Re: SQLConnect - Connection to Relational Database Already Exists

On 06-May-2016 15:54 -0500, Glenn Gundermann wrote:

If there is a problem due to a previous SQL CLI, then you might want
to try this to close things before attempting the next activity:

CALL PGM(QBRM/Q1AOLD) PARM('CLSSQLENV ')

We weren't performing what you are so this may be totally different.
When we do some BRMS activities and then call a program that tries to
use CPYTOIMPF, it failes. IBM says it is because both use SQL CLI. We
are supposed to sign off after performing the BRMS activities or call
the above mentioned program.

FWiW: I would expect the above invocation would be specific to resolving a conflict with a prior use of BRMS, and will give no joy in resolving a conflict due to prior invocation(s) of the HTTPGETCLOB [or any similar system-supplied SQL UDF that uses the SQLCLI]. I expect that specific call made to the BRMS program Q1AOLD [apparently to Close the SQL Environment (CLSSQLENV)], merely requests that the Q1ABRMS [¿or is that QBRMS or QBRMS?] activation [group (ACTGRP)] should perform a disconnect (SQLDisconnect) and close\free of the previously obtained SQLCLI handle (SQLFreeHandle) within in that activation. IMO that effect would come from having reclaimed the activation, per the activation having registered a reclaim activation exit that includes proper cleanup, such that rather than having to know some obscure CALL to effect that cleanup, one could just issue the more intuitive action of RCLACTGRP.