RE: [E] Re: x3270

[mark . paulus]

-Original Message-
> On 4/15/2018 2:48 AM, Chris Johnson wrote:
> > Hi.
> > 
> > Not sure when this happened but I now have a complaint for cygwin on 
> > exit about  x3270.sh returning exit code 1.  I found it listed in 
> > /etc/setup/install.db and remove it.  No joy.  Do I recall the x3270 
> >is defunct?  Either way, how do I get rid of this error.  It's annoying.
> >
> >Thanks.
> >
> 
> x3270 is still an active package
> 
> run /etc/postinstall/x3270.sh from bash and if you have no error rename it as 
> /etc/postinstall/x3270.sh.done
> 
> Regards
> Marco

Manually ran x3270.sh and ran into a couple of issues that looks like they 
could use addressing:
First time I ran it, got this:
$ ./x3270.sh
./x3270.sh: line 9: /usr/bin/mkfontdir: No such file or directory

Installed mkfontdir package (need a dependency maybe?), then ran it again
and got this:
$ bash -x x3270.sh
+ FONTDIR=/usr/share/fonts
+ ENCODINGSDIR=/usr/share/fonts/encodings
+ MKFONTDIR=/usr/bin/mkfontdir
+ /usr/bin/mkfontdir -e /usr/share/fonts/encodings /usr/share/fonts/misc
opendir: No such file or directory

I "think" the issue here is that the encoding directory we want is actually
in /usr/share/X11/fonts/encoding instead of /usr/share/fonts/encoding.
Solved this by creating a soft link:
/usr/share/fonts
$ ln -sf /usr/share/x11/fonts/encodings .
/usr/share/fonts
$ ll
total 93
drwxr-xr-x+ 1 v808297 Domain Users  0 Apr 16 13:23 ./
drwxr-xr-x+ 1 v808297 Domain Users  0 Apr 16 11:44 ../
drwxr-xr-x+ 1 v808297 Domain Users  0 Feb 26 08:50 dejavu/
lrwxrwxrwx  1 v808297 Domain Users 30 Apr 16 13:23 encodings -> 
/usr/share/x11/fonts/encodings/
drwxr-xr-x+ 1 v808297 Domain Users  0 Feb 26 08:57 microsoft/
drwxr-xr-x+ 1 v808297 Domain Users  0 Feb 26 08:53 misc/

After these 2 corrections the x3270.sh script ran successfully and 
I was able to rename it to x3270.sh.done

Thanks,

Mark

Sudden, unexplained failure of ssh

[mark . paulus]

Hello,

I've been running 2.9.x under windows 7 on my 16GB laptop for a long time 
(probably the 
last year), when late last week my ssh outbound quit working.  The failure is 
indicated with 
this error message:

ssh: Could not resolve hostname csu6220509.vzbi.com: Non-recoverable failure in 
name resolution

I checked the forums, and what I saw seemed to relate to connecting to VM boxes.
In my case, this is not the cause of the failure.
This machine is my company provided (and software managed) laptop, so I don't 
have a 
lot of control over (in fact, I don't have hardly any control) what gets 
"pushed" to it.
However, the one thing that I noticed that seemed to coincide with the failure 
is that
the company pushed an "Elastica Reach Agent 1.0.20930.0" to my machine.

I thought the issue might have been a version problem, so I updated to the 
latest 
Cygwin version
$ uname -a
CYGWIN_NT-6.1 TCO10106LKVTAHB 2.10.0(0.325/5/3) 2018-02-02 15:16 x86_64 Cygwin

That didn't fix the problem.  I've also tried some other utilities,  which are 
working, 
such as mobaxterm(Cygwin based), babun(Cygwin based)  and putty.  All those 
seem to work
fine.

Once I determined that it wasn't my actual network stack, I then tried running 
ssh under
strace, and now it works.  So, I can use ssh if I use it with strace, but 
natively by itself
it continues to get the failure in name resolution.

Any clues, hints or pointers on how to go about either determining what the 
problem is, or 
how to clean it up?

(I am also seeing the same issue with wget, and strace clears that up also
$ wget http://linux1.vzbi.com
--2018-02-22 09:55:38--  http://linux1.vzbi.com/
Resolving linux1.vzbi.com (linux1.vzbi.com)... failed: Non-recoverable failure 
in name resolution.
wget: unable to resolve host address 'linux1.vzbi.com'

Thanks,

Mark

--
Problem reports:   http://cygwin.com/problems.html
FAQ:   http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple

Re: Bug: Setup tool doesn't respect managed mountpoints/filesystems

[Mark Paulus]

On Mon, 19 Sep 2005 15:48:01 -0400 (EDT), Igor Pechtchanski wrote:

>On Mon, 19 Sep 2005, Brian Dessent wrote:

>> Igor Pechtchanski wrote:
>>
>> > Yes, there are issues to work out here, but they are not very different
>> > from, say, those arising when two programs try creating a file with the
>> > same name simultaneously...  In fact, the case-insensitivity of both NTFS
>> > and VFAT plays into our hands here, as you say below: creating a new file
>> > that differs only in case will always fail, so Cygwin simply has to encode
>> > the file name if its creation failed.
>>
>> It's still extra system calls that would slow things down.  And what if
>> you tried to open("readme") when README exists?  It would succeed (on
>> anything but NT with NTFS with Cygwin using the Native API), but on a
>> managed mount it's supposed to fail.

>Which was what I said in the part you snipped.

>> And what if "README" exists, and you want to create "readme"?  Okay, so
>> you have to encode "readme" as "%52%65%41%44%4d%65".  Now you delete
>> "README", leaving only "%52%65%41%44%4d%65".  A program tries to
>> open("readme") -- this fails when it should succeed.  Cygwin would have
>> to know to sometimes try "readme" and sometimes try "%52%65%41%44%4d%65"
>> when the program tries to open "readme", how would it know this without
>> looking at every file in the directory?

>Of course, it goes without saying that the encoded form should always work
>(yes, there may be clashes like "readme" created by Windows and
>"%52%45%41%44%4d%45" created by Cygwin, but those are unavoidable).  As
>for knowing which file to open, first look for "%52%45%41%44%4d%45", then
>for "readme" (case-sensitively).  No other possibilities here...

>> Repeat for "readme" vs. "Readme" and every other of billions of
>> permutation of case.  I see no way this wouldn't turn into a nightmare.

>Huh?  Why would you need to try *all* the permutations?

>> Encoding every filename seems like a win-win situation to me.  The only
>> overhead is the string processing aspect of translating "a" into "%41".

>Which may or may not be considerable (and no, I don't have measurements to
>back it up).  There is another overhead, and that is of a human trying to
>look at the directory with Windows tools.  Yes, encoding is unavoidable,
>but it shouldn't be obnoxious.

>> There are no added filesystem calls because every filename is turned
>> into a canonical unique filename.  If you let "readme" sometimes be
>> "readme" and sometimes be "%52%65%41%44%4d%65", all hell breaks loose.

>Yes, it doubles the number of calls needed to open the file.  So?  Nobody
>said managed mounts would have no overhead.

>OTOH, the issue of "readme" and "%52%45%41%44%4d%45" being the same is
>valid, and needs to be handled somehow.
>   Igor

Has anyone explored the option of an IFS that would implement the cygwin
managed filesystem semantics, but it would be transparent to all
cygwin apps.  From the brief investigation I have done, it appears there would
need to be 2 flavors, 9x/ME & NT/2K/XP, but that would relieve a big burden
from the cygwin1.dll, I would estimate.




>-- 
>   http://cs.nyu.edu/~pechtcha/
>  |\  _,,,---,,_   [EMAIL PROTECTED]
>ZZZzz /,`.-'`'-.  ;-;;,_   [EMAIL PROTECTED]
> |,4-  ) )-,_. ,\ (  `'-'  Igor Pechtchanski, Ph.D.
>'---''(_/--'  `-'\_) fLa.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

>If there's any real truth it's that the entire multidimensional infinity
>of the Universe is almost certainly being run by a bunch of maniacs. /DA

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Bug: Setup tool doesn't respect managed mountpoints/filesystems

[Mark Paulus]

My company just replaced my desktop with a laptop, and in the 
migration process, I am getting my cygwin install set back up.  
Two changes I had made before was to mount /usr/src and /usr/share
as managed filesystems (I have a directory c:\cygmanaged where
share and src live).  While doing some clean up work in there from
a previous lifetime, I noticed that some new things are being put in
the tree, but the filenames are not being "managed".  So,
Instead of having a directory
c:\cygmanaged\share\doc\%43ygwin,
the installer is creating a directory 
c:\cygmanaged\share\doc\Cygwin,
and then I have 
c:\cygmanaged\share\doc\Cygwin\stow-1.3.3.README,
instead of 
c:\cygmanaged\share\doc\%43ygwin\stow-1.3.3.%52%45%41%44%4D%45*
Which also messes up the link in
c:\cygmanaged\share\doc\stow-1.3.3\README.Cygwin, which isn't
a managed file name.

So, it appears that the installer is respecting the mount point, but
it isn't respecting the managed attribute of the filesystem/mountpoint.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: different DLL....

[Mark Paulus]

On Fri, 29 Jul 2005 16:18:37 + (UTC), Jason FU wrote:

>Dave Korn  artimi.com> writes:  
>  
>>   
>> Original Message  
>> >From: Jason FU  
>> >Sent: 29 July 2005 16:15  
>>   
>> > I just checked by dll's by "md5sum /bin/*.dll". I really wonder why the  
>> > MD5's of all files are not the same since they're of the same "ages".  
>>   
>>   MD5 is the hash of a file, not the timestamp.  RTFM!  
>>   
>> cheers,  
>>   DaveK  
>  
>  
>  
> 
>Sorry that I did not make it clear. The two set of MD5's were from two  
>different machines and when I transferred one set of these files to the other  
>one for diff'ing, diff said they were different and they are of the packages  
>of the same version.  
>  
>Please advise. Thanks.  
>  
>  
>Jason  
> 


First thing to look at is, how are your mounts defined.
md5sum will return different values depending upon
whether you are using a binary mount vs a text mount.


>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

undefined reference to `_libintl_gettext' with g++/c++

[Mark Paulus]

Hello,

I have several binaries in a project, and most of the binaries
compile.  However, one binary fails with a whole series of the
above error:

g++  -L../lib -o dselect  basecmds.o  baselist.o  basetop.o  bindings.o  curkeys
.o  main.o  methkeys.o  methlist.o  method.o  methparse.o  pkgcmds.o  pkgdepcon.
o  pkgdisplay.o  pkginfo.o  pkgkeys.o  pkglist.o  pkgsublist.o  pkgtop.o  helpms
gs.o  -lintl -liconv  -lintl -ldpkg -lncurses -L../optlib -lopt
../lib/libdpkg.a(ehandle.o): In function `push_error_handler':
/usr/src/debian-cygwin/dpkg.lappy/lib/ehandle.c:80: undefined reference to `_lib
intl_gettext'
../lib/libdpkg.a(ehandle.o): In function `print_error_cleanup':
/usr/src/debian-cygwin/dpkg.lappy/lib/ehandle.c:96: undefined reference to `_lib
intl_gettext'
../lib/libdpkg.a(ehandle.o): In function `run_cleanups':
/usr/src/debian-cygwin/dpkg.lappy/lib/ehandle.c:111: undefined reference to `_li
bintl_gettext'
../lib/libdpkg.a(ehandle.o): In function `push_cleanup':
/usr/src/debian-cygwin/dpkg.lappy/lib/ehandle.c:196: undefined reference to `_li
bintl_gettext'
../lib/libdpkg.a(ehandle.o): In function `badusage':
/usr/src/debian-cygwin/dpkg.lappy/lib/ehandle.c:265: undefined reference to `_li
bintl_gettext'
../lib/libdpkg.a(ehandle.o):/usr/src/debian-cygwin/dpkg.lappy/lib/ehandle.c:271:
 more undefined references to `_libintl_gettext' follow
Info: resolving _curscr by linking to __imp__curscr (auto-import)
Info: resolving _stdscr by linking to __imp__stdscr (auto-import)
Info: resolving _COLOR_PAIRS by linking to __imp__COLOR_PAIRS (auto-import)
collect2: ld returned 1 exit status
make: *** [dselect] Error 1

The one thing I have been able to isolate is, things that are linked using
cc/gcc link fine, but those that try to link using g++/c++ fail.

[EMAIL PROTECTED] /usr/src/debian-cygwin/dpkg.lappy/build/dselect
$ g++ -print-search-dirs
install: /usr/lib/gcc/i686-pc-cygwin/3.4.4/
programs: =/usr/lib/gcc/i686-pc-cygwin/3.4.4/:/usr/lib/gcc/i686-pc-cygwin/3.4.4/
:/usr/lib/gcc/i686-pc-cygwin/:/usr/lib/gcc/i686-pc-cygwin/3.4.4/:/usr/lib/gcc/i6
86-pc-cygwin/:/usr/libexec/gcc/i686-pc-cygwin/3.4.4/:/usr/libexec/gcc/i686-pc-cy
gwin/:/usr/lib/gcc/i686-pc-cygwin/3.4.4/:/usr/lib/gcc/i686-pc-cygwin/:/usr/lib/g
cc/i686-pc-cygwin/3.4.4/../../../../i686-pc-cygwin/bin/i686-pc-cygwin/3.4.4/:/us
r/lib/gcc/i686-pc-cygwin/3.4.4/../../../../i686-pc-cygwin/bin/
libraries: =/usr/lib/gcc/i686-pc-cygwin/3.4.4/:/usr/lib/gcc/i686-pc-cygwin/3.4.4
/:/usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../../i686-pc-cygwin/lib/i686-pc-cygwi
n/3.4.4/:/usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../../i686-pc-cygwin/lib/:/usr/
lib/gcc/i686-pc-cygwin/3.4.4/../../../i686-pc-cygwin/3.4.4/:/usr/lib/gcc/i686-pc
-cygwin/3.4.4/../../../:/lib/i686-pc-cygwin/3.4.4/:/lib/:/usr/lib/i686-pc-cygwin
/3.4.4/:/usr/lib/

[EMAIL PROTECTED] /usr/src/debian-cygwin/dpkg.lappy/build/dselect
$ g++ -print-file-name=libintl.a
/usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../libintl.a

[EMAIL PROTECTED] /usr/src/debian-cygwin/dpkg.lappy/build/dselect
$ c++ -print-file-name=libintl.a
/usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../libintl.a

[EMAIL PROTECTED] /usr/src/debian-cygwin/dpkg.lappy/build/dselect
$ c++ --version
c++ (GCC) 3.4.4 (cygming special) (gdc 0.12, using dmd 0.125)
Copyright (C) 2004 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.


