Re: Misleading warning from mount?

[luke . kendall]

On  7 Oct, Christopher Faylor wrote:
>  You did everything but the one thing that would prove that the warning 
>  message is correct: 
>   
>  bash-2.05b$ mount 
>  c: on /cygdrive/c type user (textmode,noumount) 
>  l: on /cygdrive/l type user (textmode,noumount) 
>  m: on /cygdrive/m type user (textmode,noumount) 
>  u: on /cygdrive/u type user (textmode,noumount) 
>  x: on /cygdrive/x type user (textmode,noumount) 
>  bash-2.05b$ ls /tmp 
>  ls: /tmp: No such file or directory 
>   
>  Use the -f option if this message bothers you. 
>   
>  cgf 

Okay, I'll do that, thanks.  I don't understand what you mean about
doing everything but the one thing that would prove that the warning 
message is correct.  Do you mean mount itself?  If I actually try that,
it seems to confirm that the message is *incorrect*:

bash-2.05b$ mount
c:\temp\mnt\tmp on /tmp type system (binmode)
c: on /cygdrive/c type user (textmode,noumount)
l: on /cygdrive/l type user (textmode,noumount)
m: on /cygdrive/m type user (textmode,noumount)
u: on /cygdrive/u type user (textmode,noumount)
x: on /cygdrive/x type user (textmode,noumount)
bash-2.05b$ umount /tmp
bash-2.05b$ mount
c: on /cygdrive/c type user (textmode,noumount)
l: on /cygdrive/l type user (textmode,noumount)
m: on /cygdrive/m type user (textmode,noumount)
u: on /cygdrive/u type user (textmode,noumount)
x: on /cygdrive/x type user (textmode,noumount)
bash-2.05b$ mount c:/temp/mnt/tmp /tmp
mount: warning - /tmp does not exist.
bash-2.05b$ mount
c:\temp\mnt\tmp on /tmp type system (binmode)
c: on /cygdrive/c type user (textmode,noumount)
l: on /cygdrive/l type user (textmode,noumount)
m: on /cygdrive/m type user (textmode,noumount)
u: on /cygdrive/u type user (textmode,noumount)
x: on /cygdrive/x type user (textmode,noumount)
bash-2.05b$ ls /tmp
a  b
bash-2.05b$ echo c > /tmp/c
bash-2.05b$ cat /tmp/c
c
bash-2.05b$ ls c:/temp/mnt/tmp
a  b  c

Maybe I still misunderstand.  In what sense does /tmp not exist?

luke


--
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: Spurious "You have multiple copies of cygwin1.dll on your syste m."

[luke . kendall]

On  7 Oct, Christopher Faylor wrote:
>  On Thu, Oct 07, 2004 at 02:49:29PM +1000, [EMAIL PROTECTED] wrote:
>  >After installing Cygwin by using a shell script running by executing
>  >bash from a network-installed Cygwin, the script fails when it tries to
>  >run a post-install script.  The error reported is:
>  >
> >bash-2.05b$ post-install.sh -which latest -fresh
> >c:\cygwin\bin\bash.exe (3208): *** cygheap version mismatch detected - 0x616D000
> >0/0x6178.
> >You have multiple copies of cygwin1.dll on your system.
> >Search for cygwin1.dll using the Windows Start->Find/Search facility
> >and delete all but the most recent version.  The most recent version *should*
> >reside in x:\cygwin\bin, where 'x' is the drive on which you have
> >installed the cygwin distribution.
>  >
>  >A search showed that there really is only c:\cygwin\bin\cygwin1.dll -
>  >the message is wrong.
>  >
>  >I'd say the sanity-check is detecting cygwin1.dll associated with the
>  >bash that's running from the network-mounted drive.
>  
>  You wouldn't say that if you knew what "cygheap" was.  This is an indication
>  that a child executable is detecting that its cygheap is not at the same
>  location as the parent.  Since the cygheap lives at the end of the cygwin DLL
>  it's hard to see how this could be anything other than two different versions
>  of cygwin being invoked.

Okay, and in fact there are two (possibly different) versions of
cygwin1.dll available to be loaded into memory.  I don't know why
Cygwin attempted to load the freshly-installed cygwin1.dll, though.
It wasn't on the PATH of the network-loaded shell script.

To be pedantic, the warning is wrong because it says you have multiple
copies of cygwin1.dll on your system.  There aren't, and indeed a
Search for cygwin1.dll using the Windows Start->Find/Search facility
confirms that there is only a single cygwin1.dll, freshly installed.

Try this: from a machine on your network, with Cygwin installed, share
the c:/cygwin directory (i.e. the path where you installed Cygwin).

On another PC, start a DOS window.  set PATH=this network Cygwin path
Type bash.  You get a bash prompt.

Now you have a situation where cygwin1.dll is loaded in memory from
across the network.  And if you have Cygwin installed on the local
machine, you don't get error messages about multiple Cygwin versions
installed on your PC.  You can run all the Cygwin commands.

So even knowing that the error message is imprecise and misleading in
this situation, and that it probably means that Cygwin tried to load up
cygwin1.dll from a different path to a copy that's already loaded, and
that it's incompatible with the one that's already loaded, I don't know
why Cygwin is trying to load this other DLL.

I suspect the error message should be "Attempting to load an
inconsistent version of cygwin1.dll".

I freely confess I'm doing something unusual.  Maybe I'm the first
person on the planet to attempt to automate Cygwin installation via a
shell script from an already existing and stable copy of Cygwin
installed elsewhere on the network?

Anyway, I'll keep battering away at the attempt, and see if I can find
something on cygheap so I understand more about what's going on.

Thanks,

luke


--
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: Cygserver 100% CPU

[Patrick Samson]

--- Corinna Vinschen wrote:

> On Oct  4 07:02, Patrick Samson wrote:
> > --- Corinna Vinschen wrote:
> > > I'm still hoping for a simple testcase...
> > 
> > I'm still working on it (the problem, not the
> > testcase, as it is probably a race condition).
> 
> Too bad.
> 
> I've checked in a patch to cygserver a few minutes
> ago.  Since you're
> set up for building your own version, please try
> that one.  It adds
> a tiny bit of debug output which might be
> interesting.  I don't think
> it will solve the problem, though.
> 

Still there, and 100% CPU as well.
Outputs in the log file are still intermixed
(don't know if it matters, but I don't like it)

It's getting more difficult for me to read and
understand the log file with these new debug traces.
But at a first glance, it seems that the sequence
  good morning (error=0)!
  good night!
repeats for sem 10, 12, 14, without any wakeup
between.

I killed the process to end the 100% CPU.


__
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.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/

Delivery Information

[admin]

Attention, the mail which was sent with your account information
contains either a computer virus or one of
the following attachment types: EXE, SCR, VBS, PIF, MPG, etc.
It was isolated and not delivered to the final recipients.

System Administrator 



--
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: Misleading warning from mount?

[Christopher Faylor]

On Thu, Oct 07, 2004 at 04:22:30PM +1000, [EMAIL PROTECTED] wrote:
>In the example below, mount tells me the directory doesn't exist, yet it
>does.  I can "ls" it, and create files in it, and cat them.
>
>I suspect the warning really means that "/" isn't mounted.

No.  As far as cygwin is concerned there is no such thing as '/tmp'.

>On a system without Cygwin installed:
>
>bash-2.05b$ mount
>c: on /cygdrive/c type user (textmode,noumount)
>l: on /cygdrive/l type user (textmode,noumount)
>m: on /cygdrive/m type user (textmode,noumount)
>u: on /cygdrive/u type user (textmode,noumount)
>x: on /cygdrive/x type user (textmode,noumount)
>bash-2.05b$ ls c:/temp/mnt/tmp
>a
>bash-2.05b$ mount c:/temp/mnt/tmp /tmp
>mount: warning - /tmp does not exist.
>bash-2.05b$ echo b > /tmp/b
>bash-2.05b$ ls /tmp
>a  b
>bash-2.05b$ ls c:/temp/mnt/tmp
>a  b
>bash-2.05b$ cat c:/temp/mnt/tmp/b
>b

You did everything but the one thing that would prove that the warning
message is correct:

bash-2.05b$ mount
c: on /cygdrive/c type user (textmode,noumount)
l: on /cygdrive/l type user (textmode,noumount)
m: on /cygdrive/m type user (textmode,noumount)
u: on /cygdrive/u type user (textmode,noumount)
x: on /cygdrive/x type user (textmode,noumount)
bash-2.05b$ ls /tmp
ls: /tmp: No such file or directory

Use the -f option if this message bothers you.

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: Spurious "You have multiple copies of cygwin1.dll on your system."

[Christopher Faylor]

On Thu, Oct 07, 2004 at 02:49:29PM +1000, [EMAIL PROTECTED] wrote:
>After installing Cygwin by using a shell script running by executing
>bash from a network-installed Cygwin, the script fails when it tries to
>run a post-install script.  The error reported is:
>
>bash-2.05b$ post-install.sh -which latest -fresh
>c:\cygwin\bin\bash.exe (3208): *** cygheap version mismatch detected - 0x616D000
>0/0x6178.
>You have multiple copies of cygwin1.dll on your system.
>Search for cygwin1.dll using the Windows Start->Find/Search facility
>and delete all but the most recent version.  The most recent version *should*
>reside in x:\cygwin\bin, where 'x' is the drive on which you have
>installed the cygwin distribution.
>
>A search showed that there really is only c:\cygwin\bin\cygwin1.dll -
>the message is wrong.
>
>I'd say the sanity-check is detecting cygwin1.dll associated with the
>bash that's running from the network-mounted drive.

You wouldn't say that if you knew what "cygheap" was.  This is an indication
that a child executable is detecting that its cygheap is not at the same
location as the parent.  Since the cygheap lives at the end of the cygwin DLL
it's hard to see how this could be anything other than two different versions
of cygwin being invoked.

--
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/

Misleading warning from mount?

[luke . kendall]

In the example below, mount tells me the directory doesn't exist, yet it
does.  I can "ls" it, and create files in it, and cat them.

I suspect the warning really means that "/" isn't mounted.

On a system without Cygwin installed:

bash-2.05b$ mount
c: on /cygdrive/c type user (textmode,noumount)
l: on /cygdrive/l type user (textmode,noumount)
m: on /cygdrive/m type user (textmode,noumount)
u: on /cygdrive/u type user (textmode,noumount)
x: on /cygdrive/x type user (textmode,noumount)
bash-2.05b$ ls c:/temp/mnt/tmp
a
bash-2.05b$ mount c:/temp/mnt/tmp /tmp
mount: warning - /tmp does not exist.
bash-2.05b$ echo b > /tmp/b
bash-2.05b$ ls /tmp
a  b
bash-2.05b$ ls c:/temp/mnt/tmp
a  b
bash-2.05b$ cat c:/temp/mnt/tmp/b
b

I'm about to test that I can indeed ignore this warning as incorrect.
If I'm wrong, I'll post the results.

luke


--
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 to prepend a directory to the PATH to accommodate CVSNT?

[luke . kendall]

On  6 Oct, Siegfried Heintze wrote:
>  I looked in c:\Cygwin\home\Administrator and could not find a .profile file 
>  using emacs. There was a .bash_profile file. Should I create a .profile 
>  file? 
>Thanks, 
>  Siegfried 

It's not necessary, unless you might also run other Bourne shells.

If you are, then it makes sense to use .profile instead of
..bash_profile (just don't use any bash extensions there).

BTW, if you are appearing as Administrator, and you're actually logging
into a domain, it probably means you need to run mkpasswd -d and add
the output to /etc/passwd (mkpasswd -d >> /etc/passwd).

luke


--
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/

Spurious "You have multiple copies of cygwin1.dll on your system."

[luke . kendall]

After installing Cygwin by using a shell script running by executing
bash from a network-installed Cygwin, the script fails when it tries to
run a post-install script.  The error reported is:

bash-2.05b$ post-install.sh -which latest -fresh
c:\cygwin\bin\bash.exe (3208): *** cygheap version mismatch detected - 0x616D000
0/0x6178.
You have multiple copies of cygwin1.dll on your system.
Search for cygwin1.dll using the Windows Start->Find/Search facility
and delete all but the most recent version.  The most recent version *should*
reside in x:\cygwin\bin, where 'x' is the drive on which you have
installed the cygwin distribution.

A search showed that there really is only c:\cygwin\bin\cygwin1.dll -
the message is wrong.

I'd say the sanity-check is detecting cygwin1.dll associated with the
bash that's running from the network-mounted drive.

Re-setting PATH to exclude the network Cygwin doesn't help, then doing
an "exec /bin/bash" gives the same error and of course the shell is
gone.

Do I simply have to live with this?

I can work around it by changing a driving .bat file from:

set PATH=...;network-cygwin-path
mount c:/  /
bash   my-install-script

to:

set PATH=...;network-cygwin-path
mount c:/  /
bash   my-install-script
umount /
set PATH=...;c:\cygwin\bin
bash   my-post-install-script

which is clunkier, but will work in many cases.

I don't suppose there's any option to force cygwin to ignore its
erroneous error message?

luke


--
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: Shell script ignores $PATH?!

[luke . kendall]

On  5 Oct, Christopher Faylor wrote:
>  #!/usr/bin/env perl 
>   
>  It's a standard idiom. 

I didn't know of that.  Fantastic, thanks!

luke


--
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 to prepend a directory to the PATH to accommodate CVSNT?

[Edward L. Fox]

Just Do It!

On Wed, 6 Oct 2004 11:18:31 -0600, Siegfried Heintze
<[EMAIL PROTECTED]> wrote:
> I looked in c:\Cygwin\home\Administrator and could not find a .profile file
> using emacs. There was a .bash_profile file. Should I create a .profile
> file?
>  Thanks,
>Siegfried
> 
> 
> 
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
> [EMAIL PROTECTED]
> Sent: Tuesday, October 05, 2004 5:38 PM
> To: Siegfried Heintze
> Cc: [EMAIL PROTECTED]
> Subject: RE: How to prepend a directory to the PATH to accommodate CVSNT?
> 
> On  5 Oct, Siegfried Heintze wrote:
> >  >Anyway, in your case it should be as simple as this:
> >  >
> >  >PATH="/cygdrive/c/Program Files/cvsnt/bin:$PATH"
> >  >
> >
> >  Where do I put this statement? .bashrc?
> 
> You could, but why not put it in ~/.profile so it's read just once by
> the login shell, and any child shells then have it in the environment,
> instead of having to do it themselves?
> 
> FRom memory, bash will look for ~/.bash_login, ~/.profile, ~/.login,
> (not sure if it's in that order), and use the first one it finds.
> 
> But it won't do this if $HOME was created by Cygwin "mkdir" instead of
> by Windows Explorer, due to some mysterious access permissions problem.
> 
> BTW, I use .profile because it's usable regardless of what (Bourne-like)
> shell you use.
> 
> luke
> 
> --
> 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/

Upgrade problem: Experimental package: postgresql-8.0.0cvs-1

[Reini Urban]

Reini Urban schrieb:
I've made the latest snapshot of postgresql-8.0.0 available for 
download. This is a test release to test several new features and gain 
more feedback to be fixed upstream.
Before you update or uninstall this experimental postgresql-8.0.0cvs-1 
package, please be sure to do the following steps, otherwise setup will 
just hang.

  rm /etc/preremove/postgresql.sh
It's also save to remove an existing /etc/postinstall/postgresql.sh 
script, but this is very unlikely.

