Bug in 6.1.17

[Doug]

I have noticed that this latest version does not always remove the command
prompt edit field and put up the 'Debugee is running' message when the
debugee connects. If I break into the target and start it going again, it
does remove the edit field and put up the message. Not a big deal, but it is
visually confusing when you are used to seeing the message if the target is
running.

[DieHard]

Also the .kdfiles command to download driver image cannot work on my
platform.:-( Anyone else has the same problem?

"Doug" <dha...@junk.com> wrote in message
news:eA8whxoqCHA.2172@TK2MSFTNGP12...

[Christophe]

Well, I was using 1394 link to debug my XP target machine with the
previous version of windbg...(the one before 6.1.17) Everything was
fine - even kdfiles was working - Badly, i decided to upgrade my
windbg to 6.1.0017 (6.1.17): now kdfiles hangs:
When windbg uploads my driver to the target machine using the .kdfiles
command, windbg hangs....

I WANT TO COME BACK TO THE PREVIOUS VERSION OF WINDBG, WHERE CAN I
DOWNLOAD IT?? Microsoft my friend... please... :))

Christophe

[DieHard]

Luckily I still keep the 6.1.0009 which can work. If you want, tell me where
to put it.

"Christophe" <NO_cdem...@hotmail.com> wrote in message
news:9bmq1v48jjhfa9fef...@4ax.com...

[Wendy Shi [MS]]

It's most likely user error, not the debugger itself.

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

[Wendy Shi [MS]]

The Debugger team is investigating on that.

[Phil Barila]

"Wendy Shi [MS]" <wend...@online.microsoft.com> wrote in message
news:pANXYepvCHA.1488@cpmsftngxa06...

> It's most likely user error, not the debugger itself.

No, it's not. I've seen it too, but I didn't figure you needed any more
"Me, too" posts.

Please include a little bit of text from the message to which you are
replying when responding to a post that is almost a month old.

And lastly, I can't speak for anyone else who reads this newsgroup, but I
didn't find this helpful at all. This sounds like the same attitude that
used to permeate almost all of my dealings with Microsoft in the mid-90s.
Since I started doing kernel development about the same time that Windbg was
restarted, I've found Andre, Nathan, and David (and others I'm sure I've
forgotten) to be eminently helpful and responsive, and I don't ever remember
them being this dismissive. The method they used, which ****you would do
well to emulate****, is to simply point to the proper means of doing
something, if it was known, or ask for reproduction steps. This approach
often led us to discover our own user error. Just as often, they would
respond that they were fixing the bug.

Phil
--
Philip D. Barila
Seagate Technology, LLC
(720) 684-1842
As if I need to say it: Not speaking for Seagate.
E-mail address is pointed at a domain squatter. Use reply-to instead.

[Wendy Shi [MS]]

I'm so sorry to hear that. I apologize. Actually I got that answer from
someone else and I posted it directly. I would hope that I could provide a
better one, but I just couldn't. Please take it as my personal fault, not
Microsoft. I will relay your message to the windbg team.

Once again, I apologize and thank you for your input.

--

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

"Phil Barila" <PBa...@Barila.com> wrote in message
news:tsWcnf7K9vt...@4dv.net...

[Wendy Shi [MS]]

Thanks for Phil's suggestion. The bug we talked about here is in the first
posting from Doug--

I don't see it happen on my machine. If you can repro it, please share with
us. Please also tell us your system information or any other related
information. Thanks!

--

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

"Phil Barila" <PBa...@Barila.com> wrote in message
news:tsWcnf7K9vt...@4dv.net...

[Wendy Shi [MS]]

The windbg team is working the issues about kdfiles as posted by Christophe

"When windbg uploads my driver to the target machine using the .kdfiles

command, windbg hangs.... ". Once we get it done, we will post the answer.
We apologize for any inconvenience.

--

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

"Wendy Shi [MS]" <wend...@online.microsoft.com> wrote in message
news:aZV0bfpvCHA.3284@cpmsftngxa06...

