Patch provided by David Abdurachmanov (CMSSW).

Partial fix for bug #95189: Time discrepencies between disk servers and name servers can lead to silent data loss on input
The fix is just respecting the API, without changing the current algorithm.
Note that as an internal API changed signature, this change is NON transparent.

of an index that must be done offline.
See sr #133974: Busy index canot be dropped during DB upgrade (I_SubRequest_CT_ID)

paper with the same name and therefore does not describe the tape code in this
repository any further than the paper which is in this directory.  Secondly the
poster has been copied to the following official location for CASTOR
presentations (thanks Sebastien):

http://castorold.web.cern.ch/castorold/presentations/2012/

/afs/cern.ch/project/cndoc/wwwds/HSM/CASTOR/presentations/2012/CHEP/CHEP2012CASTORTapeWriteEfficiencyPosterFinalVersion.pdf
/afs/cern.ch/project/cndoc/wwwds/HSM/CASTOR/presentations/2012/CHEP/CHEP2012CASTORTapeWriteEfficiencyPosterFinalVersion.pptx

Fixed misleading comment: the iptables rules for rfio are modified by castor-lib and not by CASTOR-client (which was dropped in 2.1.9-0)

Fixed in the trunk.

Fixed in the trunk.

Removed the code of the following deadwood classes:

castor::db::cnv::DbRepackSegmentCnv
castor::db::cnv::DbRepackSubRequestCnv

Fixed in the trunk.

Changed log level for message 'startMigrationMounts: failed migration mount creation due to lack of files' to INFO

Fixed bug introduced when fixing the handling of RequestCanceled exceptions and breaking some other cases

Handled case of a lastKnownFileName already in use when fixing lastKnownFileName in FileRenamedCleanup

Cns_bulkexist moved to the stager.
Coding by Giuseppe.