date:20040910

On Fri, Sep 10, 2004 at 09:29:35PM -0400, Pierre A. Humblet wrote:
At 12:15 AM 9/8/2004 -0400, Christopher Faylor wrote:
On Tue, Sep 07, 2004 at 09:26:02PM -0400, Pierre A. Humblet wrote:

Also, on WinME, simply holding down ^C in the bash shell will
cause a crash (thanks to Errol Smith)
~ 142 [sig] BASH 1853149 handle_threadlist_exception: 
handle_threadlist_exception called with threadlist_ix -1
1751 [sig] BASH 1853149 handle_exceptions: Exception: 
STATUS_ACCESS_VIOLATION

Any idea about what's happening? I have been unable to
make any progress.

I'll see if I can duplicate the problem with VMware.  That's the only
WinME system that I have available to me currently.

Should we appeal for the donation of a small laptop? It wouldn't
crowd your office much. BTW, I keep my Win95 in my guestroom,
so my guests can check their e-mail.

Here are 2 other small changes.

When killing the (system stressing) program I used to track the
fork bug, I noticed that it occasionally leaves stray processes
behind, as in
821 49315459   1  557307 42456518370  740 20:50:10 /c/HOME/PIERRE/A
0x821 = PID_INITIALIZING + PID_ORPHANED + PID_IN_USE
What's happening is that there is a race between the winpids
scanning in kill_pgrp and the creation of new processes.
The patch in fork.cc improves the situation by exiting
children immediately when the parent dies during a fork.

Also a ^C immediately terminates processes that are initializing.
That is not desirable for processes created by Cygwin processes. 
The patch in exceptions.cc changes that.

I don't understand.  You're removing an exit on a CTRL-C in
exceptions.cc and essentially adding it to fork.cc.  How is that an
improvement?  Why aren't both parent and child going away when they see a
CTRL-C already?  If the child process has not become fully awake yet
it should just exit now.

cgf

Re: Is it free to use

Andrew DeFaria wrote:

 Christopher Faylor wrote:
 
  This means that if you are developing software for eventual release,
  you must also make the source code available when you make binaries
  available.
 
 If I develop an app and do not wish to have a requirement to install
 Cygwin I would use MingW, right? In that case is my app still under GPL
 just because I used Cygwin's gcc to compile and link the resulting
 executable that will have no more dependencies on Cygwin? (Just curious).

The part that causes it to become GPL'd is the linking to cygwin1.dll,
not the fact that Cygwin's gcc is used.  If you use Cygwin's gcc in
mingw mode then your program does not need cygwin1.dll and the program
may be released under any license you choose, assuming there are no
other GPL libraries to which you link.  cgf meant that by default (with
no command line options to the contrary) Cygwin's gcc creates a binary
linked against the Cygwin dll.

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: Is it free to use

2004-09-10 Thread Andrew DeFaria

Brian Dessent wrote:
The part that causes it to become GPL'd is the linking to cygwin1.dll, 
not the fact that Cygwin's gcc is used. If you use Cygwin's gcc in 
mingw mode then your program does not need cygwin1.dll and the program 
may be released under any license you choose, assuming there are no 
other GPL libraries to which you link. 
That was my assumption. The parts I was a little skeptical about was: 1) 
whether using mingw was also GPL and 2) whether I created an app that 
linked in another library that Cygwin provides.

--
Real programmers don't document. If it was hard to write, it should be 
to understand.

--
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 on Cygwin Immediate Drops Connections

2004-09-10 Thread Mark Eret

Hey there again,

I ran /usr/sbin/sshd -d -d -d -e and got the follow output:

debug2: load_server_config: filename /etc/sshd_config
debug2: load_server_config: done config len = 238
debug2: parse_server_config: config /etc/sshd_config len 238
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 238
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 238
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 4733
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
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-hellma
n-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,r
[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,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-md5-96
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-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-hellma
n-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,r
[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,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-md5-96
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-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 hmac-md5 none
debug2: mac_init: found hmac-md5
debug1: kex: server-client aes128-cbc hmac-md5 none
debug1: SSH2_MSG_KEX_DH_GEX_REQUEST received
debug1: SSH2_MSG_KEX_DH_GEX_GROUP sent
debug2: dh_gen_key: priv key bits set: 117/256
debug2: bits set: 515/1024
debug1: expecting SSH2_MSG_KEX_DH_GEX_INIT
debug2: bits set: 490/1024
debug1: SSH2_MSG_KEX_DH_GEX_REPLY sent
debug2: kex_derive_keys
debug2: set_newkeys: mode 1
debug1: SSH2_MSG_NEWKEYS sent
debug1: expecting SSH2_MSG_NEWKEYS
debug2: set_newkeys: mode 0
debug1: SSH2_MSG_NEWKEYS received
debug1: KEX done
debug1: userauth-request for user eret service ssh-connection method none
debug1: attempt 0 failures 0
debug2: input_userauth_request: setting up authctxt for eret
debug2: input_userauth_request: try method none
Failed none for eret from 127.0.0.1 port 4733 ssh2
debug1: userauth-request for user eret service ssh-connection method
keyboard-interactive
debug1: attempt 1 failures 1
debug2: input_userauth_request: try method keyboard-interactive
debug1:

Re: [PATCH] /etc/profile fixes

2004-09-10 Thread John Morrison

Hi Tero

Thanks for the comments, more inline...

 I have to proposals for /etc/profile of latest Cygwin:

 1) Shut up 'tr' which produces error message at least
 with zsh ('zsh' doesn't match to [:upper:]).

It's better to fix than to silence.  Putting quotes around
the [:xxx:] strings stops zsh expanding (I presume) the
string.

 2) Set same default settings for zsh as for ksh.

Thats not a problem :)

Thanks for sending this - I don't use [zk]sh.  I'll patch
things soon.

J.


--
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 do I define a key binding for copy-to-clipboard?

David Tombs wrote:

 I'm a Windows user trying to install and configure cygwin. Thus far I've
 managed to install bash and, mostly, configure it to my preferences.
 However, there is one thing missing. I want to use the clipboard to
 transfer text to and from Windows, with the same key bindings, viz
 Ctrl-C for copy and Ctrl-V for paste. (My left finger muscles are
 hardwired to these keys and anyway, I hate using the mouse for this kind
 of action.)

The short answer is that if you're using the default Command Prompt
then cmd.exe is handling the selection and copy operations and there's
no way to tell it to use ctrl-C as you want.  The pasting is handled
entirely by  bash so as you have found it's quite easy to make it do
what you want.

My advice would be to dump cmd.exe.  Even ignoring the issue you're
having, it does not do copy and paste properly.  It does selection as a
rectange rather than as linear text, which is how every other
text-oriented application works.  rxvt is light years ahead of cmd.exe
in terms of functionality, once you get it running with some more
pleasing colors and font than the default (IMO).  FYI I do this with a
shortcut to: (no doubt it might be easier to set these as resources
rather than on the command line; I know this.)

C:\cygwin\bin\rxvt.exe -g 130x60 -bg black -fg gray -fn Lucida
Console-11 -sr -sl 5000 -j -cr white -sk -si -tn rxvt -e /bin/bash -li

And, it will automatically copy text when you select it, as in *nix,
which I find even easier than having to remember to press ctrl-C.

You *might* even be able to configure rxvt to recognise ^C as the
explicit copy-selection key, and disable the automatic copying.  But,
you would also have to mess with your tty settings because ^C is usually
the 'interrupt' keystroke and can be very, very hard to live without in
a posix-like environment.  You could this to some other keystroke,
though.  Try man stty.  Also, I think Control-Break will also send the
'interrupt' key.

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: rsync + xp sp2 failing

On Sep  9 21:00, joellists wrote:
 On Fri, 10 Sep 2004 00:09:29 +0200, Corinna Vinschen wrote
  All WSADuplicateSocket calls fail with 10045, operation is not supported
  for the type of object referenced, even thought the above created socket
  handles are referenced.  That's weird.
  
  I'm still running XP SP1 and I can't reproduce this.  I'm wondering 
  if that's a side effect of the new firewall in XP2.  Did you try 
  with switching off the firewall entirely?
 
 Yes, I did try turning off the service.  In fact, this has solved problems for
 other programs (Age of Kings, for instance) that absolutely won't work with
 the firewall enabled, even when you make what you think are the proper exceptions.
 
 I find it hard to believe no one else is having this problem... anyway...
 
 I'm guessing this may have something to do with it:
 
 A little more info on raw sockets and Windows XP SP2
 http://blogs.msdn.com/michael_howard/archive/2004/08/12/213611.aspx

No, socketpairs don't use raw sockets.


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: ERB in Ruby1.8 in Cygwin Install

On Sep 10 00:49, Dosu Mahat wrote:
 
 I have reinstalled Cygwin, selecting Ruby from the optional packages list.  
 In Cygwin, the Ruby package lists ERB as one of it's components, however 
 upon installing I get Ruby with other features like Rexml, Runit etc but no 
 ERB.
 
 I get errors in Ruby when trying to invoke ERB.
 
 Can any one suggest what I can do to fix this problem, which I think is 
 because ERB libraries are not getting downloaded along with the rest of 
 Ruby

http://cygwin.com/problems.html


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: win95 pipe problems -- report + testcase + patch

On Sep  9 21:24, Gary R. Van Sickle wrote:
 [snip]
   :  nudge our legal department again.
   
   I'll be waiting. (This will give me more time to test 
  readdir_r, once 
   I decide how to keep accesses separate.)
  
  Sic.
 
 I think you mean Ditto there Corinna.

No, I meant Sigh.

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: How do I define a key binding for copy-to-clipboard?

2004-09-10 Thread jashy

If you really hate using mouse to do copypaste, I suggest you to use device 
/dev/clipboard directly. Copy to clipboard: cat somefile  /dev/clipboard, get 
something from clipboard, cat /dev/clipboard  somewhere.

--
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 on Cygwin Immediate Drops Connections

On Sep 10 01:33, Mark Eret wrote:
 debug3: remaining preferred: 
 debug3: authmethod_is_enabled password
 debug1: Next authentication method: password
 debug3: packet_send2: adding 64 (len 59 padlen 5 extra_pad 64)
 debug2: we sent a password packet, wait for reply
 debug1: Authentication succeeded (password).

Password authentication won't work on the command line.

 debug1: permanently_set_uid: 1107/513
 setreuid 1107: Permission denied

That's what happens in that case.  Try public key authentication.
Otherwise I don't see anything wrong here.


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/

sed-4.1.2-1: backslash in 'i' and 'a' changed?

2004-09-10 Thread Jan Schormann

Hi,

try this:

 $ echo | sed -e 'a\\'

in sed-4.0.9-2 and sed-4.1.2-1: The old one produces a backslash,
the new one doesn't. Unfortunately, this breaks a makefile I use
for generating LaTeX output ...

Can this be a side effect of the change described below? Will this
be stable or is it a bug?

---
From: Corinna Vinschen
Subject: Updated: sed-4.1.2-1
Date: Sun, 22 Aug 2004 10:14:24 -0700

[...]

  when a \ is used to escape the character that would terminate an
operand
  of the s or y commands, the backslash is removed before the regex
is
  compiled.

[...]
---

Cheers,
Jan.



--
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 service won't start

On Sep  9 17:50, Alder wrote:
 Corinna Vinschen wrote:
 On Sep  8 22:37, Alder wrote:
 
 as a service.  This step seemed successful as the Windows Event Viewer 
 lists it as 'CYGWIN cygserver' and describes it as 'Cygwin shared memory 
 manager'.  This description, however, is suspiciously like the one I 
 
 
 The cygserver-config script does not add this description.  So it's from
 an old service entry.
 
 That's what I suspected, but I wonder where the string was pulled from? 
  Like I said, I cleared the Cygnus Solutions keys from LOCAL_MACHINE 
 and CURRENT_USER areas in the registry and deleted the temp folders 
 created by the previous Setup.exe.  Could it be the Service Control 
 Manager itself?

HKEY_LOCAL_MACHINE/SYSTEM/CurrentControlSet/Services/sshd

is the service key.


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: rsync + xp sp2 failing

Corinna Vinschen wrote:

All WSADuplicateSocket calls fail with 10045, operation is not supported
for the type of object referenced, even thought the above created socket
handles are referenced. That's weird.

I'm still running XP SP1 and I can't reproduce this. I'm wondering if
that's a side effect of the new firewall in XP2. Did you try with
switching off the firewall entirely?

FWIW this error happens frequently enough for Apache under windows that
it's in their FAQ:
http://httpd.apache.org/docs/misc/FAQ.html#WSADuplicateSocket

quote
Apache for Windows does not start. Error log contains this message:
[crit] (10045) The attempted operation is not supported for the type of
object referenced: Parent: WSADuplicateSocket failed for socket ###.
What does this mean?

We have seen this problem when Apache is run on systems along with
Virtual Private Networking clients like Aventail Connect. Aventail
Connect is a Layered Service Provider (LSP) that inserts itself, as a
shim, between the Winsock 2 API and Window's native Winsock 2
implementation. The Aventail Connect shim does not implement
WSADuplicateSocket, which is the cause of the failure.

The shim is not unloaded when Aventail Connect is shut down. Once
observed, the problem persists until the shim is either explicitly
unloaded or the machine is rebooted. Another potential solution (not
tested) is to add apache.exe to the Aventail Connect Exclusion List.