Cygwin Configuration Diagnostics
Current System Time: Wed Jul 20 14:18:34 2005

Windows XP Professional Ver 5.1 Build 2600 Service Pack 2

Path:   c:\cygwin\home\Mark.Paulus\bin
c:\cygwin\usr\local\bin
c:\cygwin\bin
c:\cygwin\bin
c:\cygwin\usr\sbin
c:\cygwin\usr\X11R6\bin
c:\WINNT\system32
c:\WINNT
c:\WINNT\System32\Wbem
c:\Program Files\Common Files\Adaptec Shared\System
c:\Progra~1\oracle\ora817\bin
c:\WINNTsystem32
%%43
c:\cygwin\WINNT%
%%43
c:\cygwin\WINNT%\System32\Wbem
%%43
c:\cygwin\WINNT%\system32
c:\util
c:\cygwin
c:\SQLLIB\BIN
c:\SQLLIB\FUNCTION
c:\SQLLIB\SAMPLES\REPL
c:\SQLLIB\HELP
c:\IBM\IMNNQ
.\
.\
c:\cygwin\ddrive\util
c:\cygwin\usr\x11r6\bin
c:\cygwin\usr\games

Output from c:\cygwin\bin\id.exe (nontsec)
UID: 1014(Mark.Paulus)   GID: 513(None)
0(root)  545(Users)
545(Users)   10513(Domain Users)
21826(NORTH-VA344-TPPS)  14677(West-CO117-CCM Exempt)
15038(West-CO117-Users)

Output from c:\cygwin\bin\id.exe (ntsec)
UID: 1014(Mark.Paulus)   GID: 513(None)
0(root)  545(Users)
545(Users)   10513(Domain Users)
21826(NORTH-VA344-TPPS)  14677(West-CO117-CCM Exempt)
15038(West-CO117-Users)

SysDir: C:\WINNT\system32
WinDir: C:\WINNT

USER = `Mark.Paulus'
PWD = `/usr/src/debian-cygwin/dpkg.lappy/b

Re: Problem with "set," "if," "foreach" commands

[Mark Paulus]

I would suggest you get the O'Reilly bash
book, and read your way through that.  What you
are running into are syntax problems.

On Thu, 30 Jun 2005 17:47:41 -0700 (PDT), Han Nguyen wrote:

>New to Unix, so please have patience.

>I understand that I should be able to type
>set food = pickle
>echo $food

bash doesn't use set.  And, when doing variable assignments in
the shell, no spaces are allowed:
food=pickle
echo $food




>But nothing echoes.  It just leaves a blank line, then gives another command 
>prompt.  Am I doing
>anything wrong?

>"If" also doesn't work.  This is what happens:
>$ if (5 > 0)
>> echo "Hello"
>> endif
>> 
>> 
>> 

>And it merely continues that way until I exit the shell.

if  doesn't use parens, it uses square brackets.  And, numeric tests are not
done with greater than/less than/equal symbols, but with -lt, -le, -gt, -ge, -eq
tests.  Spacing is important in the shell also.  And, you have some other syntax
issues:

if [ 5 -gt 0 ] ; then
echo Hello
fi




>Also a problem with "foreach."  Output:

>$ foreach run (1 2 3)
>bash: syntax error near unexpected token `('

>Typing "help set" and "help foreach" gives me:

>$ help foreach
>bash: help: no help topics match `foreach'.  Try `help help' or `man -k 
>foreach' or `info
>foreach'.

In Bash, it's not foreach, but rather 
for var in varlist
do
command $var ...
done



>Anything I can do to resolve this?  I appreciate the help.  Thanks.



I'm not super familiar with csh, but it almost looks to me like you are
using csh commands in a bash shell.  sh/ksh/bash/ash all sort of have
a similar look/feel.  csh is slightly different in it's approach to scripting.
If you want to use the above type syntax, then maybe you could install
the tcsh package, and get the O'Reilly csh/tcsh book.


>   
> 
>Yahoo! Sports 
>Rekindle the Rivalries. Sign up for Fantasy Football 
>http://football.fantasysports.yahoo.com

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Trouble finding the package the command "watch" is in

[Mark Paulus]

Generally you can find a certain file by going to the cygwin
website, and then selecting Setup Package Search under
the Software heading from the main menu on the left hand side.

In this particular case, watch is connected to many packages,
so from experience, I'll tell you that it is included with the 
procps package.

On Mon, 27 Jun 2005 11:10:47 -0400, John Petritis wrote:

>I am looking for the command named "watch"

>I am trying to find the package it is located in.
>I haven't had any luck finding it through various search strategies.
>My trouble is probably due to it being such a common word.
>So I have 2 questions?

>1) Which package do I need to install to get the "watch" command
>2) Is there some sort of way I can list what is in packages to find
>which package it is in?

>(I really don't want to have to download the source and grep through it)

>Thank You,

>John Petritis

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: can not compile cscope on cygwin

[Mark Paulus]

Hmmm.  Did you rerun configure after you installed
bison and flex?  I think you need to reconfigure to let
make know that those tools are now available, and which
flavor is there.


On Sat, 14 May 2005 14:13:06 + (UTC), Arvind wrote:

>Mark Paulus  mci.com> writes:

>> 
>> I just tried this, and it appears that you don't have 
>> lex/flex and/or bison/byacc installed.  Apparently
>> these aren't selected by default from the cygwin
>> setup tool.  Re-run the cygwin setup tool, and under 
>> Devel, select bison and flex, and you should be good
>> to go.
>> 

>tried that.. but got :-
>-
>make  all-recursive
>make[1]: Entering directory `/cygdrive/c/cscope/cscope-15.5'
>Making all in doc
>make[2]: Entering directory `/cygdrive/c/cscope/cscope-15.5/doc'
>make[2]: Nothing to be done for `all'.
>make[2]: Leaving directory `/cygdrive/c/cscope/cscope-15.5/doc'
>Making all in src
>make[2]: Entering directory `/cygdrive/c/cscope/cscope-15.5/src'
>bison -y   `test -f 'egrep.y' || echo './'`egrep.y
>conflicts: 2 shift/reduce
>make[2]: *** [egrep.c] Broken pipe
>make[2]: Leaving directory `/cygdrive/c/cscope/cscope-15.5/src'
>make[1]: *** [all-recursive] Error 1
>make[1]: Leaving directory `/cygdrive/c/cscope/cscope-15.5'
>make: *** [all] Error 2
>-

>Arvind.

>> On Wed, 27 Apr 2005 15:46:47 -0400 (EDT), Igor Pechtchanski wrote:
>> 
>> >On Wed, 27 Apr 2005, lin q wrote:
>> 
>> >> Hi,
>> >>  I am using latest version cygwin and I just downloaded cscope 15.5, but 
>> >> in
>> >> compiling it there is such error,
>> >>
>> >> $ make
>> >> make  all-recursive
>> >> make[1]: Entering directory `/c/bin/cscope-15.5'
>> >> Making all in doc
>> >> make[2]: Entering directory `/c/bin/cscope-15.5/doc'
>> >> make[2]: Nothing to be done for `all'.
>> >> make[2]: Leaving directory `/c/bin/cscope-15.5/doc'
>> >> Making all in src
>> >> make[2]: Entering directory `/c/bin/cscope-15.5/src'
>> >> /bin/bash ../ylwrap `test -f 'fscanner.l' || echo './'`fscanner.l .c
>> >> fscanner.c -- :
>> >> make[2]: *** [fscanner.c] Error 1
>> >> make[2]: Leaving directory `/c/bin/cscope-15.5/src'
>> >> make[1]: *** [all-recursive] Error 1
>> >> make[1]: Leaving directory `/c/bin/cscope-15.5'
>> >> make: *** [all] Error 2
>> >>
>> >> Do you know what is wrong?
>> 
>> >Please review and follow
>> >> Problem reports:   http://cygwin.com/problems.html
>> >to provide enough information about your installation for this list to
>> >help you.
>> 
>> >Without the above information, I would guess that it's a line ending issue
>> >of some sort (since you're using /c, and I suppose your cygdrive prefix is
>> >empty rather than "c:" being mounted on /c explicitly).
>> 
>> >> Is there any pre-compiled cscope for cygwin somewhere?
>> 
>> >According to <http://cygwin.com/packages/>, the only matches for "cscope"
>> >are in the vim source package, so no, there isn't an official cscope
>> >package.  You can Google for unofficial ones, I suppose.
>> >Igor
>> >-- 
>> >http://cs.nyu.edu/~pechtcha/
>> >  |\  _,,,---,,_pechtcha  cs.nyu.edu
>> >ZZZzz /,`.-'`'-.  ;-;;,_igor  watson.ibm.com
>> > |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
>> >'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!
>> 
>> >"The Sun will pass between the Earth and the Moon tonight for a total
>> >Lunar eclipse..." -- WCBS Radio Newsbrief, Oct 27 2004, 12:01 pm EDT
>> 
>> >--
>> >Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>> >Problem reports:   http://cygwin.com/problems.html
>> >Documentation: http://cygwin.com/docs.html
>> >FAQ:   http://cygwin.com/faq/
>> 
>> 





>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Can't build coreutils 5.3.0-5 - RESOLVED!!

[Mark Paulus]

Turns out that sometime ago, I had built and installed a
cygwin1.dll into /usr/local.  During the process of that 
build/install, it also installed an older types.h into
/usr/local/include/cygwin/types.h.  
Removing those fixed the problem.  Actually had to look at
.deps/fsupdate.Po to figure out where it was pulling types.h
in.

On Thu, 12 May 2005 14:59:12 +, Eric Blake wrote:

>> In file included from /usr/src/coreutils-5.3.0/lib/fsusage.c:76:
>> /usr/include/sys/statvfs.h:22: error: parse error before "fsblkcnt_t"
>> /usr/include/sys/statvfs.h:24: error: parse error before "f_bavail"
>> /usr/include/sys/statvfs.h:25: error: parse error before "f_files"
>> /usr/include/sys/statvfs.h:26: error: parse error before "f_ffree"
>> /usr/include/sys/statvfs.h:27: error: parse error before "f_favail"
>> /usr/include/sys/statvfs.h:31: error: parse error before '}' token
>> make[3]: *** [fsusage.o] Error 1
>> make[3]: Leaving directory `/usr/src/coreutils-5.3.0/.build/lib'
>> make[2]: *** [all] Error 2
>> make[2]: Leaving directory `/usr/src/coreutils-5.3.0/.build/lib'
>> make[1]: *** [all-recursive] Error 1
>> make[1]: Leaving directory `/usr/src/coreutils-5.3.0/.build'
>> make: *** [all] Error 2

>Weird - I can't reproduce that error.  Your cygcheck output shows that you 
>have cygwin-1.5.16-1, which includes the latest version of 
>/usr/include/sys/statvfs.h.  
If that is corrupt, try reinstalling cygwin.

>$ md5sum /usr/include/sys/statvfs.h
>db1fe2c8fb67478f6dc67cc1d3b642ad */usr/include/sys/statvfs.h

>--
>Eric Blake






--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Can't build coreutils 5.3.0-5

[Mark Paulus]

I am trying to build coreutils-5.3.0-5.

To do so, I requested the coreutils source via
cygwin's setup.  Then I run 
coreutils-5.3.0-5.sh prep
coreutils-5.3.0-5.sh configure
coreutils-5.3.0-5.sh build.

During the build, I get the following failure:

make  all-recursive
make[1]: Entering directory `/usr/src/coreutils-5.3.0/.build'
Making all in lib
make[2]: Entering directory `/usr/src/coreutils-5.3.0/.build/lib'
make  all-am
make[3]: Entering directory `/usr/src/coreutils-5.3.0/.build/lib'
if gcc -DHAVE_CONFIG_H -DLIBDIR=\"/usr/lib\" -I. -I/usr/src/coreutils-5.3.0/lib
-I..  -I.. -I/usr/src/coreutils-5.3.0/lib   -O2 -MT fsusage.o -MD -MP -MF ".deps
/fsusage.Tpo" -c -o fsusage.o /usr/src/coreutils-5.3.0/lib/fsusage.c; \
then mv -f ".deps/fsusage.Tpo" ".deps/fsusage.Po"; else rm -f ".deps/fsusage.Tpo
"; exit 1; fi
In file included from /usr/src/coreutils-5.3.0/lib/fsusage.c:76:
/usr/include/sys/statvfs.h:22: error: parse error before "fsblkcnt_t"
/usr/include/sys/statvfs.h:24: error: parse error before "f_bavail"
/usr/include/sys/statvfs.h:25: error: parse error before "f_files"
/usr/include/sys/statvfs.h:26: error: parse error before "f_ffree"
/usr/include/sys/statvfs.h:27: error: parse error before "f_favail"
/usr/include/sys/statvfs.h:31: error: parse error before '}' token
make[3]: *** [fsusage.o] Error 1
make[3]: Leaving directory `/usr/src/coreutils-5.3.0/.build/lib'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/usr/src/coreutils-5.3.0/.build/lib'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/src/coreutils-5.3.0/.build'
make: *** [all] Error 2

Do I have an out-of-date statvfs.h file, and if so, where/how do I 
go about getting a newer version?  Or, do I need to give some 
special argument value to the configure command?





cygcheck.out
Description: Binary data
--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Problem with ls and EACCESS

[Mark Paulus]

I have been trying out the ext2 read/write IFS driver
for WinXP, and it seems to work nicely.  However, I 
have noticed this one issue.

In /, I have a couple of symbolic links to kernels.  When
I look at them in a windows command prompt, I see the link as a 
file:  
08/12/2004  05:45 AM20 vmlinuz.knoppix
and if I try to look inside, 
L:\>type vmlinuz.knoppix
Access is denied.

So far, all is good.

However, when I try to look at the directory under cygwin:
$ ls /cygdrive/l/vmlinuz.knoppix
ls: /cygdrive/l/vmlinuz.knoppix: No such file or directory

I have pulled an strace of this, and it looks like someone
is skipping the error EACCESS or ignoring it, and somehow
is transmuting the EACCESS into ENOENT.

