Latest TCPware DRIVERS patch seems to break inbound DECnet

Heads-up to all.

I recently installed the latest TCPware DRIVERS patch,
DRIVERS_V572P081 on several systems and then discovered
that inbound DECnet-over-IP connections (both set host
and FAL) no longer work. Outbound works fine.

The symptom is after a minute or so you get a
"%SYSTEM-F-LINKEXIT, network partner exited" error
and it seems to leave an INET connection in "CLOSE-WAIT"
state on the destination system.

This bug happens on both VAX and Alpha systems (don't
have an Itanium with TCPware to try it on.) The VAXes
are VMS V7.3/TCPware V5.7-2. The Alphas are VMS V8.3
with TCPware V5.7-2.

I have one Alpha that still has the DRIVERS_V572P050
version of the patch on it, which works fine. Taking a
shot in the dark, I copied PWIPDRIVER.EXE from it to
one of the broken Alphas and rebooted it, and it now
works again. (The curious thing is PWIPDRIVER isn't
listed in any of the DRIVERS patch release notes as
a modified module, but each DRIVERS patch seems to build
a new one!)

Am forwarding this to Process Software as well.



--
John

On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:[color=blue]
> Heads-up to all...[/color]

Have just realised we have the same problem after patch upgrades at
the weekend. Will look into the fix you describe above, but was
wondering if Process have been in touch regarding this issue? I cant
see anything on their site and the patches installed were only
downloaded at the weekend.

Many thanks!

Adam

On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:[color=blue]
> Heads-up to all...[/color]

Have just realised we have the same problem after patch upgrades at
the weekend. Will look into the fix you describe above, but was
wondering if Process have been in touch regarding this issue? I cant
see anything on their site and the patches installed were only
downloaded at the weekend.

Many thanks!

Adam

You did not specify what patch you installed.

DRIVERS_V572P070 contained a bad PWIPDRIVER change.
DRIVERS_V572P081 contains the original PWIPDRIVER that shipped with
TCPware 5.7-2 and should work fine.

-----Original Message-----
From: [email]adamwilliam@gmail.com[/email] [mailto:adamwilliam@gmail.com]
Sent: Monday, December 03, 2007 6:02 AM
To: [email]info-tcpware@process.com[/email]
Subject: Re: Latest TCPware DRIVERS patch seems to break inbound DECnet

On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:[color=blue]
> Heads-up to all...[/color]

Have just realised we have the same problem after patch upgrades at
the weekend. Will look into the fix you describe above, but was
wondering if Process have been in touch regarding this issue? I cant
see anything on their site and the patches installed were only
downloaded at the weekend.

Many thanks!

Adam

Go back to the original PWIPDRIVER.EXE, which should be in the TCPware
directory as PWIPDRIVER.EXE_OLD

What architecture and version of VMS?

-----Original Message-----
From: [email]Peter.DeLaCour@interactivedata.com[/email]
[mailto:Peter.DeLaCour@interactivedata.com]
Sent: Monday, December 03, 2007 9:47 AM
To: [email]info-tcpware@process.com[/email]
Subject: RE: Latest TCPware DRIVERS patch seems to break inbound DECnet

Hi ,

I downloaded and installed DRIVERS_V572P081 this weekend and had the
problem described by Adam. Any ideas?

Regards Peter







Richard Whalen

<whalenr@process.

com>
To
[email]info-tcpware@process.com[/email]

03/12/2007 14:32
cc



Subject
Please respond to RE: Latest TCPware DRIVERS patch

Info-TCPware@proc seems to break inbound DECnet

ess.com














You did not specify what patch you installed.

DRIVERS_V572P070 contained a bad PWIPDRIVER change.
DRIVERS_V572P081 contains the original PWIPDRIVER that shipped with
TCPware 5.7-2 and should work fine.