Apache is affected in a similar way by any firewall program that isn't
correctly configured. Assure you exclude your Apache server ports
(usually port 80) from the list of ports to block. Refer to your
firewall program's documentation for the how-to.
end quote

It's probably not directly related, but there is this blurb about LSP
from the page on
http://www.microsoft.com/technet/prodtechnol/winxppro/maintain/sp2netwk.mspx#XSLTsection127121120120

quote
Winsock self-healing

Detailed description

Winsock, Windows network socket facility for applications, is
extensible by a mechanism known as a Layered Service Provider (LSP).
Winsock LSPs are available for a wide range of useful purposes,
including internet parental controls, and web content filtering. In
previous versions of Windows XP, removing a malformed (also known as
buggy) LSP could result in corruption of the Winsock catalog in the
registry, potentially resulting in a loss of all network connectivity.
Winsock now has the ability to self-heal after a user uninstalls such an
LSP.
end quote

And finally, although it's probably not relevant, the following passage
from
http://msdn.microsoft.com/security/productinfo/xpsp2/default.aspx?pull=/library/en-us/dnwxp/html/securityinxpsp2.asp
might help track down the problem:

quote
IPv4 Inbound Connections for Applications. An application that completes
a listen operation on a TCP socket or successfully binds to a UDP socket
through Winsock is covered by this scenario. Examples of these
applications include audio and video in MSN or Windows Messenger, or
hosting a multiplayer game. For this scenario, ICF can automatically
open and close ports as needed by the application. When an application
that needs to listen on a port or ports is being installed by an
administrator, it will need to ask the user if he/she wants to allow the
application to open ports in the firewall. If the user consents to this,
then the application should use the INetFwV4AuthorizedApplication API to
add itself to the AuthorizedApplications collection as enabled. If the
user does not consent, then the application should use the
INetFwV4AuthorizedApplication API to add itself to the
AuthorizedApplications collection as disabled.
end quote

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: SSH on Cygwin Immediate Drops Connections

2004-09-10 Thread Mark Eret

Same thing happens with public key authentication.

Server Log:

debug2: load_server_config: filename /etc/sshd_config
debug2: load_server_config: done config len = 238
debug2: parse_server_config: config /etc/sshd_config len 238
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 238
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 238
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 192.168.1.102 port 3033
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
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-hellma
n-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,r
[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,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-md5-96
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-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-hellma
n-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,r
[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,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-md5-96
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-96,hm
ac-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 hmac-md5 none
debug2: mac_init: found hmac-md5
debug1: kex: server-client aes128-cbc hmac-md5 none
debug1: SSH2_MSG_KEX_DH_GEX_REQUEST received
debug1: SSH2_MSG_KEX_DH_GEX_GROUP sent
debug2: dh_gen_key: priv key bits set: 124/256
debug2: bits set: 496/1024
debug1: expecting SSH2_MSG_KEX_DH_GEX_INIT
debug2: bits set: 503/1024
debug1: SSH2_MSG_KEX_DH_GEX_REPLY sent
debug2: kex_derive_keys
debug2: set_newkeys: mode 1
debug1: SSH2_MSG_NEWKEYS sent
debug1: expecting SSH2_MSG_NEWKEYS
debug2: set_newkeys: mode 0
debug1: SSH2_MSG_NEWKEYS received
debug1: KEX done
debug1: userauth-request for user eret service ssh-connection method none
debug1: attempt 0 failures 0
debug2: input_userauth_request: setting up authctxt for eret
debug2: input_userauth_request: try method none
Failed none for eret from 192.168.1.102 port 3033 ssh2
debug1: userauth-request for user eret service ssh-connection method
publickey
debug1: attempt 1 failures 1
debug2: input_userauth_request: try method publickey
debug1: test whether pkalg/pkblob are acceptable

Re: Perl module download ?

Ramneek Singh wrote:
 
 I need clarification on which modules I need for using perl on cygwin.
 I only see perl-libwin32 under system category. Is that all that's
 required ?
 Its not included in default package. I can select it, but is there
 something else
 Related to perl somewhere else. I don't see it, but I may have missed
 seeing it.

It depends on what you want to do.  Perl itself comes bundled with a
fair number of modules, but if the module you wish to use is not one of
these then you'll need to install it yourself.  This is done through
CPAN, not Cygwin setup.exe.  The only reason you see perl-libwin32 as a
package in setup.exe is because it's a particularly nasty module to
install, which you cannot typically install straight from CPAN as you
can most other modules.  See perldoc CPAN.pm for more info.  In
general you can test for the presense of module Foo::Bar with:

perl -MFoo::Bar -e1

If no error is reported then the module is installed, otherwise you will
have to use something along the lines of the following:

perl -MCPAN -e 'install Foo::Bar'

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: SSH on Cygwin Immediate Drops Connections

On Sep 10 02:19, Mark Eret wrote:
 debug1: permanently_set_uid: 1107/513
 setreuid 1107: Permission denied

What's the user and group id of the account you're running sshd under?
You can't change the account in this mode so the uid/gid must be the
same as the account you want to login to.

 Any ideas what is happening?

Other than 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: rsync + xp sp2 failing

On Sep 10 01:19, Brian Dessent wrote:
Corinna Vinschen wrote:

All WSADuplicateSocket calls fail with 10045, operation is not supported
for the type of object referenced, even thought the above created socket
handles are referenced. That's weird.

I'm still running XP SP1 and I can't reproduce this. I'm wondering if
that's a side effect of the new firewall in XP2. Did you try with
switching off the firewall entirely?

FWIW this error happens frequently enough for Apache under windows that
it's in their FAQ:
http://httpd.apache.org/docs/misc/FAQ.html#WSADuplicateSocket

Interesting.

I *pray* it's not related.

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/

Installation hangs because asks for input from terminal

2004-09-10 Thread Alexander Colesnicov

Dear Sir/Madam,
I tried to install cygwin and it hanged when executing
/etc/postinstall/post-texmf.sh . To proceed further, I marked this
script as done and all the rest was OK. Then I started cygwin and run
this script manually. I found that it was a conflict with my existing
fpTeX installation (fpTeX is a TeTeX port to Windows) that defined an
environment variable TEXMFCNF to one of Windows directories. The script
post-texmf.sh diagnosted that and asked for user's reaction. I redefined
the variable in my .bashrc as TEXMFCNF=/usr/share/texmf/web2c:, and
the script run. The bad thing is that the terminal window does not exist
during setup.exe run. Therefore, the user can not react to script's
questions. My recommedation is to reprogram post-texmf.sh in such a
manner that it will never ask for user's intervention, and will set the
TEXMFCNF variable uncoditionally to its true value to avoid the
described conflict.
Yours sincerely
Dr. Alexander Colesnicov,
Institute of Mathematics and Computer Science,
Chisinau, Rep. of Moldova

--
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: sed-4.1.2-1: backslash in 'i' and 'a' changed?

On Sep 10 10:06, Jan Schormann wrote:
 Hi,
 
 try this:
 
  $ echo | sed -e 'a\\'
 
 in sed-4.0.9-2 and sed-4.1.2-1: The old one produces a backslash,
 the new one doesn't. Unfortunately, this breaks a makefile I use
 for generating LaTeX output ...
 
 Can this be a side effect of the change described below? Will this
 be stable or is it a bug?

I'd guess this is a stable fix.  From `info sed':

  `a\'
  `TEXT'
   [...]
   Escape sequences in TEXT are processed, so you should use `\\' in
   TEXT to print a single backslash.

It looks like the old behaviour is a bug, not the new one.  Try
  echo | sed -e 'a\\\'
with the old version.  It will also print just one backslash.

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: [ANNOUNCEMENT] Updated: cygwin-1.5.11-1

2004-09-10 Thread Bjørn-Helge Mevik

Great!

Unfortunately, it doesn't seem to solve the problem reported in 

http://cygwin.com/ml/cygwin/2004-08/msg00152.html

i.e., on a NetWare volume:

14 (0) $ echo foodata  foo
15 (0) $ echo bardata  bar
16 (0) $ ls -l foo bar
-rw-r--r--1 bhme None8 Sep 10 12:39 bar
-rw-r--r--1 bhme None8 Sep 10 12:39 foo
17 (0) $ cp foo bar
cp: writing `bar': Permission denied
18 (0) $ ls -l foo bar
-rw-r--r--1 bhme None0 Sep 10 12:40 bar
-rw-r--r--1 bhme None8 Sep 10 12:39 foo

Any attemt to remove `bar' fails until I log out and back in again.

-- 
Bjørn-Helge Mevik

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

Re: rsync + xp sp2 failing

2004-09-10 Thread Chris Taylor


On Fri, September 10, 2004 9:46 am, Corinna Vinschen said:
 On Sep 10 01:19, Brian Dessent wrote:

 Corinna Vinschen wrote:


 All WSADuplicateSocket calls fail with 10045, operation is not
 supported for the type of object referenced, even thought the above
 created socket handles are referenced.  That's weird.

 I'm still running XP SP1 and I can't reproduce this.  I'm wondering
 if that's a side effect of the new firewall in XP2.  Did you try with
 switching off the firewall entirely?

 FWIW this error happens frequently enough for Apache under windows that
  it's in their FAQ:
 http://httpd.apache.org/docs/misc/FAQ.html#WSADuplicateSocket


  quote 
 Apache for Windows does not start. Error log contains this message:
 [crit] (10045) The attempted operation is not supported for the type of
  object referenced: Parent: WSADuplicateSocket failed for socket ###.
 What does this mean?


 Interesting.


 And finally, although it's probably not relevant, the following passage
  from
 http://msdn.microsoft.com/security/productinfo/xpsp2/default.aspx?pull
 =/library/en-us/dnwxp/html/securityinxpsp2.asp
 might help track down the problem:

  quote 
 IPv4 Inbound Connections for Applications. An application that completes
  a listen operation on a TCP socket or successfully binds to a UDP
 socket through Winsock is covered by this scenario. Examples of these
 applications include audio and video in MSN or Windows Messenger, or
 hosting a multiplayer game. For this scenario, ICF can automatically
 open and close ports as needed by the application. When an application
 that needs to listen on a port or ports is being installed by an
 administrator, it will need to ask the user if he/she wants to allow
 the application to open ports in the firewall. If the user consents to
 this, then the application should use the INetFwV4AuthorizedApplication
 API to
 add itself to the AuthorizedApplications collection as enabled. If the
 user does not consent, then the application should use the
 INetFwV4AuthorizedApplication API to add itself to the
 AuthorizedApplications collection as disabled.
  end quote 


 I *pray* it's not related.



 Corinna



Is it possible this is due to the changes in tcpip.sys and/or the tcp
stack? I myself am using a hacked one without these changes, so I could
test things to see if it is, if people can give me a specific set of
things to do ;-)
Unfortunately, half of the stuff in this thread went clean over my head,
so I could easily be barking up the wrong tree

Chris

-- 
And the Lord spake, saying, 'First shalt thou take out the Holy Pin. Then
shalt thou count to three, no more - no less. Three shall be the number
thou shalt count, and the number of the counting shall be three. Four
shalt thou not count, neither count thou two, excepting that thou then
proceed to three. Five is right out! Once the number three, being the
third number, be reached, then lobbest thou thy Holy Hand Grenade of
Antioch towards thy foe, who, being naughty in my sight, shall snuff it.'



--
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/

Now I see - was RE: sed-4.1.2-1: backslash in 'i' and 'a' changed?

2004-09-10 Thread Jan Schormann

Hi,

sorry, by now even I noticed that this is OT because it
depends entirely on upstream. Then again, I found out
where I was misunderstanding the documentation, and I'd
like to share:

I always thought that the '\' after 'a' was denoting a
continuation line and could be left out if you put all
the text in the same line, which of course I'm always
tempted to do when I provide expressions using the '-e'
command line switch.

Thanks for the enlightenment,
Jan.

.. Full story:

 From `info sed':
 
   `a\'
   `TEXT'
[...]
Escape sequences in TEXT are processed, so you 
should use `\\' in TEXT to print a single backslash.

Yeah well, so '\\x' should print a single backslash and an 'x',
but it doesn't print a backslash at all ... Oh I see, *if* it's
the first thing in the argument, where sed 4.1.2 reads it
as '\' + '\x' in these examples:

 $ echo | sed -e 'ix\\x'
 x\x

 $ echo | sed -e 'i\\x'
 x

While 3 bs produce the (by me) desired result in both versions:

 $ echo | sed -e 'i\\\x'
 x

OK, can live with that ;-)



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

Re: rsync + xp sp2 failing

Chris Taylor wrote:

 Is it possible this is due to the changes in tcpip.sys and/or the tcp
 stack? I myself am using a hacked one without these changes, so I could
 test things to see if it is, if people can give me a specific set of
 things to do ;-)
 Unfortunately, half of the stuff in this thread went clean over my head,
 so I could easily be barking up the wrong tree

The change he's referring to is the limit in SP2 meant to catch worms
and malware that connect to random IP addresses.  MS's fix was to
throttle the TCP/IP stack when it reaches 10 outgoing half-open
connections.  Note: that's not 10 connections total, that's 10 half-open
connections, i.e. state SYN_SENT.  This only really happens when you try
to open a bunch of simultaneous connections to non-existant / dark IP
space, which is typical of a worm.  However I doubt that rsync could
possibly trigger this state, so it's likely not relevant.  Nor are the
changes regarding raw sockets.

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: rsync + xp sp2 failing

Brian Dessent wrote:

 However I doubt that rsync could possibly trigger this state, 
 so it's likely not relevant.  

...and, when it does apply you get an event id 4226 in the event logs,
which should be easy to verify as being absent in this case.

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: [ANNOUNCEMENT] Updated: cygwin-1.5.11-1

On Sep 10 12:41, Bj?rn-Helge Mevik wrote:
 Great!
 
