At 12:33 PM 10/19/98 -0400, John Zoltak wrote:
>Jim,
>
>After re-reading the image manual I found something new to me. The
>DBXBEGIN can occur before or after the DBLOCK, but the DBUNLOCK must
>occur after the DBXEND/DBXUNDO as per image manual page 7-11. When I
>started using dynamic transactions I found that the DBUNLOCK had to be
>after the transaction so I just figured that the DBLOCK had to be before
>the start of the transaction.

As the original cause of the discussion, I'd like to thank everyone who
chimed in! The tintinabulation is truly delightful. In our case since we are
using multiple databases, and will NOT be using MR, we have chosen to use the

[DBLOCK]
[DBXBEGIN]


[DBXEND]
[DBUNLOCK]

scenario. It was John's discovery of the "As with nondynamic transactions,
the placement of the DBLOCK call either before or after the DBXBEGIN is up
to the programmer" passage, that led to my original inquiry. I simply didn't
notice the requirement that DBXEND be covered by a strong lock.

Ain't life fun when we can have such discussions over time and distances
that would have made it impossible even twenty years ago!

TOny