   37  116938 [main] ls 2268 normalize_posix_path: src 
/cygdrive/l/vmlinuz.knoppix
   37  116975 [main] ls 2268 normalize_posix_path: /cygdrive/l/vmlinuz.knoppix 
=  normalize_posix_path (/cygdrive/l/vmlinuz.knoppix)
   37  117012 [main] ls 2268 mount_info::conv_to_win32_path: 
conv_to_win32_path(/cygdrive/l/vmlinuz.knoppix)
   38  117050 [main] ls 2268 mount_info::cygdrive_win32_path: src 
'/cygdrive/l/vmlinuz.knoppix', dst 'l:\vmlinuz.knoppix'
   37  117087 [main] ls 2268 set_flags: flags: binary (0x2)
   35  117122 [main] ls 2268 mount_info::conv_to_win32_path: src_path 
/cygdrive/l/vmlinuz.knoppix, dst l:\vmlinuz.knoppix, flags 0x2A, rc 0
  172  117294 [main] ls 2268 symlink_info::check: GetFileAttributes 
(l:\vmlinuz.knoppix) failed
   44  117338 [main] ls 2268 geterrno_from_win_error: windows error 5 == errno 
13
  122  117460 [main] ls 2268 symlink_info::check: GetFileAttributes 
(l:\vmlinuz.knoppix.exe) failed
   41  117501 [main] ls 2268 geterrno_from_win_error: windows error 2 == errno 
2  
121  117622 [main] ls 2268 symlink_info::check: GetFileAttributes 
(l:\vmlinuz.knoppix.exe.lnk) failed
   93  117715 [main] ls 2268 geterrno_from_win_error: windows error 2 == errno 2
  124  117839 [main] ls 2268 symlink_info::check: GetFileAttributes 
(l:\vmlinuz.knoppix.lnk) failed
   40  117879 [main] ls 2268 geterrno_from_win_error: windows error 2 == errno 2
   34  117913 [main] ls 2268 symlink_info::check: 0 = symlink.check 
(l:\vmlinuz.knoppix, 0x22E5D0) (0x2A)



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: 1.5.16: Filename case sensitivity problem

[Mark Paulus]

Does this work require an assignment?
Is CVS having problems right now?  
(When I try to login, I get:  


$ export CVSROOT=:pserver:[EMAIL PROTECTED]:/cvs/src
[EMAIL PROTECTED]:/users/home/tibco/mpaulus/cygwin
$ cvs login
Logging in to :pserver:[EMAIL PROTECTED]:2401/cvs/src
CVS password:
cvs [login aborted]: reading from server: Connection reset by peer
[EMAIL PROTECTED]:/users/home/tibco/mpaulus/cygwin

This directly from http://cygwin.com/cvs.html

On Fri, 29 Apr 2005 10:46:27 -0400, Christopher Faylor wrote:

>On Thu, Apr 28, 2005 at 08:10:07PM -0700, Brian Dessent wrote:
>>Zhuang Jianmin wrote:
>>
>>> HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/usr/src
>>>   (default) = `e:\cygwin\usr\src'
>>>   flags = 0x080a
>>
>>This mount is a managed mount.  The MOUNT_ENC bitflag is 0x800...
>>
>>> e:\cygwin\usr\src  /usr/src   system  binmode
>>
>>...although it appears that cygcheck does not report this flag in its
>>output.  Still, if you run "mount -m" it should show the option "-o
>>managed" next to the command that would create that mount.  You can find
>>more information on the -o flags that mount takes in "man mount".
>>
>>I am not sure how this mount came to be if you did not create it.  I was
>>under the impression that neither setup.exe nor any packages enable
>>managed mounts.  It looks like that is not the case.

>Anyone want to submit a patch which ensures that all of the mount flags
>are represented correctly by cygcheck?

>cgf

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: can not compile cscope on cygwin

[Mark Paulus]

I just tried this, and it appears that you don't have 
lex/flex and/or bison/byacc installed.  Apparently
these aren't selected by default from the cygwin
setup tool.  Re-run the cygwin setup tool, and under 
Devel, select bison and flex, and you should be good
to go.


On Wed, 27 Apr 2005 15:46:47 -0400 (EDT), Igor Pechtchanski wrote:

>On Wed, 27 Apr 2005, lin q wrote:

>> Hi,
>>  I am using latest version cygwin and I just downloaded cscope 15.5, but in
>> compiling it there is such error,
>>
>> $ make
>> make  all-recursive
>> make[1]: Entering directory `/c/bin/cscope-15.5'
>> Making all in doc
>> make[2]: Entering directory `/c/bin/cscope-15.5/doc'
>> make[2]: Nothing to be done for `all'.
>> make[2]: Leaving directory `/c/bin/cscope-15.5/doc'
>> Making all in src
>> make[2]: Entering directory `/c/bin/cscope-15.5/src'
>> /bin/bash ../ylwrap `test -f 'fscanner.l' || echo './'`fscanner.l .c
>> fscanner.c -- :
>> make[2]: *** [fscanner.c] Error 1
>> make[2]: Leaving directory `/c/bin/cscope-15.5/src'
>> make[1]: *** [all-recursive] Error 1
>> make[1]: Leaving directory `/c/bin/cscope-15.5'
>> make: *** [all] Error 2
>>
>> Do you know what is wrong?

>Please review and follow
>> Problem reports:   http://cygwin.com/problems.html
>to provide enough information about your installation for this list to
>help you.

>Without the above information, I would guess that it's a line ending issue
>of some sort (since you're using /c, and I suppose your cygdrive prefix is
>empty rather than "c:" being mounted on /c explicitly).

>> Is there any pre-compiled cscope for cygwin somewhere?

>According to , the only matches for "cscope"
>are in the vim source package, so no, there isn't an official cscope
>package.  You can Google for unofficial ones, I suppose.
>   Igor
>-- 
>   http://cs.nyu.edu/~pechtcha/
>  |\  _,,,---,,_   [EMAIL PROTECTED]
>ZZZzz /,`.-'`'-.  ;-;;,_   [EMAIL PROTECTED]
> |,4-  ) )-,_. ,\ (  `'-'  Igor Pechtchanski, Ph.D.
>'---''(_/--'  `-'\_) fLa.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

>"The Sun will pass between the Earth and the Moon tonight for a total
>Lunar eclipse..." -- WCBS Radio Newsbrief, Oct 27 2004, 12:01 pm EDT

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Convertion from cygwin to linux

[Mark Paulus]

Is this a home-grown application (Created by you, or someone you
know, or your company?), or is it a package you retrieved from the 'net?
What Flavor of linux are you running?  If it's a well known app, then you 
might be able to find a pre-compiled version already packaged, depending
upon what flavor of linux you run.

If it's a home-grown app, and you are compiling it using gcc/g++, then just
copy your source tree over to your linux box, and try running make on 
the source tree.  Cygwin's environment is trying to emulate a standard 
linux environment, such that any libraries that are provided are source
code compatible (Calls to sprintf/malloc/fopen/etc are coded the same
under gcc/g++, independant of OS).  The Cygwin libraries are more of a 
Shim, in that they translate Unix/Posix type functionallity into windows 
specific behaviour.  Just so you don't get the wrong idea, however, this
is NOT trivial work.


On Thu, 07 Apr 2005 09:17:26 +0530, Mahadevaswamy.Lingaiah wrote:


>Hi,

>I have a source code compiled using cygwin(i386). I need to compile the same 
>source code in Linux(i386). cygwin has its own libraries. But how can I use in 
linux? Please put some light on it.

>Regards
>Swamy

>

>Note: The information contained in this message may be privileged and 
>confidential and thus protected from disclosure. If the reader of this message 
>is not 
the intended recipient, or an employee or agent responsible for delivering this 
message to the intended recipient, you are hereby notified that any 
dissemination, distribution or copying of this communication is strictly 
prohibited. If you have received this communication in error, please notify us 
immediately by replying to the message and deleting it from your computer. 
Thank you.

>

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: ./configure

[Mark Paulus]

On Mon, 14 Mar 2005 15:32:28 + (UTC), Donald wrote:

>Paulo Sequeira  racsa.co.cr> writes:

>> 
>> Donald wrote:
>> > I am a newbie for cygwin.When I try to typed ./configure, it shows that.
>> > 
>> > "$ ./configure"
>> > "bash: ./configure: No such file or directory"
>> > 
>> > Is there missing some packages of cygwin?
>> > 
>> > Thanks for your help
>> > 
>> 
>> where are you running that command from? Most likely, you won't find a 
>> configure file in home, unless you've deliverately put one there.
>> 
>> what are you trying to configure?
>> 


>I have typed "./configure" under /usr/src
>I try to install a program but I don't have any experience on compiling a 
>program. :(

>The readme of the program said :

>Build
>-

>Goto the "src" directory and run "./configure". When this finishs
>successfully, run "make".

>But when I typed "./configure" ,
>"bash: ./configure: No such file or directory"


The problem would be one of interpreting the readme.  When they
say "go to the src directory", the readme is referring to the source
directory of the package you are trying to build.  Which is probably 
under /usr/src.  So, if you have a tarball, foo.tar.gz, then, when you expand
it in /usr/src, it would produce some directory /usr/src/foo, or possibly
/usr/src/foo-x.y.z where x.y.z is a revision.  Then you would cd down
into /usr/src/foo, and run ./configure WITHIN /usr/src/foo.  this would
do all it's work, and then you would run make and build your package.



>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: ssh-agent and /tmp/ssh-* removal at logout

[Mark Paulus]

I think keychain as a service would be nice.  I currently
use a modified version of ssh-agent as a service, and 
it works pretty well.  I just have to remember to replace
the ssh-agent with my patched version any time the
ssh package gets updated.

And, yes, I have offered a patch upstream as well as the 
other guy who gave me the idea/pointers to where to patch
the ssh-agent code.

The issue is that the output from ssh-agent (the part that 
outputs the PID and SOCK data is put out on stdout, which
is bufferred, and does not get flushed under cygwin.  The
patch is to place a fflush statement after the output so that
when you redirect stdout to your .ssh-agent file, something 
actually appears within the file.

On Wed, 23 Feb 2005 08:41:13 -0800, Karl M wrote:



>>From: Jim Kleckner
>>Subject: ssh-agent and /tmp/ssh-* removal at logout
>>Date: Wed, 23 Feb 2005 06:18:50 -0800
>>
>>ssh-agent leaves stale directories named /tmp/ssh-
>>that contain the named pipe for authentication.
>>These left over directories come about when you log out
>>or shut down the computer without stopping ssh-agent
>>either by running keychain to shut it down or sending it
>>a SIGHUP to exit and clean up.
>>
>>Could ssh-agent catch the shutdown message and thus
>>do the proper cleanup?  What would that entail?
>>
>>Jim
>>
>>I noticed that in Karl's script to start keychain:
>>  http://sourceware.org/ml/cygwin/2004-03/msg00167.html
>>that he removes any /tmp/ssh-* pre-existing and presumed
>>stale directories left over by dead ssh-agent processes
>>and this assumes that there is only one ssh-agent per machine.
>>Not as good as actually getting rid of the source of the
>>zombie directories.
>>
>Actually, it does not assume that there is only one ssh-agent process per 
>machine. I routinely use it with ssh-agents processes for multiple users. 
>The files for other users are protected so that they can not be deleted. 
>Thus, only the current user's tmp files are deleted.

>I'm in the process of doing some clean-up work and trying out keychain 
>2.5.1. I am also adding ${HOSTNAME}.cmd file creation for use with Windows 
>shell scripts. If there is interest, perhaps I should offer to maintain 
>keychain, with additional support for launching it from a service. Launching 
>keychain from a service allows the ssh-agent process to survive logout, so 
>you only type a passphrase once per reboot instead of once per login.

>Thanks,

>...Karl



>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: insensitive case problem in latest cygwin

[Mark Paulus]

H

My understanding of Windows implementation of Filesystems 
is that they MIGHT preserve case, but they are not Case
sensitive.  Which is apparently what lead to the whole thing
about creating managed filesystems, which not only preserve
case, but are also Case Sensitive (as opposed to case 
sensitive).  I would be really shocked if your mv AAA aaa worked
on an unmanaged filesystem, and if it did, could you verify what
kind of filesystem you are working on?

On Mon, 14 Feb 2005 13:45:04 -0500 (EST), Igor Pechtchanski wrote:

>On Mon, 14 Feb 2005, Jason Taylor wrote:

>> windows xp
>> The current version of cygwin seems to ignore case in the filesystem.
>> if you open a cygwin bash shell and type
>>
>> touch AAA
>> mv AAA aaa
>>
>> it says that 'AAA' and 'aaa' are the same filename and doesn't change
>> the filename.

>That's the behavior I've gotten for as long as I can remember, which
>prompted me to write this useful script:
>--- BEGIN /usr/local/bin/mvv ---
>#!/bin/sh
>/bin/mv "${1%%/}" "${1%%/}-$$" && mv "${1%%/}-$$" "${2%%/}"
> END /usr/local/bin/mvv 

>> When I run cygwin setup and go back to the 'prev' version the mv
>> command changes the name of the file like it has in previous versions.

>Very strange.  What version was the 'prev' version?

>> Sorry I can't get version numbers right now but it's Valentine's day.
>>  - jason

>Oh, and neither 'uname -a' nor 'cygcheck -svr' run on Valentine's day?
>Why, that's just MEAN! :-)
>   Igor
>-- 
>   http://cs.nyu.edu/~pechtcha/
>  |\  _,,,---,,_   [EMAIL PROTECTED]
>ZZZzz /,`.-'`'-.  ;-;;,_   [EMAIL PROTECTED]
> |,4-  ) )-,_. ,\ (  `'-'  Igor Pechtchanski, Ph.D.
>'---''(_/--'  `-'\_) fLa.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

>"The Sun will pass between the Earth and the Moon tonight for a total
>Lunar eclipse..." -- WCBS Radio Newsbrief, Oct 27 2004, 12:01 pm EDT

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: FindNextFileA (fhandler_disk_file::readdir) Bizareness under NTFS - Resolved

[Mark Paulus]

I finally got enough time to track down this issue, and
it comes down to a perl issue.  Seems that a perl script is
creating the file, but it isn't doing a close on the file, so it's
making the file hang around a bit too long (like maybe
it's a race condition of NTFS).

