anything | 10:17 |
| ShilpaSD | So operators take care of any constraints or other issues during oflline migrations | 10:18 |
| abhishekk | yes, without affection production environment which is not possible if it is online | 10:19 |
| abhishekk | s/affection/affecting | 10:19 |
| ShilpaSD | ok, understood | 10:19 |
| ShilpaSD | so can we not just add postgreSQL support in _validate_engine()? | 10:19 |
| ShilpaSD | as i have added sqlite for funtiobnal Test cases | 10:20 |
| abhishekk | correct | 10:21 |
| abhishekk | that's not the right way | 10:21 |
| ShilpaSD | Actually calling _sync(), here migration scripts are not taking into account. | 10:23 |
| ShilpaSD | Alembic command upgrade, reads versions set inside scripts and execute the sequence | 10:23 |
| ShilpaSD | Earlier pike01/ocata01 files are consolidated files of EMC, and i have removed them in recent fix (https://review.openstack.org/#/c/433934/) | 10:24 |
| abhishekk | ShilpaSD, I doubt that for offline migration we need to run any migration scripts | 10:24 |
| abhishekk | as it does not make any sense | 10:25 |
| ShilpaSD | please elaborate this for my understanding | 10:25 |
| abhishekk | ShilpaSD, check the upgrades specs it is explained in their | 10:26 |
| *** aniketh has joined #openstack-glance | 10:27 | |
| ShilpaSD | yes | 10:28 |
| ShilpaSD | i want to understand why migration (M) is not important for offline | 10:29 |
| *** alexchadin has quit IRC | 10:31 | |
| abhishekk | ShilpaSD, do you know why migration scripts are required? | 10:33 |
| ShilpaSD | yes | 10:33 |
| abhishekk | ShilpaSD, tell me why | 10:34 |
| ShilpaSD | if we have any column altered, to sync data between old and new column, migrate scripts are required | 10:35 |
| abhishekk | yes but that is specifically in online case, | 10:36 |
| abhishekk | in case of offline migration you can directly execute the alter command as you don't need to copy data from one column to another column | 10:37 |
| abhishekk | you can simply rename that column | 10:37 |
| ShilpaSD | totally agree, but concern is earlier to support this there were two types scripts, pike01 and (pike_expand01/pike_migrate01/pikecontract01), and now going further, there won't be such scripts, so how we will provide operators such scripts which will consider alter case. | 10:41 |
| ShilpaSD | now only EMC scripts willbe there* | 10:42 |
| *** dtantsur is now known as dtantsur|bbl | 10:43 | |
| abhishekk | ShilpaSD, try to setup devstack with my patch, I can see visibility column in the images table of postgresql | 10:44 |
| abhishekk | if that was the case, there should be is_public column instead of visibility | 10:45 |
| *** abhishekk is now known as abhishekk|brb | 10:46 | |
| jokke_ | abhishekk|brb: the migration scripts are executed and will need to be executed in offline as well | 10:48 |
| abhishekk|brb | jokke_, ok | 10:48 |
| jokke_ | abhishekk|brb: lets say there is column moved to another table, the data does not just magically appear there when the column is created | 10:48 |
| jokke_ | there is migration needed to sync the old and new column location before removing the old column | 10:49 |
| abhishekk|brb | jokke_, so in my case as I stated earlier I am able to see visibility column in images table, that means it runs those scripts as well | 10:49 |
| abhishekk|brb | jokke_, agree, that is required in case of moving the column to a new table | 10:50 |
| jokke_ | it's not common for us to happen, but oversight of such case would cause horrific consequencies | 10:51 |
| abhishekk|brb | I have just created a new setup where I am running devstack again, from logs I will be able to confirm which scripts are executed | 10:51 |
| jokke_ | cool | 10:51 |
| abhishekk|brb | jokke_, Strongly agree | 10:51 |
| abhishekk|brb | i will be back shortly | 10:52 |
| jokke_ | sure | 10:52 |
| jokke_ | abhishekk|brb: when you come back, wasn't that _sync(version) the same function we used before moving everything running E-M-C? | 10:53 |
| *** ratailor has quit IRC | 10:54 | |
| *** d0ugal has quit IRC | 10:59 | |
| abhishekk|brb | jokke_, yes, that's why I thought it will be easy fix | 10:59 |
| *** abhishekk|brb is now known as abhishekk | 11:00 | |
| *** pbourke has quit IRC | 11:08 | |
| abhishekk | jokke_, as far as I know previously for db sync it was running scripts from db/sqlalchemy/migrate_repo/versions directory | 11:10 |
| abhishekk | it has nothing to do with scripts provided in alembic_migrations repo | 11:11 |
| *** d0ugal has joined #openstack-glance | 11:16 | |
| abhishekk | jokke_, I have created a new setup and the db sync command executes scripts from db/sqlalchemy/migrate_repo/versions directory and then from db/sqlalchemy/alembic_migrations/versions table ( same was the case before) | 11:21 |
| *** abhishekk has quit IRC | 11:24 | |
| *** AlexeyAbashkin has quit IRC | 11:27 | |
| *** alexchadin has joined #openstack-glance | 11:28 | |
| *** pdeore has quit IRC | 11:29 | |
| *** AlexeyAbashkin has joined #openstack-glance | 11:30 | |
| *** udesale has quit IRC | 11:35 | |
| bhagyashri_s | jokke_: Hi, I just want to ask one thing related to hook-- is it possible to make the third party lib task configurable using hook? | 11:39 |
