| dmsimard | ohwhyosa[m]: that's a tricky one. If the last few callback hooks aren't sent by Ansible, we can't tell if it's because there's something running (i.e, a long shell task) or if it was because it was interrupted. | 01:53 |
|---|---|---|
| dmsimard | We could time them out ? But that requires some adhoc process to time them out. | 01:54 |
| dmsimard | There is an issue about doing the plumbing for pruning old records: https://github.com/ansible-community/ara/issues/31 | 01:58 |
| dmsimard | We could have a similar process but instead of deleting them, we time them out | 01:58 |
| dmsimard | ¯\_(ツ)_/¯ | 01:58 |
| *** gvincent has joined #ara | 06:17 | |
| *** herald85 has joined #ara | 07:49 | |
| *** vincent--- has joined #ara | 08:12 | |
| *** vincent-- has quit IRC | 08:15 | |
| *** idir__ has joined #ara | 09:26 | |
| *** idir__ has quit IRC | 09:49 | |
| *** micisuta has joined #ara | 10:18 | |
| ohwhyosa[m] | dmsimard: is there a way to delete all the records in the database via the api? | 11:07 |
| ohwhyosa[m] | Yes, a time out could be a good idea, but it can be risky | 11:07 |
| ohwhyosa[m] | as in, sometimes the problem you want to records is ansible just hanging in there | 11:07 |
| ohwhyosa[m] | could maybe django do a system ps aux or somehting to check if there is anansible process? | 11:08 |
| ohwhyosa[m] | our office devs suggest doing a keepalived within the callback, I don't know if that is possible though | 11:12 |
| ohwhyosa[m] | django checking for the process is a no-no because it can be on different machines, for example | 11:13 |
| *** Bary has joined #ara | 11:45 | |
| Bary | Hello there! | 11:45 |
| Bary | Just finished mounting new ara server with web and realised that the failure reason or just result of tasks are not recorder | 11:46 |
| Bary | Is it the new version design? | 11:46 |
| ohwhyosa[m] | Hey Bary! | 11:47 |
| ohwhyosa[m] | dmsimard just implemented it yesterday, though it is work in progress, I tested it and it works marvels! | 11:48 |
| ohwhyosa[m] | He is going through a full rewrite, and some of the stuff is still WIP | 11:48 |
| ohwhyosa[m] | Plus I think he's pretty much doing ara all by himself | 11:48 |
| ohwhyosa[m] | https://review.opendev.org/669221 --> Here's the patch he sent here yesterday | 11:49 |
| Bary | Ah great to hear that, thanks ohwhyosa[m] | 11:49 |
| Bary | I'll follow the news then :) | 11:49 |
| ohwhyosa[m] | you can check the log here at | 11:58 |
| ohwhyosa[m] | http://eavesdrop.openstack.org/irclogs/%23ara/ | 11:59 |
| ohwhyosa[m] | Bary | 11:59 |
| Bary | Actually trying to apply the patch | 12:00 |
| *** openstackgerrit has quit IRC | 12:04 | |
| *** micisuta has quit IRC | 12:04 | |
| Bary | Looks like is not the only thing to patch heh, no worries after the weekend I can take a look into it more | 12:04 |
| ohwhyosa[m] | You can just manually add the changes (i did that and it works) | 12:08 |
| ohwhyosa[m] | then you npm build again | 12:09 |
| ohwhyosa[m] | and you serve it Bary | 12:09 |
| ohwhyosa[m] | It worked, how did you install ara-web? | 12:09 |
| Bary | Aah, we've forgotten to npm build oh dear ^^ | 12:09 |
| ohwhyosa[m] | And what is it that is failling? | 12:10 |
| ohwhyosa[m] | I forgot like a lot of times yesterday ahahahah but I'm a bit of a disaster | 12:10 |
| Bary | Hm some problems using .patch file | 12:12 |
| gvincent | dmsimard, do you want help on this one https://review.opendev.org/#/c/669221/? | 12:17 |
| Bary | ohwhyosa[m], great it works ;) So on Monday more things to discover thanks a lot | 12:20 |
| dmsimard | gvincent: sure thing! I figured I would at least try to make the data available. | 12:37 |
| gvincent | I will look at it on monday morning | 12:38 |
| dmsimard | Bary: The data is recorded and is available in the API, it's just a matter of implementing it somewhere in the web UI | 12:38 |
| dmsimard | gvincent: I'm not convinced about the modal approach fwiw | 12:39 |
| dmsimard | I first tried a full fledged page (routed under /results/<id>) and re-using the playbook summary at the top (similar to /playbooks/<id>) but the problem was that the data we use in the summary isn't available when just querying a record | 12:41 |
| dmsimard | Like playbook.item, playbook.labels, and then there was something else too.. | 12:42 |
| dmsimard | I managed to implement that patch through a mixture of what was already done in ara-web and looking at dci-ui as an example :p | 12:43 |
| dmsimard | ohwhyosa[m]: there's no specific API call for deleting everything in one shot but deleting individual playbooks should cascade to everything since everything (except labels) are children of playbooks | 12:45 |