Anyway, I patched the perl script with an appropriate close (IN)
statement, and the issue/bizareness goes away.



On Thu, 23 Dec 2004 13:40:12 -0700, Mark Paulus wrote:

>I have this really wierd/Bizarre situation, and wonder
>if anyone else has stumbled upon this:

>I have set /usr/src to be a managed link and it's located
>at c:\cygmanaged\src

>I have a directory /usr/src/cygwin-equivs/cygwin-base
>and under it there are some additional directories.

>the thing thats causing me some wierd grief is that 
>fhandler_disk_file::readdir is telling me about a file that it sees
>that DOES NOT exist.  And, it only does it when in a 
>grandparent directory.  This causes find to exit with:  

>strace --mask=all --output=strace.out find . -type f -a \( -name \#\*\#
>-o -name \*\~ -o -name DEADJOE -o -name \*.orig -o -name \*.rej -o -name \*.bak
>-o -name .\*.orig -o -name .\*.rej -o -name .SUMS -o -name TAGS -o -name core 
>-o
> \( -path \*/.deps/\* -a -name \*.P \) \) -print
>find: ./debian/README.Debian.in: No such file or directory

>I have modified fhandler_disk_file::readdir to print an additional output
>line, and here is the snippet that I'm referring to:
>   42 1666103 [main] find 6456 fhandler_disk_file::opendir: 0x100B3DA8 = 
> opendir (/usr/src/cygwin-equivs/cygwin-base/equivs/debian)
>  192 1666295 [main] find 6456 fhandler_disk_file::readdir: actually got '.'
>   49 1666344 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (.)
>   66 1666410 [main] find 6456 fhandler_disk_file::readdir: actually got '..'
>   38 1666448 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (..)
>   44 1666492 [main] find 6456 fhandler_disk_file::readdir: actually got 
> '%52%45%41%44%4D%45.%44ebian'
>   41 1666533 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (%52%45%41%44%4D%45.%44ebian)
>   41 1666574 [main] find 6456 fhandler_disk_file::readdir: actually got 
> '%52%45%41%44%4D%45.%44ebian.in'
>   40 114 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (%52%45%41%44%4D%45.%44ebian.in)
>  159 1666773 [main] find 6456 fhandler_disk_file::readdir: actually got 
> 'changelog'
>   43 1666816 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (changelog)
>   41 1666857 [main] find 6456 fhandler_disk_file::readdir: actually got 
> 'control'
>   38 1666895 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (control)
>   40 1666935 [main] find 6456 fhandler_disk_file::readdir: actually got 
> 'copyright'
>   52 1666987 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (copyright)
>   40 1667027 [main] find 6456 fhandler_disk_file::readdir: actually got 'docs'
>   38 1667065 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (docs)
>   40 1667105 [main] find 6456 fhandler_disk_file::readdir: actually got 
> 'rules'
>   38 1667143 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
> readdir (0x100B3DA8) (rules)
>   88 1667231 [main] find 6456 fhandler_disk_file::readdir: 0x0 = readdir 
> (0x100B3DA8)
>   44 1667275 [main] find 6456 fhandler_disk_file::closedir: 0 = closedir 
> (0x100B3DA8)
>  121 1667396 [main] find 6456 closedir: 0 = closedir (0x22EAA8)

>What's so bizarre is that the file README.Debian.in DOES NOT exist.
>C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian>dir
> Volume in drive C has no label.
> Volume Serial Number is B8DA-46B4

> Directory of C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian

>12/23/2004  01:26 PM  .
>12/23/2004  01:26 PM  ..
>12/23/2004  01:26 PM   770 %52%45%41%44%4D%45.%44ebian
>12/23/2004  01:26 PM   137 changelog
>12/23/2004  01:26 PM   326 control
>12/23/2004  01:26 PM   939 copyright
>12/23/2004  01:26 PM 0 docs
>12/23/2004  01:26 PM 1,100 rules
>   6 File(s)  3,272 bytes
>   2 Dir(s)  11,870,830,592 bytes free

>C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian>attrib
>A  
>C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\%52%45%41%44%4D%45.%44ebian
>A  C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\de

touch failing on network share

[Mark Paulus]

I have a network share set up on my 
OS/2 box using HPFS.  On that share, 
I have  directory:  cygmanaged, and under that, I have
a directory tmp2.

So, I can mount this directory as:

mount --system --binary -omanaged z:/cygmanaged/tmp2 /tmp2,
and it looks like everything is fine.

But, when I try to 'touch' any file/directory on that mount, I get this
error:  touch: setting times of `cygwin-equivs/': Permission denied

However, I can vi/edit/rm/mv or do just about anything else that I 
want to the file.  I just can't seem to do a touch on it.

Does anyone have any insight into touch as to what it is doing that might
be causing the permissions problem?

Thanks.

I just tested this on the raw share (/cygdrive/z/foo) and it still fails.  So,
it is not a managed mount problem/issue, but something deeper.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: is "BKDR_HACDEF.M" found in c:\cygwin\bin\cygcrypt-0.dll for real?

[Mark Paulus]

We use McAfee VirusScan Enterprise, ver8.0i, and
I just did a manual scan on cygcrypt-0.dll, and nary a 
peep


On Mon, 10 Jan 2005 12:27:01 -0500, Laurence G Esmonde wrote:


>We are using Trend Micro's OfficeScan and users at our site started
>having cygcrypt-0.dll quarantined for BKDR_HACDEF.M around 8:15am EST.
>Followed Trend's regedit procedures, but nobody was able to find the
>corresponding registry entries that BKDR_HACDEF.M should have created.  

>Does anybody @ Cygwin have a copy of Trend Micro's software to see if it
>trips up on the master TAR file?

>Anybody else who is not using Trend getting any warnings?



>Regards,

>Lar

>_
>Larry Esmonde, Jr.
>Gestalt LLC
>680 American Avenue, Suite 302
>King of Prussia, PA 19406
>Phone: (610) 768-0800
>Email: [EMAIL PROTECTED]



>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

FindNextFileA (fhandler_disk_file::readdir) Bizareness under NTFS

[Mark Paulus]

I have this really wierd/Bizarre situation, and wonder
if anyone else has stumbled upon this:

I have set /usr/src to be a managed link and it's located
at c:\cygmanaged\src

I have a directory /usr/src/cygwin-equivs/cygwin-base
and under it there are some additional directories.

the thing thats causing me some wierd grief is that 
fhandler_disk_file::readdir is telling me about a file that it sees
that DOES NOT exist.  And, it only does it when in a 
grandparent directory.  This causes find to exit with:  

strace --mask=all --output=strace.out find . -type f -a \( -name \#\*\#
-o -name \*\~ -o -name DEADJOE -o -name \*.orig -o -name \*.rej -o -name \*.bak
-o -name .\*.orig -o -name .\*.rej -o -name .SUMS -o -name TAGS -o -name core -o
 \( -path \*/.deps/\* -a -name \*.P \) \) -print
find: ./debian/README.Debian.in: No such file or directory

I have modified fhandler_disk_file::readdir to print an additional output
line, and here is the snippet that I'm referring to:
   42 1666103 [main] find 6456 fhandler_disk_file::opendir: 0x100B3DA8 = 
opendir (/usr/src/cygwin-equivs/cygwin-base/equivs/debian)
  192 1666295 [main] find 6456 fhandler_disk_file::readdir: actually got '.'
   49 1666344 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (.)
   66 1666410 [main] find 6456 fhandler_disk_file::readdir: actually got '..'
   38 1666448 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (..)
   44 1666492 [main] find 6456 fhandler_disk_file::readdir: actually got 
'%52%45%41%44%4D%45.%44ebian'
   41 1666533 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (%52%45%41%44%4D%45.%44ebian)
   41 1666574 [main] find 6456 fhandler_disk_file::readdir: actually got 
'%52%45%41%44%4D%45.%44ebian.in'
   40 114 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (%52%45%41%44%4D%45.%44ebian.in)
  159 1666773 [main] find 6456 fhandler_disk_file::readdir: actually got 
'changelog'
   43 1666816 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (changelog)
   41 1666857 [main] find 6456 fhandler_disk_file::readdir: actually got 
'control'
   38 1666895 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (control)
   40 1666935 [main] find 6456 fhandler_disk_file::readdir: actually got 
'copyright'
   52 1666987 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (copyright)
   40 1667027 [main] find 6456 fhandler_disk_file::readdir: actually got 'docs'
   38 1667065 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (docs)
   40 1667105 [main] find 6456 fhandler_disk_file::readdir: actually got 'rules'
   38 1667143 [main] find 6456 fhandler_disk_file::readdir: 0x100B3DAC = 
readdir (0x100B3DA8) (rules)
   88 1667231 [main] find 6456 fhandler_disk_file::readdir: 0x0 = readdir 
(0x100B3DA8)
   44 1667275 [main] find 6456 fhandler_disk_file::closedir: 0 = closedir 
(0x100B3DA8)
  121 1667396 [main] find 6456 closedir: 0 = closedir (0x22EAA8)

What's so bizarre is that the file README.Debian.in DOES NOT exist.
C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian>dir
 Volume in drive C has no label.
 Volume Serial Number is B8DA-46B4

 Directory of C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian

12/23/2004  01:26 PM  .
12/23/2004  01:26 PM  ..
12/23/2004  01:26 PM   770 %52%45%41%44%4D%45.%44ebian
12/23/2004  01:26 PM   137 changelog
12/23/2004  01:26 PM   326 control
12/23/2004  01:26 PM   939 copyright
12/23/2004  01:26 PM 0 docs
12/23/2004  01:26 PM 1,100 rules
   6 File(s)  3,272 bytes
   2 Dir(s)  11,870,830,592 bytes free

C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian>attrib
A  
C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\%52%45%41%44%4D%45.%44ebian
A  C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\changelog
A  C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\control
A  C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\copyright
A  C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\docs
A  C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian\rules

C:\cygmanaged\src\cygwin-equivs\cygwin-base\equivs\debian>

Does anyone have any idea where FindNextFileA is getting this extra file from?

Oh yeah, if I do this very same command down one directory, it works as 
expected, and does not
return the extra file.  Go Figure.




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: ls -F // doesn't work

[Mark Paulus]

On Thu, 23 Dec 2004 11:35:40 +0100, Corinna Vinschen wrote:


>While you're right, it's not a good thing that ls // returns these
>error messages.  I found a buglet in Cygwin which results in scanning
>the root directory accidentally in this situation.  

For us non-informed, commercial developers, would you care to 
define "buglet"?  Is it bigger than a bug, but smaller than a bugger?

Also, can you elaborate on the fix a bit, so I can find the fix and apply it
to my "sandbox"?  I'm not sure I want all of Chris' recent updates for the 
trailing dots until he releases the revised sources to fix the problem in 
it's entirety, but I am interested in seeing this issue taken care of.

Thanks.



>Corinna

>-- 
>Corinna Vinschen  Please, send mails regarding Cygwin to
>Cygwin Project Co-Leader  mailto:cygwin@cygwin.com
>Red Hat, Inc.

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

ls -F // doesn't work

[Mark Paulus]

When I do an ls -F, I get expected results:
$ ls -F /
bin/   cygwin.bat*  home/ run.groff  tmp1/  xfer/
cron_diagnose.sh*  cygwin.ico*  lib/  sbin/  usr/
cygdeb/etc/ mountem*  tmp/   var/

However, when I do ls -F //, then I get bad results:
$ ls -F //
ls: //bin: No such file or directory
ls: //cron_diagnose.sh: No such file or directory
ls: //cygdeb: No such file or directory
ls: //cygwin.bat: No such file or directory
ls: //cygwin.ico: No such file or directory
ls: //etc: No such file or directory
ls: //home: No such file or directory
ls: //lib: No such file or directory
ls: //mountem: No such file or directory
ls: //run.groff: No such file or directory
ls: //sbin: No such file or directory
ls: //tmp: No such file or directory
ls: //tmp1: No such file or directory
ls: //usr: No such file or directory
ls: //var: No such file or directory
ls: //xfer: No such file or directory

Wasn't sure if this is also intricately intertwined with the 
pathname/dots/spaces
thing, but wanted to mention it, as I am having another problem where rmdir()
is not finding a file called "//usr/share/doc/cygwin-base/README".  (should 
probably
return ENOTDIR instead of ENOENT)





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Patch to allow trailing dots on managed mounts

[Mark Paulus]

On Wed, 15 Dec 2004 21:32:47 -0500, Christopher Faylor wrote:

>On Wed, Dec 15, 2004 at 04:04:13PM -0700, Mark Paulus wrote:
>>This patch is as trivial as I could get to allow trailing
>>dots to be used on a managed file system.
>>
>>Unfortunately, my company will not sign the waiver,
>>so I cannot sign up for any significant changes to
>>the cygwin sources.  So, hopefully, this patch is
>>small enough to squeek under the limit, or it can be a
>>starting point for discussions on how to fix the original
>>problem.
>>
>>* path.cc (path_conv::check):  retain trailing dots and 
>>  spaces for managed mounts.

>Er, this patch apparently just leaves the trailing dots in the
>"converted" path, bypassing the loop which attempts to remove them.
>That's not the way to do this.  Sorry.


I am bypassing the loop which attemps to remove them, because for a
managed filesystem, I think they need to remain.