"Christophe" <NO_cdem...@hotmail.com> wrote in message
news:9bmq1v48jjhfa9fef...@4ax.com...

[Phil Barila]

"Wendy Shi [MS]" <wend...@online.microsoft.com> wrote in message

news:eeOtZiMwCHA.1624@TK2MSFTNGP10...

> I'm so sorry to hear that. I apologize. Actually I got that answer from
> someone else and I posted it directly. I would hope that I could provide a
> better one, but I just couldn't. Please take it as my personal fault, not
> Microsoft. I will relay your message to the windbg team.
>
> Once again, I apologize and thank you for your input.

I should know better than to read and post to usenet that late on a Friday
night!

Wendy, I'm sorry, too, for coming down on you so hard. It would have been
more useful for the person who gave you that unhelpful response to have
requested the steps to reproduce and all that.

Just to be absolutely clear, I'm not the original poster. I don't even
notice the behavior in question anymore, since the first time I saw it and
thought to myself, "That's odd...", and promptly forgot about it until I saw
your non-response.

I'll say this again. It's never any help to anyone to simply blame the user
without at least attempting to determine what the user is doing wrong, so
they can stop doing it. The user may have really found a bug, after all.
Responses like the one under discussion are worse than useless, they
engender user animosity and hostility.

Thank you for taking the time to read and respond,

[Doug]

I did not recognize it as you responding to my oiginal post because it has
been so long since I posted the original message that it had fallen off my
newsreader.

Anyway here is how I get it:
1) I start up WinDbg,
2) I select a stored kernel debugging session I have (which specifies the
symbol path, the source path, the com port and the speed - 115200).
3) WinDbg currently reads 'Debugee not connected' and the command window
says Opening \\.\Com2 (CRLF) waiting to reconnect...
4) Start up target (Dell 530MT Dual 1.8 Ghz) with .NET RC2 that has the
boot.ini changed to add /DEBUGPORT=COM1 & /BAUDRATE=115200
5) System comes up and displays the kernel information up to the line that
says 'System Uptime: not available'
6) Once the information is displayed the status bar usually changes to
'Debuggee is running....', but in 6.1.17 (and possibly the previous version
also) it just shows a blank command prompt.

The debuggee is running and I can break in, restart, and get the 'Debugee is
running' message. Happens EVERY time. Kind of tough for it to be a user
error. I have used WinDbg for 5 years so I am pretty familiar with it.

Doug

"Wendy Shi [MS]" <wend...@online.microsoft.com> wrote in message

news:eoacaeNwCHA.2640@TK2MSFTNGP11...

[Allan]

I have the same issue. I don't know what order of operations it is, but
sometimes the "Debugee is Running" message does not come up and an edit
window stays on this line while the debugger is actually running. Not a big
deal but can be confusing at times.

Al

"Doug" <dha...@junk.com> wrote in message

news:#rqznXVwCHA.2868@TK2MSFTNGP12...

[Doug]

Exactly the problem I am specifying.

"Allan" <aes "at" connecttech "dot" com> wrote in message
news:3e2d5da6$1...@news.sentex.net...

[Doug]

Version 6.1.17.2 prevents the KD> prompt from appearing, but the text
message is still 'Debugee is not connected'

"Wendy Shi [MS]" <wend...@online.microsoft.com> wrote in message

news:eoacaeNwCHA.2640@TK2MSFTNGP11...

[Wendy Shi [MS]]

I tested that on my machines: host XP SP3, target 2K SP3. I noticed this:
if you run the debugger (File->Kernel Debug) when the debuggee wasn't
broken in before, ie, debugee is running, then kd> command will not come
up and the msg shows "debuggee not connected". You have to Ctrl+Break to
make the connection. If your debuggee was being broken in, then whenever
you stop the kernel debuging and restart it, it should give the kd> command
window. Is this what happened to you, Doug? If you have different
scenarios, please share with us. Thank you very much.

