[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: XALLOC in program LIBHTTP/HTTPAPIR4



Hi Kim,

The only known problem with the current beta is that not all of the 
EXAMPLExx programs are there.  I started rewriting them, and never had 
time to finish.

So you might want to save the EXAMPLExx members from 1.23.

Otherwise, the beta is exactly what the 1.24 release will be.


On 2/9/2011 6:09 PM, Kim Gibson wrote:
>
>     Hello Scott,
>
>
>     Good news - I installed the newest beta into a test library, and it
>     received the large response from the web service without a problem.
>
>
>     After parsing, the output was a 79-page spooled file.
>
>
>     Now on to trouble-shooting the multiple requests to the web service.
>
>
>     Thanks for your help - I'll do more testing with the beta version of
>     HTTPAPI. Do I wait for the update of the HTTPAPI distribution before
>     updating the production HTTPAPI library?
>
>
>     Kim
>
>
>
>     Date: Mon, 07 Feb 2011 17:27:15 -0600
>
>     From: Scott Klement<[1]sk@xxxxxxxxxxxxxxxx>
>
>     Subject: Re: XALLOC in program LIBHTTP/HTTPAPIR4
>
>     To: HTTPAPI and FTPAPI Projects<[2]ftpapi@xxxxxxxxxxxxxxxxxxxxxx>
>
>     Message-ID:<[3]4D507FD3.9040009@xxxxxxxxxxxxxxxx>
>
>     Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>
>     Hi Kim,
>
>
>
>     >  Scott, I'm using version 1.23 of HTTPAPI. Is there a beta version
>
>     >  that's newer?
>
>
>     Yes...  the beta version is definitely newer:
>
>     [4]http://www.scottklement.com/httpapi/beta
>
>
>
>     >  Since I'm calling a web service that is subscriber-only, I'm not
>     sure
>
>     >  how to reproduce the problem. I can't save the response that's
>
>     >  received from the web service - that's where the error occurs.
>
>     >  However....it *is* saved in httpapi_debug.txt. Would that help? I
>
>     >  could remove the identifying information and send it to you, if it
>     would help.
>
>
>     Please try the beta first.  If that doesn't help, then try the debug
>     file.  (I don't know if the debug file will help, though.)
>
>
>
>     >  As per Dennis' idea, I've been poking around HTTPAPIR4 to see what
>
>     >  XALLOC was doing, and I've seen the TERASPACE definition. I can sure
>
>     >  take a look at enabling it&   recompiling HTTPAPI.
>
>
>     This was an experimental feature that I disabled because it didn't
>     work
>
>     well.
>
>
>     As I've already said in previous messages, enabling HTTPAPI's
>     teraspace
>
>     support will break compatibility with Expat.  It's not a question.  It
>
>     will break, believe me.
>
>
>     And I don't know what else it'll break!
>
>
>     Please only enable this *if* you are willing to debug the code
>     yourself
>
>     and get it working yourself.  If you enable this code and then post a
>
>     message saying "Scott, you have to fix Expat and HTTPAPI because I
>
>     enabled teraspace and now it crashes with memory errors that I don't
>
>     understand." then I will not be very sympathetic.  It's disabled for a
>
>     reason.
>
>
>
>     >  My boss is out of the office for the next 3 weeks, and he's the
>
>     >  SECOFR, so getting the recompiled HTTPAPI moved into production
>     might
>
>     >  be a challenge. Perhaps I could do it in a test library on our
>
>     >  iSeries that is a web server, and see if it helps.
>
>
>     Please try the beta (in your test lib) first.
>
>
>
>     >
>
>     >  I'm going to be out for the rest of the day, but will be right back
>     on
>
>     >  this first thing in the morning. I'll see about recompiling into
>     test
>
>     >  with TERASPACE enabled.
>
>
>     Please try the beta first.
>
>
>     If TERASPACE is the only way, then I'll try to help you get it
>
>     working... but right now it's purely a GUESS as to something that
>     MIGHT
>
>     help, we don't even have a clear indication of what the problem is,
>     much
>
>     less what the solution should be.
>
>
>     Fixing TERASPACE support is probably (?) within my abilities, but it's
>
>     not trivial.  I'd rather not do it until we know it's the right
>
>     direction to take, and other avenues have been exhausted.
>
> References
>
>     1. mailto:sk@xxxxxxxxxxxxxxxx
>     2. mailto:ftpapi@xxxxxxxxxxxxxxxxxxxxxx
>     3. mailto:4D507FD3.9040009@xxxxxxxxxxxxxxxx
>     4. http://www.scottklement.com/httpapi/beta
>
>
>
>
> -----------------------------------------------------------------------
> This is the FTPAPI mailing list.  To unsubscribe, please go to:
> http://www.scottklement.com/mailman/listinfo/ftpapi
> -----------------------------------------------------------------------

-----------------------------------------------------------------------
This is the FTPAPI mailing list.  To unsubscribe, please go to:
http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------