 Unfortunately, it doesn't seem to solve the problem reported in 
 
 http://cygwin.com/ml/cygwin/2004-08/msg00152.html
 
 i.e., on a NetWare volume:
 
 14 (0) $ echo foodata  foo
 15 (0) $ echo bardata  bar
 16 (0) $ ls -l foo bar
 -rw-r--r--1 bhme None8 Sep 10 12:39 bar
 -rw-r--r--1 bhme None8 Sep 10 12:39 foo
 17 (0) $ cp foo bar
 cp: writing `bar': Permission denied
 18 (0) $ ls -l foo bar
 -rw-r--r--1 bhme None0 Sep 10 12:40 bar
 -rw-r--r--1 bhme None8 Sep 10 12:39 foo
 
 Any attemt to remove `bar' fails until I log out and back in again.

AFAIK, none of the core Cygwin developers has any netware access so
you're probably the only right person to debug this problem.  Did
you try stracing that situation to see what's going on?


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: How do I define a key binding for copy-to-clipboard?

2004-09-10 Thread Buchbinder, Barry (NIH/NIAID)

At Friday, September 10, 2004 4:02 AM, jashy wrote:
 If you really hate using mouse to do copypaste, I suggest you to use
 device /dev/clipboard directly. Copy to clipboard: cat somefile 
 /dev/clipboard, get something from clipboard, cat /dev/clipboard 
 somewhere.   

or use putclip and getclip from the cygutils package.

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

Re: rsync + xp sp2 failing (solved!)

2004-09-10 Thread joellists

Thanks all!

Turns out Aventail Connect was the problem (which was installed right around
same time as SP2). And here I was blaming it on Bill Gates ... shame on me
:). Anyway, uninstalling it solved the problem and I'm back in business.

Once again, thanks all for your time and ideas!

-- Joel

On Fri, 10 Sep 2004 01:19:34 -0700, Brian Dessent wrote
Corinna Vinschen wrote:

All WSADuplicateSocket calls fail with 10045, operation is not supported
for the type of object referenced, even thought the above created socket
handles are referenced. That's weird.

I'm still running XP SP1 and I can't reproduce this. I'm wondering if
that's a side effect of the new firewall in XP2. Did you try with
switching off the firewall entirely?

FWIW this error happens frequently enough for Apache under windows that
it's in their FAQ:
http://httpd.apache.org/docs/misc/FAQ.html#WSADuplicateSocket

quote
Apache for Windows does not start. Error log contains this message:
[crit] (10045) The attempted operation is not supported for the
type of object referenced: Parent: WSADuplicateSocket failed for
socket ###. What does this mean?

Apache is affected in a similar way by any firewall program that
isn't correctly configured. Assure you exclude your Apache server ports
(usually port 80) from the list of ports to block. Refer to your
firewall program's documentation for the how-to.
end quote

It's probably not directly related, but there is this blurb about LSP
from the page on

http://www.microsoft.com/technet/prodtechnol/winxppro/maintain/sp2netwk.mspx#XSLTsection127121120120

quote
Winsock self-healing

Detailed description

Winsock, Windows network socket facility for applications, is
extensible by a mechanism known as a Layered Service Provider (LSP).
Winsock LSPs are available for a wide range of useful purposes,
including internet parental controls, and web content filtering. In
previous versions of Windows XP, removing a malformed (also known as
buggy) LSP could result in corruption of the Winsock catalog in the
registry, potentially resulting in a loss of all network
connectivity. Winsock now has the ability to self-heal after a user
uninstalls such an LSP.
end quote

And finally, although it's probably not relevant, the following passage
from

http://msdn.microsoft.com/security/productinfo/xpsp2/default.aspx?pull=/library/en-us/dnwxp/html/securityinxpsp2.asp
might help track down the problem:

quote
IPv4 Inbound Connections for Applications. An application that completes
a listen operation on a TCP socket or successfully binds to a UDP socket
through Winsock is covered by this scenario. Examples of these
applications include audio and video in MSN or Windows Messenger, or
hosting a multiplayer game. For this scenario, ICF can automatically
open and close ports as needed by the application. When an
application that needs to listen on a port or ports is being
installed by an administrator, it will need to ask the user if
he/she wants to allow the application to open ports in the firewall.
If the user consents to this, then the application should use the
INetFwV4AuthorizedApplication API to add itself to the
AuthorizedApplications collection as enabled. If the user does not
consent, then the application should use the
INetFwV4AuthorizedApplication API to add itself to the
AuthorizedApplications collection as disabled.
end quote

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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

On Fri, Sep 10, 2004 at 12:38:31AM -0400, Igor Pechtchanski wrote:
Hi,

I'm getting the following behavior consistently:

$ bash -c 'exec -a  java -h' 21 | egrep 'options?]'
Usage: java.exe [-options] class [args...]
   or  java.exe [-jar] [-options] jarfile [args...]
$

but

$ bash -c 'exec -a  bash --help' 21 | egrep 'options?]'
Usage:   [GNU long option] [option] ...
 [GNU long option] [option] script-file ...
$

In other words, invoking a Cygwin program sets argv[0] correctly, but
invoking a Windows program doesn't.  Interestingly enough, using the MinGW
execv works properly:

There is no such thing as argv[0] for a Windows program.  Cygwin uses
CreateProcess to create processes.  CreateProcess does not have the concept
of argv[0].  Perhaps Microsoft's exec functions do something funky that
is understood by other msvcrt programs.  Cygwin doesn't do that.

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: [ANNOUNCEMENT] Updated: cygwin-1.5.11-1

On Fri, Sep 10, 2004 at 12:41:18PM +0200, Bj?rn-Helge Mevik wrote:
Great!

Unfortunately, it doesn't seem to solve the problem reported in 

http://cygwin.com/ml/cygwin/2004-08/msg00152.html

I don't believe that anyone said it would.

I don't know of any cygwin developer who has Netware or is interested
in fixing Netware problems.

cgf

i.e., on a NetWare volume:

14 (0) $ echo foodata  foo
15 (0) $ echo bardata  bar
16 (0) $ ls -l foo bar
-rw-r--r--1 bhme None8 Sep 10 12:39 bar
-rw-r--r--1 bhme None8 Sep 10 12:39 foo
17 (0) $ cp foo bar
cp: writing `bar': Permission denied
18 (0) $ ls -l foo bar
-rw-r--r--1 bhme None0 Sep 10 12:40 bar
-rw-r--r--1 bhme None8 Sep 10 12:39 foo

Any attemt to remove `bar' fails until I log out and back in again.

--
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] /etc/profile fixes

2004-09-10 Thread Tero Niemela

Hi!

 It's better to fix than to silence.

Indeed. I've attached an updated patch that seems to
work with bash and zsh.

The other patch had a typo (missing |), fixed patched included.

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

zsh-quote.patch
Description: zsh-quote.patch


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

Re: rsync + xp sp2 failing (solved!)

Good catch, Brian!

Could we give Brian another gold star, please?  That was the first I thought
when reading this posting.


Corinna


On Sep 10 07:39, joellists wrote:
 Thanks all!
 
 Turns out Aventail Connect was the problem (which was installed right around
 same time as SP2).  And here I was blaming it on Bill Gates ... shame on me
 :).  Anyway, uninstalling it solved the problem and I'm back in business.
 
 Once again, thanks all for your time and ideas!
 
 -- Joel
 
 On Fri, 10 Sep 2004 01:19:34 -0700, Brian Dessent wrote
  Corinna Vinschen wrote:
  
   All WSADuplicateSocket calls fail with 10045, operation is not supported
   for the type of object referenced, even thought the above created socket
   handles are referenced.  That's weird.
   
   I'm still running XP SP1 and I can't reproduce this.  I'm wondering if
   that's a side effect of the new firewall in XP2.  Did you try with
   switching off the firewall entirely?
  
  FWIW this error happens frequently enough for Apache under windows that
  it's in their FAQ:
  http://httpd.apache.org/docs/misc/FAQ.html#WSADuplicateSocket
  
   quote 
  Apache for Windows does not start. Error log contains this message:
  [crit] (10045) The attempted operation is not supported for the 
  type of object referenced: Parent: WSADuplicateSocket failed for 
  socket ###. What does this mean?
  [...]

-- 
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: bash script doesn't wait for commands to complete

2004-09-10 Thread Dolton Tony AB

(Sorry if this doesn't appear in the right place - I've been unable to
subscribe to the list and had to bodge a reply.)

I have a problem whereby a bash script doesn't wait for foreground
commands
to complete. The script in question kicks off a number of background
processes, and processes their results as they complete.

A simplified version which exhibits the problem is attached as scr.sh. The
number of background processes and the number of iterations of the
foreground commands are supplied as parameters, (e.g. ./scr.sh 50 100).
These values, as well as the length of the sleep may need adjustment
depending on a machine's performance, to make sure that the sleeps
terminate
squarely within the foreground loop.

Cygwin currently has a hard limit of around 62 subprocesses or so before
things might get confusing.  There is a 63 handle windows limitation
which would take some effort to work around.

So, if you are starting more than around 62 subprocesses then things
probably
won't work right.

cgf

Thanks for that. However, although the problem occurs most frequently with
more subprocesses, I occasionally see it with as few as 2 subprocesses, so I
don't think the limit is coming into play. I haven't tried more than 50
subprocesses, as I knew there was a hard limit
Tony

--
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/

diff crashes with big files on Itanium machine

2004-09-10 Thread Alexei Alexandrov

Hi, All!

I would like to report a problem I have with Cygwin's diff. It crashes when I use it 
on 2 big files (the size is 100M each) on Itanium machines under 
Windows .NET server. It works OK on x86 machines. The following is the scenario of 
reproducing the bug

[EMAIL PROTECTED] ~
$ dd if=/dev/zero bs=512k count=200 | tr '\000' ' '  spaces
200+0 records in
200+0 records out

[EMAIL PROTECTED] ~
$ dd if=/dev/zero bs=512k count=200 | tr '\000' ' '  spaces.2
200+0 records in
200+0 records out

[EMAIL PROTECTED] ~
$ diff spaces spaces.2
Segmentation fault (core dumped)

[EMAIL PROTECTED] ~
$ cat diff.exe.stackdump
Exception: STATUS_ACCESS_VIOLATION at eip=610AE300
eax=00AC ebx=06401000 ecx=61057F74 edx= esi=6114E720 edi=0388
ebp=0022EBF8 esp=0022EBA0 program=C:\cygwin\bin\diff.exe, pid 1948, thread main
cs=001B ds=0023 es=0023 fs=003B gs= ss=0023
Stack trace:
Frame Function  Args
0022EBF8  610AE300  (06400018, 6114E720, 1001, 1001)
0022EC38  610AE6EB  (000E, 616E1BD4, , 61093882)
0022EC78  610AEC1B  (61115378, 0001, 0022ECC8, 610981C0)
0022ECB8  610564E8  (10010470, 06400010, 0022ECF8, 00408A04)
0022ECC8  610882AF  (10010470, 06400010, , 0001)
0022ECF8  00408A04  (0022EE50, 0004, 616E3014, )
0022ED68  00409214  (0022EE50, , , 73696E69)
0022EDA8  004098F0  (0022EE50, , 0011, 6114CF74)
0022EDF8  0040219F  (0022EE50, , , 7DA63826)
0022EFD8  00406E12  (, 616E3894, 616E38AC, )
0022F070  00404C72  (0003, 616E3834, 100100A8, 0022F0C8)
0022F0B0  61005ED4  (0022F0C8, 004243B0, 0022F574, 7DA73F78)
0022FF90  610060EB  (, , , )
End of stack trace

[EMAIL PROTECTED] ~
$ cygcheck -s -v -r

Cygwin Win95/NT Configuration Diagnostics
Current System Time: Fri Sep 10 17:46:06 2004

Windows .NET Enterprise Server Ver 5.2 Build 3763

Path:   C:\cygwin\usr\local\bin
C:\cygwin\bin
C:\cygwin\bin
C:\cygwin\usr\X11R6\bin
c:\WINDOWS\system32
c:\WINDOWS
c:\WINDOWS\System32\Wbem
c:\Program Files (x86)\Far

Output from C:\cygwin\bin\id.exe (nontsec)
UID: 500(Administrator) GID: 513(None)
513(None)

Output from C:\cygwin\bin\id.exe (ntsec)
UID: 500(Administrator) GID: 513(None)
0(root) 513(None)
544(Administrators) 545(Users)

SysDir: C:\WINDOWS\system32
WinDir: C:\Documents and Settings\Administrator\WINDOWS

HOME = `C:\cygwin\home\Administrator'
MAKE_MODE = `unix'
PWD = `/home/Administrator'
USER = `Administrator'