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

--------------------
From: "Doug" <dha...@junk.com>

[Doug]

Here is what I do:

1) Select a saved kernel debugging workspace. WinDbg has 'debugee not
connected' at bottom of command window and command display has 'Waiting to
connect...'
2) Start target
3) Command display starts showing target debug info
4) 6.1.17.0 will go from 'debugee not connected' to empty command line.
5) 6.1.17.2 will stay at 'debugee not connected'

Doug

""Wendy Shi [MS]"" <wend...@online.microsoft.com> wrote in message

news:gZ2#5Gw0CHA.3108@cpmsftngxa06...

[Wendy Shi [MS]]

Thank you for the repro, Doug. We will work on that.

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

--------------------
From: "Doug" <dha...@junk.com>

[Wendy Shi [MS]]

I have confirmed with Windbg team that it is the way it's supposed to be.
If there is no initial breakin, the bottom message will show "Debuggee not
connected" even though the connection has been established as the message
on the top shows. This is because the debugger isn't fully connected until
there's a real event, such as a breakin.

The Windbg team has fixed the problem in 6.1.17.1 that it displayed the kd>
command prompt when it shouldn't have been displayed. Maybe that's what you
saw in 6.1.17.0?

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

--------------------
From: wend...@online.microsoft.com ("Wendy Shi [MS]")
Organization: Microsoft
Date: Thu, 13 Feb 2003 19:21:02 GMT
Subject: Re: Bug in 6.1.17

[Doug]

It has always been that when the OS banner comes across that the status
message changes from 'Debugee not connected' to 'Debuggee is running'. Only
since the problems I have seen of late has this been different. If you are
saying that it is 'Working as Designed' then the design has changed.

I must also say these later versions have a hard time connecting. Often I am
left with a hung target while the communication between the two is somehow
in a wacked out state. Often I either have to kill WinDbg, restart WinDbg,
and press Ctrl-Break a number of times to get the connection reestablished
or I press Ctrl-Break a number of times followed by Alt-Ctrl-R a couple of
times, followed by Ctrl-Break again a number of times to get it
reestablished. In the latter case I sometimes get a message saying that some
?????Async routine expected a 4 and got a 64.

I think there is something buggy in the connection protocol that was
introduced in some 6.x.x.x version of WinDbg. I never had problems like
these before.

""Wendy Shi [MS]"" <wend...@online.microsoft.com> wrote in message

news:Lzh$GiJ1CHA.1652@cpmsftngxa08...

[Doug]

Actually the error is "KdReadVirtual: Asked for 4 got 64"

"Doug" <dha...@junk.com> wrote in message

news:#HffU6o1CHA.2904@TK2MSFTNGP09...
> ... In the latter case I sometimes get a message saying that some

[Wendy Shi [MS]]

Doug, I have to ask for your favor again: can you repro that problem? I
tried to repro it but didn't succeed. Thanks a lot.

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

-------------------
From: "Doug" <dha...@junk.com>

Subject: Re: Bug in 6.1.17

Date: Mon, 17 Feb 2003 09:12:35 -0500

[Phil Barila]

"Doug" <dha...@junk.com> wrote in message
news:#HffU6o1CHA.2904@TK2MSFTNGP09...

[snip]

> I must also say these later versions have a hard time connecting. Often I
am
> left with a hung target while the communication between the two is somehow
> in a wacked out state. Often I either have to kill WinDbg, restart WinDbg,
> and press Ctrl-Break a number of times to get the connection reestablished
> or I press Ctrl-Break a number of times followed by Alt-Ctrl-R a couple of
> times, followed by Ctrl-Break again a number of times to get it
> reestablished. In the latter case I sometimes get a message saying that
some
> ?????Async routine expected a 4 and got a 64.

Doug,

Has it worked for you to unplug the cable, wait a few seconds, then plug it
in again? That's been somewhat of a magic bullet for me, when windbg and
the target are both completely hung. Sometimes my target will start to
break, I can see it pause for about 10 seconds, then take off and boot up
without the connection. No amount of Ctrl+Alt+R or Ctl+Break has ever
recovered that, but the physical disconnect cycle does.