| jokke_ | bhagyashri_s: what do you mean? | 11:44 |
| bhagyashri_s | jokke_: i mean task implementation is in third party lib and that i want to make it configurable and add that task in taskflow | 11:45 |
| jokke_ | bhagyashri_s: if you look at 824badc1890206695eb335582a804d9ecabb4d67 | 11:45 |
| jokke_ | it was made to make it easier for 3rd party plugins to tap in | 11:46 |
| *** nicolasbock has joined #openstack-glance | 11:47 | |
| jokke_ | bhagyashri_s: so if you're deploying plugin that is not provided by Glance release, you drop it to plugins folder and add it to plugin_opts.py in that folder and then you need to add it to glance.image_import.plugins in the setup.cfg | 11:49 |
| jokke_ | so drop your .py file and add it to those two files and you should be able to enable it in the config | 11:50 |
| jokke_ | if it defines any config opts configgen should pick them up and add them into the glance-image-import.conf.sample | 11:51 |
| jokke_ | coffee ... brb | 11:52 |
| bhagyashri_s | bhagyashri_s: ok. will talk later | 11:54 |
| jokke_ | back | 12:05 |
| bhagyashri_s | jokke_: in that case how the inputs will be provided to that tasks to execute it successfully | 12:06 |
| jokke_ | bhagyashri_s: those deployer plugins _needs_ to be transparent for enduser, so you don't | 12:07 |
| jokke_ | they will be in the pipeline just like the metadata injection, it might get config from the deployment's config files, it may use context but you never expect the person who makes the imput call to even know about your plugin | 12:08 |
| bhagyashri_s | jokke_: for example there is one Task say TaskA which is suppose not in glance code that is located in third party lib and that task i want to make it configurable and want to add it in taskflow | 12:09 |
| bhagyashri_s | ok | 12:09 |
| jokke_ | bhagyashri_s: well you would need to make some kind of wrapper for it into the plugins | 12:10 |
| jokke_ | so if you use just random task that is not designed for these limitations, you would need to wrap it to something that can handle all that | 12:10 |
| *** alexchadin has quit IRC | 12:11 | |
| *** alexchadin has joined #openstack-glance | 12:12 | |
| bhagyashri_s | jokke_: ok. and is it possible we can give the class patch of task instead of module name of task under plugin folder in setup.cfg | 12:13 |
| bhagyashri_s | jokke_: or we can also say that giving the class name of task instead of module name in configuration to load that task | 12:16 |
| *** alexchadin has quit IRC | 12:16 | |
| *** alexchadin has joined #openstack-glance | 12:17 | |
| jokke_ | bhagyashri_s: no | 12:17 |
| bhagyashri_s | jokke_: for example no_op = glance.async.flows.plugins.no_op:_Noop in setup.cfg | 12:17 |
| bhagyashri_s | jokke_: ?? | 12:17 |
| jokke_ | the plugin framework needs that task fully initialized from the get_task function | 12:17 |
| jokke_ | so that it can be just dropped into the taskflow queue | 12:18 |
| jokke_ | so you would need to make plugins/my_task_wrapper.py that handles all that and imports and initializes the task object and returns it like the example does | 12:19 |
| jokke_ | so obviously if you have that task coming from some library you don't need to implement the task class in the my_task_wrapper.py, you just import it and implement that get_flow() function | 12:20 |
| *** alexchadin has quit IRC | 12:20 | |
| *** alexchadin has joined #openstack-glance | 12:21 | |
| bhagyashri_s | jokke_: is there any reference where i can see this | 12:21 |
| jokke_ | and then you specify it like my_task = glance.async.flows.plugins.my_task_wrapper:get_flow | 12:21 |
| jokke_ | bhagyashri_s: no | 12:22 |
| bhagyashri_s | jokke_: and plugin framework i want understand more so is there any link | 12:22 |
| jokke_ | I'm working on it, but it does not currently exist and I'm unlikely going cover importing some random task and wrapping it | 12:22 |
| jokke_ | the task framework we're using is taskflow | 12:23 |
| bhagyashri_s | i want to more understand about plugin framework | 12:23 |
| jokke_ | bhagyashri_s: the taskflow is bit like middleware in the wsgi environment ... it's pipeline of features ... taskflow takes care of that pipeline being executed without the user needing to sit there for whole time waiting | 12:25 |
| bhagyashri_s | in short setting class patch or making the task configurable using the class name is not doable directly get_flow is imp | 12:25 |
| bhagyashri_s | jokke_: ok | 12:25 |
| jokke_ | and those plugins are like the wsgi middleware ... it's a task that gets added to the pipeline, it needs to understand the inputs and limitations what info it has available and pass that info through | 12:26 |
| jokke_ | and what we do with the plugins, the no_op is kind of minimal what that task needs to handle so it can be executed | 12:27 |
| jokke_ | the no_op does not do anything but it has all the piping in place so it can be dropped into that pipeline and it won't break it | 12:28 |
| jokke_ | and the whole idea for those plugins is that, lets say cloud provider does not want a single image to be sent to nova before they have been checked by antivirus. They can implement plugin that does the viruscheck on the image data. The end user does not need to even know about this, their image import just takes bit longer | 12:29 |