Consider the following snippets from my strace.out:

   40 2418062 [main] tar 1600 normalize_posix_path: 
/tmp1/usr/share/sgml/Netscape_Comm._Corp. = normalize_posix_path 
./usr/share/sgml/Netscape_Comm._Corp.)
   40 2418102 [main] tar 1600 mount_info::conv_to_win32_path: 
conv_to_win32_path (/tmp1/usr/share/sgml/Netscape_Comm._Corp)
   43 2418145 [main] tar 1600 set_flags: flags: binary (0x2)
   47 2418192 [main] tar 1600 mount_info::conv_to_win32_path: src_path 
/tmp1/usr/share/sgml/Netscape_Comm._Corp, dst c:\cygmanaged\usr\share\sgml\%
4Eetscape_%43omm._%43orp, flags 0x80A, rc 0

In the above, conv_to_win32_path removes the trailing dot, which creates a 
directory without the trailing dot.

now, further down the strace:
  265 2419810 [main] tar 1600 normalize_posix_path: src 
./usr/share/sgml/Netscape_Comm._Corp./dtd
   41 2419851 [main] tar 1600 cwdstuff::get: posix /tmp1
   38 2419889 [main] tar 1600 cwdstuff::get: (/tmp1) = cwdstuff::get 
(0x22EA50,260, 1, 0), errno 0
   38 2419927 [main] tar 1600 normalize_posix_path: 
/tmp1/usr/share/sgml/Netscape_Comm._Corp./dtd = normalize_posix_path 
(./usr/share/sgml/Netscape_Comm._Corp./dtd)
   39 2419966 [main] tar 1600 mount_info::conv_to_win32_path: 
conv_to_win32_path (/tmp1/usr/share/sgml/Netscape_Comm._Corp./dtd)
   42 2420008 [main] tar 1600 set_flags: flags: binary (0x2)
   47 2420055 [main] tar 1600 mount_info::conv_to_win32_path: src_path 
/tmp1/usr/share/sgml/Netscape_Comm._Corp./dtd, dst c:\cygmanaged\usr\share
\sgml\%4Eetscape_%43omm._%43orp%2E\dtd, flags 0x80A, rc 0

Here, tar is going to look for a directory that cygwin clearly thinks should 
have a dot at the end (the %2E), and it's going to fail
because it's not going to find it, since the directory was created above 
without the %2E.  

Other than the way I proposed, I'm not sure how to fix this, since the issue 
seems to be that
conv_to_win32_path() needs to get the trailing dot in it's input argument, and 
check() is
stripping it out.  The only way I can see to fix this behaviour is to leave the 
trailing dots 
in the string.  Maybe conv_to_win32_path needs to deal/strip with the trailing 
dots, depending
upon whether it's a managed filesystem or not?


>cgf




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Error with managed mount point.

[Mark Paulus]

On Wed, 15 Dec 2004 19:21:51 -0500, Larry Hall wrote:

>At 03:48 PM 12/15/2004, you wrote:
>>Running strace pointed me to the proper place to look
>>for the error.
>>
>>However, the proper fix is maybe more of a philosophy
>>issue.  
>>

>No, they need to be removed.  Windows behavior is to ignore one or more
>'.'s at the end of a file so we're stuck with that in normal Cygwin-working
>mode.  I actually haven't delved into the inner workings of the managed 
>mount option code but if there's going to be a place to hang onto and 
>translate the trailing '.'s, it would be in this special case code.  Without
>looking at more than just the patch lines, I'm guessing that the patch you 
>made (submitted to the Cygwin patches list) to path_conv::check() are
>going to be in too generic of a spot to be the right thing to do for just
>managed mount mode, though I could be wrong.  I'll try to find a minute 
>tonight to comment further.

I am going to have to disagree with you on this point.  If you are going
to have a truly managed filesystem, then it needs to handle all characters,
and not arbitrarily remove/drop any characters.  
Consider the trivial tar file I've attached.  Each README file will 
contain the directory heirarchy under which it can be found.   The tar file
was created on a SunOS box, but should be portable.  
Now, if you arbitrarily drop the trailing dots, you will wind up with less than
half of the README files, which for a real application, could be a badness.


> 

>--
>Larry Hall  http://www.rfk.com
>RFK Partners, Inc.  (508) 893-9779 - RFK Office
>838 Washington Street   (508) 893-9889 - FAX
>Holliston, MA 01746 





tmpdir.tar
Description: Unix tar archive
--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Error with managed mount point.

[Mark Paulus]

Running strace pointed me to the proper place to look
for the error.

However, the proper fix is maybe more of a philosophy
issue.  

I found some code in path.cc that removes trailing dots and
spaces from a file name in path_conv::check().  The question is,
should these trailing spaces/dots actually be removed, or should
they be converted to %20/%2E, only if they are part of the file/
directory name?  Or, should they be stripped for ALL file searches,
so that something like /foo./bar will be found under the windows
path /%66oo/%62ar?


On Wed, 15 Dec 2004 12:06:54 -0700, Mark Paulus wrote:

>Hi,

>If someone can point me to the proper place, I would
>be willing to further analyze this, but this is the issue:

>Files/Directories that end in a '.'(dot) are not retaining
>the dot.  For example:

>I am attempting to untar a tarball that contains this (done on a linux box)
>drwxr-xr-x  22 mgpaulus mgpaulus 4096 Apr  7  2001 ./
>drwxr-xr-x   5 mgpaulus mgpaulus 4096 Apr  7  2001 ../
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ArborText/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Hewlett-Packard/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 IDN_faq.org/
>drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 IETF/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ISBN_82-7640-000/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ISBN_82-7640-023/
>drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 ISO_8879-1986/
>drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 ISO_8879:1986/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ISO_IEC_15445:2000/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Microsoft/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Netscape_Comm._Corp./
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Rick_Jelliffe/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Sun_Microsystems_Corp./
>drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 W3C/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 W3O/
>drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 WebTechs/
>drwxr-xr-x   2 mgpaulus mgpaulus 4096 Dec 15 11:53 declaration/
>drwxr-xr-x   2 mgpaulus mgpaulus 4096 Apr  7  2001 dtd/
>drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 entities/
>drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 html/

>When I try to explode the same tarball on a managed mountpoint under cygwin,
>I don't get the Netscape_COMM._Corp.
>or the Sun_Microsystems_Corp. directories.

>I don't know whether the issue is in cygwin1.dll, or in mount, or where I 
>should
>start to look for this.



>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Error with managed mount point.

[Mark Paulus]

Hi,

If someone can point me to the proper place, I would
be willing to further analyze this, but this is the issue:

Files/Directories that end in a '.'(dot) are not retaining
the dot.  For example:

I am attempting to untar a tarball that contains this (done on a linux box)
drwxr-xr-x  22 mgpaulus mgpaulus 4096 Apr  7  2001 ./
drwxr-xr-x   5 mgpaulus mgpaulus 4096 Apr  7  2001 ../
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ArborText/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Hewlett-Packard/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 IDN_faq.org/
drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 IETF/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ISBN_82-7640-000/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ISBN_82-7640-023/
drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 ISO_8879-1986/
drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 ISO_8879:1986/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 ISO_IEC_15445:2000/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Microsoft/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Netscape_Comm._Corp./
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Rick_Jelliffe/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 Sun_Microsystems_Corp./
drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 W3C/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 W3O/
drwxr-xr-x   3 mgpaulus mgpaulus 4096 Apr  7  2001 WebTechs/
drwxr-xr-x   2 mgpaulus mgpaulus 4096 Dec 15 11:53 declaration/
drwxr-xr-x   2 mgpaulus mgpaulus 4096 Apr  7  2001 dtd/
drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 entities/
drwxr-xr-x   4 mgpaulus mgpaulus 4096 Apr  7  2001 html/

When I try to explode the same tarball on a managed mountpoint under cygwin,
I don't get the Netscape_COMM._Corp.
or the Sun_Microsystems_Corp. directories.

I don't know whether the issue is in cygwin1.dll, or in mount, or where I should
start to look for this.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Managed mount point question(s)?

[Mark Paulus]

Thank you for that information.  That's the clue I needed.


On Wed, 15 Dec 2004 13:37:50 -0500, Christopher Faylor wrote:

>On Wed, Dec 15, 2004 at 11:19:42AM -0700, Mark Paulus wrote:
>>In general, does it make sense to mount c:/cygwin as
>>/ in managed mode?

>No.

>>Is there some reason this would not be a good idea?

>Since you are supposed to first create an empty directory, then mount it
>as managed mount, and then populate it, it doesn't make sense to do this
>with the cygwin directory unless you want to first create a normal cygwin
>directory, then create a managed directory and then move everything from
>the cygwin directory to it.

>Mounting a non-empty directory as managed is asking for trouble.

>cgf

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Managed mount point question(s)?

[Mark Paulus]

In general, does it make sense to mount c:/cygwin as
/ in managed mode?  Is there some reason this would 
not be a good idea?

I have done this, and when I do, some files "disappear", 
such as /home/Mark.Paulus, which is my home.  If I have 
a regular mount (c:\cygwin on / type system (binmode)),
I have the following files in /home:

[EMAIL PROTECTED] /home
$ ll
total 0
drwx--+   3 Mark.Pau None0 Dec 15 10:38 ./
drwx--+  12 Mark.Pau None0 Dec 15 09:51 ../
lrwxrwxrwx1 Mark.Pau None   18 Dec 15 10:38 Administrator -> 
/home/Mark.Paulus/
drwx--+  13 Mark.Pau None0 Dec 15 10:41 Mark.Paulus/
lrwxrwxrwx1 Mark.Pau None   18 Dec 15 10:38 root -> 
/home/Mark.Paulus/

when I remount c:/cygwin as a managed mount
(c:\cygwin on / type system (binmode,managed)),
I have the following in /home:

[EMAIL PROTECTED] /home
$ ll
ls: Administrator: No such file or directory
ls: Mark.Paulus: No such file or directory
total 0
drwx--+   3 Mark.Pau None0 Dec 15 11:16 ./
drwx--+  12 Mark.Pau None0 Dec 15 09:51 ../
lrwxrwxrwx1 Mark.Pau None   17 Dec 15 11:16 root -> 
/home/Mark.Paulus

ll is:  alias ll='ls -alF --color'

So, where did /home/Mark.Paulus go, or who covered it up?

Thanks for any help.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: bytes

[Mark Paulus]

And if you were to boot into something like Knoppix, does
it report the same value?

On Mon, 13 Dec 2004 22:49:13 + (UTC), David wrote:

>Yes. the "mt satus 2" is telling me that 128 bytes is the maximum

>Corinna Vinschen  cygwin.com> writes:

>> 
>> On Dec 13 17:49, David wrote:
>> > For some reason my DLT 8000 drive is telling Cygwin (or vise versa) that 
>the 
>> > maximum blocking factor it can write to DLT is 128.
>> 
>> Is that what `mt status 2' prints?
>> 
>> Corinna
>> 





>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: [Fwd: [gp@familiehaase.de: sem_* functions in cygwin]]

[Mark Paulus]

So, does that mean that if process 1 opens a semaphore,
process 2 also grabs it, then process 1 unlinks it, and then
"reconnects" to it, that process 1 and process 2 do not have
and cannot have the same semaphore anymore, even though
they are using the same IPC_KEY?

(Or am I way confused/off base here)?

  On Thu, 09 Dec 2004 17:44:42 +0100, Corinna Vinschen wrote:

>[Catching up on some older mails]

>> - Forwarded message from "Gerrit P. Haase" -
>> From: "Gerrit P. Haase" 
>> To: cygwin ML
>> Subject: sem_* functions in cygwin
>> Date: Sun, 21 Nov 2004 22:48:20 +0100
>> 
>> Hi,
>> 
>> nearly all sem_* functions are available, but sem_unlock is missing,
>> was there a problem implementing sem_unlock() or was it just missed
>> by accident?
>> 
>> 
>> Gerrit
>> - End forwarded message -

>I guess you're asking about sem_unlink().  It's not implemented so far
>since named POSIX semaphores are implemented using named Windows semaphores. 
>The SUSv3 description contains a pretty unfortunate implementation detail:

>  Calls to sem_open() to recreate or reconnect to the semaphore refer
>  to a new semaphore after sem_unlink() is called.

>There's no way I know of, which allows to implement this using named
>Windows semaphores.  At least not without adding a lot of annoying
>bookkeeping overhead, possibly involving cygserver.


>Corinna

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Clapack on Cygwin revisited

[Mark Paulus]

Sounds like you might need a -L option

gcc -o testclapack testclapack.c -L lapack.a -cblaswr -lblas.a 
tmglib.a -latlas -lg2c

You can stack multiple -L options, one per directory (no space after the -L), 
the same as the -I options.

also,
the -l is a special convention that ONLY works when the libraries 
have a physical name of lib..  For instance
-lmath means that somewhere in a known location there is a file called
libmath.a or libmath.so.  So, since the clapack libraries don't follow that
convention, you can't use the -l shortcut.  You have to give it the full name.








On Fri, 19 Nov 2004 23:29:46 +0200, [EMAIL PROTECTED] wrote:

>Hi everyone.

>My troubles with getting clapack to work on cygwin continue unfortunately. In
>particular, I completed a first run of all the suggested actions in the
>readme.install file (found on the clapack website) with some error 139's during
>that rather painful 16hour! long process.

>I downloaded a test programme from the clapack website (if it could be helpful 
>I
>can attach it next time), saved it under the name of testclapack.c and tried to
>compile it. It seemed as if finding out how to compile a programme using
>clapack was really tough.

>Nonetheless, after several attepts I tried a couple
>of compilation commands with the likes of: gcc testclapack.c -llapack,
>gcc testclapack.c -llapack -lcblaswr -lcblas -latlas -lg2c (and any permutation
>of the options (=5! I guess..:p)) and always I was getting a:

>/cygdrive/c/DOCUME~1/kele/LOCALS~1/Temp/cccMzcp0.o(.text+0x152):testclapack.c:
>undefined reference to '_dgesvd_' collect2: ld returned 1 exit status.

>The dgesvd_ is a function (the only clapack function) in the testclapack.c
>programme.

>I have installed atlas (actually I did that before installing clapack) and have
>put in the test programme the appropriate #include's (actually I #included all
>the *.h files I could find on /clapack...)

>Are there any ideas as to how I can compile a programme linking to a clapack
>routine?

>I would be grateful of your help.

>thanks,

>kele


>
>This message was sent using IMP, the Internet Messaging Program.

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: BUG: cygwin implementation of Debian package tools

[Mark Paulus]

My company is apparently working on getting my 
assignment worked out, so if/when they do, I would
be interested in taking on this package.

On Fri, 29 Oct 2004 13:27:35 -0400, Christopher Faylor wrote:

>On Fri, Oct 29, 2004 at 01:07:23PM -0400, David A. Cobb wrote:
>>Getting tired of loading disks, I structured a local mirror of the
>>Debian directory structure and tried running Cygwin
>>dpkg-scanpackages.  I expected it to take a while, so I wasn't
>>surprised - I just went and make some more coffee.  But after 20
>>minutes I suspected something was wrong and fired up ProcessExplorer.
>>   Under my console bash I discovered:
>>+perl (OK, dpkg-scanpackeages is a big perl script I guess)
>>++dpkg-deb
>>++rm.exe< getting 98% of the CPU time and not making any
>>progress.  Watching the handles display, I see an open handle on what
>>must be the first *.deb file found so I'm pretty sure of the "no
>>progress -= stuck!" diagnosis.
>>
>>Killed it all, ran my scan with the identical sh script in Debian
>>proper.  It did, indeed, take a bit of time but it completed without a
>>hitch and /bin/rm  never stayed long enough for me to see it: perl got
>>about 3 sec. CPU, dpkg-deb about 2, find about 2, everything else less
>>than 1.

