×
The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.
I run four concurrent instances of the same never-ending program. It watches a data queue and does insert/update operations to SQL Server. When an I/O fails, it writes the request to a table, and the same program will retry the request later. I only want a single instance to be reprocessing the failures at a time. I guess allocating a data area would work better.
Thanks
-----Original Message-----
From: Buck Calabro [mailto:kc2hiz@xxxxxxxxx]
Sent: Friday, March 10, 2017 9:32 AM
To: rpg400-l@xxxxxxxxxxxx
Subject: Re: Release *SHRRD lock from SQL
On 3/10/2017 9:57 AM, Justin Taylor wrote:
I have an internal procedure that reads a table using a SQL cursor. I explicitly close the cursor before returning, but it seems to be keeping a *SHRRD lock on the PF. What's the best way to eliminate this problem?
The best way is to avoid doing things restricted by *SHRRD and that's a sincere lesson I learnt the hard way. When I start using SQL on a file, I stop doing things like RGZPFM. What's happening here is a pseudo-close; the machine is trying to be efficient with future use of the same SQL statement in the same job.
Depending on how your app is constructed, you can reclaim the activation group to hard close the cursor. Or you can use ALCOBJ ...
CONFLICT(*RQSRLS) in the job that's needing the *SHRRD, and IBM i will do the hard close then.
--
--buck
Try wiki.midrange.com!
midrange.com