| dmsimard | gvincent: the issue I see with modals is that it feels awkward to permalink to that (when you want to share it with someone) | 12:48 |
| gvincent | yeah +1 | 12:48 |
| dmsimard | In 0.x, the hack was that while the content was in a modal, there was actually an iframe in the modal and the content of the iframe was a page that you could link directly to | 12:48 |
| dmsimard | So I am most definitely open to suggestions | 12:49 |
| dmsimard | Something I know would be nice is to have a page similar to /playbooks but for /hosts | 12:54 |
| Bary | Yea, I've realized it is recorder but with the different tables/columns | 12:54 |
| dmsimard | But having one for /results might not be appropriate (due to the amount of results?) | 12:55 |
| dmsimard | Bary: 1.x switches the backend from flask to django and changes the database model a lot | 12:55 |
| Bary | For sure. We had some problem figuring out this, as in readthedocs we(team mate) couldn't find the details we are working on the new 1.x version | 12:57 |
| Bary | But later discovered admin panel, the things etc | 12:57 |
| Bary | We set user password etc | 12:58 |
| Bary | I wonder if the settings like ARA_API_CLIENT should belong to the page with https://ara.readthedocs.io/en/latest/api-configuration.html | 12:58 |
| Bary | As now they are documented only in previous page | 12:59 |
| dmsimard | ARA_API_CLIENT is an Ansible plugin configuration while the other things are API server configs | 13:00 |
| dmsimard | Need to brb, meeting | 13:01 |
| *** openstackgerrit has joined #ara | 13:48 | |
| openstackgerrit | David Moreau Simard proposed recordsansible/ara master: API: Expose the task file path as task.path https://review.opendev.org/669220 | 13:48 |
| *** Bary has quit IRC | 14:41 | |
| openstackgerrit | David Moreau Simard proposed recordsansible/ara master: WIP: Implement Ansible lookup plugin to query the ARA API https://review.opendev.org/663968 | 15:40 |
| openstackgerrit | David Moreau Simard proposed recordsansible/ara master: WIP: Implement Ansible lookup plugin to query the ARA API https://review.opendev.org/663968 | 15:43 |
| openstackgerrit | Merged recordsansible/ara master: API: Expose the task file path as task.path https://review.opendev.org/669220 | 15:54 |
| *** Bary has joined #ara | 16:20 | |
| *** Bary has quit IRC | 16:37 | |
| *** Bary has joined #ara | 16:37 | |
| *** Bary has quit IRC | 16:44 | |
| *** Bary has joined #ara | 16:49 | |
| *** Baryczka has joined #ara | 17:06 | |
| *** Bary has quit IRC | 17:08 | |
| *** Baryczka has quit IRC | 17:13 | |
| *** herald85 has quit IRC | 17:22 | |
| *** Baryczka has joined #ara | 17:46 | |
| *** Baryczka has quit IRC | 18:03 | |
| *** Baryczka has joined #ara | 18:03 | |
| *** vincent--- has quit IRC | 18:08 | |
| *** Baryczka has quit IRC | 18:09 | |
| *** Baryczka has joined #ara | 18:54 | |
| *** Baryczka has quit IRC | 19:10 | |
| *** Baryczka has joined #ara | 19:10 | |
| *** Baryczka has quit IRC | 19:17 | |
| *** Baryczka has joined #ara | 19:52 | |
| *** Baryczka has quit IRC | 20:08 | |
| *** Baryczka has joined #ara | 20:08 | |
| *** Baryczka has quit IRC | 20:15 | |
| dmsimard | The approach with a modal looks terrible on mobile :p | 20:26 |
| *** vincent--- has joined #ara | 20:30 | |
| *** Baryczka has joined #ara | 20:57 | |
| *** vincent--- has quit IRC | 20:59 | |
| *** Bary has joined #ara | 21:14 | |
| *** Baryczka has quit IRC | 21:16 | |
| *** Bary has quit IRC | 21:20 | |
| *** Bary has joined #ara | 21:52 | |
| openstackgerrit | David Moreau Simard proposed recordsansible/ara-web master: WIP: Add page for displaying result details https://review.opendev.org/669221 | 21:55 |
| dmsimard | ohwhyosa[m]: thanks for the pointer for react-json-view, it's a quick win :) | 22:07 |
| *** Bary has quit IRC | 22:09 | |
| *** Bary has joined #ara | 22:09 | |
| dmsimard | https://i.imgur.com/SHotPlR.png | 22:10 |
| *** Bary has quit IRC | 22:16 | |
| openstackgerrit | David Moreau Simard proposed recordsansible/ara-web master: WIP: Add page for displaying result details https://review.opendev.org/669221 | 22:37 |
| *** Bary has joined #ara | 22:56 | |
| openstackgerrit | David Moreau Simard proposed recordsansible/ara-web master: WIP: Add page for displaying result details https://review.opendev.org/669221 | 23:02 |
| *** Baryczka has joined #ara | 23:13 | |
| *** Bary has quit IRC | 23:15 | |
| *** Baryczka has quit IRC | 23:20 | |
| *** Baryczka has joined #ara | 23:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