| jokke_ | so for the sake of interoperability those plugins must not expect any special input from the end user | 12:30 |
| *** bhagyashri_s is now known as bbs | 12:31 | |
| jokke_ | so that the user can make the same import call to 10 different clouds and at the end will have 10 images ready as long as their image is within the constrains of what the cloud provider is accepting | 12:31 |
| bbs | jokke_: ok. | 12:32 |
| *** awestin1 has quit IRC | 12:32 | |
| *** awestin1 has joined #openstack-glance | 12:33 | |
| *** dtantsur|bbl is now known as dtantsur | 12:40 | |
| bbs | jokke_: thank you :) | 12:42 |
| *** pooja-jadhav has quit IRC | 12:48 | |
| *** serverascode has quit IRC | 12:48 | |
| jokke_ | np | 12:48 |
| jokke_ | I have talk proposed to Vancouver summit about the plugins | 12:49 |
| *** serverascode has joined #openstack-glance | 12:49 | |
| jokke_ | so even if you can't be there in person for it, please vote it to be accepted as all of the talks will be recorded | 12:49 |
| *** ameade has quit IRC | 12:50 | |
| *** ameade has joined #openstack-glance | 12:51 | |
| jokke_ | and I'll try to get written documentation for the Import plugins done as well ... there has been just so much firefighting on this cycle to be even able to get everything into the release we wanted that I haven't had time for it | 12:51 |
| *** mosulica has quit IRC | 12:54 | |
| *** tesseract has quit IRC | 13:12 | |
| openstackgerrit | Merged openstack/glance master: Fix wrong usage of extend in list_image_import_opts https://review.openstack.org/542699 | 13:14 |
| bbs | jokke_: ok thank you :) for your inputs and sure i will vote | 13:16 |
| bbs | jokke_: actually my use case was to give the class patch to configure any task but it's difficult to do this | 13:17 |
| *** konan has joined #openstack-glance | 13:27 | |
| konan | jokke_: Abhishek here, sorry I was driving and didn't noticed you have also cherry-picked that patch to stable/queens | 13:28 |
| jokke_ | konan: oh, you did too? | 13:29 |
| jokke_ | no worries ... I was just following it and did the backport as soon as I saw the master merging | 13:30 |
| konan | Yes, it has uploaded new ps to the same review | 13:30 |
| konan | Will take care next time | 13:30 |
| jokke_ | konan: don't worry about it ... I was expecting you to be home already enjoying your well deserved weekend | 13:31 |
| *** konan has quit IRC | 13:31 | |
| jokke_ | konan: I do hope you compensate these late evenings with late mornings as well | 13:31 |
| *** konan has joined #openstack-glance | 13:32 | |
| konan | jokke_: thank you :) | 13:32 |
| jokke_ | The fact that I do poor work/life balance choices does not mean that others should do so too :P | 13:34 |
| *** alexchadin has quit IRC | 13:39 | |
| aniketh | jokke_: Hi | 13:42 |
| aniketh | jokke_: Just completed the whole setup to do my first commit, whole Gerrit, account setup and stuff. | 13:44 |
| aniketh | Can you point me to some smaller task that I could try to fix and get acquainted with the codebase? | 13:44 |
| *** alexchadin has joined #openstack-glance | 13:44 | |
| * aniketh reading the doc on Glance architecture. | 13:45 | |
| *** alexchadin has quit IRC | 13:47 | |
| *** alexchadin has joined #openstack-glance | 13:49 | |
| *** alexchadin has quit IRC | 13:52 | |
| *** Supun has joined #openstack-glance | 13:52 | |
| *** edmondsw has joined #openstack-glance | 13:54 | |
| *** links has quit IRC | 13:56 | |
| *** alexchadin has joined #openstack-glance | 14:01 | |
| *** rosmaita has joined #openstack-glance | 14:03 | |
| openstackgerrit | Merged openstack/glance master: Imported Translations from Zanata https://review.openstack.org/542642 | 14:05 |
| jokke_ | rosmaita: ping | 14:16 |
| rosmaita | jokke_ pong | 14:16 |
| *** kairat has quit IRC | 14:18 | |
| jokke_ | rosmaita: when you have time, couple of things around rc2 | 14:20 |
| jokke_ | let me know when you're caffeinated | 14:20 |
| rosmaita | i guess now is ok | 14:22 |
| rosmaita | i have a question about https://review.openstack.org/#/c/542814/2/glance/opts.py | 14:22 |
| jokke_ | that was one of the reviews I wanted to flag | 14:22 |
| jokke_ | so shoot | 14:22 |
| rosmaita | if list_image_import_opts() is called more than once, isn't the list going to get longer and longer because of the extend? | 14:23 |
| rosmaita | i think you'll add the plugin opts over and over | 14:23 |
| jokke_ | good point, it will and that was one of the reasons I wanted to have it in the local var, now I'm not sure if anything else than configgen actually calls that function | 14:24 |
| rosmaita | mabye we want _image_import_opts + get_plugin_opts | 14:25 |
| jokke_ | but I'm more than willing to make combination of those two to ensure that continuous expansion does not happen | 14:25 |
| rosmaita | i'm thinking your first intuition was close | 14:25 |
| rosmaita | do something like opts = importopts + plugin_opts | 14:25 |
| rosmaita | and then use that in the list comprehension in the return | 14:26 |