Hope this helps.

[Doug]

I can always get it to work again one way or another. I am just trying to
provide info so that the problem can be fixed.

"Phil Barila" <PBa...@Barila.com> wrote in message

news:rgqdneaLb68...@4dv.net...

[Doug]

I can repro it quite often. What good will that do if you cannot repro it?
Is there something you want me to look for?

""Wendy Shi [MS]"" <wend...@online.microsoft.com> wrote in message

news:u6QXqj51CHA.2372@cpmsftngxa08...

[Wendy Shi [MS]]

Yea, I was wondering if you did something uncommon which caused the
problem.

Actually, believe it or not, that problem was happening to me a minute ago.
No matter I Ctrl+Break or Alt+Ctrl+Break or clicking on the icon, it just
didn't do anything. So I killed WinDbg and restarted it. It then worked
fine. I didn't do anything special either, maybe except that I disabled a
scsi adapter in device manager. I installed the card this afternoon. I
don't think that will be the reason.

Anway, thank you Doug. I really appreciate you reporting the problem to us.
Please keep that way.

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

--------------------
From: "Doug" <dha...@junk.com>
References: <eA8whxoqCHA.2172@TK2MSFTNGP12> <pANXYepvCHA.1488@cpmsftngxa06>
<tsWcnf7K9vt...@4dv.net> <eoacaeNwCHA.2640@TK2MSFTNGP11>
<#xkeerU0CHA.1632@TK2MSFTNGP12> <gZ2#5Gw0CHA.3108@cpmsftngxa06>
<esgXDb30CHA.1624@TK2MSFTNGP11> <KwUmCU50CHA.2848@cpmsftngxa06>
<Lzh$GiJ1CHA.1652@cpmsftngxa08> <#HffU6o1CHA.2904@TK2MSFTNGP09>
<u6QXqj51CHA.2372@cpmsftngxa08>

Subject: Re: Bug in 6.1.17

Date: Tue, 18 Feb 2003 18:13:21 -0500
Lines: 15
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Message-ID: <#wFrGN61CHA.2204@TK2MSFTNGP09>
Newsgroups: microsoft.public.windbg
NNTP-Posting-Host: rdu74-163-227.nc.rr.com 24.74.163.227
Path:
cpmsftngxa08!cppssbbsa01.microsoft.com!news-out.cwix.com!newsfeed.cwix.com!n
ewsengine.sol.net!newsfeeds.sol.net!news.maxwell.syr.edu!cyclone.bc.net!msrt
rans1!TK2MSFTNGP08!TK2MSFTNGP09
Xref: cpmsftngxa08 microsoft.public.windbg:467
X-Tomcat-NG: microsoft.public.windbg

[Wendy Shi [MS]]

Doug,

That problem doesn't happen on my machine that often, and it does on yours,
so could you generate a solid repro and make note of all these things
involved in the repro?

1) host OS build and debugger version and debugger command line options
used (the debugger 'version' command will show the debugger version and
command line options)
2) target OS build and boot.ini settings (the debugger 'version' command
will give target OS build)
3) if any debugger commands are typically executed before the debugger
enters this state, what is the exact sequence of commands?
4) how long does it take to get to this state and how often does it seem to
repro?

Thank you very much!

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

--------------------

From: wend...@online.microsoft.com ("Wendy Shi [MS]")
Organization: Microsoft

Date: Wed, 19 Feb 2003 04:49:55 GMT

Subject: Re: Bug in 6.1.17

Yea, I was wondering if you did something uncommon which caused the
problem.

Actually, believe it or not, that problem was happening to me a minute ago.
No matter I Ctrl+Break or Alt+Ctrl+Break or clicking on the icon, it just
didn't do anything. So I killed WinDbg and restarted it. It then worked
fine. I didn't do anything special either, maybe except that I disabled a
scsi adapter in device manager. I installed the card this afternoon. I
don't think that will be the reason.

