In particular, in .gitlab-ci.yml I see no reason to hard-code
the version of clang tools - it seems easier to control that
only when generating the image.

Closes #411.

This is my mistake.  I was convinced this line wasn't needed
and I deleted it before merging the module.
Apparently we still have some deficiencies in the build system,
but let's defer that to migration (to meson probably).

Fixes #410.

Tomas Krizek authored 6 years ago and Vladimír Čunát committed 6 years ago

- building different version of kresd is not properly supported
  (uses current directory with checked-out code instead of git master)
- based on debian:stable instead of alpine for easier dependency
  resolution
- moved both Dockerfile and config to more appropriate locations

Anbang Wen authored 6 years ago and Vladimír Čunát committed 6 years ago

Metrics are collected and merged in http.prometheus, no need to be run
on every worker.

Marek Vavruša authored 6 years ago and Vladimír Čunát committed 6 years ago

Instead of throwing an error in the HTTP handler, server should log it.
This covers errors like client disconnecting before reading the response
body etc.

Marek Vavruša authored 6 years ago and Grigorii Demidov committed 6 years ago

This was previously hardcoded to 16. This makes it at least a compile time
define, with a default of 511 (as that's what Redis and Apache use).

Petr Špaček authored 6 years ago and Vladimír Čunát committed 6 years ago

Minor changes to be blamed on Vladimir.

Grigorii Demidov authored 6 years ago and Vladimír Čunát committed 6 years ago

Minor changes to be blamed on Vladimir.

... done in the few preceding commits.
I intentionally don't mention ::daemon_context at this point.

We were quite inconsistent here.

Grigorii Demidov authored 6 years ago and Vladimír Čunát committed 6 years ago

- answer_finalize: new layer
- kr_request: keep ::qsource.packet beyond the begin phase
- kr_request: add ::daemon_context

release 3.1.0

See merge request !692

lib/zonecut: avoid one kind of NS dependency cycles

Closes #374

See merge request !690

Vladimír Čunát authored 6 years ago and Tomas Krizek committed 6 years ago

The problem here was that we need to know which addresses are timed-out
(and not to be re-probed) much earlier than we do NS selection ATM -
that's because under some circumstances it affects the depth of NS
zone cut that we choose, i.e. if all addresses in a certain zone cut are
"bad" in a certain sense, we need to use a zone cut closer to the root,
because otherwise we'd get into a dependency cycle.

packaging: updates

See merge request !676