Reason:
I tried to save the existing db data into a tmp dump file, before the 
only version which will be able to access this data will be removed.
But this dumping is not stable enough, so setup.exe might hang.

So be sure to save away your databases with
  /usr/bin/pg_dumpall > dumpfile
with the postgresql version with which you created the database.
Otherwise you will not be able to access or save it.
If you've already done so, you have to reinstall this particular 
version, just to dump it to an ASCII SQL file. This file might get 
large, but other previously existing upgrading scripts have been pulled 
from distribution, because they didn't work.

The next experimental update will have this fixed and some more issues, 
which have been solved in between.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

--
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 finding case-sensitive filenames when running DOS batch command with bash shell

[Igor Pechtchanski]

On Wed, 6 Oct 2004, Igor Pechtchanski wrote:

> On Wed, 6 Oct 2004, DFong wrote:
>
> > Hi,
> >
> > I'm running cygwin bash shell and have set
> > the parameters
> >
> > ~/.bashrc
> >
> > shopt -s nocaseglob
> >
> > ~/.inputrc
> >
> > set completion-ignore-case on
> >
> > but I'm still having problems with running a DOS batch command that will
> > not be case-sensitive.
> >
> > x.bat
> > dir /s /b filename.txt (I want it to match with name like fileNAME.txt
> > and return filename.txt)
> >
> > If I run this x.bat under cygwin bash shell
> > the return from dir /s /b command is "file not found".
> >
> > What else do I need to set in DOS or bash shell environment to allow
> > "dir /s /b" to make the directory search for filename to be
> > case-insensitive?
> >
> > David
>
> Bash doesn't know about batch files, so when you attempt to run one, it
> interprets it as a shell script.  By coincidence, there is a "dir" command
> under Cygwin (a form of 'ls'), so what it ends up doing is looking for
> files named "s" and "b" in your root directory.
>
> You can confirm this in many ways, the simplest being to remove spaces
> between "dir" and the switches (i.e., use "dir/s/b").  This shouldn't
> affect the batch file (since cmd.exe treats "/" as an argument separator),
> but will produce an error in bash.
>
> Use "cmd /c x.bat" or "command.com /c x.bat" to run batch files.
>   Igor

Whoops, sorry, I'm wrong.  The .bat file is executed properly.  Sorry for
sending the e-mail off without checking -- don't know what came over me.

However, your example above also works for me (though not quite like you
want it to).  The case of filename.txt is adjusted to the case on the
disk, rather than the case of the argument.  If that's not the problem,
you'll have to be more specific about the error that you get and the exact
steps to reproduce it.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.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/

Re: Fix: Socket problem w/ apache & perl cgi (fwd)

[Reini Urban]

Christopher Faylor schrieb:
On Thu, Oct 07, 2004 at 03:06:02AM +0200, Reini Urban wrote:
Larry Hall schrieb:
Yup, that's a known issue.  It's not actually Cygwin that requires the 
'SYSTEMROOT' environment variable setting.  It's winsock. :-(
"Known Issue". How I hate that word!
Ok, I'll setup my private wikipage on just a wiki which will suite me.
(and which allows it).
Huh?  You weren't a participant in this thread but you are responding as
if there was an issue here that you needed to know.
Yes, astonishing, but sometimes a little bit of thinking in a broader 
context than the current thread is required. You asked me to collect the 
known issues topics, remember?

What does this have to do with anything?  Is this in reference to your
POSIX compliance thread?
Obviously. The main part of the POSIX compliance had nothing to do with 
POSIX, just with unexpected behaviour aka "Known Issues".

I will keep you informed then, so that you can discuss it on the
mailing-list.
This is important for us to know because...?
...because you wanted to keep a known issues discussion on
the mailing list. besides in the users guide on FAQ as I assume.
--
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/

[ANNOUNCEMENT] Updated: mhash-0.9.1-1 (mhash-devel removed)

[Reini Urban]

I've taken over mhash before it would be pulled from distribution, 
because the previous package was dependent on a MSVC runtime DLL
(mingw build). This is fixed now. All tests PASS.

It should arrive at your local mirrors soon.

MHash is a library of hash algorithms.

Canonical homepage:
   http://mhash.sourceforge.net/

Files included in the binary distribution:

   /usr/bin/cygmhash-2.dll
   /usr/include/mhash.h
   /usr/lib/libmhash.a
   /usr/lib/libmhash.dll.a
   /usr/lib/libmhash.la
   /usr/share/doc/Cygwin/mhash-0.9.1.README
   /usr/share/doc/mhash-0.9.1/AUTHORS
   /usr/share/doc/mhash-0.9.1/ChangeLog
   /usr/share/doc/mhash-0.9.1/COPYING
   /usr/share/doc/mhash-0.9.1/INSTALL
   /usr/share/doc/mhash-0.9.1/NEWS
   /usr/share/doc/mhash-0.9.1/README
   /usr/share/doc/mhash-0.9.1/THANKS
   /usr/share/doc/mhash-0.9.1/TODO
   /usr/share/man/man3/mhash.3.gz

Please address all questions to this Cygwin mailing list or better to 
the mhash developers list at the sf.net site, not to the maintainer. He 
is subscribed to both lists.

UPDATE
==
To update your installation, click on the "Install Cygwin now" link on
the http://cygwin.com/ web page.  This downloads setup.exe to your
system.  Save it and run setup, answer the questions and pick up
'mhash' from the 'Libs' category, if you install it for the first
time, click on the 'skip' field next to 'mhash' until '0.9.1-1' is
displayed. You might want to adjust the size of the window and the width 
of the "Current" column.


DOWNLOAD:
=
Note that downloads from sources.redhat.com (aka cygwin.com) aren't
allowed due to bandwidth limitations.  This means that you will need
to find a mirror which has this update, please choose the one nearest to
you: http://cygwin.com/mirrors.html


QUESTIONS:
==
If you want to make a point or ask a question the Cygwin mailing
list is the appropriate place.


CYGWIN-ANNOUNCE UNSUBSCRIBE INFO:
=
To unsubscribe to the cygwin-announce mailing list, look at the
"List-Unsubscribe: " tag in the email header of this message.  Send
email to the address specified there.  It will be in the format:

[EMAIL PROTECTED]


If you need more information on unsubscribing, start reading here:

http://sources.redhat.com/lists.html#unsubscribe-simple

Please read *all* of the information on unsubscribing that is available
starting at this URL.
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/


--
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: Fix: Socket problem w/ apache & perl cgi (fwd)

[Christopher Faylor]

On Thu, Oct 07, 2004 at 03:06:02AM +0200, Reini Urban wrote:
>Larry Hall schrieb:
>>Yup, that's a known issue.  It's not actually Cygwin that requires the 
>>'SYSTEMROOT' environment variable setting.  It's winsock. :-(
>
>"Known Issue". How I hate that word!
>
>Ok, I'll setup my private wikipage on just a wiki which will suite me.
>(and which allows it).

Huh?  You weren't a participant in this thread but you are responding as
if there was an issue here that you needed to know.

What does this have to do with anything?  Is this in reference to your
POSIX compliance thread?

>I will keep you informed then, so that you can discuss it on the
>mailing-list.

This is important for us to know because...?

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/

Add requirement for SYSTEMROOT to FAQ?

[Christopher Faylor]

Didn't we resolve that the need for SYSTEMROOT in the environment should
be in the FAQ at some point?

Joshua, could you add something about how SYSTEMROOT is required for proper
operation with winsock?

Or, even better yet, Corinna could we force SYSTEMROOT to be used
somehow but masked if someone removes it from the environment?

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: Mailing list redirection etiquette?

[Christopher Faylor]

On Wed, Oct 06, 2004 at 04:39:09PM -0700, Brian Dessent wrote:
>Fred Kulack wrote:
>
>> but this won't quote the arguments properly -- if you want that, you need
>> something like
>
>Sure it will - you just have to understand how the shell works.
>
>google "something like this"
>
>--> something%20like%20this
>
>google \"one two three\"
>
>--> %22one%20two%20three%22

Interesting how you seem to all be insistent on posting off-topic
messages on how to move a discussion to the mailing list where the
messages wouldn't be off-topic.

--
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: Fix: Socket problem w/ apache & perl cgi (fwd)

[Reini Urban]

Larry Hall schrieb:
Yup, that's a known issue.  It's not actually Cygwin that requires the 
'SYSTEMROOT' environment variable setting.  It's winsock. :-(
"Known Issue". How I hate that word!
Ok, I'll setup my private wikipage on just a wiki which will suite me. 
(and which allows it).
I will keep you informed then, so that you can discuss it on the 
mailing-list.

--
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: Fix: Socket problem w/ apache & perl cgi (fwd)

[Larry Hall]

Yup, that's a known issue.  It's not actually Cygwin that requires the 
'SYSTEMROOT' environment variable setting.  It's winsock. :-(

Larry

At 04:05 PM 10/6/2004, you wrote:
>A couple of months ago I reported a problem trying to use Apache and perl
>for web development.  The problem was that all attempts to use any kind of
>a socket failed.
>
>I just got this note from Eric Davis.  He's worked out the problem and a
>work-around.
>
>He asked me to forward this to the mailing list as he is not subscribed.
>
>dar
>
>
>
>
>-- Forwarded message --
>Date: Wed, 6 Oct 2004 12:18:29 -0700
>From: Eric Davis <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Cc: [EMAIL PROTECTED]
>Subject: Fix: Socket problem w/ apache & perl cgi
>
>
>Hi David,
>
>I came across your post on the Cygwin mailing list about Perl/CGI not
>working under Apache.
>
>  http://www.cygwin.com/ml/cygwin/2004-07/msg01080.html
>
>Like you, I was also pulling my hair out on this and figured out the
>problem/solution.  I verified that the problem exists and the solution
>works for both the Apache and Boa web servers.
>
>The following CGI script works from the bash prompt but fails as a
>CGI program (name - proto_real.cgi):
>
>  #!/usr/bin/perl
>
>  print "Content-Type: text/html\r\n\r\n\n\n\n";
>
>  my $num;
>  (undef, undef, $num) = getprotobyname('tcp');
>
>  if ($num)
>  {
>  print "OK: tcp($num)\n";
>  }
>  else
>  {
>  print "ERROR\n";
>  }
>
>  print "\n\n";
>
>
>The problem has to do with (good) Web servers removing all the existing
>environment variables before a fork and exec of the CGI script, and of
>course building new environment variables containing all the CGI data.
>For some reason, Cygwin requires the SYSTEMROOT environment variable to
>be set for things to work properly.
>
>The fix (i.e. without modifying any Web server CGI code) is to create
>a wrapper around the real script.  This wrapper simply sets the
>SYSTEMROOT environment variable and executes the real script.  Here
>is an example wrapper (name - proto.cgi):
>
>  #!/usr/bin/perl
>
>  $ENV{SYSTEMROOT} = "C:\\WINDOWS";
>
>  exec 'proto_real.cgi'
>
>
>That should do it.  With this, I can now write CGI code using Apache
>or Boa under the Cygwin environment.  So using DBI, DBD::mysql,
>DBD::mysqlPP, IO::Socket, etc, now works as expected... all of which
>I have used/tested under Cygwin/Apache/Perl/CGI.
>
>Lastly, could you please post this as a reply to your original question
>on the Cygwin mailing list.  I'm not on that list and don't want to
>subscribe just to post this.
>
>Good luck!
>
>- eric
>
>-- 
>   Eric Davis
>   http://www.foobargeek.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: problem with finding case-sensitive filenames when running DOS batch command with bash shell

[Larry Hall]

At 02:24 PM 10/6/2004, you wrote:

>Hi,
>
>I'm running cygwin bash shell and have set
>the parameters
>
>~/.bashrc
>
>shopt -s nocaseglob
>
>~/.inputrc
>
>set completion-ignore-case on
>
>but I'm still having problems with running a DOS batch command that will not be 
>case-sensitive.


None of the above affects DOS/cmd.exe/command.com functionality.  It controls
bash/readline functionality.


>x.bat
>dir /s /b filename.txt (I want it to match with name like fileNAME.txt and return 
>filename.txt)
>
>If I run this x.bat under cygwin bash shell
>the return from dir /s /b command is "file not found".


Works fine for me.  The obvious question is, are you in the directory with 
'fileNAME.txt'?


>What else do I need to set in DOS or bash shell environment
>to allow "dir /s /b" to make the directory search for filename to be case-insensitive?


That's a DOS/Windows question, not a Cygwin one.  So it's really off-topic
here.  However, there is no need to "set" anything for Windows to treat
file names in a case insensitive way.  It does that already (and really 
won't treat them any other way at the cmd/command prompt).  Your problem 
must actually be caused by something else on your machine.

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


--
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: cron doesnt run jobs (sometimes)!

[Reini Urban]

russ schrieb:
Hi there list!
1st off .im new to cygwin.
i have built some scripts to do tape backups
basically a find, then a cpio, then tar it to tape.
it runs from the command line.
it doesnt run from crontab, although from event viewer i see that cron ran.
also i put another line in crontab to run ls -l >> /usr/tmp/test every
minute.
this works just fine!.
ive lightly tapped all my scripts with chmod 777 still doesnt work!
not the script! the data must be readable for SYSTEM.
btw its a win2k server (dc)
any ideas?
check if it runs under okay in a "sysbash" shell. must be the permissions.
search the ml archives for "sysbash". still not a package
or run manually:
> at  /interactive bash --login -i
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/
--
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: Mailing list redirection etiquette?

[Brian Dessent]

Fred Kulack wrote:

> but this won't quote the arguments properly -- if you want that, you need
> something like

Sure it will - you just have to understand how the shell works.

google "something like this"

--> something%20like%20this

google \"one two three\"

--> %22one%20two%20three%22

Brian

--
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: Mailing list redirection etiquette?