-----Original Message-----
From: [email]adamwilliam@gmail.com[/email] [mailto:adamwilliam@gmail.com]
Sent: Monday, December 03, 2007 6:02 AM
To: [email]info-tcpware@process.com[/email]
Subject: Re: Latest TCPware DRIVERS patch seems to break inbound DECnet

On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:[color=blue]
> Heads-up to all...[/color]

Have just realised we have the same problem after patch upgrades at
the weekend. Will look into the fix you describe above, but was
wondering if Process have been in touch regarding this issue? I cant
see anything on their site and the patches installed were only
downloaded at the weekend.

Many thanks!

Adam


-----------------------------------------
This message (including any files transmitted with it) may contain
confidential and/or proprietary information, is the property of
Interactive Data Corporation and/or its subsidiaries, and is
directed only to the addressee(s). If you are not the designated
recipient or have reason to believe you received this message in
error, please delete this message from your system and notify the
sender immediately. An unintended recipient's disclosure,
copying, distribution, or use of this message or any attachments is
prohibited and may be unlawful. Interactive Data (Europe) Ltd
Registered No. 949387 England Registered Office: Fitzroy House 13-
17 Epworth Street. London. EC2A 4DL

In article
<3EF96AF20489A34296050FBD5C36ECB9BDAE0E@beacon.PSC.process.com>,
[email]whalenr@process.com[/email] says...[color=blue]
> Go back to the original PWIPDRIVER.EXE, which should be in the TCPware
> directory as PWIPDRIVER.EXE_OLD
>
> What architecture and version of VMS?
>[/color]

I had the problem on both VAX (VMS V7.3) and Alpha (VMS V8.3)
DRIVERS_V572P081 on both.

The response I got from Process led me to believe a new version
of the patch was to be released soon which would fix it. I was
starting to wonder what had happened to it...

From your response, it sounds like the "new" version is 8.1,
which AFAICT still has the problem.

If you really think V8.1 of the patch should work, I'll try
re-installing it to verify it's still broken on my system.

Checking the image idents, it looks like I installed 8.0
on our VAX at work, but 8.1 on our Alpha at work and on my
hobbyist VAX and Alpha at home. All 4 systems showed the
problem.

-- John

[color=blue]
> -----Original Message-----
> From: [email]Peter.DeLaCour@interactivedata.com[/email]
> [mailto:Peter.DeLaCour@interactivedata.com]
> Sent: Monday, December 03, 2007 9:47 AM
> To: [email]info-tcpware@process.com[/email]
> Subject: RE: Latest TCPware DRIVERS patch seems to break inbound DECnet
>
> Hi ,
>
> I downloaded and installed DRIVERS_V572P081 this weekend and had the
> problem described by Adam. Any ideas?
>
> Regards Peter
>
>
>
>
>
>
>
> Richard Whalen
>
> <whalenr@process.
>
> com>
> To
> [email]info-tcpware@process.com[/email]
>
> 03/12/2007 14:32
> cc
>
>
>
> Subject
> Please respond to RE: Latest TCPware DRIVERS patch
>
> Info-TCPware@proc seems to break inbound DECnet
>
> ess.com
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> You did not specify what patch you installed.
>
> DRIVERS_V572P070 contained a bad PWIPDRIVER change.
> DRIVERS_V572P081 contains the original PWIPDRIVER that shipped with
> TCPware 5.7-2 and should work fine.
>
> -----Original Message-----
> From: [email]adamwilliam@gmail.com[/email] [mailto:adamwilliam@gmail.com]
> Sent: Monday, December 03, 2007 6:02 AM
> To: [email]info-tcpware@process.com[/email]
> Subject: Re: Latest TCPware DRIVERS patch seems to break inbound DECnet
>
> On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:[color=green]
> > Heads-up to all...[/color]
>
> Have just realised we have the same problem after patch upgrades at
> the weekend. Will look into the fix you describe above, but was
> wondering if Process have been in touch regarding this issue? I cant
> see anything on their site and the patches installed were only
> downloaded at the weekend.
>
> Many thanks!
>
> Adam
>
>
> -----------------------------------------
> This message (including any files transmitted with it) may contain
> confidential and/or proprietary information, is the property of
> Interactive Data Corporation and/or its subsidiaries, and is
> directed only to the addressee(s). If you are not the designated
> recipient or have reason to believe you received this message in
> error, please delete this message from your system and notify the
> sender immediately. An unintended recipient's disclosure,
> copying, distribution, or use of this message or any attachments is
> prohibited and may be unlawful. Interactive Data (Europe) Ltd
> Registered No. 949387 England Registered Office: Fitzroy House 13-
> 17 Epworth Street. London. EC2A 4DL
>
>[/color]