| jokke_ | yeah, so lets rever the patch from master and redo it properly | 14:27 |
| rosmaita | ok, i can do that | 14:27 |
| *** ShilpaSD has quit IRC | 14:28 | |
| jokke_ | then there is patch from Abhishek fixing our postgre issue ... quite well I'd say if I understand it correctly | 14:28 |
| jokke_ | it just needs bunch of documentation around it | 14:28 |
| rosmaita | oh man that is fantastic | 14:28 |
| *** tesseract has joined #openstack-glance | 14:28 | |
| rosmaita | ok, well i can work up some documentation while trying to understand his patch | 14:28 |
| jokke_ | and the thrid one is the web-download I wanted to talk about | 14:29 |
| jokke_ | so I just realized that we enabled it by default | 14:29 |
| rosmaita | right | 14:29 |
| rosmaita | i was going to add a security note | 14:29 |
| jokke_ | and it's still just feature parr with the v1 as in the security issue is still there | 14:29 |
| rosmaita | that's what i was thinking | 14:30 |
| rosmaita | but we could ship with it "off" | 14:30 |
| rosmaita | either way | 14:30 |
| rosmaita | but we do need to bring it to operators attention | 14:30 |
| rosmaita | i guess the responsible thing is security note _+ "off" by default | 14:30 |
| jokke_ | I'm working on plugging that, _but_ do we want to have that backported as bugfix for it, or is it a feature that needs to land in rocky in which case we probably should drop it from the defaults | 14:30 |
| rosmaita | how complicated will it be to do? | 14:31 |
| rosmaita | if you can get it done so we can release it as part of rc-2 on thursday | 14:31 |
| rosmaita | then we have roughly a week to test | 14:31 |
| jokke_ | not so ... urllib.parse decomposes the uri quite nicely, then it's just matter of comparison if those components are in black-/whitelists that needs to be introduced as config options | 14:32 |
| jokke_ | lots of rows due to those configs and their helptexts but the logic is quite simple | 14:32 |
| rosmaita | i personally would like to see it in queens (of course) | 14:32 |
| rosmaita | let's do it as a bugfix | 14:33 |
| rosmaita | and if it gets complicated, we go with plan B | 14:33 |
| jokke_ | cool, I have meeting with my manager at the top of the hour but I'll try to get it done after so you can have a look still today | 14:33 |
| jokke_ | need to get the bug effectively copied and targeted to -rc2 then | 14:33 |
| rosmaita | man, well if we have your and abhishek's stuff working, we can do rc-2 really early and have lots of testing time | 14:34 |
| jokke_ | yes I'd like to get rc2 out Mon or Tue | 14:34 |
| jokke_ | unless something else than the once we're currently aware of pops out | 14:34 |
| jokke_ | ones | 14:34 |
| jokke_ | and preferably that without working whole weekend to make it happen :P | 14:35 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: Revert "Fix wrong usage of extend in list_image_import_opts" https://review.openstack.org/542855 | 14:38 |
| *** stephenfin is now known as finucannot | 14:38 | |
| rosmaita | jokke_ i was just about to revert https://review.openstack.org/#/c/542699/ | 14:39 |
| rosmaita | am i looking at the wrong review? | 14:39 |
| rosmaita | dumb me ... yours is the actual reversion | 14:41 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: Fix bad usage of extend in list_image_import_opts https://review.openstack.org/542860 | 14:43 |
| rosmaita | i think deepcopy is a better choice than + there! | 14:44 |
| *** prometheanfire has left #openstack-glance | 14:44 | |
| jokke_ | rosmaita: those two are revert and fix ^^ | 14:44 |
| rosmaita | cool, should i ninja them in? | 14:44 |
| jokke_ | bit less performant and memory hog but lots safer and those lists should never get big anyways | 14:45 |
| rosmaita | yeah, and pretty unlikely they'll be called more than a few times | 14:45 |
| jokke_ | indeed | 14:45 |
| jokke_ | my original patch was kind of hastely bad thinking of how that list extend works | 14:46 |
| jokke_ | tried to compress it too much ... lines are cheap :P | 14:46 |
| rosmaita | well, it looked good ... pretty sure both me and abhishek ok'd it | 14:46 |
| jokke_ | yeah ... we all thought it wrong way ... fortunately it was quick fix, the second time we overlooked it :P | 14:47 |
| jokke_ | *after the second time* | 14:47 |
| jokke_ | _really_ nice catch btw right off the bat at the morning :) | 14:48 |
| rosmaita | yeah, and only at 33% caffeination! | 14:49 |
| rosmaita | i am going to get up to 66% and then tackle abhishek's postgres patch | 14:49 |
| jokke_ | cool, fresh coffee, meeting and then back to that uri filter patch | 14:49 |
| jokke_ | I have it like half way done | 14:49 |
| *** hogepodge has quit IRC | 14:51 | |
| *** hogepodge has joined #openstack-glance | 14:52 | |
| *** Shilpa has quit IRC | 14:53 | |
| *** bbs has quit IRC | 14:54 | |
| rosmaita | cool, verified your change locally, as soon as those pass zuul check, i will ninja them in | 14:54 |
| *** Supun has quit IRC | 14:55 | |
| *** Supun has joined #openstack-glance | 14:56 | |
| *** ykarel is now known as ykarel|away | 14:56 | |
| *** r-daneel has joined #openstack-glance | 14:57 | |
| *** ykarel|away has quit IRC | 15:01 | |
