I notice that the DNS SRV namer is marked as “experimental”:
Is this because there are particular known problems with it, or is it just because it is a relatively recently added feature?
Just because we’re not aware of how thoroughly it’s been run in prod. There aren’t any known problems with it.
Grand - thanks for that. I’m likely to want to use this namer, and will be carrying out a certain amount of testing (both functional and load-testing) before committing to it. It is clear what to do if I hit problems (raise issues or provide fixes through PRs), but is there a formal mechanism for providing feedback about everything working successfully?
There’s not a formal mechanism, but if you drop a note to us (on Discourse, Slack, or via email) including some of the details of the testing, that would be greatly appreciated!
I’d recommend perhaps creating a github issue called “Graduate DNS SRV namer out of experimental” and in that issue providing supporting evidence for moving it out of experimental. e.g. testing done, amount of time in prod, scale, etc.
OK - thanks for the replies. I’ll do that at the point that I have such information.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.