[Brian Dessent]

Igor Pechtchanski wrote:

> google () {
> url="http://www.google.com/search?q=`echo -n $*|perl -mCGI -pe 
> '$_=CGI::escape($_)'`";
> echo "Starting: $url"; cygstart -o "$url" >/dev/null
> }

You can remove dependence on CGI.pm by using

google () {
url="http://www.google.com/search?q=`echo -n $* | \
  perl -pe 's/([^\w()\x27*~!.-])/sprintf "%%%02x", ord $1/eg'`";
echo "Starting: $url"; cygstart -o "$url" >/dev/null
}

Brian

(oh and you'd want -MCGI not -mCGI if you want to use the module)

--
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/

Openssh on Windows XP SP2 sshd service stops when attempting to connect

[koan]

Hello,
I have read what I can find, but am getting stuck with sshd.  I am using 
Openssh 3.9p1-2.  My service starts fine, and I am running it with -d -d 
-d -e parameters (set in registry), I used -v -v -v on the ssh command line.

My key files are owned by SYSTEM:none, and are read-write for owner 
only.  /var/empty is owned by SYSTEM:SYSTEM, and is 700,

I have run ssh-host-config, and ssh-user-config, and I have mkpasswd and 
mkgroup synced my Windows users with cygwins.

If I try ssh to localhost, 127.0.0.1 on the machine the sshd service is 
running, it negotiates for a bit, and then the connection fails, and the 
sshd service stops.

I want to use password authentication.  I have tried a seperate client 
(secureCrt)

If I run change ownership of /var/empty to my own user, and then run 
sshd in a window, I can connect ok (last time I tried).

Please find attached my cygcheck.out, the sshd.log and the ssh.log from 
the console.

If anyone can help, it would be great.
koan


debug2: load_server_config: filename /etc/sshd_config
debug2: load_server_config: done config len = 187
debug2: parse_server_config: config /etc/sshd_config len 187
debug1: sshd version OpenSSH_3.9p1
debug1: private host key: #0 type 0 RSA1
debug3: Not a RSA1 key file /etc/ssh_host_rsa_key.
debug1: read PEM private key done: type RSA
debug1: private host key: #1 type 1 RSA
debug3: Not a RSA1 key file /etc/ssh_host_dsa_key.
debug1: read PEM private key done: type DSA
debug1: private host key: #2 type 2 DSA
debug1: rexec_argv[0]='/usr/sbin/sshd'
debug1: rexec_argv[1]='-d'
debug1: rexec_argv[2]='-d'
debug1: rexec_argv[3]='-d'
debug1: rexec_argv[4]='-e'
debug2: fd 3 setting O_NONBLOCK
debug1: Bind to port 22 on 0.0.0.0.
Server listening on 0.0.0.0 port 22.
Generating 768 bit RSA key.
RSA key generation complete.
debug3: fd 4 is not O_NONBLOCK
debug1: Server will not fork when running in debugging mode.
debug3: send_rexec_state: entering fd = 7 config len 187
debug3: ssh_msg_send: type 0
debug3: send_rexec_state: done
debug1: rexec start in 4 out 4 newsock 4 pipe -1 sock 7
debug3: recv_rexec_state: entering fd = 5
debug3: ssh_msg_recv entering
debug3: recv_rexec_state: done
debug2: parse_server_config: config rexec len 187
debug1: sshd version OpenSSH_3.9p1
debug1: private host key: #0 type 0 RSA1
debug3: Not a RSA1 key file /etc/ssh_host_rsa_key.
debug1: read PEM private key done: type RSA
debug1: private host key: #1 type 1 RSA
debug3: Not a RSA1 key file /etc/ssh_host_dsa_key.
debug1: read PEM private key done: type DSA
debug1: private host key: #2 type 2 DSA
debug1: inetd sockets after dupping: 3, 3
Connection from 127.0.0.1 port 3669
debug1: Client protocol version 2.0; client software version OpenSSH_3.9p1
debug1: match: OpenSSH_3.9p1 pat OpenSSH*
debug1: Enabling compatibility mode for protocol 2.0
debug1: Local version string SSH-1.99-OpenSSH_3.9p1
debug2: fd 3 setting O_NONBLOCK
debug2: Network child is on pid 3224
debug3: preauth child monitor started
debug3: mm_request_receive entering
debug1: list_hostkey_types: ssh-rsa,ssh-dss
debug1: SSH2_MSG_KEXINIT sent
debug1: SSH2_MSG_KEXINIT received
debug2: kex_parse_kexinit: 
diffie-hellman-group-exchange-sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1
debug2: kex_parse_kexinit: ssh-rsa,ssh-dss
debug2: kex_parse_kexinit: 
aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,arcfour,aes192-cbc,aes256-cbc,[EMAIL 
PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit: 
aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,arcfour,aes192-cbc,aes256-cbc,[EMAIL 
PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL 
PROTECTED],hmac-sha1-96,hmac-md5-96
debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL 
PROTECTED],hmac-sha1-96,hmac-md5-96
debug2: kex_parse_kexinit: none,zlib
debug2: kex_parse_kexinit: none,zlib
debug2: kex_parse_kexinit: 
debug2: kex_parse_kexinit: 
debug2: kex_parse_kexinit: first_kex_follows 0 
debug2: kex_parse_kexinit: reserved 0 
debug2: kex_parse_kexinit: 
diffie-hellman-group-exchange-sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1
debug2: kex_parse_kexinit: ssh-rsa,ssh-dss
debug2: kex_parse_kexinit: 
aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,arcfour,aes192-cbc,aes256-cbc,[EMAIL 
PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit: 
aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,arcfour,aes192-cbc,aes256-cbc,[EMAIL 
PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL 
PROTECTED],hmac-sha1-96,hmac-md5-96
debug2: kex_parse_kexinit: hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL 
PROTECTED],hmac-sha1-96,hmac-md5-96
debug2: kex_parse_kexinit: none,zlib
debug2: kex_parse_kexinit: none,zlib
debug2: kex_parse_kexinit: 
debug2: kex_parse_kexinit: 
debug2: kex_parse_kexinit: first_kex_follows 0 
debug2: kex_parse_kexinit: reserved 0 
debug2: mac_init: found hmac-md5
debug1: kex: client->server aes128-cbc hm

cron doesnt run jobs (sometimes)!

[russ]

Hi there list!

1st off .im new to cygwin.

i have built some scripts to do tape backups
basically a find, then a cpio, then tar it to tape.

it runs from the command line.

it doesnt run from crontab, although from event viewer i see that cron ran.

also i put another line in crontab to run ls -l >> /usr/tmp/test every
minute.
this works just fine!.

ive lightly tapped all my scripts with chmod 777 still doesnt work!

btw its a win2k server (dc)


any ideas?

Cheers

Russ


--
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: File descriptor lossage during thread termination

[Ivan Warren]

>
> However, I am encountering a situation when another thread (not
> necessaraly
> involved in the above communication) under the same process, terminates
> (through return), some (or all) of the pipe (and possibly other) file
> descriptors become unavailable. This leads to, such symptoms as 'select'
> giving an errno of 9 (EBADF : Bad File Descriptor).
>
> I am currently running 1.5.11 (release).
>


Based on

http://cygwin.com/ml/cygwin/2004-10/msg00374.html

and

http://cygwin.com/ml/cygwin/2004-10/msg00377.html

Please disregard initial message. A new attempt will be made ASAP with a
snapshot version.

Apologies,

--Ivan


--
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/

Update: IO problems with Parrot using Cygwin1.dll-1.5.11-1

[Joshua Gatcomb]

All:
Please disregard.  I asked at PerlMonks
(http://www.perlmonks.org) if any of the Cygwin users
there were aware of any problems in 1.5.11-1 and one
said there was a major threading issue where output
was lost (my problem exactly) and that it was
corrected in recent snapshots.

I tried the latest snapshot 20041004 and it worked
great.

Sorry to have troubled anyone.

Cheers
Joshua Gatcomb



___
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.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/

Fix: Socket problem w/ apache & perl cgi (fwd)

[David A. Rogers]

A couple of months ago I reported a problem trying to use Apache and perl
for web development.  The problem was that all attempts to use any kind of
a socket failed.

I just got this note from Eric Davis.  He's worked out the problem and a
work-around.

He asked me to forward this to the mailing list as he is not subscribed.

dar




-- Forwarded message --
Date: Wed, 6 Oct 2004 12:18:29 -0700
From: Eric Davis <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: Fix: Socket problem w/ apache & perl cgi


Hi David,

I came across your post on the Cygwin mailing list about Perl/CGI not
working under Apache.

  http://www.cygwin.com/ml/cygwin/2004-07/msg01080.html

Like you, I was also pulling my hair out on this and figured out the
problem/solution.  I verified that the problem exists and the solution
works for both the Apache and Boa web servers.

The following CGI script works from the bash prompt but fails as a
CGI program (name - proto_real.cgi):

  #!/usr/bin/perl

  print "Content-Type: text/html\r\n\r\n\n\n\n";

  my $num;
  (undef, undef, $num) = getprotobyname('tcp');

  if ($num)
  {
  print "OK: tcp($num)\n";
  }
  else
  {
  print "ERROR\n";
  }

  print "\n\n";


The problem has to do with (good) Web servers removing all the existing
environment variables before a fork and exec of the CGI script, and of
course building new environment variables containing all the CGI data.
For some reason, Cygwin requires the SYSTEMROOT environment variable to
be set for things to work properly.

The fix (i.e. without modifying any Web server CGI code) is to create
a wrapper around the real script.  This wrapper simply sets the
SYSTEMROOT environment variable and executes the real script.  Here
is an example wrapper (name - proto.cgi):

  #!/usr/bin/perl

  $ENV{SYSTEMROOT} = "C:\\WINDOWS";

  exec 'proto_real.cgi'


That should do it.  With this, I can now write CGI code using Apache
or Boa under the Cygwin environment.  So using DBI, DBD::mysql,
DBD::mysqlPP, IO::Socket, etc, now works as expected... all of which
I have used/tested under Cygwin/Apache/Perl/CGI.

Lastly, could you please post this as a reply to your original question
on the Cygwin mailing list.  I'm not on that list and don't want to
subscribe just to post this.

Good luck!

- eric

-- 
   Eric Davis
   http://www.foobargeek.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/

RE: Mailing list redirection etiquette?

[Fred Kulack]

On 10/06/2004 at 01:09:39 PM, cygwin-owner wrote:
but this won't quote the arguments properly -- if you want that, you need 
something like

google () {
url="http://www.google.com/search?q=`echo -n $*|perl -mCGI -pe 
'$_=CGI::escape($_)'`";
echo "Starting: $url"; cygstart -o "$url" >/dev/null
}
--- end of excerpt ---

Sigh... You can't just put that comment out there like that can you... 8-)
That doesn't get them quite right either. 8-)
> google "oh what the hell"
Starting: http://www.google.com/search?q=oh%20what%20the%20hell

Using $@ and adding double quotes to any strings with spaces before 
running them through CGI::escape does one more step nicely.
Note that I prefer somewhat verbose perl code. 

google () {
qry=$(perl -mCGI -e 'undef $qry; for ($i=0; $i<=$#ARGV; $i++) { if 
($ARGV[$i] =~ m/ /) { $ARGV[$i]="\"$ARGV[$i]\""; } if (defined $qry) 
{$qry.=" ";} $qry.=CGI::escape($ARGV[$i]); }; print "$qry\n";' $@)
url="http://www.google.com/search?q=$qry";;
echo "Starting: $url"; cygstart -o "$url" >/dev/null
}

> google "oh what the hell" is this
Starting: 
http://www.google.com/search?q=%22oh%20what%20the%20hell%22%20is%20this

NQLGGOB...  8-)




"The stuff we call "software" is not like anything that human society 
  is used to thinking about. Software is something like a machine, and 
  something like mathematics, and something like language, and 
  something like thought, and art, and information... 
  but software is not in fact any of those other things."
Bruce Sterling - The Hacker Crackdown

Fred A. Kulack - IBM eServer iSeries - Enterprise Application Solutions
ERP, Java DB2 access, Jdbc, JTA, etc...
IBM in Rochester, MN  (Phone: 507.253.5982   T/L 553-5982)
mailto:kulack/us.ibm.com   Personal: mailto:kulack/magnaspeed.net
AIM Home:FKulack  AIM Work:FKulackWrk 
MSN Work: fakulack/hotmail.com (replace email / with @)

--
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/

IO problems with Parrot using Cygwin1.dll-1.5.11-1

[Joshua Gatcomb]

All:
I am working on Parrot (http://www.parrotcode.org), a
VM that is in development that will be running Perl6
and hopefully many other languages.  After upgrading
to the 1.5.11-1 version (from 1.5.10-3) of
cygwin1.dll, I am getting mysterious IO behavior when
doing Parrot threading.  Basically it appears that
when a thread finishes the FD is closed and subsequent
prints in a still active thread don't get printed to
the screen.

I have confirmed the behavior is with the cygwin1.dll
by swapping the two around and re-running the tests. 
Unfortunately I am admittedly a novice and have no way
of isolating the problem further.  I checked the
change log in the cygwin1.dll announcement and didn't
see anything that stood out other than very
peripherally (audio FD copying).

For anyone willing to trouble shoot the problem, I
will be more than happy to help get Parrot working
(the latest CVS checkout should build out of the box
if you use a system ICU).  The following command will
illustrate the problem from the directory where Parrot
was installed:

$ perl t/harness --gc-debug --running-make-test -b
t/pmc/threads.t

Cheers
Joshua Gatcomb



__
Do you Yahoo!?
Yahoo! Mail - Helps protect you from nasty viruses.
http://promotions.yahoo.com/new_mail

--
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: Known Issues: document missing POSIX compliance and other unexpected behaviour

[Reini Urban]

Christopher Faylor schrieb:
On Wed, Oct 06, 2004 at 06:15:10PM +0200, Reini Urban wrote:
Christopher Faylor schrieb:
On Wed, Oct 06, 2004 at 05:11:03PM +0200, Reini Urban wrote:
select():
see http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-SELECT
TODO
This points to a page which explains (in an outdated manner)
implementation details of select.  I don't see how it applies.
It applies as a TODO item. :)
TODO... what?
TODO: clarify. write about it.
processes
-
See the user-guide on fork, ipc, COFF, ...
COFF?
Yes. Imho, the topic "processes" also includes loading a process.
And handling the data structures in which the process is stored,
even if not active ("loaded").
And COFF is good to known/explain in contrast to ELF or a.out.
Also for debugging purposes.
I guess I've lost track of what you're talking about.  I don't see
any reason to know about COFF if you are debugging.
I do see a reason for it to mention COFF in contrast to ELF/a.out.
Even it's hidden behind binutils and gdb. But if you think it's 
irrelevant and no one asks about it don't mention it.