Anway, thank you Doug. I really appreciate you reporting the problem to us.
Please keep that way.

Regards,
Wendy Shi
Microsoft DDK Support

This posting is provided "AS IS" with no warranties, and confers no rights.

--------------------
From: "Doug" <dha...@junk.com>

Subject: Re: Bug in 6.1.17
Date: Tue, 18 Feb 2003 18:13:21 -0500

I can repro it quite often. What good will that do if you cannot repro it?

[David Holcomb [MS]]

Doug, I noticed in your earlier post (dtd Tuesday, January 21, 2003 6:03 AM)
that you said this was happening against a .NET RC2 target using
/debugport=com1 /baudrate=115200.

Do you have any other targets available? If so, it will be interesting to
know if the problem is only happening against the .NET RC2 target. If that
is the case, then an earlier version of the debugger may work better against
that specific target build.

BTW, there is a similar issue (can't break in) against some (?) NT4 targets.
We think the solution there is to use kd instead of windbg. You can launch
kd -server npipe:pipe=mypipe and then still connect to kd with
windbg -remote npipe:pipe=mypipe,server=myserver.

--Dave

--
--

This posting is provided "AS IS" with no warranties, and confers no rights.

""Wendy Shi [MS]"" <wend...@online.microsoft.com> wrote in message

news:eVY0IZI2CHA.1656@cpmsftngxa08...

[Doug]

On the target I use mainly XP Pro & XP Pro SP1 to which the same problem
occurs. Host is XP SP1.

Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message
news:3e5469a6$1...@news.microsoft.com...

[Doug]

Where can I find a 5.x.x.x version of WinDbg? Is it on the original XP DDK?

Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message
news:3e5469a6$1...@news.microsoft.com...

[Doug]

I will let you know...

""Wendy Shi [MS]"" <wend...@online.microsoft.com> wrote in message

news:eVY0IZI2CHA.1656@cpmsftngxa08...

[David Holcomb [MS]]

Did you ever check to verify that the target machine does not have Legacy
USB support enabled in the BIOS?

The previous version of the debugger was 4.x, there was never a 5.x (except
for what shipped in Windows itself, but you don't want to use that
debugger). The XP DDK shipped the 3.0.20.0 debugger which is really really
old. The XPSP1 DDK shipped with the 6.0.17.0 debugger.

Are you thinking to try some older debuggers to see if any of them work for
the same target?

You should be able to append any of the following (e.g.
dbg_x86_6.1.17.2.exe) to this url:

http://msdl.microsoft.com/download/symbols/debuggers/

02/12/2001 dbg_ia64_2023.exe
02/13/2001 dbg_x86_2023.exe
05/7/2001 dbg_ia64_3.0.10.0.exe
05/7/2001 dbg_x86_3.0.10.0.exe
07/19/2001 dbg_ia64_3.0.20.0.exe
07/19/2001 dbg_x86_3.0.20.0.exe
10/13/2001 dbg_ia64_4.0.11.0.exe
10/13/2001 dbg_x86_4.0.11.0.exe
12/17/2001 dbg_ia64_4.0.18.0.exe
12/17/2001 dbg_x86_4.0.18.0.exe
12/20/2001 dbg_ia64_4.0.18.0r.exe
12/20/2001 dbg_x86_4.0.18.0r.exe
03/7/2002 dbg_ia64_6.0.7.0.exe
03/7/2002 dbg_x86_6.0.7.0.exe
06/7/2002 dbg_ia64_6.0.17.0.exe
06/7/2002 dbg_x86_6.0.17.0.exe
10/7/2002 dbg_ia64_6.1.9.0.exe
10/7/2002 dbg_x86_6.1.9.0.exe
10/17/2002 dbg_amd64_6.1.11.1.exe
12/16/2002 dbg_amd64_6.1.17.0.exe
12/16/2002 dbg_ia64_6.1.17.0.exe
12/16/2002 dbg_x86_6.1.17.0.exe
02/3/2003 dbg_ia64_6.1.17.2.exe
02/3/2003 dbg_x86_6.1.17.2.exe

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:egaqcaO2CHA.2668@TK2MSFTNGP12...