>Since Robert Collins is no longer actively involved with the Cygwin
>community, this package is now unmaintained.

>If no one wants to claim it within the next week or so, I'll probably
>remove it from the distribution.

>cgf

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygutils cygstart eats program arguments

[Mark Paulus]

Have you tried using a -- to indicate end of arguments
to cygstart:

cygstart -- tail --version

On Mon, 25 Oct 2004 20:50:05 +0200, Robert Schmidt wrote:

>Hi!

>I'd like to use cygstart to launch a couple of log windows to monitor 
>server activity, much like the Windows START command.  However, cygstart 
>eats *all* the arguments, not only the ones before the executable.  Is 
>there something I've missed?

>The usage suggests otherwise:
>Usage: cygstart [OPTION]... FILE [ARGUMENTS]

>Sample session:

>/cygdrive/l/bin $ cygstart tail -f /var/log/fetchmail.log
>Unable to start 'tail': There is no application associated with the 
>given file name extension.

>/cygdrive/l/bin $ cygstart tail
>[opens a new window waiting for stdin (as it should).]

>/cygdrive/l/bin $ cygstart tail --version
>[displays cygstart's version, not tail's]
>cygstart version 1.0, by Michael Schaap

>Let Windows start a program or open a file or URL.

>/cygdrive/l/bin $ cygstart "tail --version"
>Unable to start 'tail --version': The specified file was not found.


>Cheers,
>Rob


>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: problem with find/grep

[Mark Paulus]

Thanks.  I was not aware of the -print0 option.  This is 
what I love about *n*x:  Always something new to learn, 
or another way to find a solution.


On Tue, 12 Oct 2004 13:14:12 -0400, Christopher Faylor wrote:

>On Tue, Oct 12, 2004 at 11:00:35AM -0600, Mark Paulus wrote:
>>Another issue I've run into with find, but typically when I pipe
>>it to xargs is that directories/files with spaces in the names cause
>>all kind of disruption.

>So use "find -print0 | xargs -0".

>cgf

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: problem with find/grep

[Mark Paulus]

Another issue I've run into with find, but typically when I pipe
it to xargs is that directories/files with spaces in the names cause
all kind of disruption.

On Tue, 12 Oct 2004 12:25:15 -0400 (EDT), Igor Pechtchanski wrote:

>On Tue, 12 Oct 2004, Daniel Miller wrote:

>> On linux systems, I typically to global searches with a command such as:
>>
>>  find . -name "*.[ch]*" -exec grep -H -n stuff {} \;
>>
>> and this works nicely.  However, when I try the same command under Cygwin
>> (from a 4NT prompt, not Bash), I get "find: missing argument to '-exec' ".
>> I tried a variety of modifications to the command but nothing makes this
>> work.  What am I missing??

>I'd bet it's the fact that 4NT uses different quoting mechanisms than
>bash.  Try quoting the semicolon as ";" instead of \;.  You may also need
>to quote the braces.  CGF uses 4NT, AFAIK, so he could probably provide
>better hints.  I won't ask the implied "Why not use bash?" question...
>   Igor
>-- 
>   http://cs.nyu.edu/~pechtcha/
>  |\  _,,,---,,_   [EMAIL PROTECTED]
>ZZZzz /,`.-'`'-.  ;-;;,_   [EMAIL PROTECTED]
> |,4-  ) )-,_. ,\ (  `'-'  Igor Pechtchanski, Ph.D.
>'---''(_/--'  `-'\_) fLa.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

>"Happiness lies in being privileged to work hard for long hours in doing
>whatever you think is worth doing."  -- Dr. Jubal Harshaw

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: child thread termination causes main thread error termination

[Mark Paulus]

I tried compiling this, and it failed with:

$ make lab4_1
gcc lab4_1.c   -o lab4_1
lab4_1.c:13:32: ../inc/local_mutex.h: No such file or directory
lab4_1.c:49: error: `PTHREAD_COND_INITIALIZER' undeclared here (not in a functio
n)
lab4_1.c:50: error: `PTHREAD_MUTEX_INITIALIZER' undeclared here (not in a functi
on)
lab4_1.c:52: error: `PTHREAD_COND_INITIALIZER' undeclared here (not in a functio
n)
lab4_1.c:53: error: `PTHREAD_MUTEX_INITIALIZER' undeclared here (not in a functi
on)
lab4_1.c:55: error: `PTHREAD_MUTEX_INITIALIZER' undeclared here (not in a functi
on)
make: *** [lab4_1] Error 1

after changing the include to be #include , it compiled fine.
And it seems to run to completion, as expected

$ ./lab4_1
waiting do_one_thing
waiting do_another_thing
<5 second pause>
Ready to start threads .
Starting ... do_one_thing
Starting ... do_another_thing
&+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-@(!(

Running on a self-built cygwin 1.5.11 kernel, on WinXP / SP1.

Hmmm, some more slightly interesting results.  Moving this to a Debian Sarge
box, and trying it gives a slightly different output:

make lab4_1
cc lab4_1.c   -o lab4_1
/tmp/ccbFl3T5.o(.text+0x2a0): In function `main':
: undefined reference to `pthread_create'
/tmp/ccbFl3T5.o(.text+0x2c3): In function `main':
: undefined reference to `pthread_create'
/tmp/ccbFl3T5.o(.text+0x3d2): In function `main':
: undefined reference to `pthread_join'
/tmp/ccbFl3T5.o(.text+0x3e5): In function `main':
: undefined reference to `pthread_join'
collect2: ld returned 1 exit status
make: *** [lab4_1] Error 1
[EMAIL PROTECTED]:/home/mgpaulus
$ make "LDFLAGS=-lpthread" lab4_1
cc   -lpthread  lab4_1.c   -o lab4_1
[EMAIL PROTECTED]:/home/mgpaulus
$ ./lab4_1
waiting do_one_thing
waiting do_another_thing
Ready to start threads .
&Starting ... do_one_thing
++Starting ... do_another_thing
-++--++-+++--++-
+++--+--
+@(#&!(-@)#^
All threads complete
!)
wrap up: doing another 100, another 100, total 200

So, maybe it's not working quite as expected.  If I get some time, I'll
look into it some more from my side.




On Sat, 09 Oct 2004 14:49:15 -0400, [EMAIL PROTECTED] wrote:

>I have a sample thread application that runs under Ferdora Core 2 latest but exits on 
>an error condition under cygwin.  If I add sleeps to the end of the child 
threads then the program will go to completion.  While using "insight" the error can 
be tracked into the library code.  I have attached the source code and 
output of cygcheck.  Is this a know problem and if so is there any other workaround.









--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Patch for fhandler_tty.cc

[Mark Paulus]

Corinna,

Thanks for applying this, and I apologize for not following
protocol.  Subsequent patches will go to cygwin-patches,
and will be accompanied by ChangeLog entries as well.

On Thu, 07 Oct 2004 10:01:49 +0200, Corinna Vinschen wrote:

>On Oct  6 12:10, Mark Paulus wrote:
>> --- fhandler_tty_old.cc 2004-10-06 12:05:54.196444800 -0600
>> +++ fhandler_tty.cc 2004-10-06 12:06:13.465645500 -0600
>> @@ -766,7 +766,7 @@ fhandler_tty_slave::read (void *ptr, siz
>>w4[0] = signal_arrived;
>>w4[1] = input_available_event;
>> 
>> -  DWORD waiter = !ptr ? 0 : INFINITE;
>> +  DWORD waiter = time_to_wait;
>>while (len)
>>  {
>>rc = WaitForMultipleObjects (2, w4, FALSE, waiter);

>Looks good.  I've applied it.  Just a matching ChangeLog entry would
>have been nice.


>Thanks,
>Corinna

>-- 
>Corinna Vinschen  Please, send mails regarding Cygwin to
>Cygwin Project Co-Leader  mailto:[EMAIL PROTECTED]
>Red Hat, Inc.

>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Patch for fhandler_tty.cc

[Mark Paulus]

Hi,

While trying to port xtermcontrol, I found a problem
where it would hang.  Turns out, there is a small
one-liner that appears to be something that was
missed when the code either written, or later
enhanced.  The code currently breaks the handling
of the case of MIN==0, TIME > 0.  The call never 
returns because waiter is incorrectly being set to infinite.

Patch is included below:

--- fhandler_tty_old.cc 2004-10-06 12:05:54.196444800 -0600
+++ fhandler_tty.cc 2004-10-06 12:06:13.465645500 -0600
@@ -766,7 +766,7 @@ fhandler_tty_slave::read (void *ptr, siz
   w4[0] = signal_arrived;
   w4[1] = input_available_event;

-  DWORD waiter = !ptr ? 0 : INFINITE;
+  DWORD waiter = time_to_wait;
   while (len)
 {
   rc = WaitForMultipleObjects (2, w4, FALSE, waiter);




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: rsync local dir copy hang - solved for me

[Mark Paulus]

Would it be possible to get the testcase?  I
have an idea for a fix, and would like to try it
out on the testcases that I can see.

Thanks.

On Sat, 25 Oct 2003 12:17:46 +0200, Corinna Vinschen wrote:

>On Fri, Oct 24, 2003 at 07:41:54AM -0700, Bakken, Luke wrote:
>> It never got past cygwin_select().
>> [...]
>> > The bottom point: *don't do that* ;-)
>> >  i.e. leave the msleep() call there.
>> 
>> If I leave msleep there, it hangs every time. So I'm taking it out. Now,
>
>But Hannu is right.  Why not replace msleep(2) by usleep(2)?  AFAICS,
>it should fix the hang as well.  Would you mind to try it?
>
>>  I'm now going to see if
>> I can isolate the heart of the problem with a simple program that forks
>> and sleeps in the same fashion as with rsync.
>
>No worries, I have a working testcase which shows the hang.  But thanks
>for the offer.
>
>Corinna
>
>-- 
>Corinna Vinschen  Please, send mails regarding Cygwin to
>Cygwin Developermailto:[EMAIL PROTECTED]
>Red Hat, Inc.
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: How about a TWIKI page?

[Mark Paulus]

That could very well be the page I want(ed).  But,
I think it should have a link off of the Cygwin Home Page,
If it were there, then maybe others would more
willing to contribute to it.  For Example, I do believe that 
Mark J. de Jong's instructions on how to set up an SSHD
server would be good for the TWIKI.  As well as perhaps, 
"How to set up an RXVT title with bash", etc, etc.  Searching
the mail archives is nice, if you have the time, know the key
words, and are willing to run down all the little rabbit trails
that get started.

I also believe that using the TWIKI could relieve the FAQ 
maintainers jobs some.  Much of what's in the FAQ could
be put in the TWIKI, and the FAQ can become a FAQ, not
a container for all the distilled knowledge about Cygwin.

On Thu, 25 Sep 2003 09:12:27 -0400, Dawson, David W wrote:

>Just wondering, but
>
>Is the current "home of TWiki.Cygwin." at 
><<http://www.doc.ic.ac.uk/~ccj00/twiki/bin/view/Cygwin/WebHome>>
>what you are looking for?
>
>Or are you looking for something different?
>
>   -David.
>-
>David Dawson
>[EMAIL PROTECTED]
>703-367-3885
>
>
>-Original Message-
>From: Marco Carnut [mailto:[EMAIL PROTECTED] 
>Sent: Wednesday, September 24, 2003 10:47 PM
>To: [EMAIL PROTECTED]
>Subject: RE: How about a TWIKI page?
>
>
>Hi folks,
>
>Mark Paulus wrote:
>
>> Perhaps, instead of creating/adding all the issues that
>> are coming up to the FAQ, how about running a
>> TWIKI application, and linking it to the cygwin home page, 
>> and then using that as cygwin's "expert system"?
>> 
>> Just a thought.
>
>I'm for it. I can even volunteer to host it, or, if someone else has a
>better place, install/maintain it.
>
>Although I'm involved in writing a PKI-Enabled version of TWiki, I'd
>reccomend using name-and-password for most users. Client certs, although
>much more secure, are a pain to use in most browsers.
>
>--Marco "Kiko" Carnut, CISSP
>--Tempest Security Technologies -- www.tempest.com.br
>
>
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

How about a TWIKI page?

[Mark Paulus]

Perhaps, instead of creating/adding all the issues that
are coming up to the FAQ, how about running a  
TWIKI application, and linking it to the cygwin home page, 
and then using that as cygwin's "expert system"?

Just a thought.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Like to start dialog about dpkg package

[Mark Paulus]

Hello,

I would like to start a dialog about the dpkg package.
I see that it's included in the cygwin package list,
and is installable via the cygwin setup/installer.

However.

I also see that the package doesn't initialize correctly 
(when you run dselect, things don't happen as I expect).

Anyway, I would like to see this aspect of cygwin 
enhanced and taken forward, but I would also like to
get some more information (such as, is there a debian
package repository somewhere for cygwin packages), 
what is the agreed upon architecture for the debian
packages, etc, etc.





--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygipc semaphores not persistent?

[Mark Paulus]

Are you running the ipcdaemon process?  At one time
the cygipc libraries required the ipcdaemon process
to manage the global objects.  I thought I read something
that the ipc-daemon is going away, and being incorporated
into the cygwin architecture, but I did not gather that it had
happened yet.  That may be a misunderstanding on my
part however.

On Thu, 19 Jun 2003 11:49:30 -0400, Joe Buehler wrote:

>I am attempting to use cygipc for its semaphore functionality
>in porting some existing UNIX code.
>
>It would appear that semaphores are not persistent?
>
>The code depends on the UNIX semantic of a semaphore being a global
>system object that exists regardless of whether any processes are
>currently using it.
>
>I have processes coming and going, using the semaphore to mediate
>access to a disk-based queue.  I have been debugging the code,
>and it appears that the semaphore is destroyed when no process has
>it open?
>-- 
>Joe Buehler
>
>
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Problem reports:   http://cygwin.com/problems.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

winsock error 10048 (errno 112) from ssh

[Mark Paulus]

I am running 
CYGWIN_NT-5.0 WS117V0509 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown
Cygwin
under Windows 2000 w/latest service pack.  I have noticed this little odd
behaviour for a while, and I have tried a couple of upgrades to ssh to see if that
will alleviate the problem, but to no avail.