Sure, but where?
In the list or elsewhere?
the list
As I suggested in a wiki for some while until it is in the docs.  Seems
to work out good, even if it should be "rigorously maintained", as you
said.
Yes, you can make any method of communication work.  I would never advocate
going to some other site and working via a wiki when communicating via a
mailing list and having decisions archived in one place has been the way
we've been doing things for years.
ok. but, seems unrealistic IMHO.
--
Reini Urban
--
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: Mailing list redirection etiquette?

[Fred Kulack]

Dang...  Had a "multiple code iterations while posting and trying to beat 
the next guy" bug. 
$qry.=" " should have been posted as $qry.="%20"
Sigh
Is this thread dead yet?

Fixed:

google () {
qry=$(perl -mCGI -e 'undef $qry; for ($i=0; $i<=$#ARGV; $i++) { if 
($ARGV[$i] =~ m/ /) { $ARGV[$i]="\"$ARGV[$i]\""; } if (defined $qry) 
{$qry.="%20";} $qry.=CGI::escape($ARGV[$i]); }; print "$qry\n";' $@)
url="http://www.google.com/search?q=$qry";;
echo "Starting: $url"; cygstart -o "$url" >/dev/null
}

> google "oh what the hell" is this
Starting: 
http://www.google.com/search?q=%22oh%20what%20the%20hell%22%20is%20this



--
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 finding case-sensitive filenames when running DOS batch command with bash shell

[Igor Pechtchanski]

On Wed, 6 Oct 2004, DFong wrote:

> Hi,
>
> I'm running cygwin bash shell and have set
> the parameters
>
> ~/.bashrc
>
> shopt -s nocaseglob
>
> ~/.inputrc
>
> set completion-ignore-case on
>
> but I'm still having problems with running a DOS batch command that will
> not be case-sensitive.
>
> x.bat
> dir /s /b filename.txt (I want it to match with name like fileNAME.txt
> and return filename.txt)
>
> If I run this x.bat under cygwin bash shell
> the return from dir /s /b command is "file not found".
>
> What else do I need to set in DOS or bash shell environment to allow
> "dir /s /b" to make the directory search for filename to be
> case-insensitive?
>
> David

Bash doesn't know about batch files, so when you attempt to run one, it
interprets it as a shell script.  By coincidence, there is a "dir" command
under Cygwin (a form of 'ls'), so what it ends up doing is looking for
files named "s" and "b" in your root directory.

You can confirm this in many ways, the simplest being to remove spaces
between "dir" and the switches (i.e., use "dir/s/b").  This shouldn't
affect the batch file (since cmd.exe treats "/" as an argument separator),
but will produce an error in bash.

Use "cmd /c x.bat" or "command.com /c x.bat" to run batch files.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.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/

Re: Calls to system() blocks return of SIGCHLD signal

[Christopher Faylor]

On Wed, Oct 06, 2004 at 12:24:48PM -0400, Remy Gauthier wrote:
>We manage synchronous events through calls to system(), and
>asynchronous events via fork() and a SIGCHLD handler.  We have noticed
>(on V1.5.10-1 and V1.5.11-1) that after a call to system(), the
>handlers were not being called after the child process stopped.  This
>program has this behaviour (removing the call to system() will restore
>correct SIGCHLD handling):

SYSTEM(3)  Linux Programmer's Manual SYSTEM(3)



NAME
   system - execute a shell command

SYNOPSIS
   #include 

   int system(const char *string);

DESCRIPTION
   system()  executes  a command specified in string by calling /bin/sh -c
   string, and returns after the command has been completed.  During  exe-
   cution  of the command, SIGCHLD will be blocked, and SIGINT and SIGQUIT
   will be ignored.


--
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: Known Issues: document missing POSIX compliance and other unexpected behaviour

[Christopher Faylor]

On Wed, Oct 06, 2004 at 06:15:10PM +0200, Reini Urban wrote:
>Christopher Faylor schrieb:
>>On Wed, Oct 06, 2004 at 05:11:03PM +0200, Reini Urban wrote:
>>>I found no document where missing POSIX functionality is listed
>>>and where other non-POSIX but expected typical UNIX functionality and/or
>>>behaviour are different or missing. The user guide is a bit short on that.
>>
>>The only way such a document would be useful would be if it was
>>rigorously maintained.
>
>ok.
>
>>You show the reason why below when you quote out-of-date parts of the
>>existing documentation.
>>It is certainly worthwhile to sweep through the docs and get everything
>>up-to-date, though.
>>
>>
>>>select():
>>>see http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-SELECT
>>>TODO
>>
>>This points to a page which explains (in an outdated manner)
>>implementation details of select.  I don't see how it applies.
>
>It applies as a TODO item. :)

TODO... what?

>>>processes
>>>-
>>>See the user-guide on fork, ipc, COFF, ...
>>
>>COFF?
>
>Yes. Imho, the topic "processes" also includes loading a process.
>And handling the data structures in which the process is stored,
>even if not active ("loaded").
>And COFF is good to known/explain in contrast to ELF or a.out.
>Also for debugging purposes.

I guess I've lost track of what you're talking about.  I don't see
any reason to know about COFF if you are debugging.

>>>fork:
>>>see http://cygwin.com/faq/faq_3.html#SEC74 and 
>>>http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS
>>>
>>>PID's:
>>>Unfortunately cygwin must use seperate PID's than the
>>>underlying windows PID's.
>>>http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS
>>
>>The documentation is out of date.  This is not true.
>>
>>I think I'll stop here.  You're showing that it would be worthwhile to
>>go through the document and look for stuff that is out of date.
>>
>>I guess that would be a job for Corinna, Pierre, and me, since
>>no one else seems to understand these issues.  If someone else wants
>>to take a stab at it, however, please feel free.
>
>Sure, but where?
>In the list or elsewhere?

the list

>As I suggested in a wiki for some while until it is in the docs.  Seems
>to work out good, even if it should be "rigorously maintained", as you
>said.

Yes, you can make any method of communication work.  I would never advocate
going to some other site and working via a wiki when communicating via a
mailing list and having decisions archived in one place has been the way
we've been doing things for years.

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: problem with finding case-sensitive filenames when running DOS batch command with bash shell

[Vince Hoffman]

On Wed, 6 Oct 2004 [EMAIL PROTECTED] wrote:
Hi,
I'm running cygwin bash shell and have set
the parameters
~/.bashrc
shopt -s nocaseglob
~/.inputrc
set completion-ignore-case on
but I'm still having problems with running a DOS batch command that will not be 
case-sensitive.
x.bat
dir /s /b filename.txt (I want it to match with name like fileNAME.txt and return 
filename.txt)
If I run this x.bat under cygwin bash shell
the return from dir /s /b command is "file not found".
What else do I need to set in DOS or bash shell environment
to allow "dir /s /b" to make the directory search for filename to be case-insensitive?
have a read of the options at
http://cygwin.com/cygwin-ug-net/using-cygwinenv.html
specificly check_case and maybe glob as well.
it says its relaxed by default but it may be worth setting it explicitly 
as well.

Vince
David

--
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/

problem with finding case-sensitive filenames when running DOS batch command with bash shell

[DFong]

Hi,

I'm running cygwin bash shell and have set
the parameters

~/.bashrc

shopt -s nocaseglob

~/.inputrc

set completion-ignore-case on

but I'm still having problems with running a DOS batch command that will not be 
case-sensitive.

x.bat
dir /s /b filename.txt (I want it to match with name like fileNAME.txt and return 
filename.txt)

If I run this x.bat under cygwin bash shell
the return from dir /s /b command is "file not found".

What else do I need to set in DOS or bash shell environment
to allow "dir /s /b" to make the directory search for filename to be case-insensitive?

David



--
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: subscribe dfongs3graphicscom

[Igor Pechtchanski]

On Wed, 6 Oct 2004, DFong wrote:

> subscribe dfongs3graphicscom

.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.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/

subscribe dfong@s3graphics.com

[DFong]

subscribe [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/

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: Mailing list redirection etiquette?

[Igor Pechtchanski]

On Wed, 6 Oct 2004, Dave Korn wrote:

> > -Original Message-
> > From: cygwin-owner On Behalf Of Fred Kulack
> > Sent: 06 October 2004 17:29
>
> > From your command prompt, use:
> > > google cygwin acronym list
> >
> > If that doesn't work, setup this function:
> > > typeset -f google
> > google ()
> > {
> > url="http://www.google.com";;
> > separator="/search?q=";
> > for i in $@;
> > do
> > url="${url}${separator}${i}";
> > separator="+";
> > done;
> > echo "Starting: $url";
> > ~/bin/firefox "$url" &
> > }
>
>   Nice one!  Can I just suggest one little change:
>
> ==
> echo "Starting: $url";
> -~/bin/firefox "$url" &
> +cygstart "$url" &
> }
>
> which should be more generic across different browsers etc.

FWIW,

typeset -f google
google () {
IFS=+; url="http://www.google.com/search?q=$*";;
echo "Starting: $url"; cygstart -o "$url" >/dev/null
}

but this won't quote the arguments properly -- if you want that, you need
something like

google () {
url="http://www.google.com/search?q=`echo -n $*|perl -mCGI -pe 
'$_=CGI::escape($_)'`";
echo "Starting: $url"; cygstart -o "$url" >/dev/null
}

instead.  HTH,
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.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/

RE: Mailing list redirection etiquette?

[Fred Kulack]

On 10/06/2004 at 12:26:21 PM, cygwin-owner wrote:
-~/bin/firefox "$url" &
+cygstart "$url" &
[...snip...]
which should be more generic across different browsers etc.
--- end of excerpt ---

Right.  Changed that at the last minute and wanted to get my firefox plug 
in...

I actually use "browse" which is aliased to cygstart when I use
these startup files on Windows, but when I use them on Linux or AIX, then 
it
gets aliased to something else.

You know, oh crap, here it comes already...
NQLGGOB (Nothing Quite Like Geeks Going Over Board) 
8-).



"The stuff we call "software" is not like anything that human society 
  is used to thinking about. Software is something like a machine, and 
  something like mathematics, and something like language, and 
  something like thought, and art, and information... 
  but software is not in fact any of those other things."
Bruce Sterling - The Hacker Crackdown

Fred A. Kulack - IBM eServer iSeries - Enterprise Application Solutions
ERP, Java DB2 access, Jdbc, JTA, etc...
IBM in Rochester, MN  (Phone: 507.253.5982   T/L 553-5982)
mailto:kulack/us.ibm.com   Personal: mailto:kulack/magnaspeed.net
AIM Home:FKulack  AIM Work:FKulackWrk 
MSN Work: fakulack/hotmail.com (replace email / with @)

--
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: Mailing list redirection etiquette?

[Dave Korn]

> -Original Message-
> From: cygwin-owner On Behalf Of Fred Kulack
> Sent: 06 October 2004 17:29

> From your command prompt, use:
> > google cygwin acronym list
> 
> If that doesn't work, setup this function:
> > typeset -f google
> google ()
> {
> url="http://www.google.com";;
> separator="/search?q=";
> for i in $@;
> do
> url="${url}${separator}${i}";
> separator="+";
> done;
> echo "Starting: $url";
> ~/bin/firefox "$url" &
> }

  Nice one!  Can I just suggest one little change:

==
echo "Starting: $url";
-~/bin/firefox "$url" &
+cygstart "$url" &
}

which should be more generic across different browsers etc.

cheers, 
  DaveK
-- 
Can't think of a witty .sigline today


--
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: Known Issues: document missing POSIX compliance and other unexpected behaviour

[Igor Pechtchanski]

On Wed, 6 Oct 2004, Reini Urban wrote:

> Christopher Faylor schrieb:
> > On Wed, Oct 06, 2004 at 05:11:03PM +0200, Reini Urban wrote:
> > > I found no document where missing POSIX functionality is listed
> > > and where other non-POSIX but expected typical UNIX functionality and/or
> > > behaviour are different or missing. The user guide is a bit short on that.
> >
> > The only way such a document would be useful would be if it was
> > rigorously maintained.
>
> ok.
>
> > You show the reason why below when you quote out-of-date parts of the
> > existing documentation.
> > It is certainly worthwhile to sweep through the docs and get everything
> > up-to-date, though.
> >
> >[snip]
> > I think I'll stop here.  You're showing that it would be worthwhile to
> > go through the document and look for stuff that is out of date.
> >
> > I guess that would be a job for Corinna, Pierre, and me, since
> > no one else seems to understand these issues.  If someone else wants
> > to take a stab at it, however, please feel free.
>
> Sure, but where?
> In the list or elsewhere? As I suggested in a wiki for some while until
> it is in the docs. Seems to work out good, even if it should be
> "rigorously maintained", as you said.

Volunteer to become a (rigorous) maintainer of these docs sections.  Send
patches to the cygwin list.  After a while, you may get commit privileges
so you can commit your own docs patches.
.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.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/

RE: How to prepend a directory to the PATH to accommodate CVSNT?

[Siegfried Heintze]

I looked in c:\Cygwin\home\Administrator and could not find a .profile file
using emacs. There was a .bash_profile file. Should I create a .profile
file?
  Thanks,
Siegfried

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
[EMAIL PROTECTED]
Sent: Tuesday, October 05, 2004 5:38 PM
To: Siegfried Heintze
Cc: [EMAIL PROTECTED]
Subject: RE: How to prepend a directory to the PATH to accommodate CVSNT?

On  5 Oct, Siegfried Heintze wrote:
>  >Anyway, in your case it should be as simple as this: 
>  > 
>  >PATH="/cygdrive/c/Program Files/cvsnt/bin:$PATH" 
>  > 
>   
>  Where do I put this statement? .bashrc? 

You could, but why not put it in ~/.profile so it's read just once by
the login shell, and any child shells then have it in the environment,
instead of having to do it themselves?

FRom memory, bash will look for ~/.bash_login, ~/.profile, ~/.login,
(not sure if it's in that order), and use the first one it finds.

But it won't do this if $HOME was created by Cygwin "mkdir" instead of
by Windows Explorer, due to some mysterious access permissions problem.

BTW, I use .profile because it's usable regardless of what (Bourne-like)
shell you use.

luke


--
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/

File descriptor lossage during thread termination

[Ivan Warren]

Hello,

I am encountering the following problem :

I have a multithreaded application. Some of the inter-thread communication
is performed by pipes (pipe()).. One thread issues a 'select' and the other
one writes to it..

However, I am encountering a situation when another thread (not necessaraly
involved in the above communication) under the same process, terminates
(through return), some (or all) of the pipe (and possibly other) file
descriptors become unavailable. This leads to, such symptoms as 'select'
giving an errno of 9 (EBADF : Bad File Descriptor).

