I do a lot of one-timer SQLs, some for testing before embedding in an
application, some for data interrogation to determine problems.
Some of the SELECT queries I do, run for hours. Of course, RUNSQLSTM
doesn't support SELECT, otherwise I'd use that to submit to batch.
The one that I have running now is set up to send output to a data file
(F13=Services, 1.Change session attributes, SELECT output: 3=File).
I suppose the way to do this is to create the output table with the fields
specified in the SELECT, then submit to batch a RUNSQLSTM with the same
query, but prefaced with a "INSERT INTO xxxxxxxxxx (" It seems to me that
this would give the same results. Any disagreement there?
But what a PITA! First, I have to manually create the output table, then I
have to put the query in a source member for the RUNSQLSTM, and then submit
the RUNSQLSTM. There's gotta be an easier way.
FWIW, here's the query that's now in its third hour:
select DISTINCT CVCVCD, CVCVLC, CVDEDC, count(*)
from cvgdetl, cusdetl_DE
where CVCUSN = cdCUSN
and CVRENN = cdRENN
and CVSEQN = cdSEQN
and cdstat = 'DE'
group by CVCVCD, CVCVLC, CVDEDC
Any ideas would be greatly appreciated!