GitLab

This change applies to the Memfile backend and is intended to significantly
improve its performance.

This change should seriously rise the performance of the server as the
server seem to spend a lot of time executing the validate function.

Note: This change fulfills both 2699 (lease4) and 2703 (lease6).

ignoring the result fixes a scan-build error.  besides, ignore() should be
more appropriate choice for what it intends to do anyway.

test cases aren't incomplete, so it's revised to cover all combinations.

Explicit allocation failures cause the return of packet to the client
indicating that the address allocation has been refused.  Also added
was a "catch" for other exceptions - this causes the packet from the
client that generated the exception to be ignored.

It insists signature match between declarations and definitions even with such
unnecessary qualifiers, and fails to link due to a false-positive
"undefined symbol" at the link time.

with one minor regression fix.

Actually the support was already there.
Renamed the OpenBSD class to NetBSD (since it was first)
and then use it for OpenBSD too (and NetBSD).

This is for ticket #2692. It was reviewed on jabber.

This is for ticket #2128. It was reviewed and discussed on jabber.

This extends SysInfoPOSIX to get the hostname and the load average.
Then as a fallback for unknown systems it uses the SysInfoPOSIX.

(By the way, some information is looked up twice, but this
happened before this patch, but now happens for a couple more things.)

http://git.bind10.isc.org/~tester/builder//BIND10-cppcheck/20130129171501-FreeBSD8-amd64-GCC/logs/unittests.out
http://git.bind10.isc.org/~tester/builder//BIND10-cppcheck/20130205115001-FreeBSD8-amd64-GCC/logs/unittests.out

As the above results, on this platform, it doesn't mostly return an
accurate float of seconds which is less than 1.0. The above failure
happened at line 203 in counters_test.py , but also at line 142 there
is probably the same problem.  This fix is similar to
89fd37be.

(reported by emhn: left over after a distclean)

Some of the makefiles clear that but not that one. We do
automate the distclean (via distcheck) as part of our
buildfarm. But I think this __pycache__ was introduced in python3.2 and
those build systems don't have it.

This is trivial so didn't get review.

wouldn't matter for the usage of this script, but this should be slightly
more efficient.

that's probably simpler than using dummy files and directories.