I am currently running 1.5.11 (release).

The following test case demonstrates the issue. This is invoked with 0 or n
arguments. With n arguments, a new thread is created and then terminates
after 4 seconds. When this happens, select receives rc<0 and the program
terminates (via exit()).

Also, this same testcase does not exhibit that particular behaviour on other
posix systems (tried on linux anyways).

The application that is being affected by this issue doesn't seem to have
any problem running under 1.5.10 (haven't tried this particular test case
under 1.5.10 though).


#include 
#include 
#include 
#include 
#include 
void *thr(void *);
void *thr1(void *);

int fds[2];

int main(int ac,char **av)
{
int rc;
pthread_t   pt;
fd_set  rfd;

av=av;

pipe(fds);

if (ac>1)
pthread_create(&pt,NULL,thr1,NULL);

if(pthread_create(&pt,NULL,thr,NULL))
{
perror("CT");
exit(1);
}
while(1)
{
FD_ZERO(&rfd);
FD_SET(fds[0],&rfd);
FD_SET(0,&rfd);
rc=select(fds[0]+1,&rfd,NULL,NULL,NULL);
if(rc<0)
{
perror("select");
exit(1);
}
if(FD_ISSET(fds[0],&rfd))
{
char c[256];
rc=read(fds[0],c,256);
printf("Read %d chars from pipe\n",rc);
}
if(FD_ISSET(0,&rfd))
{
char c[256];
rc=read(0,c,256);
printf("Read %d chars from kbd\n",rc);
}
}
}

void *thr(void *x)
{
x=x;
while(1)
{
sleep(1);
write(fds[1],"X",1);
}
return(NULL);
}

void *thr1(void *x)
{
x=x;
sleep(4);
return(NULL);
}


Thank you,

--Ivan


--
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: Mailing list redirection etiquette?

[Igor Pechtchanski]

On Wed, 6 Oct 2004, Fred Kulack wrote:

> On 10/06/2004 at 04:30:02 AM, cygwin-owner wrote:
> it's shouting "TITTTL" at the top of your voice
> while running round the room flapping your arms like a chicken!
> --- end of excerpt ---
>
> 8-)  Oh yeah  That'll help, near as I can tell there's 95 acronyms
> right now.
> there's just
> "Nothing Quite Like Geeks Going Over Board" (I think I'd pronounce that
> one 'nickel gob')

Use that one a few times, and you'll get it in. :-D

> And just to make this post more on topic (which topic, I don't have a
> clue), here's one for you that you don't want to live without.
>
> >From your command prompt, use:
> > google cygwin acronym list
>
> If that doesn't work, setup this function:
> > typeset -f google
> google ()
> {
> url="http://www.google.com";;
> separator="/search?q=";
> for i in $@;
> do
> url="${url}${separator}${i}";
> separator="+";
> done;
> echo "Starting: $url";
> ~/bin/firefox "$url" &
> }

Or, from your command prompt, use:

> wtf OLOCA

If that doesn't work, install the 'wtf' package.

(and as soon as I find the time to update the 'wtf' package, you'll be
able to also say "wtf TITTTL").
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.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/

RE: Mailing list redirection etiquette?

[Fred Kulack]

On 10/06/2004 at 04:30:02 AM, cygwin-owner wrote:
it's shouting "TITTTL" at the top of your voice
while running round the room flapping your arms like a chicken!
--- end of excerpt ---

8-)  Oh yeah  That'll help, near as I can tell there's 95 acronyms 
right now.
there's just 
"Nothing Quite Like Geeks Going Over Board" (I think I'd pronounce that 
one 'nickel gob')

And just to make this post more on topic (which topic, I don't have a 
clue), 
here's one for you that you don't want to live without.

>From your command prompt, use:
> google cygwin acronym list

If that doesn't work, setup this function:
> typeset -f google
google ()
{
url="http://www.google.com";;
separator="/search?q=";
for i in $@;
do
url="${url}${separator}${i}";
separator="+";
done;
echo "Starting: $url";
~/bin/firefox "$url" &
}






"The stuff we call "software" is not like anything that human society 
  is used to thinking about. Software is something like a machine, and 
  something like mathematics, and something like language, and 
  something like thought, and art, and information... 
  but software is not in fact any of those other things."
Bruce Sterling - The Hacker Crackdown

Fred A. Kulack - IBM eServer iSeries - Enterprise Application Solutions
ERP, Java DB2 access, Jdbc, JTA, etc...
IBM in Rochester, MN  (Phone: 507.253.5982   T/L 553-5982)
mailto:kulack/us.ibm.com   Personal: mailto:kulack/magnaspeed.net
AIM Home:FKulack  AIM Work:FKulackWrk 
MSN Work: fakulack/hotmail.com (replace email / with @)

--
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/

Calls to system() blocks return of SIGCHLD signal

[Remy Gauthier]

We manage synchronous events through calls to system(), and asynchronous events via 
fork() and a SIGCHLD handler. We have noticed (on V1.5.10-1 and V1.5.11-1) that after 
a call to system(), the handlers were not being called after the child process 
stopped. This program has this behaviour (removing the call to system() will restore 
correct SIGCHLD handling):

#include 
#include 
#include 
#include 
typedef void (*SH)(int);
static volatile sig_atomic_t iStop = 0;
static void Handler(int iSigno)
{
 printf("in Handler\n");
 iStop = 1;
}
int main(void)
{
 SH shOld;
 int i;
 int j = 20;

 system ("ls");

 shOld = signal(SIGCHLD, Handler);
 if (shOld == (SH)SIG_ERR)
   goto LErr;
 i = fork ();
 if (i < 0)
   goto LErr;
 if (!i) /* I'm the fork'ed child */
 {
  char *argv [3];
  argv[0] = "/usr/bin/perl";
  argv[1] = "--version";
  argv[2] = NULL;
  execv(argv[0], argv);
  perror("tsig child");
  exit(EXIT_SUCCESS);
 }
 printf ("Child has pid %d\n", i);
 while (!iStop && j)
 {
  printf("Doing something for 1 second\n");
  sleep(1);
  j--;
 }
 printf(j > 0 ? "Child has sent its SIGCHLD signal\n":"Timer has expired\n");
 exit(EXIT_SUCCESS);
 LErr:
  perror("tsig");
  exit(EXIT_FAILURE);
}

We managed to trace this down to spawn.cc in two areas of code:

  if (mode == _P_SYSTEM)
{
  sigset_t child_block;
  cleanup.oldint = signal (SIGINT, SIG_IGN);
  cleanup.oldquit = signal (SIGQUIT, SIG_IGN);
  sigemptyset (&child_block);
  sigaddset (&child_block, SIGCHLD);
  (void) sigprocmask (SIG_BLOCK, &child_block, &cleanup.oldmask);
}

where it seems SIGCHLD is explicitly blocked out on a call to system(), but with a 
memory of previously blocked signals (cleanup.oldmask). Then, when the child has 
completed, a cleanup process is done:

static void
do_cleanup (void *args)
{
# define cleanup ((pthread_cleanup *) args)
  if (cleanup->oldint)
signal (SIGINT, cleanup->oldint);
  if (cleanup->oldquit)
signal (SIGQUIT, cleanup->oldquit);
  if (cleanup->oldmask)
sigprocmask (SIG_SETMASK, &(cleanup->oldmask), NULL);
# undef cleanup
}

It makes sense to re-init the SIGINT and SIGQUIT signals to their previous non default 
handlers, but when the signal mask was previously empty (no blocked signals), the 
sigprocmask call will not be done, therefore leaving SIGCHLD blocked.

We have done a workaround by blocking SIGUSR2 (which we are not using) as part of the 
init process of the application, and this seems to have cured the behaviour problem.

Thank you for looking into this.

Rémy Gauthier.

---

Output of cygcheck -srv is:


Cygwin Configuration Diagnostics
Current System Time: Tue Oct 05 13:25:40 2004

Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 4

Path:   C:\cygwin\usr\local\bin
C:\cygwin\bin
C:\cygwin\usr\X11R6\bin
.\

Output from C:\cygwin\bin\id.exe (nontsec)
UID: 11693(remyg) GID: 10513(Domain Users)
10513(Domain Users)

Output from C:\cygwin\bin\id.exe (ntsec)
UID: 11693(remyg) GID: 10513(Domain Users)
0(root)   11774(Altitude Group)
17163(Comite Directeurs)  10513(Domain Users)
11121(Exceed Users)

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

CYGWIN = `ntsec binmode tty'
HOME = `C:\home\remyg'
PWD = `/home/remyg'

Use `-r' to scan registry

c:  hd  NTFS   11509Mb  88% CP CS UN PA FC 
d:  cd   N/AN/A
h:  net NTFS   341656Mb  84% CP CSPA.
p:  net NTFS2048Mb  58% CP CS UN PA FC DATA

\\scicily\altcm/home/altcm   userbinmode
\\scicily\altdata  /home/altdata userbinmode
\\scicily\altmr/home/altmr   userbinmode
\\scicily\alt  /misc/alt userbinmode
\\everest\cplex-8.0/misc/cplex-8.0   userbinmode
C:\Opt /opt  userbinmode
c:\temp/temp userbinmode
C:\cygwin  / system  binmode
C:\home/home system  binmode
C:\cygwin/bin  /usr/bin  system  binmode
C:\cygwin/lib  /usr/lib  system  binmode
C:\cygwin\usr\X11R6\lib\X11\fonts  /usr/X11R6/lib/X11/fonts  system  binmode
.  /cygdrive system  binmode,cygdrive

Found: C:\cygwin\bin\awk.exe
Found: C:\cygwin\bin\bash.exe
Found: C:\cygwin\bin\cat.exe
Found: C:\cygwin\bin\cp.exe
Found: C:\cygwin\bin\cpp.exe
Found: C:\cygwin\bin\find.exe
Found: C:\cygwin\bin\gcc.exe
Found: C:\cygwin\bin\gdb.exe
Found: C:\cygwin\bin\grep.exe
Found: C:\cygwin\bin\ld.exe
Found: C:\cygwin\bin\ls.exe
Found: C:\cygwin\bin\make.exe
Found: C:\cygwin\bin\mv.exe
Found: C:\cygwin\bin\rm.exe
Found: C:\cygwin\bin\sed.exe
Found: C:\cygwin\bin\sh.exe
Found: C:\cygwin\bin\tar.exe

   92k 2004/06/

Re: Known Issues: document missing POSIX compliance and other unexpected behaviour

[Reini Urban]

Christopher Faylor schrieb:
On Wed, Oct 06, 2004 at 05:11:03PM +0200, Reini Urban wrote:
I found no document where missing POSIX functionality is listed
and where other non-POSIX but expected typical UNIX functionality and/or
behaviour are different or missing. The user guide is a bit short on that.
The only way such a document would be useful would be if it was
rigorously maintained.
ok.
You show the reason why below when you quote out-of-date parts of the
existing documentation.
It is certainly worthwhile to sweep through the docs and get everything
up-to-date, though.

select():
see http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-SELECT
TODO
This points to a page which explains (in an outdated manner)
implementation details of select.  I don't see how it applies.
It applies as a TODO item. :)
processes
-
See the user-guide on fork, ipc, COFF, ...
COFF?
Yes. Imho, the topic "processes" also includes loading a process.
And handling the data structures in which the process is stored,
even if not active ("loaded").
And COFF is good to known/explain in contrast to ELF or a.out.
Also for debugging purposes.
fork:
see http://cygwin.com/faq/faq_3.html#SEC74 and 
http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS

PID's:
Unfortunately cygwin must use seperate PID's than the
underlying windows PID's.
http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS
The documentation is out of date.  This is not true.
I think I'll stop here.  You're showing that it would be worthwhile to
go through the document and look for stuff that is out of date.
I guess that would be a job for Corinna, Pierre, and me, since
no one else seems to understand these issues.  If someone else wants
to take a stab at it, however, please feel free.
Sure, but where?
In the list or elsewhere? As I suggested in a wiki for some while until 
it is in the docs. Seems to work out good, even if it should be 
"rigorously maintained", as you said.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

--
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: Request for a version/ revision/ release number for the whole cygwin release/ distribution

[Corinna Vinschen]

On Oct  6 11:09, Christopher Faylor wrote:
> On Wed, Oct 06, 2004 at 02:47:15PM +0200, Reini Urban wrote:
> >Go with the Redhat scheme and use "Cygwin 1.6.0".
> 
> Please don't do this.  Red Hat is using the even numbers.  I don't know
> if there is already a 1.6.0 [...]

There is.

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/

(De)activating network connection [SOLVED]

[Thiers Botelho]

However off-topic, let me make this last posting in order to report my 
[very positive and interesting] findings on this matter.


1. Microsoft KB article 262265 explains that NETSH.EXE returns an error 
when trying do disable an Ethernet connection:

"Dedicated interfaces cannot be connected, disconnected, enabled, 
or disabled".


2. Microsoft KB article 311272 introduces DEVCON.EXE .

"The DevCon utility is a command line utility that acts as an 
alternative to Device Manager. Using DevCon, you can enable, disable, 
restart, update, remove, and query individual devices or groups of 
devices."


3. I've successfully tested DEVCON.EXE (Win XP SP1) to disable the 
Ethernet connection. As a matter of fact, the entire hardware device 
disappeared from Device Manager. I suppose the procedure can be refined 
(with further testing) to disable just the WAN miniport instead of the 
whole device. As follows:

devcon hwids =net   # query hardware id of all network 
interfaces (print the log and pinpoint which id belongs to the network 
interface)

devcon disable 
PCI\VEN_8086&DEV_1050&SUBSYS_02851014&REV_02\4&3A322F38&0&40F0  # puff ! 
there it goes (fireworks abound if you're connected anywhere)

devcon enable 
PCI\VEN_8086&DEV_1050&SUBSYS_02851014&REV_02\4&3A322F38&0&40F0  # presto ! 
here's it back again (no reboot required)


4. BTW, got this thru 

 
http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&group=microsoft.public.windowsxp.network_web


Cheers


Thiers Botelho


--
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: Known Issues: document missing POSIX compliance and other unexpected behaviour

[Christopher Faylor]

On Wed, Oct 06, 2004 at 05:11:03PM +0200, Reini Urban wrote:
>I found no document where missing POSIX functionality is listed
>and where other non-POSIX but expected typical UNIX functionality and/or
>behaviour are different or missing. The user guide is a bit short on that.

The only way such a document would be useful would be if it was
rigorously maintained.

You show the reason why below when you quote out-of-date parts of the
existing documentation.

It is certainly worthwhile to sweep through the docs and get everything
up-to-date, though.

>select():
>  see http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-SELECT
>  TODO

This points to a page which explains (in an outdated manner)
implementation details of select.  I don't see how it applies.