ALLUSERSPROFILE = `C:\Documents and Settings\All Users'
APPDATA = `C:\Documents and Settings\Administrator\Application Data'
CLIENTNAME = `NNWAALEXAND'
CLUSTERLOG = `C:\WINDOWS\Cluster\cluster.log'
COMMONPROGRAMFILES(X86) = `C:\Program Files (x86)\Common Files'
COMMONPROGRAMFILES = `C:\Program Files (x86)\Common Files'
COMMONPROGRAMW6432 = `C:\Program Files\Common Files'
COMPUTERNAME = `ITANIUM110'
COMSPEC = `C:\WINDOWS\system32\cmd.exe'
CVS_RSH = `/bin/ssh'
FARLANG = `English'
HOMEDRIVE = `C:'
HOMEPATH = `\Documents and Settings\Administrator'
HOSTNAME = `itanium110'
INFOPATH = 
`/usr/local/info:/usr/info:/usr/share/info:/usr/autotool/devel/info:/usr/autoto
ol/stable/info:'
LOGONSERVER = `\\ITANIUM110'
MANPATH = `/usr/local/man:/usr/man:/usr/share/man:/usr/autotool/devel/man:'
NUMBER_OF_PROCESSORS = `2'
OLDPWD = `/usr/bin'
OS = `Windows_NT'
PATHEXT = `.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH'
PROCESSOR_ARCHITECTURE = `x86'
PROCESSOR_ARCHITEW6432 = `IA64'
PROCESSOR_IDENTIFIER = `ia64 Family 31 Model 0 Stepping 7, GenuineIntel'
PROCESSOR_LEVEL = `31'
PROCESSOR_REVISION = `0007'
PROGRAMFILES(X86) = `C:\Program Files (x86)'
PROGRAMFILES = `C:\Program Files (x86)'
PROGRAMW6432 = `C:\Program Files'
PROMPT = `$P$G'
PS1 = `\[\033]0;\w\007
[EMAIL PROTECTED] \[\033[33m\w\033[0m\]
$ '
SESSIONNAME = `RDP-Tcp#1'
SHLVL = `1'
SYSTEMDRIVE = `C:'
SYSTEMROOT = `C:\WINDOWS'
TEMP = `c:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\1'
TERM = `cygwin'
TMP = `c:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\1'
USERDOMAIN = `ITANIUM110'
USERNAME = `Administrator'
USERPROFILE = `C:\Documents and Settings\Administrator'
WINDIR = `C:\WINDOWS'
_ = `/usr/bin/cygcheck'
POSIXLY_CORRECT = `1'

HKEY_CURRENT_USER\Software\Cygnus Solutions
HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin
HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\mounts v2
HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\Program Options
HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions
HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin
HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2
  (default) = `/cygdrive'
  cygdrive flags = 0x0020
HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/
  (default) = `C:\cygwin'
  flags = 0x0008
HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/usr/bin
  (default) = `C:\cygwin/bin'
  flags = 0x0008
HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/usr/lib
  (default)

Re: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

On Fri, 10 Sep 2004, Christopher Faylor wrote:

 On Fri, Sep 10, 2004 at 12:38:31AM -0400, Igor Pechtchanski wrote:
 Hi,
 
 I'm getting the following behavior consistently:
 [snip]
 In other words, invoking a Cygwin program sets argv[0] correctly, but
 invoking a Windows program doesn't.  Interestingly enough, using the MinGW
 execv works properly:

 There is no such thing as argv[0] for a Windows program.  Cygwin uses
 CreateProcess to create processes.  CreateProcess does not have the concept
 of argv[0].  Perhaps Microsoft's exec functions do something funky that
 is understood by other msvcrt programs.  Cygwin doesn't do that.

Allow me to disagree.  CreateProcess takes two separate arguments: the
image to run, and the command line (yes, as one long line).  The actual
executable being run is the image parameter, but the value of argv[0] is
extracted from the command line one (well, unless the image name is NULL,
in which case the image name and argv[0] are the same).  That's the value
of argv[0] that I meant.

Now, spawn_guts() *seems* to retain the original (correct) argv[0] when
constructing the command line string, but it apparently replaces it with
the prog_arg parameter somewhere for Windows programs...

Hmm, wait a minute, now that I phrased it this way, I think it enabled me
to find the culprit.  Lines 513-515 of spawn.cc say:

  /* Replace argv[0] with the full path to the script if this is the
 first time through the loop. */
  newargv.replace0_maybe (prog_arg);

Is there a particular reason why this is done for .exe files?
Thanks,
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: Is it free to use

2004-09-10 Thread Andrew DeFaria

Christopher Faylor wrote:
I also have to wonder how, as a long time reader of this list, you 
could *possibly* have missed the tedious discussion on this point that 
crops up on an almost weekly basis.
I have to wonder why you think just because you may or may not talk 
about it ever so often that I would spend my time reading that 
discussion faithfully every week. But to answer your (implied) question: 
consider the possibility that I just so happened to have not been 
interested in the topic until now (then let your imagination fly)

:-)
--
It may be that your sole purpose in life is simply to serve as a warning 
to others.

--
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/

fork fails

2004-09-10 Thread Tero Niemela

I installed Cygwin over the net in 2004-09-10 to WinXP
where I am using Administrator account and everything
seems to work just fine. However, with the same
installation procedure on Win2000 where I'm just in
the Users group I get:

440 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 1960, Win32 error 487
/etc/profile:33: fork failed: resource temporarily
unavailable
  92177 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 316, Win32 error 487
/home/ttn/.zshenv:9: fork failed: resource temporarily
unavailable
 246057 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 852, Win32 error 487
/home/ttn/.zshrc:23: fork failed: resource temporarily
unavailable
 290951 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 2644, Win32 error 487
/home/ttn/.zlogin:6: fork failed: resource temporarily
unavailable

This happen when I replaced bash --login -i in
cygwin.bat with zsh -l and ran it. Shell (zsh)
built- in commands work, but everything else fails
with same symptoms. Mysteriously, using bash --login
-i works just fine and also if I start zsh from bash,
also zsh is working then!

:-O



__
Do you Yahoo!?
Take Yahoo! Mail with you! Get it on your mobile phone.
http://mobile.yahoo.com/maildemo 

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

Re: Is it free to use

2004-09-10 Thread Bobby McNulty

Andrew DeFaria wrote:
Christopher Faylor wrote:
I also have to wonder how, as a long time reader of this list, you 
could *possibly* have missed the tedious discussion on this point 
that crops up on an almost weekly basis.

I have to wonder why you think just because you may or may not talk 
about it ever so often that I would spend my time reading that 
discussion faithfully every week. But to answer your (implied) 
question: consider the possibility that I just so happened to have not 
been interested in the topic until now (then let your imagination fly)

:-)
Personally, I'm bored with the free verses copyrighted material.
GPL means you should include your source code with your binary code. Simple.
Redhat does it, Mandrake does it, Cygnus does it.
Cygwin's sources are included in the download section.
Next time, think before you speak.

--
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/

Cygwin Cron on Windows 2003 Server Issues

Folks - any assistance would be greatly appreciated...

Regards,

Paul J. Ghosh

 


--
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: bash script doesn't wait for commands to complete -- me too

2004-09-10 Thread Eric Hanchrow

 Dolton == Dolton Tony AB [EMAIL PROTECTED] writes:
 
 So, if you are starting more than around 62 subprocesses then
 things probably won't work right.
 
 cgf

Dolton Thanks for that.  However, although the problem occurs
Dolton most frequently with more subprocesses, I occasionally see
Dolton it with as few as 2 subprocesses, so I don't think the
Dolton limit is coming into play.  I haven't tried more than 50
Dolton subprocesses, as I knew there was a hard limit Tony

I've seen similar behavior in my .bash_profile -- one of the last
things it does is invoke keychain, which almost always winds up
running in the background.  It's true that my .bash_profile starts a
large number of processes, but I don't start any of them in the
background intentionally -- that is, none of the commands in
.bash_profile nor .bashrc end with a single `'.

I've also occasionally seen commands that I've started by hand at the
command line execute as if they were in the background.  I've never
been able to come up with a simple reliable recipe to reproduce the
problem, though.

-- 
If you can't change your underwear, can you be sure you have any?


--
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: [ANNOUNCEMENT] Updated: cygwin-1.5.11-1

2004-09-10 Thread Bjørn-Helge Mevik

Christopher Faylor writes:

 On Fri, Sep 10, 2004 at 12:41:18PM +0200, Bj?rn-Helge Mevik wrote:
Great!

Unfortunately, it doesn't seem to solve the problem reported in 

http://cygwin.com/ml/cygwin/2004-08/msg00152.html

 I don't believe that anyone said it would.

I don't believe that I claimed that anyone had said it would.

-- 
Bjørn-Helge Mevik

--
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 on Cygwin Immediate Drops Connections

2004-09-10 Thread Cary Lewis

On server 2003 I had to add the sshd user into the Administrators group
to allow auto logon, the -r option did not work.

Big side effect, with auto logon, the user logging in does not have
their domain groups because sshd can not impersonate them like it could
on w2k, so you will have to fiddle with group permissions.

This had the effect that I could not access my tape drive because I am
normally a domain admin, but when I logged into via ssh with authorized
keys, I was no longer a domain admin.

Logging in with keyboard password did not have this problem.

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Mark Eret
Sent: Friday, September 10, 2004 4:19 AM
To: [EMAIL PROTECTED]
Subject: RE: SSH on Cygwin Immediate Drops Connections

Same thing happens with public key authentication.

Server Log:

debug2: load_server_config: filename /etc/sshd_config
debug2: load_server_config: done config len = 238
debug2: parse_server_config: config /etc/sshd_config len 238
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 238
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 238
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 192.168.1.102 port 3033
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
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-he
llma
n-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-c
bc,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,arcfour,aes192-cbc,aes256-c
bc,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-9
6,hm
ac-md5-96
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-9
6,hm
ac-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-he
llma
n-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-c
bc,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,arcfour,aes192-cbc,aes256-c
bc,r
[EMAIL PROTECTED],aes128-ctr,aes192-ctr,aes256-ctr
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-9
6,hm
ac-md5-96
debug2: kex_parse_kexinit:
hmac-md5,hmac-sha1,hmac-ripemd160,[EMAIL PROTECTED],hmac-sha1-9
6,hm
ac-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 hmac-md5 none
debug2: mac_init: found hmac-md5
debug1: kex: server-client aes128-cbc hmac-md5 none
debug1: SSH2_MSG_KEX_DH_GEX_REQUEST received
debug1: SSH2_MSG_KEX_DH_GEX_GROUP sent
debug2: dh_gen_key: priv key bits set: 124/256
debug2: bits

Re: cygserver service won't start

2004-09-10 Thread Alder

Corinna Vinschen wrote:
On Sep  9 17:50, Alder wrote:
Corinna Vinschen wrote:
HKEY_LOCAL_MACHINE/SYSTEM/CurrentControlSet/Services/sshd
is the service key.
Thanks, Corinna.  Any more ideas on why cygserver won't run as a 
service.  Did the permissions I posted look correct?

Cheers,
Alder
--
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: fhandler_base::dup failed error after upgrade to 1.5.11-1

2004-09-10 Thread Mark Bohlman

Christopher Faylor wrote:
On Thu, Sep 09, 2004 at 07:16:22PM -0400, Matthew Moss wrote:
I ran into a problem after installing updates. I removed cygwin and
reinstalled and now have a fresh installation of 1.5.11-1 using
setup.exe 2.427.
I tried the following command in bash:

tidy contrib.html  con.html
Which gave the following error:
479975353 [main] bash 4024 fhandler_base::dup: dup(/home/Matthew
Moss/con.html) failed, handle 2F, Win32 error 87
bash: con.html: Invalid argument

con is a reserved word on Windows.  Cygwin 1.5.10 erroneously let you
produce a filename named con but 1.5.11 no longer allows this.
This error message should not occur and that should be fixed but fixing
things to provide a more unix-like error message is, unfortunately, not
going to solve your problem.
Sorry.
cgf
See also http://cygwin.com/ml/cygwin/2004-06/msg01235.html if you have 
no way to get rid of any create con file.
-- Mark

--
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 service won't start

2004-09-10 Thread Dave Korn

 -Original Message-
 From: cygwin-owner On Behalf Of Corinna Vinschen
 Sent: 10 September 2004 09:08

  That's what I suspected, but I wonder where the string was 
 pulled from? 
   Like I said, I cleared the Cygnus Solutions keys from 
 LOCAL_MACHINE 
  and CURRENT_USER areas in the registry and deleted the temp folders 
  created by the previous Setup.exe.  Could it be the Service Control 
  Manager itself?
 
 HKEY_LOCAL_MACHINE/SYSTEM/CurrentControlSet/Services/sshd
 
 is the service key.

  Nitpick: that's the sshd service key.  The cygserver one, which is created by
the line

  if ! cygrunsrv -I cygserver -d CYGWIN cygserver -p /usr/sbin/cygserver

from /usr/bin/cygserver-config would have to be 

  HKLM/SYSTEM/CCS/Services/cygserver 

... this would also be consistent with the 'net start' usage info 

  echo To start it, call \`net start cygserver' or \`cygrunsrv -S
cygserver'.


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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

2004-09-10 Thread Dave Korn

 -Original Message-
 From: cygwin-owner On Behalf Of Igor Pechtchanski
 Sent: 10 September 2004 14:57

 On Fri, 10 Sep 2004, Christopher Faylor wrote:

  There is no such thing as argv[0] for a Windows program.  
 Cygwin uses
  CreateProcess to create processes.  CreateProcess does not 
 have the concept
  of argv[0].  Perhaps Microsoft's exec functions do 
 something funky that
  is understood by other msvcrt programs.  Cygwin doesn't do that.
 
 Allow me to disagree.  CreateProcess takes two separate arguments: the
 image to run, and the command line (yes, as one long line).  
 The actual
 executable being run is the image parameter, but the value of 
 argv[0] is
 extracted from the command line one (well, unless the image 
 name is NULL,
 in which case the image name and argv[0] are the same).  
 That's the value
 of argv[0] that I meant.

  Allow me to disagree!  The Win32 API specifies precisely the form that the entry
function of a win32 application must conform to:

int WINAPI WinMain(
  HINSTANCE hInstance,  // handle to current instance
  HINSTANCE hPrevInstance,  // handle to previous instance
  LPSTR lpCmdLine,  // command line
  int nCmdShow  // show state
);

  There is no concept of argv in the windows OS itself.  The application gets the
entire commmand line as one unparsed string.