| *** Supun has quit IRC | 15:02 | |
| aniketh | rosmaita: Ping | 15:04 |
| aniketh | Could we talk? Sorry for being late. Just time line issue :( | 15:04 |
| rosmaita | aniketh how about at 15:30 utc? | 15:04 |
| aniketh | rosmaita: Yes, That works perfectly fine :D | 15:05 |
| rosmaita | olk | 15:05 |
| *** alexchadin has quit IRC | 15:08 | |
| *** ratailor has joined #openstack-glance | 15:11 | |
| aniketh | rosmaita: Hey, you there? | 15:32 |
| rosmaita | hello | 15:32 |
| aniketh | Just repeating what I said earlier today in the channel | 15:33 |
| aniketh | Just completed the whole setup to do my first commit, whole Gerrit, account setup and stuff. | 15:33 |
| aniketh | Can you point me to some smaller task that I could try to fix and get acquainted with the codebase? | 15:33 |
| rosmaita | sorry, i was offline and didn't read the scrollback | 15:34 |
| rosmaita | give me a few min to locate something | 15:34 |
| aniketh | Yes, I know it is fine :) | 15:34 |
| aniketh | Also, what I'm now unaware of is how to reproduce a bug, that I chose to fix and if I made some changes in the code base regarding that - how could I test that, to verify if the fix is right or not? I guess this is because - I have cloned the repo of Glance into my localhost, I don't know how to run that or build it. Knowing this, would help the above I assume? | 15:36 |
| aniketh | rosmaita: ^ | 15:36 |
| rosmaita | aniketh so the bug should have some description of how it happens | 15:37 |
| rosmaita | at that point, you have 2 choices: | 15:37 |
| aniketh | yes | 15:37 |
| rosmaita | reproduce it in your devstack | 15:37 |
| rosmaita | write a test that fails (that is, catches the bug) in your local repo | 15:38 |
| rosmaita | it's usually best to do both | 15:38 |
| aniketh | ok | 15:38 |
| rosmaita | reproducing in devstack is helpful so you can see what's going on | 15:38 |
| rosmaita | and you'll probably be asked to have a test along with the bugfix anyway | 15:38 |
| aniketh | Yeah! -- oh, okay! Alright. | 15:39 |
| rosmaita | but on your local machine, you should be able to do tox -e py27 to run the unit tests | 15:39 |
| rosmaita | if you just do tox | 15:39 |
| rosmaita | it will run everything | 15:39 |
| *** AlexeyAbashkin has quit IRC | 15:40 | |
| rosmaita | and you can also run just specific tests | 15:40 |
| rosmaita | so that you don't have to wait forever to see what happens | 15:41 |
| *** AlexeyAbashkin has joined #openstack-glance | 15:42 | |
| aniketh | tox -e py27 -- project_name_here/tests/folder_name1_here/folder_name2_here/test_file_name_here.py:TestClassName.test_method_name | 15:42 |
| aniketh | ? | 15:42 |
| rosmaita | yeah, something like that ... i have to admit, i do it the hard way (activate the appropriate tox virtualenv and then run python -m testtools.run <fully-qualified-test-name> | 15:43 |
| rosmaita | aniketh were you able to get devstack running? | 15:44 |
| aniketh | rosmaita: Oh, which is better to be followed then? | 15:44 |
| aniketh | rosmaita: Yes | 15:44 |
| rosmaita | aniketh tox , if it works for you | 15:44 |
| aniketh | okay! I haven't tried it yet. Will try and see while writing tests. | 15:45 |
| *** AlexeyAbashkin has quit IRC | 15:46 | |
| *** Supun has joined #openstack-glance | 15:48 | |
| aniketh | rosmaita: So, after cloning the Glance repo in the local system, I don't have to build it or do anything else right? | 15:48 |
| rosmaita | right, you should be able to run all the commands out of tox without actually building and/or installing glance | 15:49 |
| aniketh | okay, alright :D | 15:50 |
| rosmaita | (still looking for something for you) | 15:50 |
| aniketh | also, btw, you mentioned about functional testing as well inside Glance itself. Could you help me with that? | 15:50 |
| aniketh | rosmaita: Thanks :) | 15:51 |
| rosmaita | sure | 15:51 |
| rosmaita | when you look at the glance repo, you'll find glance/tests/functional | 15:51 |
| rosmaita | that's where they live | 15:51 |
| rosmaita | you can look at the base classes, they set up servers for the tests to interact with | 15:51 |
| rosmaita | you run them with tox -e functional or tox -e functional-py35 | 15:52 |
| aniketh | oh okay. | 15:54 |
| aniketh | Btw, all the tests to be written and which are already written exist in glance/tests/ right? | 15:55 |
| rosmaita | right, they're organized into sub directories for unit, integration, and functional | 15:55 |
| rosmaita | and then subdirectories under that for various components | 15:56 |
| aniketh | Yes - Neat! | 15:56 |
| aniketh | I believe - if I get some small bug to fix, I could get acquainted with the codebase a bit as well as understand more of this structure in Glance even more and even better. | 15:59 |
| *** r-daneel has quit IRC | 15:59 | |
| rosmaita | you are right about that! | 15:59 |
| *** Supun has quit IRC | 16:00 | |
| aniketh | :D | 16:01 |
| aniketh | rosmaita: Also, if there is some documentation that I should be reading as a beginner or which I should be visiting quite often(Like devs manual), please do point to me that as well :D | 16:01 |
| *** pcaruana has quit IRC | 16:01 | |