When running ssh, I notice that after a variable amount of time, my ssh session
locks up.  I have also noticed this using scp with large transfers.  The amount
of time between locks can vary, but lately it seems to be happening more
frequently.

I have run an strace on the session, and I can include it if needed, but it's 4+ MB,
so I won't unless asked for. 

However, here is the part that is wierd:

  123 181642461 [main] ssh 3220 set_bits: me 0x100FFF60, testing fd 4 (/dev/tty0)
   66 181642527 [main] ssh 3220 set_bits: ready 1
   68 181642595 [main] ssh 3220 select_stuff::wait: gotone 1
   60 181642655 [main] ssh 3220 select_stuff::wait: returning 0
   75 181642730 [main] ssh 3220 select_stuff::cleanup: calling cleanup routines
   60 181642790 [main] ssh 3220 socket_cleanup: si 0x10155A70 si->thread 0x610D1C20
   67 181642857 [main] ssh 3220 socket_cleanup: connection to si->exitsock 0x3E45
   91 181643448 [main] ssh 3220 __set_winsock_errno: socket_cleanup:1356 - winsock 
error 10048 -> errno 112
  169 181643617 [main] ssh 3220 socket_cleanup: connect failed
89231 181732848 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0
  226 181733074 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0
   74 181733148 [ttyin] ssh 3220 fhandler_pty_master::accept_input: about to write 1 
chars to slave
2624201 184357349 [ttyin] ssh 3220 fhandler_termios::line_edit: char ]
  246 184357595 [ttyin] ssh 3220 fhandler_termios::line_edit: char ]
   79 184357674 [ttyin] ssh 3220 fhandler_pty_master::accept_input: about to write 1 
chars to slave
3306761 187664435 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  500 187664935 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
   89 187665024 [ttyin] ssh 3220 fhandler_pty_master::accept_input: about to write 1 
chars to slave
207901 187872925 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  255 187873180 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  137 187873317 [ttyin] ssh 3220 fhandler_pty_master::accept_input: about to write 1 
chars to slave
185715 188059032 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  890 188059922 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  107 188060029 [ttyin] ssh 3220 fhandler_pty_master::accept_input: about to write 1 
chars to slave
169991 188230020 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  255 188230275 [ttyin] ssh 3220 fhandler_termios::line_edit: char 0x1B
  101 188230376 [ttyin] ssh 3220 fhandler_pty_master::accept_input: about to write 1 
chars to slave

up there at the line headed 91, there is the winsock error 10048, which I don't 
understand.
The pattern of the file shows that the next file should be either 
fhandler_termios::line_edit: char0x1B
or
select_stuff::wait: woke up.

But I don't get either.  just a hard lock up.

Has anyone else seen this, or have any ideas as to why this might be happening??

Thanks.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Problem with mailing list??

[Mark Paulus]

This is the second message today that I have received with this 
problem.  Is it my machine/client, or is there a small glitch in the
list?

It appears that the Date: line does not have a CR/LF at the end, and
the To: is appended directly to the end, and so my mail client does not
find the To: and does not filter it appropriately.

==BEGIN FORWARDED MESSAGE==
>Status:  U
>Return-Path: <[EMAIL PROTECTED]>
>Received: from sources.redhat.com ([209.249.29.67])
>   by merlin (EarthLink SMTP Server) with SMTP id 1716ju47L3NZFlq0
>   for <[EMAIL PROTECTED]>; Fri, 26 Apr 2002 07:03:04 -0700 (PDT)
>Received: (qmail 8821 invoked by alias); 26 Apr 2002 14:02:23 -
>Mailing-List: contact [EMAIL PROTECTED]; run by ezmlm
>Precedence: bulk
>List-Unsubscribe: 
>List-Subscribe: 
>List-Archive: 
>List-Post: 
>List-Help: , 
>Sender: [EMAIL PROTECTED]
>Mail-Followup-To: [EMAIL PROTECTED]
>Delivered-To: mailing list [EMAIL PROTECTED]
>Received: (qmail 8736 invoked from network); 26 Apr 2002 14:02:17 -
>Received: from unknown (HELO ligarius-fe0.ultra.net) (146.115.8.189)
>  by sources.redhat.com with SMTP; 26 Apr 2002 14:02:17 -
>Received: from enterprise-e.rfk.com (207-172-102-11.s773.apx1.sbo.ma.dialup.rcn.com 
>[207.172.102.11]) by ligarius-fe0.ultra.net (8.8.8/ult/n26500/mtc.v2) with ESMTP id 
>KAA18472; Fri, 26 Apr 
2002 10:02:12 -0400 (EDT)
>Message-Id: <[EMAIL PROTECTED]>
>X-Sender: [EMAIL PROTECTED]
  Offending line below  
**
>Date: Fri, 26 Apr 2002 10:00:11 -0400To: terrificskyfox <[EMAIL PROTECTED]>, 
>[EMAIL PROTECTED]
  Offending line above  
**
>From: "Larry Hall (RFK Partners, Inc)" <[EMAIL PROTECTED]>
>Subject: Re: How can I get all permissions when I login to cygwin
>In-Reply-To: <[EMAIL PROTECTED]>
>Mime-Version: 1.0
>Content-Type: text/plain; charset="us-ascii"
>

At 11:31 PM 4/25/2002, terrificskyfox wrote:
>When I login to cygwin using administrator, I found that I did not have 
>all permissions. How can I get them.


Close your eyes and turn around three times while saying "I need all 
permissions, I need all permissions..." ;-)

Seriously, you need to be more specific about what you're doing and 
the issues you see before someone can attempt to offer you advice.


>Is there some parameters should I use when I login to cygwin?


I don't know.  What are you using to "login"?



Larry Hall  [EMAIL PROTECTED]
RFK Partners, Inc.  http://www.rfk.com
838 Washington Street   (508) 893-9779 - RFK Office
Holliston, MA 01746 (508) 893-9889 - FAX


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/


===END FORWARDED MESSAGE===




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Nevermind, fixed: Re: Having lots of troubles building/getting cygwin base source!!!

[Mark Paulus]

Disregard the previous note.  I apologizing for cluttering up the
"airwaves" with this trivial matter.


On Fri, 19 Apr 2002 10:03:37 -0600, Mark Paulus wrote:

>Ok,
>
>I seem to be having some difficulties getting cygwin sources.
>When I pull the latest source for cygwin (cygwin-1.3.10-1) using
>setup, I cannot build it because the winsup/w32api directory is empty,
>and make fails with a "no  for target 'all'.
>When I try to pull the 20020409 snapshots from either planetmirror.com
>or mirrors.rcn.net, I keep getting a 'no such file or directory', and it
>looks like they have some wierd symbolic link to a file that doesn't
>exist within the machine.
>
>SO, how do I get a complete & buildable snapshot image of 
>the cygwin base?
>
>Any help appreciated.
>
>Thanks
>
>
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting: http://cygwin.com/bugs.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Having lots of troubles building/getting cygwin base source!!!

[Mark Paulus]

Ok,

I seem to be having some difficulties getting cygwin sources.
When I pull the latest source for cygwin (cygwin-1.3.10-1) using
setup, I cannot build it because the winsup/w32api directory is empty,
and make fails with a "no  for target 'all'.
When I try to pull the 20020409 snapshots from either planetmirror.com
or mirrors.rcn.net, I keep getting a 'no such file or directory', and it
looks like they have some wierd symbolic link to a file that doesn't
exist within the machine.

SO, how do I get a complete & buildable snapshot image of 
the cygwin base?

Any help appreciated.

Thanks



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

How do I build strace.exe?

[Mark Paulus]

Hi,

I am attempting to build a debuggable version of strace.exe,
but am having some compile difficulties.  I'm hoping it's as
simple as giving some configure parms, but it looks like 
my build is getting confused trying to define or use
__uid16_t & __gid16_t.

I just updated my system to the latest system,
and grabbed the source for cygwin-1.3.10-1.

I then went into winsup/utils, and did a './configure,',
and then 'make strace.exe'

Do I need to do any additional steps to build strace.exe?

Thanks.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygwin randomly pauses under Win2k

[Mark Paulus]

Chris J (Not to be confused with Chris F),

How about sending along the source code for your small program,
so I can give it a wiz-bang.  So far, running strace on my system
seems to make the problem I have go away, so if you can readily 
reproduce the behaviour, I would like to try it on my system.

Thanks.


On Thu, 18 Apr 2002 16:31:56 +0100, Chris January wrote:

>> I have this wierd behaviour, and it is causing some stats that I
>> run to whack out.
>>
>> Every now and again (every 15-30 minutes or so), my cygwin apps
>> pause for 1-2 MINUTES   I have written a simple script to capture
>> this behaviour:
>
>> What's interesting is that between seconds 42 & 45,
>> time shows an elapsed time of 1:00.21, but the system clock has
>> actually elapsed 2 minutes & 50 seconds, (unless there is some
>> major hangups in the invoking of the echo and/or `date` commands).
>>
>> I am working on this machine, so I know that there are no 2-3 minute
>> pauses in user responsiveness
>>
>> Has anyone else experienced this kind of behaviour??
>
>Yes, I've seen this behaviour. It's something to do with file I/O IIRC. 100%
>reproducible in a small program I have. Here's an strace demonstrating the
>problem:
>
>   97 7911229 [main] base 2872 fhandler_base::open:
>(C:\cygwin\opt\kde2\lib\cygkatecore.dll, 0x11)
>*** big long pause ***
>11237820 19149049 [main] base 2872 fhandler_base::open: 0x19C = CreateFileA
>(C:\cygwin\opt\kde2\lib\cygkatecore.dll, 0x8
>000, 0x7, 0x22F764, 0x3, 0x280, 0)
> 1287 19150336 [main] base 2872 fhandler_base::open: filemode set to binary
>  192 19150528 [main] base 2872 fhandler_base::open: 1 = fhandler_base::open
>(C:\cygwin\opt\kde2\lib\cygkatecore.dll, 0x
>11)
>   89 19150617 [main] base 2872 fhandler_disk_file::open: 1 =
>fhandler_disk_file::open (C:\cygwin\opt\kde2\lib\cygkateco
>re.dll, 0x11)
>  136 19150753 [main] base 2872 fhandler_disk_file::fstat_helper: 1 =
>GetFileInformationByHandle (C:\cygwin\opt\kde2\lib
>\cygkatecore.dll, 412)
>  424 19151177 [main] base 2872 get_nt_attribute: file:
>C:\cygwin\opt\kde2\lib\cygkatecore.dll
>  208 19151385 [main] base 2872 read_sd: file =
>C:\cygwin\opt\kde2\lib\cygkatecore.dll
> 1385 19152770 [main] base 2872 read_sd: file =
>C:\cygwin\opt\kde2\lib\cygkatecore.dll: len=128
>
>Regards
>Chris
>
>
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting: http://cygwin.com/bugs.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygwin randomly pauses under Win2k

[Mark Paulus]

Ah yes.  My instructor for our Advanced Unix course
was always fond of stating:  "It is perfectly acceptable
to smother your performance problems with hardware."


On Wed, 17 Apr 2002 12:59:10 -0400, Larry Hall (RFK Partners, Inc) wrote:

>Yep.  That's the occasional detriment of strace.  It hides the problem you're
>looking to track.  I wouldn't dismiss it's potential as a solution for your
>problem so quickly though.  strace with a much larger disk and a top-of-the-
>line processor is a clear win! ;-)
>
>Sorry it wasn't more help to you.
>
>Larry Hall  [EMAIL PROTECTED]
>RFK Partners, Inc.  http://www.rfk.com
>838 Washington Street   (508) 893-9779 - RFK Office
>Holliston, MA 01746     (508) 893-9889 - FAX
>
>
>
>At 12:56 PM 4/17/2002, Mark Paulus wrote:
>>Great!
>>
>>I set up a strace on the 'doit script', and the very act of
>>running strace seems to make everything behave as 
>>expected.  But, I can't run this thing all the time, as 
>>strace generates a small quantity of output.
>>Oh well, I'll keep monitoring it for a while and see what
>>comes out.
>>
>>
>>
>>
>>On Tue, 16 Apr 2002 12:17:11 -0400, Larry Hall (RFK Partners, Inc) wrote:
>>
>> >At 12:03 PM 4/16/2002, Mark Paulus wrote:
>> >>I have this wierd behaviour, and it is causing some stats that I 
>> >>run to whack out.
>> >>
>> >>Every now and again (every 15-30 minutes or so), my cygwin apps
>> >>pause for 1-2 MINUTES   I have written a simple script to capture
>> >>this behaviour:
>> >
>> >
>> >If you're interested in trying to track down why you see this behavior, you
>> >might want to try running with strace.  With it, you should be able to 
>> >pinpoint any Cygwin internals that are in force when you see this pause.
>> >
>> >
>> >Larry Hall  [EMAIL PROTECTED]
>> >RFK Partners, Inc.  http://www.rfk.com
>> >838 Washington Street   (508) 893-9779 - RFK Office
>> >Holliston, MA 01746 (508) 893-9889 - FAX
>> >
>> >
>> >--
>> >Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>> >Bug reporting: http://cygwin.com/bugs.html
>> >Documentation: http://cygwin.com/docs.html
>> >FAQ:   http://cygwin.com/faq/
>> >
>>
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygwin randomly pauses under Win2k

[Mark Paulus]

Great!

I set up a strace on the 'doit script', and the very act of
running strace seems to make everything behave as 
expected.  But, I can't run this thing all the time, as 
strace generates a small quantity of output.
Oh well, I'll keep monitoring it for a while and see what
comes out.