  Now, it may well be the case that most applications link against the msvcrt
library, which contains an implementation of WinMain that parses lpCmdLine into
separate tokens, puts them into an array, and then calls out to a function named
'main' in the user's app that is linked with the library, passing it argc and argv
parameters.  But it's fair to point out that the argv concept is not part of the
OS, it's part of a compiler support library; it's not handled by the OS, it's not
passed to applications, it's not defined in the API, and there are plenty of
applications that implement their own (or a predefined) WinMain in a different
fashion and do _not_ tokenize and parse the command line; 99% of all MFC apps, for
instance.


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/

TCSH 6.13 shell script can't re-write to a file

2004-09-10 Thread Myers, Paul R (Research)

I have a number of CSH/TCSH shell scripts that work with text files.
In several cases I want to remove/erase a file before I start writing to it
again. 

I simplified the problem to this example:
  #!/usr/bin/tcsh

  echo Hello  stuff
  rm -f stuff  
  echo Goodbye  stuff

After a recent tcsh upgrade (from 6.12.00-7 to 6.13.00-2), the 2nd echo
command will produce the following error:
stuff:  permission denied

The problem doesn't appear when run line-by-line from the command line, and
it doesn't appear when in a bash script.

Can anyone else verify this behavior?  
It seems like a windows sharing violation, does that make sense?


Thanks for your help,

Paul


 cygcheck_new.out 

begin 600 cygcheck_new.out
M#0I#6=W:[EMAIL PROTECTED])A=EO;B!$:6%G;F]S=ECPT*0W5RF5N=!3
M[EMAIL PROTECTED][EMAIL PROTECTED])I(%-E`Q,`Q,CHQ-#HS.`R,#`T#0H-E=I;F1O
M=W,@,C`P,!0F]F97-S:[EMAIL PROTECTED](#4N,!=6EL9`R,3DU(%-EG9I
[EMAIL PROTECTED]%C:R`S#0H-E!A=@Z4,Z7-Y9W=I;EQUW);]C86Q8FEN#0H)
M0SI8WEG=VEN7)I;@T*4,Z7-Y9W=I;EQB:6X-@E#.EQC6=W:6Y=7-R
[EMAIL PROTECTED],5(V7)I;@T*6,Z7]R86-L95QOF$X,5QB:6X-@EC.EQ0F]GF%M
M($9I;5S7$]R86-L95QJF5,2XQ+C=8FEN#0H)8SI=VEN;G1WES=5M
M,S(-@EC.EQW:6YN=`T*6,Z7'=I;FYT7%-YW1E;3,R7%=B96T-@EC.EQ0
MF]GF%M($9I;5S7$-O;6UO;B!:6QEUQ!9%P=5C(%-H87)E9%Q37-T
M96T-@EC.EQ%-E960W7$%C8V5SV]R:65S7`T*4,Z7-Y9W=I;EQB:6X-
M@T*3W5T'5T(9R;[EMAIL PROTECTED]8WEG=VEN7)I;EQI9YE[EMAIL PROTECTED]YO;G1S96,I
M#0I5240Z(#$X,3`V*'5S97(I($=)[EMAIL PROTECTED];6EN:7-TF%T;W)S*0T*
M-30T*$%D;6EN:7-TF%T;W)S*0T*#0I/[EMAIL PROTECTED])O;2!#.EQC6=W:6Y
M8FEN7ED+F5X92`H;G1S96,I#0I5240Z(#$X,3`V*'5S97(I($=)[EMAIL PROTECTED]
M*$%D;6EN:7-TF%T;W)S*0T*,AR;V]T*2`@(`@(`@(`@([EMAIL PROTECTED]
M;6EN:7-TF%T;W)[EMAIL PROTECTED]G,I#0H-E-YT1I[EMAIL PROTECTED]5TE.
M3E1WES=5M,S(-E=I;D1I[EMAIL PROTECTED]5TE.3E0-@T*0UE'5TE.([EMAIL PROTECTED]'1T
M2!N='-E8R-DA/344@/[EMAIL PROTECTED]8WEG=VEN7AO;65=7-EB-DU!2T5?
M34]$12`](!U;FEX)PT*#0I!3$Q54T524U!23T9)3$4@/[EMAIL PROTECTED]1]C=6UE
M;G1S(%N9!3971T:6YGUQ!;[EMAIL PROTECTED]G,G#0I#3TU-3TY04D]'4D%-1DE,
M15,@/[EMAIL PROTECTED]4')O9W)A;2!:6QEUQ#;VUM;[EMAIL PROTECTED],G#0I#3TU34$5#
M([EMAIL PROTECTED],Z7%=)3DY47'-YW1E;3,R7-M9YE4G#0I#55-41$E2([EMAIL PROTECTED],Z
M7$-UW1PF5P)PT*1%)61$E2([EMAIL PROTECTED],Z7$125B-D=23U50([EMAIL PROTECTED];6EN
M:7-TF%T;W)S)PT*2$]-1412259%([EMAIL PROTECTED],Z)PT*2$]35%194$4@/2!@:3,X
M-B-DU!0TA)3D5?5%E012`](!$15-+5$]0)PT*34%#2%194$4@/2!@:3,X
M-B-DU!3E!!5$@@/[EMAIL PROTECTED]W(OW-L+VUA;B-DY534)%4E]/1E]04D]#
M15-33U)3([EMAIL PROTECTED](G#0I/4S),24)0051(([EMAIL PROTECTED],Z7%=)3DY47'-YW1E;3,R
M7]S,EQD;P[)PT*3U,@/[EMAIL PROTECTED]]WU].5-D]35%E012`](!P;W-I
M-E!!5$A%6%0@/[EMAIL PROTECTED]/33LN15A%.RY050[+D--1#LN5D)3.RY60D4[
M+DI3.RY*4T4[+E=31CLN5U-()PT*4%)/0T534T]27T%20TA)5$5#5%5212`]
M(!X.#8G#0I04D]#15-33U)?241%3E1)1DE%4B`]([EMAIL PROTECTED]:6QY(#$U
M($UO95L(#(@4W1E'!I;F@[EMAIL PROTECTED]5L)PT*4%)/0T534T]2
M7TQ%5D5,([EMAIL PROTECTED])PT*4%)/0T534T]27U)%5DE324].([EMAIL PROTECTED],#4G#0I0
M4D]'4D%-1DE,15,@/[EMAIL PROTECTED]4')O9W)A;2!:6QER-E-(14Q,([EMAIL PROTECTED]]B
M:6XO=-S:-E-(3%9,([EMAIL PROTECTED]([EMAIL PROTECTED],Z7%=)
M3DY4)PT*4U-(7U1462`](`O95V+W1T3(G#0I365-414U$4DE612`](!#
M.B-E-94U1%35)/3U0@/[EMAIL PROTECTED]5TE.3E0G#0I415)-([EMAIL PROTECTED]-Y9W=I;B-
ME194$4@/[EMAIL PROTECTED]')E8VES:6]N)PT*5%H@/[EMAIL PROTECTED],2XP+S(L
M33$P+C4N,\R)PT*5D5.1$]2([EMAIL PROTECTED]EN=5L)PT*5TE.1$E2([EMAIL PROTECTED],Z7%=)
M3DY4)PT*4$]325A,65]#3U)214-4([EMAIL PROTECTED]DA+15E?0U524D5.5%]5
M4T527%-O9G1W87)E7$-Y9VYUR!3;VQU=EO;G,-DA+15E?0U524D5.5%]5
M4T527%-O9G1W87)E7$-Y9VYUR!3;VQU=EO;G-0WEG=VEN#0I(2T597T-5
M4E)%3E1?55-%4EQ3;V9T=V%R95Q#6=N=7,@4V]L=71I;VYS7$-Y9W=I;EQM
M;W5N=',@=C(-DA+15E?0U524D5.5%]54T527%-O9G1W87)E7$-Y9VYUR!3
M;VQU=EO;G-0WEG=VEN7%!R;[EMAIL PROTECTED]:6]NPT*2$M%65],3T-!3%]-
M04-(24Y%7%-/1E1705)%7$-Y9VYUR!3;VQU=EO;G,-DA+15E?3$]#04Q?
M34%#2$E.15Q33T945T%215Q#6=N=7,@4V]L=71I;VYS7$-Y9W=I;@T*2$M%
M65],3T-!3%]-04-(24Y%7%-/1E1705)%7$-Y9VYUR!3;VQU=EO;G-0WEG
M=VEN7UO=6YTR!V,@T*(`H95F875L=D@/[EMAIL PROTECTED]:79E)PT*(!C
M6=DFEV92!F;%GR`](#!X,#`P,#`P,C`-DA+15E?3$]#04Q?34%#2$E.
M15Q33T945T%215Q#6=N=7,@4V]L=71I;VYS7$-Y9W=I;EQM;W5N=',@=C)
M+PT*(`H95F875L=D@/[EMAIL PROTECTED]8WEG=VEN)PT*(!F;%GR`](#!X,#`P
M,#`P,[EMAIL PROTECTED]DA+15E?3$]#04Q?34%#2$E.15Q33T945T%215Q#6=N=7,@4V]L
M=71I;VYS7$-Y9W=I;EQM;W5N=',@=C)+W5SB]B:6X-[EMAIL PROTECTED]1E9F%U;'0I
M([EMAIL PROTECTED],Z7-Y9W=I;B]B:6XG#0H@(9L86=S(#T@,'@P,#`P,#`P.`T*2$M%
M65],3T-!3%]-04-(24Y%7%-/1E1705)%7$-Y9VYUR!3;VQU=EO;G-0WEG
M=VEN7UO=6YTR!V,[EMAIL PROTECTED](`H95F875L=D@/[EMAIL PROTECTED]8WEG
M=VEN+VQI8B-[EMAIL PROTECTED],@/2`P#`P,#`P,#`X#0I(2T597TQ/0T%,7TU!
M0TA)3D54T]5%=!4D50WEG;G5S(%-O;'5T:6]NUQ#6=W:6Y;6]U;G1S
M('8R7]UW(O6#$Q4C8O;[EMAIL PROTECTED],2]F;VYTPT*(`H95F875L=D@/2!@
M0SI8WEG=VEN7'5SEQ8,3%2-EQL:6)6#$Q79O;G1S)PT*(!F;%GR`]
M(#!X,#`P,#`P,$-DA+15E?3$]#04Q?34%#2$E.15Q33T945T%215Q#6=N
M=7,@4V]L=71I;VYS7$-Y9W=I;EQ0F]GF%M($]P=EO;G,-@T*83H@(9D
M(`@(`@(`@(!.+T$@([EMAIL PROTECTED](`@(`@(`@(`@(`@(`@(`-F,Z
M(!H9[EMAIL PROTECTED]4R`@(#$T,#`Q,4UB(`Q.[EMAIL PROTECTED]@0U,@[EMAIL

Re: TCSH 6.13 shell script can't re-write to a file

2004-09-10 Thread Isaac Foraker

I just tried this and had the same problem:
 echo Hello  Stuff
 rm -f Stuff
 echo Goodbye  Stuff
Stuff: Permission denied.
 ls -l Stuff
ls: Stuff: No such file or directory
 tcsh --version
tcsh 6.13.00 (Astron) 2004-05-19 (i386-intel-posix) options 
8b,nls,dl,al,kan,rh,color,dspm
 uname -a
CYGWIN_NT-5.1 xcoisaacf20 1.5.11(0.116/4/2) 2004-09-04 23:17 i686 
unknown unknown Cygwin

IF
Myers, Paul R (Research) wrote:
I have a number of CSH/TCSH shell scripts that work with text files.
In several cases I want to remove/erase a file before I start writing to it
again. 

I simplified the problem to this example:
 #!/usr/bin/tcsh
 echo Hello  stuff
 rm -f stuff  
 echo Goodbye  stuff

After a recent tcsh upgrade (from 6.12.00-7 to 6.13.00-2), the 2nd echo
command will produce the following error:
stuff:  permission denied
The problem doesn't appear when run line-by-line from the command line, and
it doesn't appear when in a bash script.
Can anyone else verify this behavior?  
It seems like a windows sharing violation, does that make sense?

Thanks for your help,
Paul
cygcheck_new.out 
 


--
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 service won't start

2004-09-10 Thread Alder

Corinna,
I don't know if this is relevant, but I have to add the /bin directory 
to the cygrunsrv command line in the /bin/cygserver-config file in order 
to avoid Windows failing to find the 'cygwin1.dll.  So, the line I have is:

if ! cygrunsrv -I cygserver -d CYGWIN cygserver \
  -p /usr/sbin/cygserver -f Shared memory manager \
  -2 /var/log/cygserver.err -e PATH=/bin:$PATH
Regards,
Alder
--
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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

On Fri, 10 Sep 2004, Dave Korn wrote:

  -Original Message-
  From: cygwin-owner On Behalf Of Igor Pechtchanski
  Sent: 10 September 2004 14:57

  On Fri, 10 Sep 2004, Christopher Faylor wrote:

   There is no such thing as argv[0] for a Windows program.  Cygwin
   uses CreateProcess to create processes.  CreateProcess does not have
   the concept of argv[0].  Perhaps Microsoft's exec functions do
   something funky that is understood by other msvcrt programs.
   Cygwin doesn't do that.
 
  Allow me to disagree.  CreateProcess takes two separate arguments: the
  image to run, and the command line (yes, as one long line).  The
  actual executable being run is the image parameter, but the value of
  argv[0] is extracted from the command line one (well, unless the image
  name is NULL, in which case the image name and argv[0] are the same).
  That's the value of argv[0] that I meant.