[Doug]

What does Legacy USB support do to WinDbg?

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e55a5a1$1...@news.microsoft.com...

[David Holcomb [MS]]

I don't think it does anything to windbg. I think it just affects the
serial communication stability of the target machine. I'll double-check.

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:#orfVYb2...@TK2MSFTNGP09.phx.gbl...

[David Holcomb [MS]]

Okay, here is the official word...

"It's purely a software issue. Legacy USB support is implemented using a
periodic system-management interrupt. The SMI occurs frequently enough that
it interferes with the target machine's ability to read bytes from the
serial port without data loss."

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e563d8a$1...@news.microsoft.com...

[Doug]

I will check it out. Why did this not affect earlier versions?

Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e566de1$2...@news.microsoft.com...

[Doug]

1) Host = WinXP SP1 on dual 400Mhz IBM Intellistation. No command line
options.
2) Target = WinXP on 700Mhz Pentium III. Boot.ini = COM1,115200

Version Info=

Windows XP Kernel Version 2600 UP Free x86 compatible
Built by: 2600.xpclient.010817-1148
Kernel base = 0x804d0000 PsLoadedModuleList = 0x80545b28
System Uptime: not available
Remote KD: COM:Port=\\.\com1,Baud=115200,Timeout=4000
command line: '"D:\WinDbg\windbg.exe" ' Debugger Process 0xD38
dbgeng: image 6.1.0017.2, built Wed Jan 29 17:10:44 2003
[path: D:\WinDbg\dbgeng.dll]
dbghelp: image 6.1.0017.2, built Wed Jan 29 17:10:30 2003
[path: D:\WinDbg\dbghelp.dll]
DIA version: 2254
Extension DLL search Path:
D:\WinDbg\winext;D:\WinDbg\WINXP;D:\WinDbg\pri;D:\WinDbg;D:\Perl\bin\;C:\WIN
DOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;D:\DDK\bin;c:\vslick\win;d
:\vc98\bin
Extension DLL chain:
exts: image 6.1.0017.0, API 1.0.0, built Fri Dec 13 01:46:07 2002
[path: D:\WinDbg\WINXP\exts.dll]
kext: image 6.1.0017.0, API 1.0.0, built Fri Dec 13 01:46:08 2002
[path: D:\WinDbg\winext\kext.dll]
kdexts: image 5.2.3692.0, API 1.0.0, built Tue Dec 03 13:57:57 2002
[path: D:\WinDbg\WINXP\kdexts.dll]
Driver verifier enabled, level 0xbb
dbghelp: image 6.1.0017.2, API 6.0.6, built Wed Jan 29 17:10:30 2003
[path: D:\WinDbg\dbghelp.dll]
ext: image 6.1.0017.0, API 1.0.0, built Fri Dec 13 01:46:07 2002
[path: D:\WinDbg\winext\ext.dll]

3) I usually have active debugger going and target shutsdown. While shutdown
I close/open the debug output file. I just wait for the target to come up.
Here is the output:

Waiting to reconnect...
Closing open log file d:\debug1.out
Opened log file 'd:\debug1.out'
Connected to Windows XP 2600 x86 compatible target, ptr64 FALSE
Kernel Debugger connection established.
Symbol search path is:
d:\symbols\xp;d:\drivers\bmiwdm\libxp\i386\chk;d:\drivers\netwdm\libxp\i386\
chk;d:\drivers\smwdm\libxp\i386\chk;d:\drivers\vdwdm\libxp\i386\chk
Executable search path is:
Loading symbols for 804d0000 ntoskrnl.exe -> ntoskrnl.exe
ModLoad: 804d0000 806b3f00 ntoskrnl.exe

