| opendevreview | OpenStack Proposal Bot proposed openstack/oslo.log master: Imported Translations from Zanata https://review.opendev.org/c/openstack/oslo.log/+/861550 | 03:12 |
|---|---|---|
| opendevreview | OpenStack Proposal Bot proposed openstack/oslo.policy master: Imported Translations from Zanata https://review.opendev.org/c/openstack/oslo.policy/+/861556 | 03:48 |
| opendevreview | Merged openstack/oslo.db master: tests: Define a primary key https://review.opendev.org/c/openstack/oslo.db/+/856837 | 05:04 |
| opendevreview | Merged openstack/oslo.config master: Make tests backward compat with cliff' older versions https://review.opendev.org/c/openstack/oslo.config/+/859786 | 05:46 |
| damani | hi | 15:07 |
| damani | sorry i think we will skip the meeting today | 15:08 |
| damani | i'm busy with the ptg | 15:08 |
| damani | but if you want we can do it another day in the week | 15:08 |
| damani | or we can maybe do a little call this week if you want | 15:08 |
| tobias-urdin | ok, I can do quick written update about NATS. We explored the path of using the official nats.py library that is using asyncio by testing https://review.opendev.org/q/topic:asyncio-nats this path | 15:12 |
| tobias-urdin | unfortunately it's a massive effort required to get projects to support that, but I don't think it's neccesarily a bad idea to support asyncio in the common libraries in the future | 15:12 |
| tobias-urdin | Right now I've been exploring another path to get a supported/maintained NATS python library without asyncio, I was contacted by Synadia which works with NATS and where the current maintainer of nats.py is employed and had a quick Zoom meeting with them regarding the possibility of having that, will follow up on that. Apparently they had contact | 15:14 |
| tobias-urdin | with other parties which was interested in using NATS for OpenStack as well. | 15:14 |
| tobias-urdin | So next on my list is to explore that path, the other suggestion from them was to potentially breakout the minimal functionality required from nats.py and into the driver layer in oslo. | 15:15 |
| tobias-urdin | I will update when I have more | 15:15 |
| tobias-urdin | hberaud: yoctozepto ^ fyi | 15:16 |
| hberaud | ack | 15:18 |
| hberaud | tobias-urdin: thanks for your feedback | 15:18 |
| tobias-urdin | however I guess a longterm path forward to support asyncio and the new ASGI standard would be beneficial for projects that would want to skip eventlet-based themselves, but they would need to use for example an API framework that supported that as well (e.g not pecan but something like quartz, or potentially Flask in worst case, as demonstrated in | 15:22 |
| tobias-urdin | https://review.opendev.org/c/openstack/blazar/+/858938 ) | 15:22 |
| yoctozepto | thanks tobias-urdin, and I agree but we all know this requires a tremendous effort to achieve without breaking the world | 15:31 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!