On Tue, 16 Apr 2002 12:17:11 -0400, Larry Hall (RFK Partners, Inc) wrote:

>At 12:03 PM 4/16/2002, Mark Paulus wrote:
>>I have this wierd behaviour, and it is causing some stats that I 
>>run to whack out.
>>
>>Every now and again (every 15-30 minutes or so), my cygwin apps
>>pause for 1-2 MINUTES   I have written a simple script to capture
>>this behaviour:
>
>
>If you're interested in trying to track down why you see this behavior, you
>might want to try running with strace.  With it, you should be able to 
>pinpoint any Cygwin internals that are in force when you see this pause.
>
>
>Larry Hall  [EMAIL PROTECTED]
>RFK Partners, Inc.  http://www.rfk.com
>838 Washington Street   (508) 893-9779 - RFK Office
>Holliston, MA 01746 (508) 893-9889 - FAX
>
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting: http://cygwin.com/bugs.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

cygwin randomly pauses under Win2k

[Mark Paulus]

I have this wierd behaviour, and it is causing some stats that I 
run to whack out.

Every now and again (every 15-30 minutes or so), my cygwin apps
pause for 1-2 MINUTES   I have written a simple script to capture
this behaviour:

$ cat doit
#!/usr/bin/bash
while [ 1 ]
do
echo `date` Starting >> doit.log
/usr/bin/time -o doit.log -a sleep 60
echo `date` Ending >> doit.log
done

I have also written a quick sed script to filter the output from time:
$ cat gettimes
#!/usr/bin/sed -f
s/^.* \([^ ]*\)elapsed .*$/\1 elapsed/
/^0inputs.*$/d

And here is some of the offending output
Tue Apr 16 09:40:53 2002 Starting
1:00.13 elapsed
Tue Apr 16 09:41:54 2002 Ending
Tue Apr 16 09:41:54 2002 Starting
1:00.11 elapsed
Tue Apr 16 09:42:55 2002 Ending
Tue Apr 16 09:42:55 2002 Starting
1:00.21 elapsed
Tue Apr 16 09:45:45 2002 Ending
Tue Apr 16 09:45:46 2002 Starting
1:00.16 elapsed
Tue Apr 16 09:46:46 2002 Ending
Tue Apr 16 09:46:47 2002 Starting
1:00.16 elapsed
Tue Apr 16 09:47:47 2002 Ending

What's interesting is that between seconds 42 & 45, 
time shows an elapsed time of 1:00.21, but the system clock has
actually elapsed 2 minutes & 50 seconds, (unless there is some
major hangups in the invoking of the echo and/or `date` commands).

I am working on this machine, so I know that there are no 2-3 minute
pauses in user responsiveness

Has anyone else experienced this kind of behaviour??

Thanks.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Wierdness of WSASendTo()??

[Mark Paulus]

Thanks Corinna for that insight.  However it didn't
work.  Further research has shown me this:

The program opens one socket in RAW mode, which 
the developer uses to create an ICMP packet.  This
packet then get's sent out (He has 2 segments in his 
packet.  An IP segment and an ICMP segment.  He sets
the protocol of the IP Header to IPPROTO_ICMP).
Then he opens a second socket with IPPROTO_ICMP,
to listen for the returning events.

Because his is using his own header in his packet, he is
able to manually send out an ICMP packet.  However,
under cygwin, I am not able to seemingly set the sending
socket to ICMP mode, and so the protocol for these packets
is defaulted to 0xFF, which the queried host says, "What
the heck", and fails the packet.

I have gotten around this by ifdef'ing around the code, and by
basically creating one socket in IPPROTO_ICMP mode, and
using it to send & receive the packets.  Now I CAN use the
setsockopt option to change the ttl, and the ICMP socket 
takes care of setting the protocol for the IP packet header.

IP_HDRINCL does NOT exist when I only use .
Further searches only show it defined in .
And when I manually define it as 2, then I get those packets
with an extra 20 bytes, and a protocol of 0xFF.

My system is slightly old, but not that old:
$ uname -a
CYGWIN_NT-5.0 CSP00509 1.3.10(0.51/3/2) 2002-02-25 11:14 i686 unknown

Has this socket code updated recently, or do we have a lack?
(I would love to be able to dig a bit deeper and provide any
patches, but I can't seem to get my company's Legal Eagles
off their "royal" cans to provide/sign a release... :(:(:(  )



On Fri, 12 Apr 2002 13:41:38 +0200, Corinna Vinschen wrote:

>On Thu, Apr 11, 2002 at 10:33:53AM -0600, Mark Paulus wrote:
>> Ok, I have done some browsing, and now I'm totally 
>> confused.
>> 
>> According to Microsoft, there are 2 headers and libraries:
>> winsock.h
>> Ws2tcpip.h
>> 
>> Since I can seem to use the IP_HDRINCL macro, I am picking
>> ws2tcpip.h file.
>
>Don't include winsock.h and friends if you're using Cygwin sockets.
>Just include  and all these POSIX schtuff.
>
>If you don't get IP_HDRINCL included that way, it's a leak in
>our headers (shouldn't that be in in.h?).
>
>For testing, just define IP_HDRINCL as 2 in your source file.
>
>Corinna
>
>-- 
>Corinna Vinschen  Please, send mails regarding Cygwin to
>Cygwin Developermailto:[EMAIL PROTECTED]
>Red Hat, Inc.
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting: http://cygwin.com/bugs.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Wierdness of WSASendTo()??

[Mark Paulus]

Ok, I have done some browsing, and now I'm totally 
confused.

According to Microsoft, there are 2 headers and libraries:
winsock.h
Ws2tcpip.h

Since I can seem to use the IP_HDRINCL macro, I am picking
ws2tcpip.h file.

However, I can't seem to figure out how to set the option.
When I invoke the following code fragment:

#ifdef IP_HDRINCL
  /*  FreeBSD wants this to avoid sending out packets with protocol type RAW
  to the network.  */
  if(setsockopt(sendsock, SOL_IP, IP_HDRINCL, &trueopt, sizeof(trueopt)))
  {
perror("setsockopt(IP_HDRINCL,1)");
return -1;
  }
#endif

I get the following from strace, which indicates that maybe cygwin doesn't
support IP_HDRINCL???  

  202  280647 [main] mtr 2576 cygwin_socket: socket (2, 3, 255)
 4034  284681 [main] mtr 2576 wsock_init: res 0
  210  284891 [main] mtr 2576 wsock_init: wVersion 514
  281  285172 [main] mtr 2576 wsock_init: wHighVersion 514
  177  285349 [main] mtr 2576 wsock_init: szDescription WinSock 2.0
  174  285523 [main] mtr 2576 wsock_init: szSystemStatus Running
  171  285694 [main] mtr 2576 wsock_init: iMaxSockets 0
  170  285864 [main] mtr 2576 wsock_init: iMaxUdpDg 0
  372  286236 [main] mtr 2576 wsock_init: lpVendorInfo 0
64252  350488 [main] mtr 2576 fdsock: not setting socket inheritance since 
winsock2_active 1
  272  350760 [main] mtr 2576 dtable::build_fhandler: fd 3, fh 0x615607C0
  169  350929 [main] mtr 2576 fdsock: fd 3, name '/dev/udp', soc 0x27C
  168  351097 [main] mtr 2576 cygwin_socket: 3 = socket (2, 3, 255)
 5043  356140 [main] mtr 2576 cygwin_setsockopt: setsockopt optval=1
  352  356492 [main] mtr 2576 cygwin_setsockopt: 0 = setsockopt (3, 0, 1 (SO_DEBUG), 
22FE40, 4)


Do I need to maybe restructure my code so that I do not create a header???


On Thu, 11 Apr 2002 17:07:40 +0200, Corinna Vinschen wrote:

>On Thu, Apr 11, 2002 at 08:20:59AM -0600, Mark Paulus wrote:
>> MTR believes it's sending out a good ICMP packet of 64 bytes (0x40).
>> cygwin_sendto() also thinks it's sending out 64 bytes (as reported by strace).
>> 
>> However, tcpdump on my linux box is receiving 84 bytes.  
>> 
>> What it looks like to me is that somehow WSASendTo is re-formatting 
>> the packet that is going out.  But, this
>> turns it into an invalid packet, and causes a "network unreachable" error
>> on the linux box, and the response never comes back.
>> 
>> Has anyone seen this behaviour before??
>
>http://msdn.microsoft.com/library/default.asp?url=/library/en-us/winsock/wsanxref_8xo2.asp
>
>Look for IP_HDRINCL.
>
>
>Corinna
>
>-- 
>Corinna Vinschen  Please, send mails regarding Cygwin to
>Cygwin Developermailto:[EMAIL PROTECTED]
>Red Hat, Inc.
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting: http://cygwin.com/bugs.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/
>




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Wierdness of WSASendTo()??

[Mark Paulus]

I am attempting to port MTR (Matt's TraceRoute), and after getting
the minires library, I can get it all to compile and it seems to run, except
for one small problem.  I never see any output.  So, after further
analysis, and also using iptraf & tcpdump on my linux box, I have this
to report:

MTR believes it's sending out a good ICMP packet of 64 bytes (0x40).
cygwin_sendto() also thinks it's sending out 64 bytes (as reported by strace).

However, tcpdump on my linux box is receiving 84 bytes.  

What it looks like to me is that somehow WSASendTo is re-formatting 
the packet that is going out.  But, this
turns it into an invalid packet, and causes a "network unreachable" error
on the linux box, and the response never comes back.

Has anyone seen this behaviour before??

Here is a synopsis of the data traffic:

What cygwin/MTR creates/sends out:

0x   4500 0040   0601 f40d  E..@
0x0010   c0a8 0008 0800 1ef7 d408 0500   
0x0020          
0x0030          

Strace Result:

  285 10960006 [main] mtr 1944 cygwin_sendto: 64 = sendto (3, 22ED84, 40, 0)
 showing a 64 byte packet going out to WSASendTo()

Here is what my linux box/tcpdump sees:

20:33:36.661467 gairloch.recluce.org > seti2.recluce.org:  ip-proto-255 64
0x   4500 0054 6fa8  80ff 48a8 c0a8 0002E..To.H.
0x0010   c0a8 0008 4500 0040   0601 f40dE..@
0x0020     c0a8 0008 0800 1ef7 d408 0500
0x0030          
0x0040          
0x0050      ..

As can be seen above, the original packet is buried within the new packet, but
an additional 20 byte header has been prepended to this packet, which skews 
everything and invalidates the packet.

Here is how my linux box responds:

20:33:36.661506 seti2.recluce.org > gairloch.recluce.org: icmp: seti2.recluce.or
g protocol 255 unreachable [tos 0xc0]
0x   45c0 0070 5207  ff01 e76a c0a8 0008E..pR..j
0x0010   c0a8 0002 0302 fcfd   4500 0054E..T
0x0020   6fa8  80ff 48a8 c0a8 0002 c0a8 0008o.H.
0x0030   4500 0040   0601 f40d  E..@
0x0040   c0a8 0008 0800 1ef7 d408 0500  
0x0050      ..




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Who is supposed to set SIGURG for OOB in rlogin/tcp/rcmd??

[Mark Paulus]

Would you like some help, or what kind of patch are you
looking for? 

On Sat, 5 Jan 2002 22:29:03 +0100, Corinna Vinschen wrote:

>On Thu, Jan 03, 2002 at 02:47:17PM -0700, Mark Paulus wrote:
>> Hi,
>> 
>> I am using rlogin to talk to a sun box, and have noticed that my rows &
>> columns are not being set, so I'm not getting good terminal 
>> characteristics.  After digging down into the rlogin code (and cross
>> comparing with some debian/linux code), it seems as though someone
>> is not passing the SIGURG (Urgent Out Of Band data) signal back
>> to rlogin, and so rlogin does not know that it needs to send the 
>> terminal characteristics back to the remote session.
>> 
>> My question is, who is supposed to be passing on the SIGUSR
>> signal?  rlogin is setting the appropriate handler, but it is never
>> being called.
>
>SIGURG is sort of a problem for Cygwin due to the way Winsock
>handles messages.  In contrast to SIGURG on U*X systems, which
>is sent always when OOB messages arrive, the corresponding Winsock
>FD_OOB message is only generated when async IO is used (as is on
>U*X sytems with SIGIO).  Unfortunately, using async IO on sockets
>has some unwanted side-effects.  I began to implement something
>appropriate but my current solution breaks ssh so that's not going
>to help very much.
>
>The current Cygwin doesn't handle SIGURG at all and unless I can
>find a practical workaround for my async IO problem or unless
>somebody else contributes a patch this stae will not change that
>quickly.
>
>Corinna
>
>-- 
>Corinna Vinschen  Please, send mails regarding Cygwin to
>Cygwin Developermailto:[EMAIL PROTECTED]
>Red Hat, Inc.
>
>--
>Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting: http://cygwin.com/bugs.html
>Documentation: http://cygwin.com/docs.html
>FAQ:   http://cygwin.com/faq/




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Did I hear you folks might be interested in POSIX threads?

[Mark Paulus]

Where do we stand with pthreads support?  I ran into
this on the debian hurd list, and thought we (cygwin)
might be interested in this as well.  Unfortunately, I'm
not sure where we are on licensing issues concerning
glibc, etc.


On Tue, 11 Dec 2001 11:00:31 -0500, Bill Abt wrote:

>
>I'm the maintainer of NGPT.  NGPT is a POSIX compliant (95%) replacement 
>library for LinuxThreads that is fully integrated with glibc and is also 
>binary compatible where possible (POSIX wins ;-).  I was wondering you folks 
>might be interested in a port to the Hurd?  Thanks.
>
>-- 
>Regards,
> Bill Abt
> Senior Software Engineer
> Next Generation POSIX Threading for Linux
> IBM Cambridge, MA, USA 02142
> Ext: +(00)1 617-693-1591
> T/L: 693-1591 (M/W/F)
> T/L: 253-9938 (T/Th/Eves.)
> Cell: +(00)1 617-803-7514
> [EMAIL PROTECTED] or [EMAIL PROTECTED]
> http://oss.software.ibm.com/developerworks/opensource/pthreads
>
>
>-- 
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>







--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Who is supposed to set SIGURG for OOB in rlogin/tcp/rcmd??

[Mark Paulus]

Hi,

I am using rlogin to talk to a sun box, and have noticed that my rows &
columns are not being set, so I'm not getting good terminal 
characteristics.  After digging down into the rlogin code (and cross
comparing with some debian/linux code), it seems as though someone
is not passing the SIGURG (Urgent Out Of Band data) signal back
to rlogin, and so rlogin does not know that it needs to send the 
terminal characteristics back to the remote session.

My question is, who is supposed to be passing on the SIGUSR
signal?  rlogin is setting the appropriate handler, but it is never
being called.

Thanks.



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/