Conference smaug::snagwy

Daniele,

What kind of problem is your customer having? DTF should not install on V7.1.

ECO 02 is in the process of being submitted to TIMA as we speak and is the 
V7.1 certified release.

Linas

    
    Linas
    
    
    Thanks for reply,
    
    Actually the probel is that when the customer send to IBM files, put in
    saturation the IBM MVS buffers and the Host goes in hung
    Seems that it does not matter if the host can receive data or not.
    Of course with a little files, this does not happens, but
    unfornutally this customer send out a lot of MB of files a time.
    Customer IBM people claims that it seems that it ignore the PACING
    mechanism.
    Actually he works with DTF 3.1 and 3.2 over OpenVMS 6.2
    
    
    Tomorrow I will go on site and I will take more info.
    
    
    
    But anyway You give me a great info, that V3.2 is not yet certified for 
    V7.1 !!
    Is that correct ?
    
    How can obtain the ECO 02, is it already released for customer ?
    In our Tima I even did not found the V3.2 ECO 01.
    
    
    Thanks a lot 
    Daniele
    

Daniele,

What is the severity of this problem to the customer?

ECO 02 contains several fixes for pacing problems. I suggest you 
obtain the ECO 02 kit from TIMA. (I submitted it to them yesterday and
do not know how soon it will be available)

ECO 01 was created but never submitted to TIMA. 

ECO 02 is the release for V7.1.
 
Linas

    Linas,
    
       I went to the customer this morning, the problem is very big, becouse 
    yesterday during some customer test, they blocked all VTAM, so they were 
    stopped for 4 hour. 
    The problem seems to be dued by pacing.
    They are tried to stop DTF IBM application, and the AXP instead of stop
    trasmission it continue to transmit wasted the VTAM I/O buffer.
    The same kind of test issue from AXP OpenVMS 6.1 and DTF 3.1 works
    well.
    The negotiated pacing is 4, the same of the others DTF session.
    
    I have got some VTAM/DTF trace from IBM side, and I have checked the
    DTF IBM PTF level, they have installed SD31079.
    
    Would you like to have a quick look to this trace ?
    
    Do you also need to install IBM DTF v3.2 for compatibility or V3.1 with
    those PTF level is supported for V3.2 with ECO 02 ?
    
    Could you give me a direct pointer to this patch for examples on CSCPT2" ?
    
    Thanks a lot.
    Daniele
    
    

RE:                     <<< Note 7092.4 by TOSSUB::SABBADIN >>>
    
>       I went to the customer this morning, the problem is very big, becouse 
>    yesterday during some customer test, they blocked all VTAM, so they were 
>    stopped for 4 hour. 

	I don't know what you mean by "blocked all VTAM". This is almost
	imposssible. VTAM has a lot of defensive code in it to prevent
	buffer starvation.

>    The problem seems to be dued by pacing.

	Can you please give us more detail as to why you think that?
	Also please give us a diagram of the network configuration and
	the version/release of all the IBM components.

>    They are tried to stop DTF IBM application, and the AXP instead of stop
>    trasmission it continue to transmit wasted the VTAM I/O buffer.

	How did you try to "stop the DTF IBM application"? Please be exact.
	What do you mean when you say "wasted the VTAM I/O buffer"?
	Do you have output from the VTAM command D NET,BFRUSE?
    
>    I have got some VTAM/DTF trace from IBM side, and I have checked the
>    DTF IBM PTF level, they have installed SD31079.

	Please open a IPMT and copy the trace(I am assuming it is a file, not
	a printout) to SMAUG::[deposit] and give it a file name that is the
	same as the IPMT number.
    
>    Would you like to have a quick look to this trace ?

	See above.
    
>    Do you also need to install IBM DTF v3.2 for compatibility or V3.1 with
>    those PTF level is supported for V3.2 with ECO 02 ?

	DTF/MVS V3R1 will work just fine with DTF/OVMS V3R2.
    
Craig    

    Please install ECO-01 or ECO-02 of DTF if it is available.
    ECO-01 fixes pacing problem that sounds a lot like the
    symptoom here. ECO-02 will include ECO-01 fixes.
    
    Bill