>processes
>-
>See the user-guide on fork, ipc, COFF, ...

COFF?

>fork:
>see http://cygwin.com/faq/faq_3.html#SEC74 and 
>http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS
>
>PID's:
>Unfortunately cygwin must use seperate PID's than the
>underlying windows PID's.
>http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS

The documentation is out of date.  This is not true.

I think I'll stop here.  You're showing that it would be worthwhile to
go through the document and look for stuff that is out of date.

I guess that would be a job for Corinna, Pierre, and me, since
no one else seems to understand these issues.  If someone else wants
to take a stab at it, however, please feel free.

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: compilig error: storage size of `st' isn't known

[Edward L. Fox]

Hi Jari:

Of course your code cannot be compiled. You should declare "struct stat"
before you use it. The declaration of "struct stat" may be in one of these
files.

#include 
#include 
#include 

Good luck.


Edward L. Fox

2004-10-06


From: [EMAIL PROTECTED]


On Fri, 03 Sep 2004 10:29:45 +0300, Jari Aalto <[EMAIL PROTECTED]> wrote:
> 
> I'm trying to compile Linux program that starts like this:
> 
>int main(int argc, char **argv)
>{
>  char **dirname = NULL;
>  int i,j,n,p,q,dtotal,ftotal,colored = FALSE;
>  struct stat64 st;
> 
> But the compiler says:
> 
>tree.c: In function `main':
>tree.c:204: error: storage size of `st' isn't known
>tree.c:204: warning: unused variable `st'
> 
> What needs to be done in Cygwin to see "struct stat" ?
> 
> Jari
> 
> --
> 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: compilig error: storage size of `st' isn't known (FAQ alert)

[Christopher Faylor]

On Fri, Sep 03, 2004 at 10:29:45AM +0300, Jari Aalto wrote:
>I'm trying to compile Linux program that starts like this:
>
>int main(int argc, char **argv)
>{
>  char **dirname = NULL;
>  int i,j,n,p,q,dtotal,ftotal,colored = FALSE;
>  struct stat64 st;
>
>But the compiler says:
>
>tree.c: In function `main':
>tree.c:204: error: storage size of `st' isn't known
>tree.c:204: warning: unused variable `st'
>
>What needs to be done in Cygwin to see "struct stat" ?

There is no stat64 in cygwin.  Use stat.

This.  Is a recording.

I guess it should be a FAQ.

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/

Known Issues: document missing POSIX compliance and other unexpected behaviour

[Reini Urban]

I found no document where missing POSIX functionality is listed
and where other non-POSIX but expected typical UNIX functionality and/or 
behaviour are different or missing. The user guide is a bit short on that.

This should help UNIX developers and should go the the user guide
(or FAQ?) then.
Unsolved issues (such as some Ghanshyam items) should go to bugzilla then.
I'd like to use a wiki page for that. (bugzilla not applicable)
Which wiki?
I'd suggest http://www.mega-tokyo.com/osfaq2/index.php/Cygwin
(CygwinIssues page) because this runs on my wiki engine :)
We could also abuse a wikipedia subpage of 
http://en.wikipedia.org/wiki/Cygwin.

Before adding your fixes below in a reply, please please let's talk 
first where to do that actually. Mailinglist not preferred.

Known Issues:
=
First see the user-guide http://cygwin.com/cygwin-ug-net/highlights.html
and http://cygwin.com/cygwin-ug-net/using.html ff. for the definitive 
answers.

strictly NON-POSIX:
===
fopen(): Ghanshyam listed several NON-POSIX issues starting with
  http://www.cygwin.com/ml/cygwin/2004-03/msg00357.html
sysconf():
  http://www.cygwin.com/ml/cygwin/2004-03/msg01193.html
strftime():
  http://www.cygwin.com/ml/cygwin/2004-03/msg01192.html
sigaction():
  http://www.cygwin.com/ml/cygwin/2004-03/msg01194.html
TODO: find and document more problems.
Unexpected Behaviour:

filesystem related:
---
case-insensitive search:
  Windows filesystems are usually case-insensitive in search, but
  case-preservant (I know of 5 windows fs in general use, NTFS5 as
  best, a new one on the horizon). For POSIX compatibility there's no
  optional option or flag to restrict searches
  (open,fopen,readdir,unlink,...) to be case-sensitive.
  You can only tell your filesystem to be case-sensitive (somewhere
  in the registry, which is not recommended and unrealistic), use
  special cygwin managed mounts (mount -o managed) or check
  case-sensitivity explicitly in your code. Best is to avoid such
  issues beforehand, not allowing only case-different filenames.
  But with user-created plugins it will get complicated.
unlink():
  * unlink isn't atomic (as on other platforms also)
=> while (!unlink(..) or end-of-waiting()) {sleep}
On cygwin unlink() is deferred to the end of the process,
when unlink fails because of locks or delete on close
(ERROR_SHARING_VIOLATION).
  * unlink a file and subsequent creation of a new file
in the same loction is not supported.
  * "delete on close" (unlinking an open file) is supported.
passing file descriptors:
  * you cannot pass file descriptors between seperate processes.
See http://www.cygwin.com/ml/cygwin-patches/2002-q3/msg3.html
for description and workarounds.
(There's more. clamav came up with a patch recently, which has
a good workaround as easy to use code-snippet)
hardlinks:
  * hardlinking directories is only available to the superuser
'ln -d', ...
  * junctions are almost equivalent to symlinks, but symlinks are
implemented in cygwin differently and slower.
junctions on filenames do not exists to my knowledge, only
native hardlinks, which cygwin supports (for filenames at
least, see above).
(TODO: native symlinks)
mount:
  mounting filesystems and devices are not supported.
  TODO
device limitations:
  explain about our device mapping, special block/char devices,
  other magic devices like sockets, fifo's, ...
  TODO
special filenames:
http://cygwin.com/cygwin-ug-net/using-specialnames.html
* using (esp. unlinking) "NUL" and other special filenames is
  problematic (see FAQ and windows-special "\\.\" and "\??\" paths).
  "DEL \\.\c:\somedir\nul"
getfacl/setfacl:
  see user guide 
  cygwin ACL's are not 1:1 mapped to windows ACL's,
  esp. for inherited directory permissions. (?)
  TODO
select():
  see http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-SELECT
  TODO
sync():
  TODO SUSv3
security:
-
See http://cygwin.com/cygwin-ug-net/ntsec.html, 
http://cygwin.com/faq/faq_3.html#SEC78, sysbash,
network shares, samba, windows profiles, ...

setuid/su/login/cron:
 * UNIX root uid is 0, windows SYSTEM uid is 18.
 * cygwin cron usually runs as user SYSTEM and therefore ...
   (network shares, file permissions, ...) TODO
 * TODO: setuid,setgid file permission flag and workarounds.
 * setuid()/setgid() process workarounds.
see http://cygwin.com/cygwin-ug-net/ntsec.html#NTSEC-SETUID
 * TODO: setsid()/setpgid(): ??
 * TODO: su,sudo vs runas ...
 * TODO: login limitations
processes
-
See the user-guide on fork, ipc, COFF, ...
fork:
see http://cygwin.com/faq/faq_3.html#SEC74 and 
http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS

PID's:
Unfortunately cygwin must use seperate PID's than the
underlying windows PID's.
http://cygwin.com/cygwin-ug-net/highlights.html#OV-HI-PROCESS
IPC:
* requires the cygserver daemon
* not fully supported on Win95 yet. TODO
threads:
* -lpthread

Re: Request for a version/ revision/ release number for the whole cygwin release/ distribution

[Christopher Faylor]

On Wed, Oct 06, 2004 at 02:47:15PM +0200, Reini Urban wrote:
>Karl M schrieb:
>>What about calling it B21? ":>
>
>As already discussed on cygwin-talk and as officially described on the 
>webpage, "B" stood for Beta that times (up to 1998).
>
>We are already stable since a few years, though we use uneven version 
>numbers, marking it as developer releases.
>
>
>So we could use "S1511" (stable 1.5.11)
>or "SS1511" (standalone stable 1.5.11)
>  - I obviously watched a lot of world war movies.
>
>
>Go with the Redhat scheme and use "Cygwin 1.6.0".

Please don't do this.  Red Hat is using the even numbers.  I don't know
if there is already a 1.6.0 or not but there's certainly no reason to
irritate the organization which is providing the network bandwidth for
the cygwin release.

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/

compilig error: storage size of `st' isn't known

[Jari Aalto]

I'm trying to compile Linux program that starts like this:

int main(int argc, char **argv)
{
  char **dirname = NULL;
  int i,j,n,p,q,dtotal,ftotal,colored = FALSE;
  struct stat64 st;

But the compiler says:

tree.c: In function `main':
tree.c:204: error: storage size of `st' isn't known
tree.c:204: warning: unused variable `st'

What needs to be done in Cygwin to see "struct stat" ?

Jari


--
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: Request for a version/ revision/ release number for the whole cygwin release/ distribution

[Christopher Faylor]

On Tue, Oct 05, 2004 at 10:19:29PM -0700, Karl M wrote:
>Hi All...
>
>What about calling it B21? ":>

It's as good a name for vaporware as any.

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: Building plugin that is used from multiple executables

[Reini Urban]

Jani Tiainen schrieb:
Maarten Boekhold wrote:
Suppose I have a plugin library that contains calls that reside in the
executable that dlopen()'s that plugin. You can link such a plugin by
using a .def file that contains something like:
IMPORTS
symbol = executable.exe.symbol
Now if I need to be able to use the *same* plugin from *two*
executables, how do I resolve that??? Can I use a .def file that
contains:
IMPORTS
symbol = executable_1.exe.symbol
symbol = executable_2.exe.symbol
My gut feeling says 'no', but hope against hope...

Well, shortly... You can't do that. Since windows exports are binded to 
name of library (application) and symbol...

Long way: You can do it, but you need to take alternative route.
You need one extra DLL that has imports symbol from executable and 
ex-exports symbol you need forward to plugin:

for plugin you need to do following:
IMPORT
symbol = helper.dll.symbol
and for both apps you create own helper.dll:
app1.exe create helper.dll:
IMPORT
symbol = app1.exe.symbol
EXPORT
symbol
app2.exe create helper.dll:
IMPORT
symbol = app2.exe.symbol
EXPORT
symbol
And finally...
Or even better, DTRT in windoze world is to pass needed symbol pointers 
as arguments to plugin function, since windows libraries and 
applications doesn't allow unresolved symbols in linktime (so called 
backlinking issue)...
hmm,
stubs:
  (intermediate dll, which dlopen's dependent on
  various issues another dll. platform, release, required feature se: 
pro/basic, language)

lazy linking:
  resolve at run-time. as soon as it will be added to ld.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/
--
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: Request for a version/ revision/ release number for the whole Cygwin release/ distribution

[Reini Urban]

Karl M schrieb:
What about calling it B21? ":>
As already discussed on cygwin-talk and as officially described on the 
webpage, "B" stood for Beta that times (up to 1998).

We are already stable since a few years, though we use uneven version 
numbers, marking it as developer releases.


So we could use "S1511" (stable 1.5.11)
or "SS1511" (standalone stable 1.5.11)
  - I obviously watched a lot of world war movies.

Go with the Redhat scheme and use "Cygwin 1.6.0".
That's a unique name (unlike Yggdrasil, John Kerry or Roman Catholic) 
and would only clash with Redhat's supported GNUPro tools.
Which is good IMHO, since it has comparable versions,
we would just add much more packages.

BTW: The link from http://sources.redhat.com/cygwin/
to "Red Hat support contracts" 
http://www.redhat.com/software/tools/cygwin/ fails.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

--
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: Did man break recently?

[Brian Dessent]

Ashok Vadekar wrote:
> 
> After a recent upgrade of what appears to be tetex and x11 stuff, I see the
> following generated on stderr prior to any normal response from a man query:

Read the announcement of the last "man" release.  It explains this.

Brian

--
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] cygrunsrv --recovery

[Corinna Vinschen]

On Oct  6 13:16, Dave Korn wrote:
>   Well, of course, if you're using the API routine rather than going behind
> its back, then when M$ change the registry format, the code will continue to
> work...  that is the whole idea for an API.
> [...]
>   See, I think that the best generic solution to this situation is to write
> code that dynamically links to ChangeServiceConfig2 if available, and falls
> back to writing hard-formatted registry keys/values only if that is not
> available.  That's both backwardly *and* forwardly compatible, as opposed to
> the current solution, which is fragile.  

Sure, that's fine (and done that way in the Cygwin DLL itself).
So now just http://cygwin.com/acronyms/#SHTDI , right? ;-)

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/

Did man break recently?

[Ashok Vadekar]

After a recent upgrade of what appears to be tetex and x11 stuff, I see the
following generated on stderr prior to any normal response from a man query:

Unrecognized line in config file (ignored)
JNROFF  LANG=ja_JP.UTF-8 /usr/bin/groff -Tnippon -mandocj
Unrecognized line in config file (ignored)
KNROFF  /usr/bin/groff -Tkorean -mandoc
Unrecognized line in config file (ignored)
JNEQN   /usr/bin/eqn -Tnippon
Unrecognized line in config file (ignored)
KNEQN   /usr/bin/eqn -Tkorean

Is there a setup dependancy issue perhaps?  I know it is not an issue with the
mks toolkit stuff also instaled (as seen below) because I renamed the
directory to get it out of the path and there was no change in behaviour.

