Otto Moerbeek

I believe the failing test has a race condition.

Looking in history, I see #11592, which is not backported.

Note that dnsdist and recursor define suppression entries for similar cases. The idea being that atomic updates are quite expensive and we don't care too much about these races for...

Nice, but it would be even better if it also would print the right dir if chroot is used or running under systemd (maybe it does already for systemd, I...

With the latest commit I think this is good to go. Please convert to non-draft.

I'm inclined to merge this, even with the regression failure, as it is unrelated and fixed in master.

Do we want this?

> I'm not very comfortable with the idea of inflating the content for every HTTP query. Perhaps we could cache the inflated content, so we do that operation at most...

> if we assume modern browsers, you could also assume they can decompress gzip... To do that properly, I would need to interpret the `Accept-Encoding` request header. At the time...

Mostly obsoleted by #11859