*****>>>> Frozen here <<<***** Ctrl-Break followed by Alt-Ctrl-R causes it
to restore.

Windows XP Kernel Version 2600 UP Free x86 compatible
Built by: 2600.xpclient.010817-1148
Kernel base = 0x804d0000 PsLoadedModuleList = 0x80545b28
System Uptime: not available
Loaded exts extension DLL
Loaded kext extension DLL
Loaded kdexts extension DLL
Loaded dbghelp extension DLL
Loaded ext extension DLL
Force unload of ntoskrnl.exe
ModLoad: 804d0000 806b3f00 ntoskrnl.exe
ModLoad: 806b4000 806cd080 hal.dll

....

4) It happens frequently enough, but the Ctrl-Break/Ctrl-Alt-R causes things
to come back.

""Wendy Shi [MS]"" <wend...@online.microsoft.com> wrote in message

news:eVY0IZI2CHA.1656@cpmsftngxa08...

[David Holcomb [MS]]

I have no idea. What earlier versions of the debugger or OS are you
referring to?

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:OWm25dd2...@TK2MSFTNGP10.phx.gbl...

[Doug]

I guess the 4.x versions...

Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e56928f$1...@news.microsoft.com...

[Doug]

So far the disabling of USB Mouse and USB keyboard emulation has stopped the
hangs. I stil contend that earlier versions recovered from the
miscommunication problems much more gracefully than the 6.x versions.

Thanks for the help.

"Doug" <dha...@junk.com> wrote in message

news:OWm25dd2...@TK2MSFTNGP10.phx.gbl...

[David Holcomb [MS]]

To be clear...are you saying you had W2K installed earlier, you have since
upgraded, and you don't recall having to disable Legacy USB support with the
4.x debuggers when you were running against a W2K target?

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:ughWs$g2CH...@TK2MSFTNGP12.phx.gbl...

[Doug]

The communication failure is always a result of the target machine, not the
host machine. I have two targets where one always hangs (before the USB
disabling) and one rarely if ever hangs (Dell 530MT). I am not sure if the
hanging is more prevelant if the target is Win2K or WinXP. I have not looked
at it that far. Are you saying the host OS may cause a difference also?

Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e5717b1$1...@news.microsoft.com...

[Doug]

BTW, I cannot access these old debuggers. I get a 'not authorized' error.

Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e55a5a1$1...@news.microsoft.com...

[David Holcomb [MS]]

I tried a couple of older packages and had no problems. Perhaps one or two
of them are not downloadable.

If you let me know specifically which one you had a problem with I'll try
it.

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:emX2T1p2...@TK2MSFTNGP11.phx.gbl...

[David Holcomb [MS]]

I'm not suggesting anything. I don't really think the host OS has anything
to do with the problem you were having getting the target to break in.

Based on what I thought I remembered from your earlier postings, I seemed to
recall you had upgraded your target from W2K to XP before you noticed the
problems, and I thought you were implying you were running the 4.x debuggers
against the W2K target and had not noticed such problems before. I'm only
going from memory though and trying to confirm whether there is a pattern of
behavior you have witnessed.

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:eVY9g0p2...@TK2MSFTNGP10.phx.gbl...

[Doug]

You can get it because you are in the MS domain. I am not. I was trying to
get the latest 4.0.18 version.

Thanks.
Doug

"David Holcomb [MS]" <dhol...@microsoft.com> wrote in message

news:3e58199b$1...@news.microsoft.com...

[David Holcomb [MS]]

Doug and I got this resolved. To be 100% clear, you have to append the URL
below with the filename for the download to make a complete URL.

For example: http://msdl.microsoft.com/download/symbols/debuggers/ +
dbg_x86_6.1.17.2.exe
=
http://msdl.microsoft.com/download/symbols/debuggers/dbg_x86_6.1.17.2.exe

--
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Doug" <dha...@junk.com> wrote in message

news:eJpUI4A3...@TK2MSFTNGP10.phx.gbl...