Cygcheck -s reports:

Cygwin Configuration Diagnostics
Current System Time: Wed Oct 06 08:24:12 2004

Windows XP Professional Ver 5.1 Build 2600 Service Pack 2

Path:   D:\cygwin\usr\local\bin
D:\cygwin\bin
D:\cygwin\bin
D:\cygwin\usr\X11R6\bin
.\
D:\cygwin\bin
D:\mks\mksnt
D:\mks\mkssi
C:\PROGRAM FILES\THINKPAD\UTILITIES
C:\WINDOWS\system32
C:\WINDOWS
C:\WINDOWS\System32\Wbem
C:\Program Files\ATI Technologies\ATI Control Panel
C:\Program Files\PC-Doctor for Windows\services

Output from D:\cygwin\bin\id.exe (nontsec)
UID: 1004(avadekar) GID: 513(None)
513(None)

Output from D:\cygwin\bin\id.exe (ntsec)
UID: 1004(avadekar) GID: 513(None)
544(Administrators)  545(Users)

SysDir: C:\WINDOWS\system32
WinDir: C:\WINDOWS

HOME = `E:\'
MAKE_MODE = `UNIX'
PWD = `/home/avadekar'
USER = `avadekar'

Use `-r' to scan registry

c:  hd  NTFS   2Mb  34% CP CS UN PA FC IBM_PRELOAD
d:  hd  NTFS   10003Mb  89% CP CS UN PA FC DEVELOPMENT
e:  hd  NTFS8002Mb  51% CP CS UN PA FC WORK
f:  cd   N/AN/A
p:  hd  NTFS   12004Mb  46% CP CS UN PA FC PACKAGES
t:  hd  NTFS3861Mb  28% CP CS UN PA FC Temp

D:\cygwin  / system  binmode
C: /csystem  binmode
D: /dsystem  binmode
E: /esystem  binmode
F: /fsystem  binmode
G: /gsystem  binmode
d:\sdks\eracom\fm-sdk  /gcc  system  textmode
P: /packages system  binmode
T: /tsystem  binmode
D:\cygwin/bin  /usr/bin  system  binmode
D:\cygwin/lib  /usr/lib  system  binmode
D:\cygwin\usr\X11R6\lib\X11\fonts  /usr/X11R6/lib/X11/fonts  system  binmode
V: /vsystem  binmode
.  /cygdrive system  binmode,cygdrive

Found: D:\mks\mksnt\awk.exe
Found: D:\cygwin\bin\bash.exe
Found: D:\cygwin\bin\cat.exe
Found: D:\mks\mksnt\cat.exe
Found: D:\cygwin\bin\cp.exe
Found: D:\mks\mksnt\cp.exe
Found: D:\cygwin\bin\cpp.exe
Found: D:\cygwin\bin\find.exe
Found: D:\mks\mksnt\find.exe
Found: D:\cygwin\bin\gcc.exe
Found: D:\cygwin\bin\gdb.exe
Found: D:\cygwin\bin\grep.exe
Found: D:\mks\mksnt\grep.exe
Found: D:\cygwin\bin\ld.exe
Found: D:\cygwin\bin\ls.exe
Found: D:\mks\mksnt\ls.exe
Found: D:\cygwin\bin\make.exe
Found: D:\mks\mksnt\make.exe
Found: D:\cygwin\bin\mv.exe
Found: D:\mks\mksnt\mv.exe
Found: D:\cygwin\bin\rm.exe
Found: D:\mks\mksnt\rm.exe
Found: D:\cygwin\bin\sed.exe
Found: D:\mks\mksnt\sed.exe
Found: D:\cygwin\bin\sh.exe
Found: D:\mks\mksnt\sh.exe
Found: D:\cygwin\bin\tar.exe
Found: D:\mks\mksnt\tar.exe

   55k 2004/09/14 D:\cygwin\bin\cygbz2-1.dll
   18k 2004/07/06 D:\cygwin\bin\cygcharset-1.dll
7k 2003/10/19 D:\cygwin\bin\cygcrypt-0.dll
  839k 2004/09/27 D:\cygwin\bin\cygcrypto-0.9.7.dll
   22k 2004/02/10 D:\cygwin\bin\cygcygipc-2.dll
  831k 2003/09/20 D:\cygwin\bin\cygdb-4.1.dll
  895k 2004/04/28 D:\cygwin\bin\cygdb-4.2.dll
 1080k 2003/09/20 D:\cygwin\bin\cygdb_cxx-4.1.dll
 1156k 2004/04/28 D:\cygwin\bin\cygdb_cxx-4.2.dll
  155k 2004/01/07 D:\cygwin\bin\cygexpat-0.dll
  129k 2004/03/11 D:\cygwin\bin\cygfontconfig-1.dll
   45k 2001/04/25 D:\cygwin\bin\cygform5.dll
   35k 2002/01/09 D:\cygwin\bin\cygform6.dll
   48k 2003/08/09 D:\cygwin\bin\cygform7.dll
  361k 2003/10/25 D:\cygwin\bin\cygfreetype-6.dll
   28k 2003/07/20 D:\cygwin\bin\cyggdbm-3.dll
   30k 2003/08/11 D:\cygwin\bin\cyggdbm-4.dll
   19k 2003/03/22 D:\cygwin\bin\cyggdbm.dll
   15k 2003/07/20 D:\cygwin\bin\cyggdbm_compat-3.dll
   15k 2003/08/11 D:\cygwin\bin\cyggdbm_compat-4.dll
  107k 2004/07/06 D:\cygwin\bin\cyggettextlib-0-14-1.dll
   17k 2004/07/06 D:\cygwin\bin\c

Re: Building plugin that is used from multiple executables

[Jani Tiainen]

Maarten Boekhold wrote:
Hi,
Suppose I have a plugin library that contains calls that reside in the
executable that dlopen()'s that plugin. You can link such a plugin by
using a .def file that contains something like:
IMPORTS
symbol = executable.exe.symbol
Now if I need to be able to use the *same* plugin from *two*
executables, how do I resolve that??? Can I use a .def file that
contains:
IMPORTS
symbol = executable_1.exe.symbol
symbol = executable_2.exe.symbol
My gut feeling says 'no', but hope against hope...
Well, shortly... You can't do that. Since windows exports are binded to 
name of library (application) and symbol...

Long way: You can do it, but you need to take alternative route.
You need one extra DLL that has imports symbol from executable and 
ex-exports symbol you need forward to plugin:

for plugin you need to do following:
IMPORT
symbol = helper.dll.symbol
and for both apps you create own helper.dll:
app1.exe create helper.dll:
IMPORT
symbol = app1.exe.symbol
EXPORT
symbol
app2.exe create helper.dll:
IMPORT
symbol = app2.exe.symbol
EXPORT
symbol
And finally...
Or even better, DTRT in windoze world is to pass needed symbol pointers 
as arguments to plugin function, since windows libraries and 
applications doesn't allow unresolved symbols in linktime (so called 
backlinking issue)...

--
Jani Tiainen
--
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] cygrunsrv --recovery

[Dave Korn]

> -Original Message-
> From: cygwin-owner On Behalf Of Corinna Vinschen
> Sent: 06 October 2004 13:06

> On Oct  6 13:52, Rainer Hochreiter wrote:
> > On Wed, 06 Oct 2004 12:17:49 +0200, Corinna Vinschen wrote:
> > > Erm?!?  Of course you should write directly to the 
> registry.  That's how
> > > writing the description field is implemented to stay NT4 
> compatible.  So
> > > just copy the behaviour for writing the actions parameter.
> > 
> > but who guarantees that writing directly to the registry will be 
> > compliant to upcoming windows version?
> 
> Nobody.  But in that case we change cygrunsrv to write the parameters
> differently into the registry if newer NT versions require that, no?
> 

  Well, of course, if you're using the API routine rather than going behind
its back, then when M$ change the registry format, the code will continue to
work...  that is the whole idea for an API.

> - Please don't use ChangeServiceConfig2.  It will break running cygrunsrv
>   on NT4.  That's the reason the description is written directly to the
>   registry instead of using ChangeServiceConfig2.

  See, I think that the best generic solution to this situation is to write
code that dynamically links to ChangeServiceConfig2 if available, and falls
back to writing hard-formatted registry keys/values only if that is not
available.  That's both backwardly *and* forwardly compatible, as opposed to
the current solution, which is fragile.  


cheers, 
  DaveK
-- 
Can't think of a witty .sigline today


--
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] cygrunsrv --recovery

[Corinna Vinschen]

On Oct  6 13:52, Rainer Hochreiter wrote:
> On Wed, 06 Oct 2004 12:17:49 +0200, Corinna Vinschen wrote:
> > Erm?!?  Of course you should write directly to the registry.  That's how
> > writing the description field is implemented to stay NT4 compatible.  So
> > just copy the behaviour for writing the actions parameter.
> 
> but who guarantees that writing directly to the registry will be 
> compliant to upcoming windows version?

Nobody.  But in that case we change cygrunsrv to write the parameters
differently into the registry if newer NT versions require that, no?

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/

Re: [PATCH] cygrunsrv --recovery

[Rainer Hochreiter]

On Wed, 06 Oct 2004 12:17:49 +0200, Corinna Vinschen wrote:

> On Oct  6 11:57, Rainer Hochreiter wrote:
>> well, are there any guidelines how to submit patches and describing
>> ChangeLog syntax available?
> 
> http://cygwin.com/contrib.html, section "When you have finalized your changes"

thanks for the link - i'll have a look at it!

> Just ignore the hint to send patches to the cygwin-patches list.
> 
>> right! but if using ChangeServiceConfig2() is prohibited, why not writing 
>> also directly to registry?
> 
> Erm?!?  Of course you should write directly to the registry.  That's how
> writing the description field is implemented to stay NT4 compatible.  So
> just copy the behaviour for writing the actions parameter.

but who guarantees that writing directly to the registry will be 
compliant to upcoming windows version? this will also be a problem
for all the other config parameters written by cygrunsrv!

>> ok, maybe there is no public documentation
>> available for that. what i found out is, that SCM writes into a key named
>> 'FailureActions' of type REG_BINARY. anyone out here, who knows what to write
>> there? best would be the source of the sc.exe command ;-)
> 
> I guess the easy way is to use ChangeServiceConfig2 with different
> arguments and look what it creates as result.

hm, reverse-engineering? sounds great ;-)

rainer



--
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: (De)activating network connection

[Jani Tiainen]

Brian Dessent wrote:
Thiers Botelho wrote:

I wish to use some command within a script to activate / deactivate an
Ethernet connection.
In WinXP this would normally be done thru Settings / Network Connections /
Local Connection / right-click .
I found _rasdial.exe_ but it seems to deal only with dial-up connections.
There's got to be some simple cmdline on/off switch somewhere (hope that
won't be regedit . . .)

Starting with XP Microsoft realized that it sure would be nice to be
able to set networking parameters (interfaces, addresses, routing, etc)
from the command line or scripts.  [Gee, what an idea, huh?  Took them
long enough to think of that.]
Try the following or something like it:
netsh int set interface "Local Area Connection" admin=ENABLED
Replace ENABLED with DISABLED as appropriate.  I'm not entirely sure of
the syntax but you can just run netsh and play around with ? and help to
get it down.
At least I have this netsh.exe available in my win2k system.. Never 
heard about it before this but seems very nice tool indeed.

--
Jani Tiainen
--
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/

Building plugin that is used from multiple executables

[Maarten Boekhold]

Hi,

Suppose I have a plugin library that contains calls that reside in the
executable that dlopen()'s that plugin. You can link such a plugin by
using a .def file that contains something like:

IMPORTS
symbol = executable.exe.symbol

Now if I need to be able to use the *same* plugin from *two*
executables, how do I resolve that??? Can I use a .def file that
contains:

IMPORTS
symbol = executable_1.exe.symbol
symbol = executable_2.exe.symbol

My gut feeling says 'no', but hope against hope...

Maarten

--
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 + CVS But no login.

[Corinna Vinschen]

On Oct  6 06:27, [EMAIL PROTECTED] wrote:
> >Urgh!  SYSTEM is the account sshd is running under.  Don't remove that.
> 
> Various Guest/System accounts that weren't the SYSTEM account, I removed
> IUSR_MOTHER and IWAM_MOTHER, the ASP.NET user etc.

Ah, ok.  That makes it clearer.

> Without an entry in the cygwin passwd file a user with a valid windows
> login can not authenticate into cygwin.
> 
> Was this wrong?

I just jumped up when I read the word "system".  I'd remove only additional
accounts which are not used anyway (as the above noted IUSR_foo accounts). 
Personally I'm in favor of keeping these accounts in the file and just do
one or many of

- Disable the account in the Windows user management console
- Set the login shell in /etc/passwd to /bin/false
- Use the AllowUsers and DenyUsers entries in sshd_config


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/

Re: [PATCH] cygrunsrv --recovery

[Corinna Vinschen]

On Oct  6 11:22, Dave Korn wrote:
>   Oh.  I should have said "which includes a broken link.  Try
> 
> http://www.gnu.org/prep/standards/html_node/Change-Logs.html#Change-Logs 
> 
> instead".  Ping webmaster!

Changed.  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/

Re: SSH + CVS But no login.

[[EMAIL PROTECTED]]

>Urgh!  SYSTEM is the account sshd is running under.  Don't remove that.

Various Guest/System accounts that weren't the SYSTEM account, I removed
IUSR_MOTHER and IWAM_MOTHER, the ASP.NET user etc.

>I'm wondering why you *first* removed entries from passwd and *then*
>asked if that's right instead of vice versa (which would be the natural
>order if one don't know, wouldn't it?)

I have a vague memory of having done this previously in an older cygwin
instance on another box. I did of course back up my passwd file first, as
always, and have managed to log into the box fine as myself. But also:

>> Ah, yeah. Um. I have a nasty head cold and was in a rush.

I don't feel too bright at the moment.

Under cygwin, I understood that the passwd file essentialy maps windows
users to cygwin settings.
username:unused:uid:grpid:windows account:home:shell

Without an entry in the cygwin passwd file a user with a valid windows
login can not authenticate into cygwin.

Was this wrong?

Mike


mail2web - Check your email from the web at
http://mail2web.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/

RE: [PATCH] cygrunsrv --recovery

[Dave Korn]

> -Original Message-
> From: cygwin-owner On Behalf Of Rainer Hochreiter
> Sent: 06 October 2004 10:57

> well, are there any guidelines how to submit patches and describing
> ChangeLog syntax available?

  Check out http://cygwin.com/contrib.html, which includes a link to
http://www.gnu.org/prep/standards_42.html for the changelog syntax.

  Oh.  I should have said "which includes a broken link.  Try

http://www.gnu.org/prep/standards/html_node/Change-Logs.html#Change-Logs 

instead".  Ping webmaster!

> > That's the change from direct writing of registry keys to using
> > that ChangeServiceConfig2 function.
> 
> right! but if using ChangeServiceConfig2() is prohibited, why 
> not writing 
> also directly to registry? ok, maybe there is no public documentation
> available for that. what i found out is, that SCM writes into 
> a key named
> 'FailureActions' of type REG_BINARY. anyone out here, who 
> knows what to write
> there? best would be the source of the sc.exe command ;-)

  Looks to me like it's a struct _SERVICE_FAILURE_ACTIONS, with a mysterious extra
DWORD between the cActions member and the SC_ACTION array.


cheers, 
  DaveK
-- 
Can't think of a witty .sigline today


contrib-patch.diff
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/

Re: [PATCH] cygrunsrv --recovery

[Corinna Vinschen]

On Oct  6 11:57, Rainer Hochreiter wrote:
> well, are there any guidelines how to submit patches and describing
> ChangeLog syntax available?

http://cygwin.com/contrib.html, section "When you have finalized your changes"

Just ignore the hint to send patches to the cygwin-patches list.

> right! but if using ChangeServiceConfig2() is prohibited, why not writing 
> also directly to registry?