   Allow me to disagree!  The Win32 API specifies precisely the form that
 the entry function of a win32 application must conform to:

 int WINAPI WinMain(
   HINSTANCE hInstance,  // handle to current instance
   HINSTANCE hPrevInstance,  // handle to previous instance
   LPSTR lpCmdLine,  // command line
   int nCmdShow  // show state
 );

   There is no concept of argv in the windows OS itself.  The application
 gets the entire commmand line as one unparsed string.

Fair enough.  Thanks for the clarification.

What I meant was that the first token of the command line was set to the
image name when invoking Windows applications from Cygwin.  So, even if a
Cygwin app sets argv[0] explicitly (which should be incorporated into the
lpCmdLine), that first token would be the same as the image name, and not
what was specified.

   Now, it may well be the case that most applications link against the
 msvcrt library, which contains an implementation of WinMain that parses
 lpCmdLine into separate tokens, puts them into an array, and then calls
 out to a function named 'main' in the user's app that is linked with the
 library, passing it argc and argv parameters.  But it's fair to point
 out that the argv concept is not part of the OS, it's part of a compiler
 support library; it's not handled by the OS, it's not passed to
 applications, it's not defined in the API, and there are plenty of
 applications that implement their own (or a predefined) WinMain in a
 different fashion and do _not_ tokenize and parse the command line; 99%
 of all MFC apps, for instance.

True, but irrelevant.  My complaint was that the first token of the
command line, as specified by Cygwin's execv(), was ignored in favor of
the image name.
Hope this clarifies things,
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: Installation hangs because asks for input from terminal (Attn: tetex maintainer, setup maintainer)

On Fri, 10 Sep 2004, Alexander Colesnicov wrote:

 Dear Sir/Madam,

 I tried to install cygwin and it hanged when executing
 /etc/postinstall/post-texmf.sh . To proceed further, I marked this
 script as done and all the rest was OK. Then I started cygwin and run
 this script manually. I found that it was a conflict with my existing
 fpTeX installation (fpTeX is a TeTeX port to Windows) that defined an
 environment variable TEXMFCNF to one of Windows directories. The script
 post-texmf.sh diagnosted that and asked for user's reaction. I redefined
 the variable in my .bashrc as TEXMFCNF=/usr/share/texmf/web2c:, and
 the script run. The bad thing is that the terminal window does not exist
 during setup.exe run. Therefore, the user can not react to script's
 questions. My recommedation is to reprogram post-texmf.sh in such a
 manner that it will never ask for user's intervention, and will set the
 TEXMFCNF variable uncoditionally to its true value to avoid the
 described conflict.

This is a bug -- postinstall scripts should *not* be interactive, and
certainly not via the TUI (since their output is not seen by the user
while installing).  Thanks for reporting it.

As I see it, there are two places where this should be fixed:
- setup.exe should run postinstall scripts with stdin redirected from
  /dev/null or equivalent (so that errant scripts don't hang the install)
- the postinstall scripts themselves should have reasonable defaults on
  encountering EOF in the input (as a consequence of the first fix).

Hope this helps,
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/

What Cygwin is

2004-09-10 Thread Brian Ford

On Thu, 9 Sep 2004, Corinna Vinschen wrote:

 On Sep  8 11:31, Brian Ford wrote:
  No, now we continually have to explain why you can't install said
  x86 Linux binary/package and run it under our Linux emulator.  Heck, you
  can't even *compile* said Linux firewall/router/proxy server etc. under
  our Linux emulator.  Doesn't sound like much of an emulation to me.

 You don't think that's the fault of our project page, do you?  Did you
 read it lately?  For about three months, there's a pretty precise text
 with the headline What Isn't Cygwin?

Sorry, no; I had not read it within that time period.  I was going
on recollection; my bad.  It is much better now.

However, if Cygwin is a Linux emulation layer providing substantial Linux
API functionality, then shouldn't Linux APIs and behavior now trump POSIX
and the Single Unix Specification?  Is that what we really want?

I'm just not comfortable with the removal of all all references to POSIX
and UNIX from that description.

-- 
Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
the best safety device in any aircraft is a well-trained pilot...

--
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/

CISCO SPECIALS

2004-09-10 Thread Mark Levy



ALL EQUIPMENT IS SMARTNETABLE


CISCO2621 (48D/16F)   $950 

CISCO3661-AC  10/100E Cisco 3660 6-slot   $4,750

CISCO3662-AC  Dual 10/100E Cisco 3660 6-slot Dual AC  $4,895
Cisco3725 CISCO 3700 SERIES 2-SLOT$5,250

Cisco6160 DSlam System (64D/32F) Full Feature Package  $1500 
GLC-SX-MM GE SFP LC CONNECTOR SX  $195
NM-1A-T3  CISCO 3600 1-PORT DS3 ATM   $3500
NM-16AINTERFACE MODULE 16-PORT SERIAL $300
PA-2T32PORT T 3 SERIAL ADAPTER$300
WS-C2924XL-EN SWITCH 24PORT 10/100$450


SWITCHES  BLADES

WS-C4006   CAT 4000 Chasis,fan and 2 ps$1040 
WS-C4006-S2CAT 4000 with supII $2600 
WS-C4006-S3CAT 4000 with supIII$8450 
WS-X4148-RJCAT 4000 10/100 Blade   $1430 
WS-X4148-RJ45-VCAT 4000 10/100 BLADES  $2340 
WS-X4013   CAT 4000 sup eng II $1430 
WS-X4013   CAT 4000 sup eng II NEW $4290 
WS-X4014   CAT 4000 SUP END III US $6990 
Cisco7206VXR-CH - (New)   7206VXR 6-SLOT CHASSIS 1 $5900
Cisco7206VXR-CH - (Ref)   7206VXR 6-SLOT CHASSIS 1 $4030 
NPE-G1 (New)   ENGINE WITH 3 GE/FE/E PORTS $9500
NPE-G1 (Ref)   ENGINE WITH 3 GE/FE/E PORTS $9200
NPE-400 -(New) NETWORK PROCESSING ENGINE 400   $6200
NPE-400 -(Ref) NETWORK PROCESSING ENGINE 400   $4550
NPE-300 -(Ref) PROCESSING ENGINE 300   $1495 
C7200-I/O-FE -(Ref)FAST ETHERNET I/O   $975
C7200-I/O-FE-2FE/E - (Ref)7200 I/O CTRL 2PT 10/100 ETH SPARE $2340

If you have any questions or would like further information, please contact me at the 
numbers below.

Best Regards,

Mark Levy
JAMES RIVER SYSTEMS CORP.
757-833-3409 Main
757-833-3410 Fax
757-287-9902 Mobile
Buy*Sell*Rent*Lease

*To UNSUBSCRIBE from these mailings, please reply with REMOVE.*
--
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: TCSH 6.13 shell script can't re-write to a file

2004-09-10 Thread neal somos

If you use /bin/echo the problem goes away.

I can reproduce the problem with tcsh and
when using /bin/echo instead of the builtin, 
I get the expected behavior.

Apparently the built-in echo functions differently.

neal



___
Do you Yahoo!?
Shop for Back-to-School deals on Yahoo! Shopping.
http://shopping.yahoo.com/backtoschool

--
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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

2004-09-10 Thread David A. Rogers

On Fri, 10 Sep 2004, Dave Korn wrote:

   There is no concept of argv in the windows OS itself.  The application gets the
 entire commmand line as one unparsed string.


Actually, this is not true.  Even in a true win32 app, argc and argv are
available, they are just not as obvious. The following program will print
the location of the exe.

---
#include stdafx.h
#include stdlib.h  // for __argc and __argv
#include stdio.h  // for sprintf


int APIENTRY WinMain(HINSTANCE hInstance,
 HINSTANCE hPrevInstance,
 LPSTR lpCmdLine,
 int   nCmdShow)
{
char msg[300];

sprintf(msg, %d args.  Arg0 is %s, __argc, __argv[0]);
MessageBox(NULL, msg, argv test, MB_OK);

return 0;
}

---


--
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: [ANNOUNCEMENT] Updated: cygwin-1.5.11-1

On Fri, Sep 10, 2004 at 05:52:38PM +0200, Bj?rn-Helge Mevik wrote:
Christopher Faylor writes:

 On Fri, Sep 10, 2004 at 12:41:18PM +0200, Bj?rn-Helge Mevik wrote:
Great!

Unfortunately, it doesn't seem to solve the problem reported in 

http://cygwin.com/ml/cygwin/2004-08/msg00152.html

 I don't believe that anyone said it would.

I don't believe that I claimed that anyone had said it would.

I've rotated the tires on the car.

Unfortunately, this does not solve the problem of the oil leak.

Well, duh.

--
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/

make --help typo

2004-09-10 Thread Dave Korn


 Hi all,

  I reckoned the tail of the make --help output looked a bit wrong...

---snip!---
[EMAIL PROTECTED] /artimi make --help | tail -10
  -w, --print-directory   Print the current directory.
  --no-print-directoryTurn off -w, even if it was turned on implicitly.
  -W FILE, --what-if=FILE, --new-file=FILE, --assume-new=FILE
  Consider FILE to be infinitely new.
  --warn-undefined-variables  Warn when an undefined variable is referenced.
Run in UNIX mode (use sh.exe subshell).
Run in Win32 mode (use Win32 subshell).

This program built for i686-pc-cygwin
Report bugs to bug-makePCYMTNQREAIYRgnu.org
---snip!---

 and it particularly hit home when I needed to know whether the w was capitalized
in --win32 or not (it isn't).  Here's what surely counts as a trivial patch to fix
it:


--- main-3.80-1.c   2004-09-10 18:21:12.444251700 +0100
+++ main.c  2004-09-10 18:22:04.163663700 +0100
@@ -341,8 +341,10 @@ static const char *const usage[] =
 N_(\
   --warn-undefined-variables  Warn when an undefined variable is referenced.\n),
 #ifdef __CYGWIN__
-N_(Run in UNIX mode (use sh.exe subshell).\n),
-N_(Run in Win32 mode (use Win32 subshell).\n),
+N_(\
+  --unix  Run in UNIX mode (use sh.exe subshell).\n),
+N_(\
+  --win32 Run in Win32 mode (use Win32 subshell).\n),
 #endif
 NULL
   };

[ Also attached to avoid linewrap]

  CGF, you released the last version of make, and I can't see that there's likely
to be any great need to roll another in the foreseeable future, but maybe you'd
like to drop this into your personal source tree just in case sometime you do end
up doing another release?


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


make-main.c-docs-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: rsync + xp sp2 failing (solved!)

Done.
Igor

On Fri, 10 Sep 2004, Corinna Vinschen wrote:

 Good catch, Brian!

 Could we give Brian another gold star, please?  That was the first I
 thought when reading this posting.

 Corinna

  On Fri, 10 Sep 2004 01:19:34 -0700, Brian Dessent wrote
   Corinna Vinschen wrote:
  
All WSADuplicateSocket calls fail with 10045, operation is not supported
for the type of object referenced, even thought the above created socket
handles are referenced.  That's weird.
   
I'm still running XP SP1 and I can't reproduce this.  I'm wondering if
that's a side effect of the new firewall in XP2.  Did you try with
switching off the firewall entirely?
  
   FWIW this error happens frequently enough for Apache under windows
   that it's in their FAQ:
   http://httpd.apache.org/docs/misc/FAQ.html#WSADuplicateSocket
  
    quote 
   Apache for Windows does not start. Error log contains this message:
   [crit] (10045) The attempted operation is not supported for the
   type of object referenced: Parent: WSADuplicateSocket failed for
   socket ###. What does this mean?
   [...]

-- 
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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

2004-09-10 Thread Chuck McDevitt

argv and argc are concepts from the C runtime, not the Windows OS.

The actual entry point to your program is to a routine that calls the 
initialization routines of the C library, then calls winMain.

Those initialization routines get the command line via Win32 call, 
allocates memory for argv, and parses the command line.

Windows itself has no requirement that an application support argv and 
argc, and in fact programs in other languages (VB etc) don't have any such 
concept.

Cygwin, when launching an application, just needs to make sure the 
CreateProcess call has the command line passed to it.
Everything else is handled by the the receiving program (via C runtime, if 
a C program).


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

gv and ghostscript-x11 broken out of the box

2004-09-10 Thread Phillips, James R

Hi,

I'm using gv to view postscript files, but it is broken following a clean install.  It 
works after some hand
configuration, but that shouldn't be necessary.

First off, the installation of ghostscript-x11 should probably uninstall ghostscript, 
otherwise the version of gs picked
up by gv will probably be /usr/bin/gs instead of /usr/bin/X11R6/gs.  In the 
alternative, gv configuration should specify
the full path to ghostscript-x11 out of the box.

Secondly, ghostscript-x11 requires installation of XFree86-lib-compat in order to have 
liblCE.dll available.  This
should be enforced in setup.exe.  It isn't, so ghostscript-x11 is broken after a clean 
install.


James R. Phillips
South Bend, Indiana


--
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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

On Fri, 10 Sep 2004, Chuck McDevitt wrote:

 argv and argc are concepts from the C runtime, not the Windows OS.

 The actual entry point to your program is to a routine that calls the
 initialization routines of the C library, then calls winMain.

 Those initialization routines get the command line via Win32 call,
 allocates memory for argv, and parses the command line.

 Windows itself has no requirement that an application support argv and
 argc, and in fact programs in other languages (VB etc) don't have any such
 concept.

