While it could be argued that EBF nomenclature provides this facility
already, given the number of regressions I have seen over the years using
ASE and earlier, something that goes beyond the top level of an EBF # easily
is worthy of investigation for implementation.

Similar to package management on most Unix and Windows variants ASE should
be able to enumerate the CRs applied to the dataserver binary via:
Virtual or persisted table; and
Command-line option that doesn't require the ASE binary to be running
live e.g. `$SYBASE\$ASE_DIR\EBFnnnnnTest\bin\dataserver -CRLIST`

A command line utility to enumerate the CR information for shared
libraries/utilities/non-dataserver server binaries and an XP proc to wrap it
to present in the same way as the ASE virtual table.

The reason I prefer these to be documented in the binary they relate to, is
to make them more atomic - i.e. it would be impossible to mix the CR
library/file and the dataserver it relates to. Ofcourse, checksums etc would
prevent this too, but it would impact build and distribution/installation
more.

A virtual table is preferred and one that is not dependant on monitoring or
CIS loopback being configured. The table should list at a minimum four
columns the ProductArea (ASE|CIS|RSE|OPT|OCS|etc), CR#, CR# dependency, CR#
replaced, "traceflag to enable/disable".

This could be supplemented with a bcp file that contains:
CR#s, CR short problem/fix descriptions, traceflag to enable/disable,
default status {Enabled|disabled}
for people to be able to load these.

This would be helpful to DBAs and support folks alike.

Thoughts?
Grant Queenin