Fixed h/rfcntl.h which missed multiple inclusion guards entirely.

Added cleanup in the pending list of the c++ version of the library (for cases when dlf_init is not called.

     ${MAJOR_CASTOR_VERSION}

to:

    ${MAJOR_CASTOR_VERSION}.${MINOR_CASTOR_VERSION}

Fixed wrong COMPILE_SERVER default (now 1).
Added many man pages installations.
Created an empty placeholder libshift.c to prevent the confusion of cmake about this target (libshift is created by merely linking all of castor's libraries).
Moved more configuration files to the (debatable) config diretory (it is also used for compilation configuration).

I have put back the CASTOR man page sections in the master branch.

The good news is that I have been able to remove the perl processing of the
CASTOR manual pages that was changing the simple numerical section numbers
into CASTOR specific ones.  Developers now just write their manual pages and
the Imake system will not change the man page section within them.  Therefore
if a manual page should be part of the 8castor man page section then the
developer should explicitly put 8castor in the .TH line and not just 8.  Being
explicit means the reader of the man source know what they will get.

In theory the tape library and its dependents are now build my cmake.  This is a small drop in a very large ocean.

Dropped publication of messages when they are first used (infamous MSGTEXT= in the /var/log/massage file)

integration of the monitoring of diskserver statuses in the diskmanagerd/transfermanagerd infrastrucutre. In particular rmmasterd, rmnoded and all the shared memory infrastructure have now disappeared

Cleanup some more defects and especially got rid of lot of useless code around network releated functions

Changed the handling of logmasks. Now the configuration is always queried and the source is respected

This involved moving the definition of basic exceptions to the libcastorcommon and thus adding dependencies to this library in a couple of places that depended only on libcastorclient

Moved all of the DLF parameter type 'defines' into the same file (h/dlf_api.h) and fixed freeing of a parameter name for raw parameters that don't have a name.

Used it to put back the migration related messages in the tapegatewayd log

Fixed timestamp issues with the dedicated cleanup logging facility.
Extended the DLF api with a method taking a message string rather than a message number. This should be the only used from now on.

Implemented a possibility to log to syslog with a given timestamp. Also packaged the example rsyslog conf files

thread specific key named 'reg_key' rather than the associated pthread_once
control variable named 'reg_key_once'.  The comments in the function were clear
so it was easy to correct the code.

I have replaced C++ comments in C code with C code comments.

I have moved/changed variable declarations to not bream the ISO C90 rule
against mixed declarations and code.

  - lot's of renaming. jobs are now transfers, client commands are named listtransfers, killtransfers and draintransfers, daemons are called diskmanagerd and transfermanagerd, packages are renamed accordingly, ...
  - a drain mode has been added to the transferdaemon to ease the retirement of machines running it
  - ports used are not colliding anymore with LSF so that we can let LSF run and only stop/start the jobmanagerd and the transfermanagerd when we switch from one mode to the other and back
  - the badmin command has dropped in favour of automatic, regular reload of all configuration files
  - client commands have been extended to allow restriction to a given diskpool and user
  - the listtransfer commands has been extended to display more parameters, depending on options (e.g. number of unique pending jobs in a pool, per protocols values)
  - reconnection to the ORACLE DB have been fixed
  - DLF logging has been fixed (both the wrapping of it through enums and the insertion of new facilities in the facility table)

Fixed a function leaving VID set when moving a tapecopy to MIG_RETRY.