home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!paladin.american.edu!news.univie.ac.at!hp4at!mcsun!sun4nl!orcenl!nl.oracle.com!nroworth
- From: nroworth@nl.oracle.com (Nick Roworth)
- Newsgroups: comp.databases.oracle
- Subject: Re: Oracle 7.0.9.3, XA, and Encina
- Message-ID: <3198@nlsun1.oracle.nl>
- Date: 6 Nov 92 09:52:05 GMT
- References: <19921104.071527.486@almaden.ibm.com> <3190@nlsun1.oracle.nl>
- Sender: news@nl.oracle.com
- Organization: Oracle Europe
- Lines: 33
- Nntp-Posting-Host: nlsu23
-
- In article <3190@nlsun1.oracle.nl>, kverruyt@nl.oracle.com (Kees Verruyt) writes:
- |> Alan Beal writes:
- |> > Has anyone been successful using Oracle 7.0.9.3 with Transarc Encina
- |> > 1.0.1a using the XA interface? Using embedded SQL, we get data
- |> > retrieved on the first Encina transaction and then get invalid
- |> > cursor after that (sqlcode = 1001). Oracle Support will not help
- |> > because it is not a production product. Any ideas?
- |> >
- |> > Alan Beal alanb@owgvm0.vnet.ibm.com
- |>
- |> Alan,
- |>
- |> I struggled with this myself -- just compile your Pro*C program with
- |>
- |> EXEC ORACLE OPTION (RELEASE_CURSOR=YES);
- |>
- |> Otherwise the Pro*C cursor cache gets in the way: Even if you CLOSE a cursor,
- |> Pro*C tries to keep it open.
- |>
- |> It should work fine after that.
- |>
- |> Note: Since this is ORACLE7, closing a cursor in the front-end isn't so bad since
- |> the parsed SQL is kept around in the ORACLE7 server SGA.
-
- An easier way to do this is to set the flag when using proc to compile the program
- Just set release_cursor=yes in the makefile for the proc options.
-
-
- --
- Nick Roworth, ORACLE Europe "Bugs Mr. Rico! Zillions of 'em!"
- Rijnzathe 6, NL-3454 PV De Meern Starship Troopers - Robert A. Heinlein
- Phone: +31-3406-94211
- nroworth@nl.oracle.com
-