The PPID should be an argument of the print format, not of the date
command.

Otherwise the VCL prefix would be subject to the -p option.

When the killmode is 'process', only the mgt process is ever signalled
by systemd. If that fails to exit within TimeoutStopSec, we may leave
unresponsive processes till tearing themselves down. During a restart of
the service, that could lead to the old cache process still holding onto
resources like bound ports, preventing the new service from starting.

With 'mixed' mode instead, first SIGTERM is sent to the mgt process only
(found through its PID file), but all the processes in the group will be
sent SIGKILL after TimeoutStopSec, and systemd will wait for them all to
exit.
Signed-off-by: Dridi Boukelmoune <dridi.boukelmoune@gmail.com>

Signed-off-by: Dridi Boukelmoune <dridi.boukelmoune@gmail.com>

RPM macros for systemd used to be part of the main systemd package but
on Fedora they moved to their own subpackage. The %{systemd_requires}
macro is no longer guaranteed to be present in a default build root.

This change enables Fedora Rawhide builds, for a current Rawhide target
of Fedora 36.

Closes #153

But keep the current pcre dependency, they can both coexist.

Refs varnishcache/varnish-cache#3616

On a system where the active VCL's role is merely to switch to a
different label, each tenant might reload its VCL with a dedicated
prefix. Discarding only has an effect on same-prefix reloads.

It can also simply be used for cosmetic reasons to simply replace the
'reload' prefix with something else.

This will match the default hitch configuration, making interaction
between it and varnish easier.

Otherwise host resolution may not be available during boot resulting in
startup failure of varnish.service for no good reason.

weekly is now dead, we use master for weeklies now

There is only python2 and python3 packages

to avoid unnecessary errors when running outside jenkins

Starting with el8 annobin is used to annotate ELF binaries and this is
set up via the %configure RPM macro. As a result annobin becomes a
requirement of the default cc_command in varnishd but it relies on GCC
spec files that are provided by redhat-rpm-config.

So, similar to how we explicitly require GCC at run time we now also
need to bring the annobin setup.

There was already a hint in b2ea64fd
that we lack some packaging stability from the platforms we support
and this is another occurrence. The configure script of a supported
Varnish release will find the tools we need.

This enables build support for el8 in addition to el7.

We don't care when a pipeline does not read the complete varnishadm
output because it found what it needed way before. However, we don't
want noise in the standard error with spurious broken pipe messages.

Closes #134

Slightly better diff with the --ignore-all-space option.

Refs #134