Ok, so I didn't use quite the right phrasing, but I believe it's been
explained adequately since.

 Cygwin, when launching an application, just needs to make sure the
 CreateProcess call has the command line passed to it.
 Everything else is handled by the the receiving program (via C runtime, if
 a C program).

Exactly, with one small clarification: Cygwin, when launching an
application, just needs to make sure the CreateProcess call has the
*right* command line passed to it..  Currently, spawn_guts() munges the
first token in that command line to match the image name, even if
instructed otherwise (via the args array).
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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

2004-09-10 Thread David A. Rogers

On Fri, 10 Sep 2004, Chuck McDevitt wrote:

 argv and argc are concepts from the C runtime, not the Windows OS.

 The actual entry point to your program is to a routine that calls the
 initialization routines of the C library, then calls winMain.


Yes, certainly.  The point I was making was that the information was available
in a true win32 app.  David Korn implied that the WinMain call was the only
access to the calling arguments that were available to a win32 app.  That is
not true, as I showed.

dar


--
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: TCSH 6.13 shell script can't re-write to a file

2004-09-10 Thread Myers, Paul R (Research)


Thanks Neal, 

I had assumed the issue was with the rm statement, and was experimenting
there.

Some experimentation shows the problem is with the first echo statement.
If I replace the first echo statement with /bin/echo or even printf
the second echo statement works fine. 

So, as you said, the TCSH echo statement has the strange behavior.
I guess it doesn't release the file, or close a pipe or something.

Thanks again, this is a good workaround.

Paul

-Original Message-
From: neal somos [mailto:[EMAIL PROTECTED]
Sent: Friday, September 10, 2004 1:15 PM
To: [EMAIL PROTECTED]
Subject: Re: TCSH 6.13 shell script can't re-write to a file


If you use /bin/echo the problem goes away.

I can reproduce the problem with tcsh and
when using /bin/echo instead of the builtin, 
I get the expected behavior.

Apparently the built-in echo functions differently.

neal



___
Do you Yahoo!?
Shop for Back-to-School deals on Yahoo! Shopping.
http://shopping.yahoo.com/backtoschool

--
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: fork fails

2004-09-10 Thread Larry Hall

At 10:19 AM 9/10/2004, you wrote:
I installed Cygwin over the net in 2004-09-10 to WinXP
where I am using Administrator account and everything
seems to work just fine. However, with the same
installation procedure on Win2000 where I'm just in
the Users group I get:

440 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 1960, Win32 error 487
/etc/profile:33: fork failed: resource temporarily
unavailable
  92177 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 316, Win32 error 487
/home/ttn/.zshenv:9: fork failed: resource temporarily
unavailable
 246057 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 852, Win32 error 487
/home/ttn/.zshrc:23: fork failed: resource temporarily
unavailable
 290951 [main] zsh 1432 fork_copy: linked dll data/bss
pass 0 failed,
0x542000..
0x5427F0, done 0, windows pid 2644, Win32 error 487
/home/ttn/.zlogin:6: fork failed: resource temporarily
unavailable

This happen when I replaced bash --login -i in
cygwin.bat with zsh -l and ran it. Shell (zsh)
built- in commands work, but everything else fails
with same symptoms. Mysteriously, using bash --login
-i works just fine and also if I start zsh from bash,
also zsh is working then!

:-O


Apparently you don't remember this error on the Windows
2000 machine and what you did to fix it then. ;-)

I should point out that this is a perfect question for 
the email archives.  You would have seen plenty of references
and the likely solution.

Install the 'rebase' package.  Run 'rebaseall'.  A simpler 
fix you couldn't ask for. :-)


--
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: Cygwin Cron on Windows 2003 Server Issues

2004-09-10 Thread Larry Hall

At 10:43 AM 9/10/2004, you wrote:
Folks - any assistance would be greatly appreciated...


And what about Igor's response would not qualify as assistance?
If you're looking for something specific, you should ask the specific
question.  I thought Igor's reply looked quite helpful and was certainly
an excellent attempt at assistance.  If it's not for you, then at the
very least you need to be more specific about the kind of assistance
you're looking for.  Or maybe I just missed a portion of this thread.
I tried to go back and recap from the web but it looks like your replies
(at least) break the thread every time.  I got tired of searching for some
new branch...



--
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: Cygwin Cron on Windows 2003 Server Issues

Thank you all for your help.  We pursued all that was recommended and
still ran into the error: cron : PID 3784 : starting service `cron'
failed: execv: 128, Transport endpoint is not connected.

Since invoking it command line was successful we put a wrapper in place
that executes cron and added the wrapper as a service.  cron is now
working.  For the issue where crontab assumes that SYSTEM is the owner
of the cron service and leaves the contents of the /var/cron/tabs as 640
with a group of SYSTEM we also created a wrapper that changes the
permission to 644.

Here is the detail of what we have put into place and provided to the
release engineer:

overview:
- create local user called 'svccron'
- set up permissions for local user
- create wrappers for cron/crontab
- add wrapped cron as service
- set up schedules

execute the following within a bash shell:

- create local user called 'svccron'

net user svccron passwd /add /yes
net localgroup administrators_group_name svccron /add
editrights -a SeAssignPrimaryTokenPrivilege -u svccron
editrights -a SeCreateTokenPrivilege -u svccron
editrights -a SeIncreaseQuotaPrivilege -u svccron
editrights -a SeServiceLogonRight -u svccron
mkpasswd -l -u svccron  /etc/passwd

for security reasons:
editrights -a SeDenyInteractiveLogonRight -u svccron
editrights -a SeDenyNetworkLogonRight -u svccron
editrights -a SeDenyRemoteInteractiveLogonRight -u svccron

- create wrapper for cron - attached 'xcron'/contents below
- install in /usr/sbin
- change permissions for xcron
$ chmod 744 /usr/sbin/xcron

$ cat /usr/sbin/xcron
#!/usr/bin/bash
/usr/sbin/cron -D

$ ls -l /usr/sbin/xcron
-rwxr-xr-x1 devadm   Domain U   34 Sep 10 14:03
/usr/sbin/xcron

- create wrapper for crontab - attached 'xcrontab'/contents
below
- install in /usr/bin
- change permissions for xcrontab
$ chmod 744 /usr/bin/xcrontab

$ cat /usr/bin/xcrontab
#!/usr/bin/bash
/usr/bin/crontab $@;
xval=$?;
chmod 644 /var/cron/tabs/$USER;
exit $xval;

$ ls -l /usr/bin/xcrontab
-rwxr-xr-x1 devadm   Domain U   90 Sep 10 13:58
/usr/bin/xcrontab

- add wrapped cron as service
$ cygrunsrv -I xcron -p /usr/sbin/xcron -a -u svccron -w
passwd

- execute following as user for which tasks need to be scheduled
$ xcrontab -e

- for details
$ man 5 crontab


Regards,

Paul J. Ghosh


xcron
Description: Binary data


xcrontab
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: Cygwin Cron on Windows 2003 Server Issues

On Fri, 10 Sep 2004, Paul J. Ghosh wrote:

 Thank you all for your help.  We pursued all that was recommended and
 still ran into the error: cron : PID 3784 : starting service `cron'
 failed: execv: 128, Transport endpoint is not connected.

 Since invoking it command line was successful we put a wrapper in place
 that executes cron and added the wrapper as a service.  cron is now
 working.  For the issue where crontab assumes that SYSTEM is the owner
 of the cron service and leaves the contents of the /var/cron/tabs as 640
 with a group of SYSTEM we also created a wrapper that changes the
 permission to 644.

 Here is the detail of what we have put into place and provided to the
 release engineer:

 overview:
   - create local user called 'svccron'
   - set up permissions for local user
   - create wrappers for cron/crontab
   - add wrapped cron as service
   - set up schedules

 execute the following within a bash shell:

   - create local user called 'svccron'

   net user svccron passwd /add /yes
   net localgroup administrators_group_name svccron /add
   editrights -a SeAssignPrimaryTokenPrivilege -u svccron
   editrights -a SeCreateTokenPrivilege -u svccron
   editrights -a SeIncreaseQuotaPrivilege -u svccron
   editrights -a SeServiceLogonRight -u svccron
   mkpasswd -l -u svccron  /etc/passwd

   for security reasons:
   editrights -a SeDenyInteractiveLogonRight -u svccron
   editrights -a SeDenyNetworkLogonRight -u svccron
   editrights -a SeDenyRemoteInteractiveLogonRight -u svccron

   - create wrapper for cron - attached 'xcron'/contents below
   - install in /usr/sbin
   - change permissions for xcron
   $ chmod 744 /usr/sbin/xcron

   $ cat /usr/sbin/xcron
   #!/usr/bin/bash
   /usr/sbin/cron -D

   $ ls -l /usr/sbin/xcron
   -rwxr-xr-x1 devadm   Domain U   34 Sep 10 14:03 /usr/sbin/xcron

   - create wrapper for crontab - attached 'xcrontab'/contents below
   - install in /usr/bin
   - change permissions for xcrontab
   $ chmod 744 /usr/bin/xcrontab

   $ cat /usr/bin/xcrontab
   #!/usr/bin/bash
   /usr/bin/crontab $@;
   xval=$?;
   chmod 644 /var/cron/tabs/$USER;
   exit $xval;

   $ ls -l /usr/bin/xcrontab
   -rwxr-xr-x1 devadm   Domain U   90 Sep 10 13:58 /usr/bin/xcrontab

   - add wrapped cron as service
   $ cygrunsrv -I xcron -p /usr/sbin/xcron -a -u svccron -w passwd

   - execute following as user for which tasks need to be scheduled
   $ xcrontab -e

   - for details
   $ man 5 crontab

Well, Larry is right -- it's very hard to follow a thread where the
replies neither quote the original message nor have the correct
References: or In-Reply-To: fields.  However, I've just reviewed the whole
thread.  I don't believe you (or Anna Goldberg) ever said what the exact
cygrunsrv invocation of the cron service was.

Had you said that, it would have been rather obvious that you were simply
missing the -D argument to cron, and that adding that would have most
likely solved your problem.

FWIW, I don't believe the cygrunsrv invocation above works for you, since
the -a flag requires an argument, and will grab the -u.  You probably
didn't cut-and-paste the exact command.

In any case, you can try simply using

$ cygrunsrv -I cron -p /usr/sbin/cron -a -D -u svccron -w passwd

and see if it works.  The crontab trick, on the other hand, might be
necessary.
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: Cygwin Cron on Windows 2003 Server Issues

2004-09-10 Thread Pierre A. Humblet

On Fri, Sep 10, 2004 at 04:45:07PM -0500, Paul J. Ghosh wrote:
 Thank you all for your help.  We pursued all that was recommended and
 still ran into the error: cron : PID 3784 : starting service `cron'
 failed: execv: 128, Transport endpoint is not connected.
 
 Since invoking it command line was successful we put a wrapper in place
 that executes cron and added the wrapper as a service.  cron is now
 working.  For the issue where crontab assumes that SYSTEM is the owner
 of the cron service and leaves the contents of the /var/cron/tabs as 640
 with a group of SYSTEM we also created a wrapper that changes the
 permission to 644.
 
 Here is the detail of what we have put into place and provided to the
 release engineer:
 
 overview:
   - create local user called 'svccron'
   - set up permissions for local user
   - create wrappers for cron/crontab
   - add wrapped cron as service
   - set up schedules
 
 execute the following within a bash shell:
 
   - create local user called 'svccron'
 
   net user svccron passwd /add /yes
   net localgroup administrators_group_name svccron /add
   editrights -a SeAssignPrimaryTokenPrivilege -u svccron
   editrights -a SeCreateTokenPrivilege -u svccron
   editrights -a SeIncreaseQuotaPrivilege -u svccron
   editrights -a SeServiceLogonRight -u svccron
   mkpasswd -l -u svccron  /etc/passwd
 
   for security reasons:
   editrights -a SeDenyInteractiveLogonRight -u svccron
   editrights -a SeDenyNetworkLogonRight -u svccron
   editrights -a SeDenyRemoteInteractiveLogonRight -u svccron
 
   - create wrapper for cron - attached 'xcron'/contents below
   - install in /usr/sbin
   - change permissions for xcron
   $ chmod 744 /usr/sbin/xcron
 
   $ cat /usr/sbin/xcron
   #!/usr/bin/bash
   /usr/sbin/cron -D
 
   $ ls -l /usr/sbin/xcron
   -rwxr-xr-x1 devadm   Domain U   34 Sep 10 14:03
 /usr/sbin/xcron
 
   - create wrapper for crontab - attached 'xcrontab'/contents
 below
   - install in /usr/bin
   - change permissions for xcrontab
   $ chmod 744 /usr/bin/xcrontab
 
   $ cat /usr/bin/xcrontab
   #!/usr/bin/bash
   /usr/bin/crontab $@;
   xval=$?;
   chmod 644 /var/cron/tabs/$USER;
   exit $xval;
 
   $ ls -l /usr/bin/xcrontab
   -rwxr-xr-x1 devadm   Domain U   90 Sep 10 13:58
 /usr/bin/xcrontab
 
   - add wrapped cron as service
   $ cygrunsrv -I xcron -p /usr/sbin/xcron -a -u svccron -w
 passwd
 
   - execute following as user for which tasks need to be scheduled
   $ xcrontab -e
 
   - for details
   $ man 5 crontab
 

I am wondering if the additional steps below would not remove
the need for xcrontab:

1) aftermkpasswd -l -u svccron  /etc/passwd
edit /etc/passwd and set the uid of svccron to 0

2) Change the command in xcron to
su -c exec /usr/sbin/cron -D system

Pierre


--
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/

[Mirror] new french mirror

2004-09-10 Thread Ludovic Bellier

Hello,

I've added a new mirror for cygwin, it is available via http
and ftp.

