GitLab

modules/http: fix too early renewal of ephemeral certificates

See merge request knot/knot-resolver!808

http.config() fixes

See merge request knot/knot-resolver!807

Formerly http.config({tls=false}, 'doh') also removed all endpoints for
'doh', which was unexpected.

With this change configurations are merged as expected and user does not
have to repeat endpoints and other options from builtin template.

http sockets

Closes #85

See merge request knot/knot-resolver!803

Output buffering in tests made debugging much harder because crash in
middle of operation discarded while buffer and left us without a trace
what was going on.

We still need to somehow solve socket closure etc.

Add interface for providing different named configuration sets.
This will be handy once we start getting named sockets from systemd.

Now we hopefully won't need to touch it for a long time.

- allow omitting port number
- sync docs
- fix memory leak
- fix with kind != NULL

Originally it's from another branch.

The complication is that we need to work with addresses and
just file-descriptors passed from some parent process.
The former approach lead to logical duplication of some steps;
now we add a step converting addresses to file-descriptors.
Thanks to that we always do bind() without touching libuv,
so the problem with forking disappears :-)

That was against docs.  Fortunately, nil is falsy.

lua net.interfaces(): fix mac addresses

See merge request knot/knot-resolver!804

Only the first byte was being shown since 3ab77332.
I can't see at all why this part was changed;
the buffer is (and was) way overlong for this,
so writing one zero byte just after the end is OK.

lib/generic/lru: try to resolve alignof warnings

See merge request knot/knot-resolver!802

We run meson with -std=gnu11, but apparently some compiler still
complained about it.  Unfortunately it wouldn't be easy to use
standard C11 in this case.

parts of !797: lua light userdata WIP

See merge request knot/knot-resolver!800

The layer functions are supposed to return the new state.
We've been lucky that errors are negative numbers and thus
they typically passed the (status & KR_STATE_FAIL) test.

... for layers and props.  This breaks C module API+ABI.

It seemed weird to repeatedly call a function that returns a pointer
to a structure in which we find the function we want to actually call.
We've never used changing these functions AFAIK, and the target
functions could easily be written to change their behavior instead
(i.e. move the indirection *inside* the function).

When breaking this, I also removed these two (_layers and _props)
from the dynamic symbols (to be) exported from the C modules.
They always pointed to memory belonging inside the module,
and they seem quite sensible to be set up by the _init symbol instead.

dnstap: fix compilation on Turris Omnia

Closes #458

See merge request knot/knot-resolver!805

Fixes: #458

It gets handy for debugging, especially when looking at dufferent
callback "instances".