GitLab

When the script does not exist, because it was not generated due to some
configure switch or environmental condition, don't try to check if it
gets renamed. That would fail even when the situation is completely OK
and expected.

Also fix configure argument name to match help documentation.

Conflicts:
	src/bin/Makefile.am

In case it is turned off, don't link against stuff requiring shared
memory, in particular mapped segment creator in tests. It is not used,
but causes link errors, as it is in a virtual method.

Disable building resolver by default

Conflicts:
	configure.ac

Do the condition at one level higher.

Detect old versions of boost and suggest either disabling shared memory
for them or upgrading.

... as we only use it in an automake condition.

This has caused issues in the past (see other error messages for example),
and escaping with \ doesn't seem to help in those cases.

So don't disable it by default.

Don't build the resolver and its libraries (lib/resolve and lib/libnsas)
in case the --enable-experimental-resolver is not set.

This reverts commit 2092528a.

It should not go to master yet. Reverting until it is through
review.

The switch doesn't have an effect yet.

Just ifdef out the whole makefiles.

Tell more details about what the problem is.

The bug is in offset pointer, not in the red-black tree.

Conflicts:
	ChangeLog
	src/bin/dhcp6/dhcp6_srv.cc

D2UpdateMgr::checkFinishedTransaction was issuing an invalid read
when attempting to prefix increment an iterator used in an std::map erase
call.  This was flagged by valgrind and also core dumped under FreeBSD10.