Always re-create missing links in sites-enabled

https://projects.bytemark.co.uk/issues/12768

See merge request !3

Exit if no domains have been detected

See https://projects.bytemark.co.uk/issues/12783

It seems that during periods of memory pressure, sometimes `getpwuid` fails and `symbiosis/domains.rb` just dies mid-script. Nothing is returned to `symbiosis-httpd-configure`, so the `configurations` array remains empty indicating that there aren't any domains configured. All sites in `sites-enabled` subsequently get deleted.

The only situation in which the `configurations` array should be empty is when `/srv` has no directories, but in that case there's no real reason to run `symbiosis-httpd-configure`. As I understand it, every customer using Symbiosis+Apache should have at least one directory in `/srv`, so it's probably better to just exit if `configurations` is empty rather than continue to delete all sites.

See merge request !4

We no longer log errors during the normal course of events, such
as:

* Virtual Hosts that don't exist.
* Malformed input-lines.

Add -v, or --verbose, to see them.

This is better than using a path.

We've moved the opening of the logfiles into a central routine,
such that we can add some sanity checking, and ensure that we're
not going to append to files which are symlinks.

As part of this work we've also made the --max-files|-f command-line
argument function, with a default of 100.  When more than the specified
number of logfiles are open they are all closed so they can slowly
re-grow.

This is required, and achieved via a symlink

Some versions use:

dh_prep
dh_auto_install
dh_install

Some use:

dh_build
dh_binary

The former will work.  The latter will clean debian/tmp as part of
the process, which means installing the binary to debian/tmp will
fail.

I tested this too.

This code is functionally identical to the Ruby version, all being
well, and barring bugs.

The intention is that we can allow users to install this package
if they're at all suffering with issues with RAM usage.  This is
deliberately taken as a safe middle-ground before rewriting a core
component of symbiosis and forcing all users to upgrade at once.

Once this package is installed the old ruby-based logger will be
divered, via dpkg-divert, such that it is no longer in use and
can be replaced.

Removign this package undoes the diversion to restore the system
to that state it was previously in.

This updates #12663.

email: futher fix to makefile to make tests pass



See merge request !2

email: Makefile fix for patched exim4



See merge request !1

The auth program should only return -1 if it does not make sense to try
any further authentication methods.

In fact the only time we don't want to continue is if the user's
password is wrong,

This just makes the redirect a bit more reliable.

This removes a needless redirect when the HTTP_HOST is not the same as
the SERVER_NAME.

Fixes #12383

Although the diff is quite large, the changes are as follows:

* Allow a backup2l.conf to be specified as an argument
* A reguar expression is used to find the BACKUP_DIR definition
* More error checking around the running of the backup2l estimate
* The two regexes parsing the output have been squashed into one.
* When the estimate was just Bytes, the parsing failed, as a space was
  in between the number and the letter B.

This leads to a cron email every day.  Oops!

Fixes #12473

It has to check all the directories in sets/ to pick the next, not just
use the next name of the "last" available set.

The defaults are as follows:

* If the current set is available

** If it is due to expire inside the threshold

*** generate a new set if there is no set more recent (unless instructed
otherwise)
*** roll over to the new set if one has been generated (unless
instructed not to)

** Otherwise

*** do not generate a new set (unless instructed otherwise)
*** do not roll over (unless instructed to)

* If the "current" set is missing, but other sets are available

** If the most recent set is due to expire inside the threshold

*** generate a new set (unless instructed otherwise)
*** roll over to the new set if one has been generated (unless
instructed not to)

** If the most recent set is not due to expire soon

*** do not generate a new set (unless instructed otherwise)
*** roll over to the latest set (unless instructed not to)

* If there are no certificate sets

** generate a new one (unless instructed otherwise)
** roll over to the new set if one has been generate (unless instructed
not to)

Apparently the solution is to "retry a reasonable number of times".

  https://github.com/letsencrypt/boulder/issues/1217
  https://community.letsencrypt.org/t/getting-the-client-sent-an-unacceptable-anti-replay-nonce/9172