--
John

[email]adamwilliam@gmail.com[/email] wrote:[color=blue]
> On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:
>[color=green]
>>Heads-up to all...[/color]
>
>
> Have just realised we have the same problem after patch upgrades at
> the weekend. Will look into the fix you describe above, but was
> wondering if Process have been in touch regarding this issue? I cant
> see anything on their site and the patches installed were only
> downloaded at the weekend.
>
> Many thanks!
>
> Adam[/color]

From a response from Process on vmsnet.networks.tcp-ip.tcpware, it
seems they thought it was fixed in V8.1, but I'm 99% sure it was
broken there too, and interpret the response I received when I
reported the problem to them that they were going to fix it in a
new (post V8.1) version of the patch.

So I've been waiting for a patch, but Process thought they had already
fixed it. Hopefully, they now realize it is still an issue.



--
John Santos
Evans Griffiths & Hart, Inc.
781-861-0670 ext 539

[email]adamwilliam@gmail.com[/email] wrote:[color=blue]
> On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:
>[color=green]
>>Heads-up to all...[/color]
>
>
> Have just realised we have the same problem after patch upgrades at
> the weekend. Will look into the fix you describe above, but was
> wondering if Process have been in touch regarding this issue? I cant
> see anything on their site and the patches installed were only
> downloaded at the weekend.
>
> Many thanks!
>
> Adam[/color]

From a response from Process on vmsnet.networks.tcp-ip.tcpware, it
seems they thought it was fixed in V8.1, but I'm 99% sure it was
broken there too, and interpret the response I received when I
reported the problem to them that they were going to fix it in a
new (post V8.1) version of the patch.

So I've been waiting for a patch, but Process thought they had already
fixed it. Hopefully, they now realize it is still an issue.



--
John Santos
Evans Griffiths & Hart, Inc.
781-861-0670 ext 539

In article <3fm5j.2250$UG1.300@trnddc01>, [email]john@egh.com[/email] says...[color=blue]
> [email]adamwilliam@gmail.com[/email] wrote:[color=green]
> > On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:
> >[color=darkred]
> >>Heads-up to all...[/color]
> >
> >
> > Have just realised we have the same problem after patch upgrades at
> > the weekend. Will look into the fix you describe above, but was
> > wondering if Process have been in touch regarding this issue? I cant
> > see anything on their site and the patches installed were only
> > downloaded at the weekend.
> >
> > Many thanks!
> >
> > Adam[/color]
>
> From a response from Process on vmsnet.networks.tcp-ip.tcpware, it
> seems they thought it was fixed in V8.1, but I'm 99% sure it was
> broken there too, and interpret the response I received when I
> reported the problem to them that they were going to fix it in a
> new (post V8.1) version of the patch.
>
> So I've been waiting for a patch, but Process thought they had already
> fixed it. Hopefully, they now realize it is still an issue.[/color]

I tried patch drivers_v572p081 again, it definitely still shows the
problem on Alpha VMS V8.3.