| rosmaita | aniketh i put something together about that for the last summit, i should be able to find it real quickly | 16:01 |
| aniketh | Right now, I am going through the following | 16:01 |
| aniketh | https://docs.openstack.org/glance/pike/contributor/index.html | 16:02 |
| aniketh | https://docs.openstack.org/infra/manual/developers.html | 16:02 |
| aniketh | https://docs.openstack.org/glance/latest/ | 16:02 |
| aniketh | rosmaita: oh, that would be great, If I can get my hand to it ;) | 16:02 |
| rosmaita | those are all good | 16:04 |
| rosmaita | this might have some good tips, too: https://www.openstack.org/videos/boston-2017/how-to-train-an-openstack-dragon | 16:04 |
| *** ratailor has quit IRC | 16:04 | |
| *** e0ne has quit IRC | 16:06 | |
| * aniketh going through the video :D | 16:07 | |
| rosmaita | this has some info about contributing: https://etherpad.openstack.org/p/SYD-forum-glance-onboarding | 16:09 |
| *** itlinux has joined #openstack-glance | 16:10 | |
| jokke_ | aniketh: also review lots of patches submitted to gerrit. That makes all the difference understanding how people approaches the issues and how the code is constructed | 16:16 |
| aniketh | jokke_: Yes, Thanks for the suggestion. I was doing that earlier. Even though, I wasn't able to understand much :P | 16:19 |
| aniketh | jokke_: I will continue again. :D | 16:19 |
| *** konan has quit IRC | 16:21 | |
| aniketh | rosmaita: Thanks for the links and the video, will dig into those today and probably tomorrow (IST) as well. | 16:28 |
| aniketh | rosmaita: Also, please do link me up with some bug which I could try on and help me with a confidence boost maybe :D | 16:28 |
| rosmaita | aniketh : i'm still looking, got sidetracked while looking at some of these bugs | 16:29 |
| aniketh | rosmaita: Oh, it is fine :D | 16:30 |
| rosmaita | aniketh i was just about to close this one as "invalid", it would be helpful if you could confirm that it is not an issue: https://bugs.launchpad.net/glance/+bug/1733500 | 16:30 |
| openstack | Launchpad bug 1733500 in Glance "image-show interface with Chinese characters return 500 Internal Server Error" [Undecided,Incomplete] - Assigned to weikeyou (weikeyou) | 16:30 |
| rosmaita | there are some notes in there about how to test | 16:31 |
| rosmaita | you can use curl and your devstack to see what happens | 16:31 |
| aniketh | okay, So it basically asks to write test for it, I guess. I will see to it :D | 16:31 |
| rosmaita | you know, we should have unit tests already that make sure utf-8 is handled correctly | 16:32 |
| rosmaita | you could check to verify that | 16:32 |
| rosmaita | and actually, here is a bug for you: https://bugs.launchpad.net/glance/+bug/1737738 | 16:33 |
| openstack | Launchpad bug 1737738 in Glance "Unable to scrub image " [Undecided,New] | 16:33 |
| rosmaita | you'll have to read up on the delayed-delete feature and how to turn it on in your devstack | 16:34 |
| rosmaita | aniketh ^^ | 16:35 |
| aniketh | rosmaita: Yes, I wil take a look into both of those. | 16:36 |
| rosmaita | cool! thanks | 16:36 |
| aniketh | Thank you :D | 16:37 |
| rosmaita | aniketh: is your launchpad id also 'aniketh' ? | 16:46 |
| openstackgerrit | Matthew Treinish proposed openstack/glance master: Specify a python executable for prlimits when using uwsgi https://review.openstack.org/531498 | 16:49 |
| *** links has joined #openstack-glance | 16:49 | |
| *** danpawlik has quit IRC | 16:59 | |
| *** Supun has joined #openstack-glance | 17:03 | |
| *** AlexeyAbashkin has joined #openstack-glance | 17:12 | |
| *** links has quit IRC | 17:15 | |
| *** konan has joined #openstack-glance | 17:18 | |
| *** konan is now known as abhishek | 17:29 | |
| abhishek | jokke_, rosmaita just finished with your conversation about extend issue | 17:30 |
| rosmaita | abhishek hope that made sense? | 17:30 |
| abhishek | Yes, indeed | 17:31 |
| rosmaita | the revert is in the gate now | 17:31 |
| rosmaita | if you want to +2 https://review.openstack.org/#/c/542860/ | 17:31 |
| *** Supun has quit IRC | 17:31 | |
| rosmaita | i will +2A it when the reversion is merged | 17:31 |
| abhishek | Saw that as well, some problem with my mail client (not receiving mails), so I logged in and read all the discussion | 17:32 |
| abhishek | rosmaita: definitely | 17:32 |
| rosmaita | abhishek nice work on the postgres thing | 17:32 |
| rosmaita | i'll be looking more closely after lunch, but that's a big load off my mind | 17:32 |
| rosmaita | i was worried that it would be much uglier | 17:33 |
| abhishek | I am hoping this will solve our purpose, | 17:33 |
| abhishek | Yes need some careful review | 17:33 |
| jokke_ | abhishek: rosmaita: would you have time to help me out a bit? | 17:34 |
| abhishek | If something is needed, drop a mail on konan.abhi@gmail.com I will look at them tomorrow morning time | 17:34 |
| rosmaita | ok, thanks | 17:34 |
| abhishek | jokke_: yes | 17:35 |
| rosmaita | jokke_ i will help, i think it's like 11pm where abhishek is | 17:35 |
| jokke_ | ok, let me push the wip to gerrit, easier to explain | 17:35 |
| rosmaita | sounds good | 17:35 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: WIP: Uri Filtering for web-download https://review.openstack.org/542956 | 17:35 |
