It sounds like you have programs that allow user intervention between
the read (with lock) and the update?
Please consider programs that have two reads: one read (no lock) to get
the needed data, then do the user stuff, and compare the new user data
to the original data. If different, do an update routine where you read
the record (with lock), check if the original data is the same as the
just retrieved data (in case of changes from another workstation), and
if still ok, update the record.
On 9/12/2015 11:41 AM, Raul A Jager W wrote:
Sometimes a record gets locked, DSPRCDLCK shows the job that is holding
the record, but not the program that caused the lock.
It is likely that a program returned (or failed) without releasing all
records it has read for update, but I don't know which one.
DSPJOB just shows in the stack the http server ready to receive data and
the log shows only errors or info messages.
How can I find out which program left the record locked?
-- Este e-mail fue enviado desde el Mail Server del diario ABC Color --
-- Verificado por Anti-Virus Corporativo Symantec --