--
John

In article <3fm5j.2250$UG1.300@trnddc01>, [email]john@egh.com[/email] says...[color=blue]
> [email]adamwilliam@gmail.com[/email] wrote:[color=green]
> > On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:
> >[color=darkred]
> >>Heads-up to all...[/color]
> >
> >
> > Have just realised we have the same problem after patch upgrades at
> > the weekend. Will look into the fix you describe above, but was
> > wondering if Process have been in touch regarding this issue? I cant
> > see anything on their site and the patches installed were only
> > downloaded at the weekend.
> >
> > Many thanks!
> >
> > Adam[/color]
>
> From a response from Process on vmsnet.networks.tcp-ip.tcpware, it
> seems they thought it was fixed in V8.1, but I'm 99% sure it was
> broken there too, and interpret the response I received when I
> reported the problem to them that they were going to fix it in a
> new (post V8.1) version of the patch.
>
> So I've been waiting for a patch, but Process thought they had already
> fixed it. Hopefully, they now realize it is still an issue.[/color]

I tried patch drivers_v572p081 again, it definitely still shows the
problem on Alpha VMS V8.3.


--
John

John Santos wrote:[color=blue]
> In article <3fm5j.2250$UG1.300@trnddc01>, [email]john@egh.com[/email] says...
>[color=green]
>>adamwilliam@gmail.com wrote:
>>[color=darkred]
>>>On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:
>>>
>>>
>>>>Heads-up to all...
>>>
>>>
>>>Have just realised we have the same problem after patch upgrades at
>>>the weekend. Will look into the fix you describe above, but was
>>>wondering if Process have been in touch regarding this issue? I cant
>>>see anything on their site and the patches installed were only
>>>downloaded at the weekend.
>>>
>>>Many thanks!
>>>
>>>Adam[/color]
>>
>> From a response from Process on vmsnet.networks.tcp-ip.tcpware, it
>>seems they thought it was fixed in V8.1, but I'm 99% sure it was
>>broken there too, and interpret the response I received when I
>>reported the problem to them that they were going to fix it in a
>>new (post V8.1) version of the patch.
>>
>>So I've been waiting for a patch, but Process thought they had already
>>fixed it. Hopefully, they now realize it is still an issue.[/color]
>
>
> I tried patch drivers_v572p081 again, it definitely still shows the
> problem on Alpha VMS V8.3.[/color]

Today I received a test version of the next DRIVERS patch and it
fixes the problem on both VAX and Alpha. I expect it will be
generally available Real Soon Now.


--
John Santos
Evans Griffiths & Hart, Inc.
781-861-0670 ext 539

John Santos wrote:[color=blue]
> In article <3fm5j.2250$UG1.300@trnddc01>, [email]john@egh.com[/email] says...
>[color=green]
>>adamwilliam@gmail.com wrote:
>>[color=darkred]
>>>On Nov 13, 2:12 am, John Santos <john.san...@post.harvard.edu> wrote:
>>>
>>>
>>>>Heads-up to all...
>>>
>>>
>>>Have just realised we have the same problem after patch upgrades at
>>>the weekend. Will look into the fix you describe above, but was
>>>wondering if Process have been in touch regarding this issue? I cant
>>>see anything on their site and the patches installed were only
>>>downloaded at the weekend.
>>>
>>>Many thanks!
>>>
>>>Adam[/color]
>>
>> From a response from Process on vmsnet.networks.tcp-ip.tcpware, it
>>seems they thought it was fixed in V8.1, but I'm 99% sure it was
>>broken there too, and interpret the response I received when I
>>reported the problem to them that they were going to fix it in a
>>new (post V8.1) version of the patch.
>>
>>So I've been waiting for a patch, but Process thought they had already
>>fixed it. Hopefully, they now realize it is still an issue.[/color]
>
>
> I tried patch drivers_v572p081 again, it definitely still shows the
> problem on Alpha VMS V8.3.[/color]

Today I received a test version of the next DRIVERS patch and it
fixes the problem on both VAX and Alpha. I expect it will be
generally available Real Soon Now.


--
John Santos
Evans Griffiths & Hart, Inc.
781-861-0670 ext 539