| rosmaita | but abhishek if you are not sleepy yet, by all means stick around | 17:35 |
| rosmaita | :) | 17:35 |
| abhishek | Yes I will, just having my dinner | 17:36 |
| jokke_ | images.py line 101 | 17:36 |
| jokke_ | at which point we parse that body? do I need to do it there and if so what should I call to get the needed info out of that body part? | 17:37 |
| *** harlowja has joined #openstack-glance | 17:37 | |
| jokke_ | that line fails on py27 tests with None not having ['method'] | 17:37 |
| jokke_ | tried with .get() as well, which obviously failed :P | 17:38 |
| abhishek | body is list or dict? | 17:40 |
| jokke_ | it should be dict after it's parsed from json iirc | 17:41 |
| jokke_ | but if I change it to .get("method") I get "NoneType" object has no attribute 'get' | 17:42 |
| abhishek | Need to start my laptop, to have clear look | 17:46 |
| *** abhishekk has joined #openstack-glance | 17:48 | |
| *** r-daneel has joined #openstack-glance | 17:48 | |
| *** AlexeyAbashkin has quit IRC | 17:48 | |
| rosmaita | i'm looking at _validate_import_body(), we prob need to add some stuff there to make sure the web-download also has a uri in it (doesn't address your question, just something i'm noticing) | 17:49 |
| jokke_ | rosmaita: it doesn't matter, really as the check I wrote already covers that case | 17:50 |
| jokke_ | rosmaita: obviously if we do it in the deserialization phase, then we don't need to do it in validation phase | 17:51 |
| rosmaita | jokke_ what does the request look like? | 17:52 |
| jokke_ | rosmaita: the request should have all the headers etc. | 17:52 |
| abhishekk | body looks like, {u'method': {u'name': u'glance-direct'}} | 17:54 |
| jokke_ | this would have been so much easier if I did it in the task, but that would be just wrong as then the import call would go through and the import task would just fail ... horribly | 17:54 |
| jokke_ | abhishek: and with web-download it should look like {u'method': {u'name': u'web-download', u'uri': u'insert-uri-here'}} | 17:55 |
| abhishekk | yes | 17:55 |
| openstackgerrit | Merged openstack/glance master: Revert "Fix wrong usage of extend in list_image_import_opts" https://review.openstack.org/542855 | 17:55 |
| rosmaita | take a look at line 831 | 17:55 |
| abhishekk | just a minute let me reopen it | 17:56 |
| rosmaita | i think maybe there is one more layer of indirection? | 17:56 |
| jokke_ | oh, gr8 ... that actually may be it, let me try | 17:57 |
| abhishekk | i guess so | 17:57 |
| rosmaita | so maybe import_method = body['body']['method']['name'] | 17:57 |
| *** Supun has joined #openstack-glance | 17:59 | |
| *** AlexeyAbashkin has joined #openstack-glance | 18:00 | |
| jokke_ | with body['body'] I get KeyError: 'body' | 18:00 |
| rosmaita | yuck | 18:00 |
| jokke_ | just like with method I get KeyError: 'method' | 18:00 |
| abhishekk | but then if import-method is glance-direct then it should show body like {'body': {u'method': {u'name': u'glance-direct'}}}, but its not | 18:01 |
| rosmaita | maybe put in a print("%r" % dir(body)) and let's see what it is | 18:04 |
| rosmaita | because i for one am confused | 18:04 |
| *** AlexeyAbashkin has quit IRC | 18:04 | |
| abhishekk | just a minute | 18:08 |
| *** AlexeyAbashkin has joined #openstack-glance | 18:08 | |
| abhishekk | rosmaita, ['__class__', '__cmp__', '__contains__', '__delattr__', '__delitem__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getitem__', '__gt__', '__hash__', '__init__', '__iter__', '__le__', '__len__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__setitem__', '__sizeof__', '__str__', '__subclasshook__', 'clear', 'copy', 'fromkeys', 'get', 'has_key', 'items', 'iteritems', | 18:13 |
| abhishekk | 'iterkeys', 'itervalues', 'keys', 'pop', 'popitem', 'setdefault', 'update', 'values', 'viewitems', 'viewkeys', 'viewvalues'] | 18:13 |
| jokke_ | so it looks like dict | 18:14 |
| abhishekk | yes, its a dict | 18:15 |
| rosmaita | i guess print body.keys() and see what we get | 18:16 |
| abhishekk | give me 5-10 minutes, network speed is suddenly down here | 18:22 |
| aniketh | rosmaita: Sorry for the late reply. Yes, my launchpad id is aniketh01 | 18:22 |
| rosmaita | aniketh np, thanks | 18:23 |
| abhishekk | Erno that test is failing because we are passing empty dictionary from the test | 18:34 |
| abhishekk | jokke_, | 18:34 |
| abhishekk | jokke_, https://github.com/openstack/glance/blob/master/glance/tests/unit/v2/test_images_resource.py#L638 | 18:35 |
| jokke_ | abhishekk: I was looking, that it might be the case, but that just means that our deserializer is actually not running the validation ofr it | 18:35 |
| jokke_ | it should have thrown bad request already from the deserializer | 18:36 |
| abhishekk | the reason is from test we are calling direct controller's import_image method | 18:37 |
| jokke_ | aargh | 18:37 |
| abhishekk | it doesn't call deserializer from tests | 18:37 |
| rosmaita | arrgh | 18:37 |
| rosmaita | what jokke_ said | 18:37 |
| jokke_ | _This_ is why I hate unittests! :P | 18:37 |
| abhishekk | so instead of empty dictionary now you need to pass actual body from their | 18:38 |