URL: ftp://mirror1.finix.eu.org/cygwin 
http://mirror1.finix.eu.org/cygwin/
Server location: France [fr], Paris
Band-width: more than 10Mbps
Organisation: Finix (French Linux User Group, city Brest,
http://finix.eu.org) and Zyrianes team (http://qeo.zyrianes.net)
Daily updated.

Someone know how being added to the official mirror list
(http://cygwin.com/mirrors.html) and the cygwin setup mirror list?
I've tried sourcemaster mail without success

Thank's.

-- 
ludovic Bellier

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

Re: Error Launching zsh

2004-09-10 Thread Peter A. Castro

On Wed, 8 Sep 2004, Mike Dillinger wrote:

 Hello,

Hi Mike,

 I am running Win2K SP4.  I have a shortcut to run zsh on my desktop that
 simply runs C:\cygwin\bin\zsh.exe.

 After powering on the PC or rebooting, I get the following error when I run
 the zsh shortcut:
  20 [main] zsh 688 fork_copy: linked dll data/bss pass 0 failed, 
 0x442000..0x4427F0, done 0, windows pid 1532, Win32 error 87
 /etc/zprofile:6: fork failed: resource temporarily unavailable
  306356 [main] zsh 688 fork_copy: linked dll data/bss pass 0 failed, 
 0x442000..0x4427F0, done 0, windows pid 1528, Win32 error 87
 /cygdrive/c/Documents and Settings/miked/.zshrc:41: fork failed: resource 
 temporarily unavailable

Hmmm... I haven't seen this before.  I do use Win2kSP4 as a build 
testing machine, along with the shortcut regularly.  I suppose it might
be necessary to rebase your Cygwin environment, but that's just a WAG.

 The second error (zprofile:6) is the 'id -un' command.

 I looked up Win32 error 87, and here is the text for it:
 87  The parameter is incorrect. ERROR_INVALID_PARAMETER

 Generally, if I close the errant zsh window and run the shortcut again,
 everything is fine.  bash works just fine - no problems.

Quite strange...  What happens if you, after a fresh reboot, open a
Command Prompt and type:
C:
cd c:\cygwin\bin
zsh -i -l

Do you get the same error or something else?

 Any suggestions why this is happening?  I've Googled for a while now and
 can't find anything.

Not a clue.  You'll need to provide details about your Cygwin environment
as well as anything extra you might have installed.

 Thanks,
 -MikeD

-- 
Peter A. Castro [EMAIL PROTECTED] or [EMAIL PROTECTED]
Cats are just autistic Dogs -- Dr. Tony Attwood

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

Re: Is it free to use

2004-09-10 Thread Andrew DeFaria

Bobby McNulty wrote:
Personally, I'm bored with the free verses copyrighted material.
Then why are you here (I mean in this thread)?
GPL means you should include your source code with your binary code. 
Simple. Redhat does it, Mandrake does it, Cygnus does it. Cygwin's 
sources are included in the download section. Next time, think before 
you speak.
Contrary to your opinion I did think before I spoke! I don't think the 
people who would use my application would want, know, care about or 
understand the source and I didn't want to go through the additional 
effort to make them available, place proper attribution in the sources 
and require them to download an install Cygwin (if I were to use 
cygwin1.dll).
--
Question: Why do people always seem to find things in the last place 
that they look? Answer: Because most people stop looking after they find it!

--
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: Cygwin Cron on Windows 2003 Server Issues

Larry/Igor - well - this is the first time I am participating in the
cygwin mailing list and from what I have read at http://ezmlm.org/ it is
supposed to add to threads if the subject is identical.  I have been
using the same subject line.  There was not an initial email that I
could have replied to nor was there a References: that I could have
manipulated when I started.  I read the responses by going to the site
http://cygwin.com/ml/cygwin/2004-09/ which just lists items
chronologically.  But, since it was not available night before last I
discovered the site http://www.mail-archive.com/[EMAIL PROTECTED]/ that
actually does have all my responses threaded correctly.  In fact, the
presentation is prefered since it is by thread/subject as opposed to by
date/time.  So, tell me what I am supposed to do such that responses are
correctly threaded and I will comply...

Igor - even though we did not state the command executed to add cron as
a service it did include the -D option as described in the message
http://www.mail-archive.com/[EMAIL PROTECTED]/msg31538.html.  Yes, we
have tried running it as you have described and it still fails with the
error: cron : PID 3784 : starting service `cron' failed: execv: 128,
Transport endpoint is not connected.  Also yes, you have correctly
pointed out that the -a is erroneous in my summation of steps.  I am
including an updated set of instructions just to ensure that any future
readers will not be mislead.

Pierre - your suggestion - will it not lead to root privileges for
svccron?  If yes, then that is not desirable.  However, I will try it
out and let all know the outcome.  In a bash shell with a simple command
it prompts for a password.

$ su -c exec echo foobar system
Password:
su: incorrect password

Thank you all once again.  Hopefully one of you can get it to work on
W2K3 without wrappers.


Regards,

Paul J. Ghosh

overview:
- create local user called 'svccron'
- set up permissions for local user
- create wrappers for cron/crontab
- add wrapped cron as service
- set up schedules

execute the following within a bash shell:

- create local user called 'svccron'

net user svccron passwd /add /yes
net localgroup administrators_group_name svccron /add
editrights -a SeAssignPrimaryTokenPrivilege -u svccron
editrights -a SeCreateTokenPrivilege -u svccron
editrights -a SeIncreaseQuotaPrivilege -u svccron
editrights -a SeServiceLogonRight -u svccron
mkpasswd -l -u svccron  /etc/passwd

for security reasons:
editrights -a SeDenyInteractiveLogonRight -u svccron
editrights -a SeDenyNetworkLogonRight -u svccron
editrights -a SeDenyRemoteInteractiveLogonRight -u svccron

- create wrapper for cron - attached 'xcron'/contents below
- install in /usr/sbin
- change permissions for xcron
$ chmod 744 /usr/sbin/xcron

$ cat /usr/sbin/xcron
#!/usr/bin/bash
/usr/sbin/cron -D

$ ls -l /usr/sbin/xcron
-rwxr-xr-x1 devadm   Domain U   34 Sep 10 14:03
/usr/sbin/xcron

- create wrapper for crontab - attached 'xcrontab'/contents
below
- install in /usr/bin
- change permissions for xcrontab
$ chmod 744 /usr/bin/xcrontab

$ cat /usr/bin/xcrontab
#!/usr/bin/bash
/usr/bin/crontab $@;
xval=$?;
chmod 644 /var/cron/tabs/$USER;
exit $xval;

$ ls -l /usr/bin/xcrontab
-rwxr-xr-x1 devadm   Domain U   90 Sep 10 13:58
/usr/bin/xcrontab

- add wrapped cron as service
$ cygrunsrv -I xcron -p /usr/sbin/xcron -u svccron -w passwd

- execute following as user for which tasks need to be scheduled
$ xcrontab -e

- for details
$ man 5 crontab


xcrontab
Description: Binary data


xcron
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: cygwin Cron on Windows 2003 Server Issues

On Fri, Sep 10, 2004 at 08:27:05PM -0500, Paul J. Ghosh wrote:
Larry/Igor - well - this is the first time I am participating in the
cygwin mailing list and from what I have read at http://ezmlm.org/ it is
supposed to add to threads if the subject is identical.

URL?  This is the first that I've heard that ezmlm is broken on
sourceware.org.

I have been using the same subject line.  There was not an initial
email that I could have replied to nor was there a References: that I
could have manipulated when I started.  I read the responses by going
to the site http://cygwin.com/ml/cygwin/2004-09/ which just lists items
chronologically.

Look closer.  There are a variety of listing options available.

Regardless, this has nothing to do with how some random web archiver
lists email.

You're reading a mailing list.  Respond to the email using any modern
email client and you will preserve threading.

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: cygwin 1.5.11: execv doesn't set argv[0] on Windows programs

On Fri, Sep 10, 2004 at 02:19:40PM -0500, David A. Rogers wrote:
On Fri, 10 Sep 2004, Chuck McDevitt wrote:

 argv and argc are concepts from the C runtime, not the Windows OS.

 The actual entry point to your program is to a routine that calls the
 initialization routines of the C library, then calls winMain.


Yes, certainly.  The point I was making was that the information was available
in a true win32 app.  David Korn implied that the WinMain call was the only
access to the calling arguments that were available to a win32 app.  That is
not true, as I showed.

This is what Dave Korn said:

But it's fair to point out that the argv concept is not part of the OS,
it's part of a compiler support library;

You don't seem to be getting the concept that this is not part of the
OS.  It is an extension imposed by msvcrt.dll.  If the application
doesn't use msvcrt.dll then there is no guarantee that argv[0] will be
set as expected.

You can write 100 programs showing that you are seeing argv and argc.
That doesn't mean that it is part of the standard Windows API.

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: cygwin Cron on Windows 2003 Server Issues

Christopher - you lost me.  How am I supposed to reply to your message
from the site http://cygwin.com/ml/cygwin/2004-09/msg00505.html?  My
current response is being composed by clicking on the button at the
bottom of the page
http://www.mail-archive.com/[EMAIL PROTECTED]/msg44535.html which I am
assuming will contain the correct reference.  Help me out here - I have
web site blindness!

Yes, I have now looked closer and discovered the 'Thread Index' option
on http://cygwin.com/ml/cygwin/2004-09/.  Thank you.

Regards,

Paul J. Ghosh

 


--
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: What Cygwin is

2004-09-10 Thread Gary R. Van Sickle

 On Thu, 9 Sep 2004, Corinna Vinschen wrote:
 
  On Sep  8 11:31, Brian Ford wrote:
   No, now we continually have to explain why you can't install said
   x86 Linux binary/package and run it under our Linux 
 emulator.  Heck, 
   you can't even *compile* said Linux firewall/router/proxy server 
   etc. under our Linux emulator.  Doesn't sound like much 
 of an emulation to me.
 
  You don't think that's the fault of our project page, do 
 you?  Did you 
  read it lately?  For about three months, there's a pretty 
 precise text 
  with the headline What Isn't Cygwin?
 
 Sorry, no; I had not read it within that time period.  I was 
 going on recollection; my bad.  It is much better now.
 
 However, if Cygwin is a Linux emulation layer providing 
 substantial Linux API functionality, then shouldn't Linux 
 APIs and behavior now trump POSIX and the Single Unix 
 Specification?  Is that what we really want?


No.
 
 I'm just not comfortable with the removal of all all 
 references to POSIX and UNIX from that description.
 

Nor I, nor apparently many others, especially when the replacement is simply
not correct.  And anyway, if it's going to be a Linux emulation layer, it
better be changed to GNU/Linux emulation layer or Stallman will throw a
hissy fit.

-- 
Gary R. Van Sickle
 


--
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: diff crashes with big files on Itanium machine

2004-09-10 Thread Gary R. Van Sickle

 Hi, All!
 
 I would like to report a problem I have with Cygwin's diff. 
 It crashes when I use it on 2 big files (the size is 100M 
 each) on Itanium machines under Windows .NET server. It works 
 OK on x86 machines.

There's a reason they call it Itanic.

HHehehehehee!  Man I love that one!

-- 
Gary R. Van Sickle


--
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: Cygwin Cron on Windows 2003 Server Issues

2004-09-10 Thread Pierre A. Humblet

On Fri, Sep 10, 2004 at 08:27:05PM -0500, Paul J. Ghosh wrote:
 
 Pierre - your suggestion - will it not lead to root privileges for
 svccron?

You mean setting its uid to 0? It won't change anything for Windows
ntsec. The advantage is that su won't prompt for passwd.
The reason for the su is that cron will run as system, which is what 
crontab expects.

  If yes, then that is not desirable.  However, I will try it
 out and let all know the outcome.  In a bash shell with a simple command
 it prompts for a password.
Except if the uid is 0 !

If it works you could also try to get rid of xcron and run something like
$ cygrunsrv -I cron -p /usr/bin/su -a -c 'exec /usr/sbin/cron -D' system 
-u svccron -w passwd
(my quoting may be off).
Also don't forget to delete /var/run/cron.pid when changing cron daemon uid.

Pierre

--
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: OT: RE: filesystem encoding

2004-09-10 Thread Gary R. Van Sickle

 Gary R. Van Sickle wrote:
 
 Welcome to the 21st century, where computers can't even 
 unambiguously 
 represent written text.
   
 
 Isn't this something unicode was meant to solve?

Yes, and if implemented properly, it mostly does.

 or does 
 unicode still need a codepage to map to glyphs?

No, not a system-wide one anyway.  IIRC (and I'm no Unicode expert), the
verbage in the specs talk about codepages (or something like that) a lot,
but it's a few bits in each character that specify something similar to the
ASCII/ISO-style codepages, the end result being that each individual Unicode
character is unambiguously represented[1].

-- 
Gary R. Van Sickle
 
[1] Gross oversimplification alert.  Some of the Asian languages have
characters who's precise glyph depends on the previous character, and I
think that introduces some context sensitivity.  Ah well, we have to leave
*something* for the 31st century folks to fix. :-(.


--
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: cd in bash script not being executed

2004-09-10 Thread Aaron Stephanic

I'm having the same problem as in this thread:
http://www.cygwin.com/ml/cygwin/2004-07/msg00197.html
cd in bash scripts is not working.  Is this a bug?
Aaron Stephanic
--
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: cd in bash script not being executed