Re: The pointer parameter passed to free or realloc is not valid.

To expand on what Brad said, from http://www-912.ibm.com/, select "fixes" on the
left side of the screen, then select "PTF Cover Letter Database" (4th item from
the bottom on the next screen).  This should take you to
http://www-912.ibm.com/supporthome.nsf/document/24825731 where you can select
which cover letters to look at.  I usually select V5R1, then select "search,"
then enter my search arguments.

Regarding the PTF not fixing the problem, most HTTP server PTFs require that the
server instance be ended then restarted.  Perhaps this was not done.  The
following comes from the cover letter of PTF SI03023:

  This PTF will be activated for a HTTP server when that server
    instance is ended and restarted, or at the next IPL.  If a
    HTTP server is already active at the time this PTF is applied,
    the HTTP server must be ended and restarted in order to
    activate this PTF for that HTTP server instance.

    Do the following to activate this PTF for all HTTP server
    instances:
       1.  ENDTCPSVR SERVER(*HTTP) HTTPSVR(*ALL)
       2.  Restart the HTTP server(s) using:
              STRTCPSVR SERVER(*HTTP) HTTPSVR(<server instance name>)

Regarding GET, I suspect that the problem had nothing to do with the request
method and that the fix should work for either GET or POST.

Mel Rothman


"Bartell, Aaron L. (TC)" wrote:
>
> I forwarded that to our AS/400 admin and he told me that we had already
> applied that a couple of weeks ago because of a different problem we were
> having.  And I am actually doing a GET not a POST so I wonder if there is a
> different PTF out there for that?
>
> How do I find the info you found and where do I go to look?  Please educate
> me:-)
>
> Thanks,
> Aaron Bartell
>
> -----Original Message-----
> From: Mel Rothman [mailto:mel@rothmanweb.com]
> Sent: Wednesday, May 08, 2002 3:44 PM
> To: web400@midrange.com
> Subject: Re: [WEB400] The pointer parameter passed to free or realloc is
> not valid.
>
> Aaron, this is fixed for V5R1 in PTF SI03023 for 5722DG1.
>
> The PTF cover letter says:
>
>   DESCRIPTION OF PROBLEM FIXED FOR APAR SE04099
> :
>
> -----------------------------------------------
>     C2M1212 message logged on CGI POST
>
>   CORRECTION FOR APAR SE04099
> :
>
> -----------------------------
>     The logic was not setting a pointer properly when a POST form
>     had a blank entry field.  The logic has been corrected.
>
> Aside from the messages appearing in the job log, no harm is done.  The
> server
> was attempting to free storage that had not been allocated.
>
> Mel Rothman