| jokke_ | abhishekk: I'm not any happier that it's Friday past midnight for you and you're still working on this (well I'm happy that you figured it out, but feel super bad about it at the same time) | 18:39 |
| abhishekk | jokke_, no issues | 18:39 |
| abhishekk | so would you like me to fix those tests? | 18:40 |
| jokke_ | abhishekk: so I just add body={'method': {'name': 'glance-direct'}} as attribute to that get_fake_request? | 18:41 |
| jokke_ | and that should sort us out? | 18:41 |
| abhishekk | just give me a minute | 18:41 |
| * rosmaita is being no help at all and is going to get lunch | 18:42 | |
| abhishekk | jokke_, unfortunately that will not help, you need to replace {} with {'method': {'name': 'glance-direct'}} | 18:43 |
| abhishekk | 5 tests from test_image_resources.py | 18:44 |
| jokke_ | abhishekk: ohh, you mean in the asset()? | 18:44 |
| abhishekk | yes | 18:44 |
| abhishekk | test_image_import_raises_conflict_for_invalid_status_change | 18:45 |
| jokke_ | cool ... thanks buddy, will do that right now and see how it behaves | 18:45 |
| *** dtantsur is now known as dtantsur|afk | 18:45 | |
| abhishekk | test_image_import_not_allowed | 18:45 |
| abhishekk | test_image_import_raises_conflict | 18:45 |
| abhishekk | test_image_import_raises_bad_request | 18:46 |
| abhishekk | and test_image_import | 18:46 |
| abhishekk | let me know if any other failures are their | 18:46 |
| abhishekk | after this change py27 and py35 tests will pass | 18:49 |
| jokke_ | it worked! | 18:50 |
| abhishekk | jokke_, I will be around for next 30 minutes, please let me know if anything is required | 18:50 |
| abhishekk | great | 18:50 |
| jokke_ | you, my friend, are testing rockstar :D | 18:50 |
| abhishekk | :D | 18:50 |
| abhishekk | Anyway, I should have figured it out earlier, but I was under impression that something is wrong with the actual code | 18:51 |
| jokke_ | because I was looking that we are throwing empty fake request at it, but didn't realize it bypassed the deserializer | 18:51 |
| abhishekk | we have separate class to test deserializer part | 18:52 |
| jokke_ | ok, let me put a commit message to it and it just might be good to go :D | 18:52 |
| abhishekk | great, I will have a look at entire patch tomorrow | 18:52 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: URI filtering for web-download https://review.openstack.org/542956 | 18:57 |
| jokke_ | gr8 thanks again abhishekk | 18:57 |
| abhishekk | jokke_, its pleasure, I will take your leave now | 18:59 |
| abhishekk | rosmaita, if you have any doubts/suggestion for postgresql patch, drop me a mail on konan.abhi@gmail.com (I will check tomorrow) | 19:00 |
| *** abhishekk has quit IRC | 19:00 | |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: URI filtering for web-download https://review.openstack.org/542956 | 19:14 |
| jokke_ | rosmaita: abhishek ^^ works properly with configgen as well after the 542860 gets merged | 19:15 |
| *** lbragstad has quit IRC | 19:18 | |
| rosmaita | great! | 19:20 |
| jokke_ | that should merge any second now and I'll propose the backport for it | 19:20 |
| *** AlexeyAbashkin has quit IRC | 19:21 | |
| rosmaita | cool | 19:21 |
| openstackgerrit | Merged openstack/glance master: Fix bad usage of extend in list_image_import_opts https://review.openstack.org/542860 | 19:21 |
| jokke_ | there we go :P | 19:21 |
| rosmaita | !!! | 19:21 |
| openstack | rosmaita: Error: "!!" is not a valid command. | 19:21 |
| abhishek | Great | 19:23 |
| jokke_ | rosmaita: https://review.openstack.org/542985 | 19:24 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: URI filtering for web-download https://review.openstack.org/542956 | 19:29 |
| *** idlemind has quit IRC | 19:37 | |
| *** idlemind has joined #openstack-glance | 19:38 | |
| *** AlexeyAbashkin has joined #openstack-glance | 19:49 | |
| *** abhishek has quit IRC | 19:49 | |
| *** AlexeyAbashkin has quit IRC | 19:53 | |
| *** tesseract has quit IRC | 20:00 | |
| *** dalgaaf has quit IRC | 20:04 | |
| *** Supun has quit IRC | 20:04 | |
| *** dalgaaf has joined #openstack-glance | 20:05 | |
| *** e0ne has joined #openstack-glance | 20:18 | |
| *** e0ne has quit IRC | 20:20 | |
| *** e0ne has joined #openstack-glance | 20:47 | |
| *** e0ne has quit IRC | 21:13 | |
| *** DuncanT has quit IRC | 21:49 | |
| *** DuncanT has joined #openstack-glance | 21:49 | |
| *** r-daneel has quit IRC | 22:09 | |
| *** aniketh has quit IRC | 22:14 | |
| openstackgerrit | Brian Rosmaita proposed openstack/glance master: Use configured value for import-methods header https://review.openstack.org/543039 | 22:20 |
| openstackgerrit | Erno Kuvaja proposed openstack/glance master: URI filtering for web-download https://review.openstack.org/542956 | 22:37 |
| jokke_ | rosmaita: Abhishek's postgre fix worked ... the Manila postgre tests passed :D | 22:44 |
| jokke_ | makes me happy ... how much couple of days of working gate makes for productivity :D | 22:44 |
| *** cargonza has quit IRC | 22:53 | |
| *** cargonza has joined #openstack-glance | 22:54 | |
| *** edmondsw has quit IRC | 22:59 | |
| *** esberglu has quit IRC | 23:37 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!