Nils Goroll
Nils Goroll
bugwash: @walid-git wants to roll a release by the end of next week, thank you
> The global counter is a quick way to tell if looking into backends make sense. I am not entirely convinced that the double accounting pays off, but I can...
@simonvik good point. Are there any established alternatives to handling `%00` like throwing an error? If yes, I think we should rather review any `IDENTITY` encoding to `STRING`.
This triggered #4255
bugwash (phk): Introduce VSL query "macros" for such tasks
> We are currently evaluating different design ideas to improve the dynamics between h2 sessions and streams. I think we should use the new request disembark and handle all h2...
Irrespective of what I think a proper solution is, I would like to understand what you are saying, @dridi > The reason for this arbitrary 500ms timeout is `h2_window_timeout`. When...
#4247 contains the necessary change to make this work. The test case there contains a different approach to implementing the same functionality avoiding the restart and cache key modification, but...
@sbraz yes I would recommend the `Vary` way, because it avoids the complications with `restart`. I would be hesitant to add this particular and quite special use case to the...
To make progress here, I'd suggest the following: * turn all VCL and out-of-workspace related errors into `500` because this is really what they are. [503](https://httpwg.org/specs/rfc9110.html#status.503) makes specific reference to...