Collator Protocol Revamp: Testing tools

[alindima]

Tools for testing:

1. connection exhaustion. We need to be able to spawn hundreds of connections ideally on a single machine. Let's first try by spawning a large number of regular test collators. If it doesn't work, we may need a custom stripped-down tool with minimal overhead.
2. Providing invalid collations. We don't need such a large scale for this.
3. Advertising but timing out during fetching (or slow loris). We also don't need very large scale for this.

[MrishoLukamba]

I am taking it @alindima @skunert

[alindima]

@MrishoLukamba sorry, all of these collator protocol revamp issues are quite complex and not ready to be taken over by someone else yet

[MrishoLukamba]

@MrishoLukamba sorry, all of these collator protocol revamp issues are quite complex and not ready to be taken over by someone else yet

Sure but I wanna get involved In solving it piece by piece. As I do understand enough to start on it.

[alindima]

The design of the project is still work in progress, these are still draft tickets.

[bkchr]

@alindima why are all of these individual tickets, if they are all in "draft" stage?

[alindima]

@alindima why are all of these individual tickets, if they are all in "draft" stage?

They are legitimate issues that will be worked on, but the implementation details are not all fleshed out, therefore not immediately actionable by someone else (not without significant guidance)

[bkchr]

Maybe then you should keep them as draft :)