Erm?!?  Of course you should write directly to the registry.  That's how
writing the description field is implemented to stay NT4 compatible.  So
just copy the behaviour for writing the actions parameter.

> ok, maybe there is no public documentation
> available for that. what i found out is, that SCM writes into a key named
> 'FailureActions' of type REG_BINARY. anyone out here, who knows what to write
> there? best would be the source of the sc.exe command ;-)

I guess the easy way is to use ChangeServiceConfig2 with different
arguments and look what it creates as result.


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/

Re: DLL rebase required?

[Jason Tishler]

Maarten,

Please post instead of sending private email.

On Wed, Oct 06, 2004 at 01:10:53PM +0400, Maarten Boekhold wrote:
> I wanted to run this by you before I hose my system with a potentially
> unnecessary rebaseall...
> 
> I'm porting XFCE to cygwin/X, most of it is running perfectly now. One
> of the executables loads plugins (g_module_open etc.), and it fails on
> two of the plugins with:
> 
> dlopen, Win32 error 193
> 
> Some of the references to this I found in google suggest I need to run
> rebaseall to fix this. However, your README for rebase implies that:
> 
> 1) there will be a fairly specific error message if rebase is required
> (eg "unable to remap C:\cygwin\bin\cygssl.dll to same address as
> parent")
> 2) it is only relevant for fork(), and this executable doesn't fork...
> 
> Do you think a rebase would help

No, because we have the following:

$ fgrep 193L /usr/include/w32api/winerror.h 
#define ERROR_BAD_EXE_FORMAT 193L

> or should I look for the solution somewhere else?

Yes, it sounds like a build problem.

Jason

-- 
PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers
Fingerprint: 7A73 1405 7F2B E669 C19D  8784 1AFD E4CC ECF4 8EF6

--
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 + CVS But no login.

[Corinna Vinschen]

On Oct  6 07:52, Michael Jervis wrote:
> >Do you Google: ?
> 
> Ah, yeah. Um. I have a nasty head cold and was in a rush. No idea why I
> didn't google but that works just lovely thank you.
> 
> Next question:
> 
> I've just deleted the various Guest/System accounts from my passwd file.
> This wont' break anything will it? And it's all I need to do to stop people
> logging in as guest?

Urgh!  SYSTEM is the account sshd is running under.  Don't remove that.

I'm wondering why you *first* removed entries from passwd and *then*
asked if that's right instead of vice versa (which would be the natural
order if one don't know, wouldn't it?)

Don't mess around with passwd and group files, except you know what
you're doing.


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/

[SOLVED] Re: How to make setup not look for external mirrors

[Carlo Florendo]

Hello Dave and Bertrand,
Great advice!  Worked well.
Thank you very much!
Best Regards,
Carlo
Dave Korn wrote:
-Original Message-
From: cygwin-owner On Behalf Of Carlo Florendo
Sent: 06 October 2004 08:51
   

 

Hello great cygwin people!
I have mirrored cygwin and made the installation accessible 
via http on 
our local web server.  There are some boxes that may access only LAN 
resources and I want those boxes to have cygwin.

When I start running setup on those boxes, I select download from 
internet (since I have the entire distro on the local http server).  
Thus, I assume that the local web server accessible via http is the 
"internet" site where cygwin will be installed from.  

Everything is fine until I get to the point of downloading 
the list of 
mirrors.  Since the boxes do not have Internet connections, 
they won't 
be able to download the mirror list.  Installation stops from 
that point. 

Is there any way via setup to install cygwin from my local 
http server?
   

 Build it from the source.  First, read the instructions at
http://sources.redhat.com/cygwin-apps/setup.html 

to see how to get the source from cvs and what configure options you'll need.
Edit the res.rc file.  Find the resource given by IDS_MIRROR_LST and change it to
point to your local mirror, then put your own mirrors.txt file on your local
mirror with only one entry in it.  This patch will be b0rked by line wrapping, but
it shows you the idea.
Index: res.rc
===
RCS file: /cvs/cygwin-apps/setup/res.rc,v
retrieving revision 2.56
diff -p -w -u -r2.56 res.rc
--- res.rc  29 Aug 2004 16:59:40 -  2.56
+++ res.rc  6 Oct 2004 09:42:32 -
@@ -462,7 +462,7 @@ STRINGTABLE DISCARDABLE
BEGIN
IDS_ROOT_SLASH  "Warning: we recommend you do NOT use the root of y
our hard drive as the cygwin root.  Proceed anyway?"
IDS_ROOT_SPACE  "You should not choose a root path that include spa
ces in directory names.  Proceed anyway?"
-IDS_MIRROR_LST  "http://sources.redhat.com/cygwin/mirrors.lst";
+IDS_MIRROR_LST  "http://your.local.server/cygwin/mirrors.lst";
IDS_DIALOG_FAILED   "Unable to create Dialog Box"
IDS_CYGWIN_FUNC_MISSING "Error: unable to find function `%s' in %s"
IDS_DOWNLOAD_SHORT  "Download error: %s too short (%d, wanted %d)"
 You may also need to delete the following lines from autoload.c if you get
link-time errors.
Index: autoload.c
===
RCS file: /cvs/cygwin-apps/setup/autoload.c,v
retrieving revision 2.5
diff -p -w -u -r2.5 autoload.c
--- autoload.c  30 Aug 2004 16:17:52 -  2.5
+++ autoload.c  6 Oct 2004 09:42:32 -
@@ -48,12 +48,6 @@ Auto (wininet, HttpSendRequestA, 20);
DLL (advapi32);
-Auto (advapi32, AddAccessAllowedAce, 16);
-Auto (advapi32, AllocateAndInitializeSid, 44);
-Auto (advapi32, FreeSid, 4);
-Auto (advapi32, InitializeAcl, 12);
-Auto (advapi32, OpenProcessToken, 12);
-Auto (advapi32, SetTokenInformation, 16);
Auto (advapi32, OpenSCManagerA, 16);
Auto (advapi32, CloseServiceHandle, 4);
Auto (advapi32, OpenServiceA, 16);

   cheers, 
 DaveK
 


--
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: Request for a version/ revision/ release number for the whole Cygwin release/ distribution

[Corinna Vinschen]

On Oct  5 22:19, Karl M wrote:
> Hi All...
> 
> What about calling it B21? ":>

*cough*

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/

Re: [PATCH] cygrunsrv --recovery

[Rainer Hochreiter]

On Thu, 30 Sep 2004 18:15:45 +0200, Corinna Vinschen wrote:

> On Sep 30 12:06, Igor Pechtchanski wrote:
>> On Thu, 30 Sep 2004, Corinna Vinschen wrote:
>> And when you resubmit, please use "diff -u" instead of "diff -c" -- the
>> patch is much more readable that way.
> 
> ACK.

well, are there any guidelines how to submit patches and describing
ChangeLog syntax available?

>> One immediate change I noticed was that the description parameter migrated
>> from calls to install_registry_keys to calls to install_service, but you
>> haven't provided any changes to either of those functions.  Any reason for
>> this?
> 
> That's the change from direct writing of registry keys to using
> that ChangeServiceConfig2 function.

right! but if using ChangeServiceConfig2() is prohibited, why not writing 
also directly to registry? ok, maybe there is no public documentation
available for that. what i found out is, that SCM writes into a key named
'FailureActions' of type REG_BINARY. anyone out here, who knows what to write
there? best would be the source of the sc.exe command ;-)

-- rainer



--
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 to make setup not look for external mirrors

[Dave Korn]

> -Original Message-
> From: cygwin-owner On Behalf Of Carlo Florendo
> Sent: 06 October 2004 08:51

> Hello great cygwin people!
> 
> I have mirrored cygwin and made the installation accessible 
> via http on 
> our local web server.  There are some boxes that may access only LAN 
> resources and I want those boxes to have cygwin.
> 
> When I start running setup on those boxes, I select download from 
> internet (since I have the entire distro on the local http server).  
> Thus, I assume that the local web server accessible via http is the 
> "internet" site where cygwin will be installed from.  
> 
> Everything is fine until I get to the point of downloading 
> the list of 
> mirrors.  Since the boxes do not have Internet connections, 
> they won't 
> be able to download the mirror list.  Installation stops from 
> that point. 
> 
> Is there any way via setup to install cygwin from my local 
> http server?

  Build it from the source.  First, read the instructions at

http://sources.redhat.com/cygwin-apps/setup.html 

to see how to get the source from cvs and what configure options you'll need.
Edit the res.rc file.  Find the resource given by IDS_MIRROR_LST and change it to
point to your local mirror, then put your own mirrors.txt file on your local
mirror with only one entry in it.  This patch will be b0rked by line wrapping, but
it shows you the idea.

Index: res.rc
===
RCS file: /cvs/cygwin-apps/setup/res.rc,v
retrieving revision 2.56
diff -p -w -u -r2.56 res.rc
--- res.rc  29 Aug 2004 16:59:40 -  2.56
+++ res.rc  6 Oct 2004 09:42:32 -
@@ -462,7 +462,7 @@ STRINGTABLE DISCARDABLE
 BEGIN
 IDS_ROOT_SLASH  "Warning: we recommend you do NOT use the root of y
our hard drive as the cygwin root.  Proceed anyway?"
 IDS_ROOT_SPACE  "You should not choose a root path that include spa
ces in directory names.  Proceed anyway?"
-IDS_MIRROR_LST  "http://sources.redhat.com/cygwin/mirrors.lst";
+IDS_MIRROR_LST  "http://your.local.server/cygwin/mirrors.lst";
 IDS_DIALOG_FAILED   "Unable to create Dialog Box"
 IDS_CYGWIN_FUNC_MISSING "Error: unable to find function `%s' in %s"
 IDS_DOWNLOAD_SHORT  "Download error: %s too short (%d, wanted %d)"


  You may also need to delete the following lines from autoload.c if you get
link-time errors.

Index: autoload.c
===
RCS file: /cvs/cygwin-apps/setup/autoload.c,v
retrieving revision 2.5
diff -p -w -u -r2.5 autoload.c
--- autoload.c  30 Aug 2004 16:17:52 -  2.5
+++ autoload.c  6 Oct 2004 09:42:32 -
@@ -48,12 +48,6 @@ Auto (wininet, HttpSendRequestA, 20);

 DLL (advapi32);

-Auto (advapi32, AddAccessAllowedAce, 16);
-Auto (advapi32, AllocateAndInitializeSid, 44);
-Auto (advapi32, FreeSid, 4);
-Auto (advapi32, InitializeAcl, 12);
-Auto (advapi32, OpenProcessToken, 12);
-Auto (advapi32, SetTokenInformation, 16);
 Auto (advapi32, OpenSCManagerA, 16);
 Auto (advapi32, CloseServiceHandle, 4);
 Auto (advapi32, OpenServiceA, 16);



cheers, 
  DaveK
-- 
Can't think of a witty .sigline today


--
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: Mailing list redirection etiquette?

[Dave Korn]

> From: cygwin-owner On Behalf Of Brian Ford
> Sent: 05 October 2004 23:46

> On Tue, 5 Oct 2004, Christopher Faylor wrote:
> > On Tue, Oct 05, 2004 at 04:37:35PM -0500, Fred Kulack wrote:
> > >Getting way off topic here,
> >
> > ...which makes it a better topic for the cygwin-talk mailing list...
> 
> So what is the recommended Cygwin method for redirecting a 
> thread to the
> appropriate list?


  In the case of the talk list, it's shouting "TITTTL" at the top of your voice
while running round the room flapping your arms like a chicken!

cheers, 
  DaveK
-- 
Can't think of a witty .sigline today


--
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 to make setup not look for external mirrors

[bertrand marquis]

Hello,

there is an option to the setup in command line specifying the mirror
site, perhaps could you try that.

see http://sources.redhat.com/ml/cygwin-apps/2003-03/msg00526.html for
more info.

bertrand

Le mer 06/10/2004 Ã 09:50, Carlo Florendo a Ãcrit :
> Hello great cygwin people!
> 
> I have mirrored cygwin and made the installation accessible via http on 
> our local web server.  There are some boxes that may access only LAN 
> resources and I want those boxes to have cygwin.
> 
> When I start running setup on those boxes, I select download from 
> internet (since I have the entire distro on the local http server).  
> Thus, I assume that the local web server accessible via http is the 
> "internet" site where cygwin will be installed from.  
> 
> Everything is fine until I get to the point of downloading the list of 
> mirrors.  Since the boxes do not have Internet connections, they won't 
> be able to download the mirror list.  Installation stops from that point. 
> 
> Is there any way via setup to install cygwin from my local http server?
> 
> Thanks!
> 
> Best Regards,
> 
> Carlo
> --
> Carlo Florendo
> Astra Philippines Inc.
> www.astra.ph
> 
> 
> 
> 
> 
> --
> 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 to make setup not look for external mirrors

[Carlo Florendo]

Hello great cygwin people!
I have mirrored cygwin and made the installation accessible via http on 
our local web server.  There are some boxes that may access only LAN 
resources and I want those boxes to have cygwin.

When I start running setup on those boxes, I select download from 
internet (since I have the entire distro on the local http server).  
Thus, I assume that the local web server accessible via http is the 
"internet" site where cygwin will be installed from.  

Everything is fine until I get to the point of downloading the list of 
mirrors.  Since the boxes do not have Internet connections, they won't 
be able to download the mirror list.  Installation stops from that point. 

Is there any way via setup to install cygwin from my local http server?
Thanks!
Best Regards,
Carlo
--
Carlo Florendo
Astra Philippines Inc.
www.astra.ph


--
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: g++ 3.4.x library issue

[Gerrit P. Haase]

Hallo Steven,

Am Mittwoch, 6. Oktober 2004 um 02:23 schriebst du:

> Greetings!

> I tried downloading g++ version 3.4.1 (which I've been
> eagerly awaiting) from setup.exe.  It had trouble with
> the following (minimal) program because the 
> header could not find  where it expected to
> find size_t defined.  Trying to compile this one line
> program using 'g++ test.cpp' produced 500 lines of
> errors :)

> #include 
> int main(){return 0;}

Works for me(tm):

[EMAIL PROTECTED] /gcc-bugs
$ g++ -o g++-bug.o g++-bug.C 

[EMAIL PROTECTED] /gcc-bugs
$ 

Gerrit
-- 
=^..^=



--
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: setsockopt SO_REUSEADDR

[Rainer Hochreiter]

On Tue, 05 Oct 2004 09:54:39 -0500, Brian Ford wrote:


> BTW to the OP, why are you printing the address of the socket function
> out?  I think you meant to print "s" instead of "socket"?

oops - you are right! i wanted to print the socket and not the 
address of the socket() function!

;-) rainer


--
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/