RE: Please upload new version of curl (7.10.2-1)
No, but thanks for checking. The source maintainer upgraded his versions of autotools, and since then I haven't been able to cleanly build without rebuilding the Makefile.in's first. I don't have the knowledge or expertise needed to actually find the specific problem, so I just rebuilt all of them. I am running the latest versions of gcc and autotools as distributed with cygwin. If this is excessively large, I could attempt just rebuilding the specific one-or-two Makefile.in's that are affected. Do you want me to do that? Or, since this is just the source package, could we leave it as is for now, and revisit when the next curl upgrade is released? If anyone wants a crack at the autotools issue, try building the official source package, as downloadable from http://curl.haxx.se/download.html. After running a configure followed by a make, and get an error during building in the ./lib sub-directory about WinMain@12 something-or-another. --Kevin -Original Message- From: Corinna Vinschen [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 3:32 AM To: [EMAIL PROTECTED] Subject: Re: Please upload new version of curl (7.10.2-1) On Tue, Nov 19, 2002 at 04:20:32PM -0500, Roth, Kevin P. wrote: The files are available as follows: Binary: http://curl.haxx.se/download/curl-7.10.2-1-cygwin.tar.bz2 Source: http://curl.haxx.se/download/curl-7.10.2-1-cygwin-src.tar.bz2 Devel: http://curl.haxx.se/download/curl-devel-7.10.2-1-cygwin.tar.bz2 Do you have a packaging error, perhaps? The old source file is 680K, the new one is 1.1M. Looking into them, there's an interesting difference: $ tar tvIf curl-7.10.1-1-src.tar.bz2 [...] -rw-r--r-- kproth/None5664 2002-10-11 15:03:00 curl-7.10.1-1/CYGWIN-PATCHES/curl-7.10.1-1.patch ^^^ $ tar tvIf curl-7.10.2-1-src.tar.bz2 [...] -rw-r--r-- kproth/None 1686363 2002-11-19 12:14:14 curl-7.10.2-1/CYGWIN-PATCHES/curl-7.10.2-1.patch ^^^ Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc.
Re: Setup's Choose a download source screen...
On Thu, 2002-11-21 at 07:28, Max Bowsher wrote: The principal difficulty is that there is no good framework in place for persisting setup's configuration. Yep. In fact I've commented on this before - that we need a common persistence mechanism to deal with the moving requirements for where to store settings, and to do the storage for us. Patches happily reviewed. Rob -- --- GPG key available at: http://users.bigpond.net.au/robertc/keys.txt. --- signature.asc Description: This is a digitally signed message part
RE: Errors when building setup.exe
From: Robert Collins Sent: Wednesday, 20 November 2002 7:49 pm To: Max Bowsher Cc: Thomas Chadwick; [EMAIL PROTECTED] Subject: Re: Errors when building setup.exe On Thu, 2002-11-21 at 07:19, Max Bowsher wrote: PS to Robert Collins: Perhaps the setup source tarball should be removed from the mirrors? Its so old as to be impractical to use now. Can't do that - it's the source code that must be offered (GPL :}). Rob Followup to cygwin-apps please. I think he means replace, not remove ;) J.
RE: Errors when building setup.exe
On Thu, 2002-11-21 at 08:05, John Morrison wrote: I think he means replace, not remove ;) Even so. The source downloadable via setup is that for the same version we offer on cygwin.com. To replace it with a newer version's tarball means we need a new version of setup available there. Rob -- --- GPG key available at: http://users.bigpond.net.au/robertc/keys.txt. --- signature.asc Description: This is a digitally signed message part
RE: Errors when building setup.exe
From: Robert Collins [mailto:[EMAIL PROTECTED]] On Thu, 2002-11-21 at 08:05, John Morrison wrote: I think he means replace, not remove ;) Even so. The source downloadable via setup is that for the same version we offer on cygwin.com. To replace it with a newer version's tarball means we need a new version of setup available there. *shrug* I know. How's it going anyway? J.
Re: Fw: Detecting NetworkSimplicity in setup.exe
On Thu, 2002-11-21 at 05:18, Mark Bradshaw wrote: I've decided to call it quits on the package. Announcement at http://www.networksimplicity.com/openssh/. I hope we can make a minimal cygwin w/ssh as easy to use and install as your package was. Good luck with your next endeavour(s). Cheers, Rob -- --- GPG key available at: http://users.bigpond.net.au/robertc/keys.txt. --- signature.asc Description: This is a digitally signed message part
RE: Errors when building setup.exe
Even so. The source downloadable via setup is that for the same version we offer on cygwin.com. To replace it with a newer version's tarball means we need a new version of setup available there. *shrug* I know. How's it going anyway? Massive overhauls. Much new functionality. Not enough hours in the day. You know, the usual. ;-) -- Gary R. Van Sickle Brewer. Patriot.
RE: Detecting NetworkSimplicity in setup.exe
Nyah, just make a command option to ignore 'insane mount tables'. That gives the know-it-alls a knob to twist. Rob, PROMISE me that this option gets the long name --ignore-insane-mount-tables. PROMISE ME! How about this: Whoever implements it gets to name it :}. There's always a catch. ;-) -- Gary R. Van Sickle Brewer. Patriot.
Re: Please upload new version of curl (7.10.2-1)
On Wed, Nov 20, 2002 at 02:50:15PM -0500, Roth, Kevin P. wrote: If this is excessively large, I could attempt just rebuilding the specific one-or-two Makefile.in's that are affected. Do you want me to do that? Or, since this is just the source package, could we leave it as is for now, and revisit when the next curl upgrade is released? It's your choice. I've just uploaded the package. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc.
re: Direction of xwinclip
I now remember what it was that I did not like: Ownership of the X selection is grabbed when XWin.exe loses focus. Is that correct? No. It only takes ownership selection when XWin is gaining activation (and not from another window). When Xwin looses activation (and therefore focus) it merely grabs whatever is in the selections, if anything, and puts it into the windows clipboard. Robert's method of watching for changes in XA_CUT_BUFFER0 might be hack, but I think it is probably closer to what other X Servers for Windows are doing, as those other X Server do not grab ownership of the X selection when they lose the focus. Neither does my code, but see later... Have you got any objections to watching for changes in XA_CUT_BUFFER0 as a signal that we should request another copy of the X selection? Yes I have, the ICCM standard says don't use it in new applications and the rest of the x development community says don't use it in new applications. Seemingly only a few do. Motif apps have some wierd logic for working out wether to try and use it but they don't use it directly (they've got their own wierd and wonderful way of doing things for CLIPBOARD). Basically as the ICCM says it's obsolete. You can write a compliant X selection app that works with PRIMARY without problems but never uses the cut buffers. How many applications have you tried it with, sorry I can't try it at mine, gcc's specs don't seem to work for my box (or the work box I have it on). If we decide to go with the XA_CUT_BUFFER0 hack, then there will probably need to be only one more release of xwinclip. After that we can concentrate on integrating the clipboard support into XWin.exe. I think that it would be usefull (possibly as thats the way xceed works, when it does, to make it an xt/xm app as well) so that we can get the motif clipboard if it's being used. However within XWin it may not be the best approach either way, as all messages go through the xserver anyway should not a simple hook into the server code be better then making a client within a server (and having to look at threading it or more complicated setjmp's). By the way... I like the setjmp/longjmp stuff... I thought it was two assembly language instructions that were used. Imagine my surprise when I saw setjmp.h and realized that setjmp and longjmp were (used like) functions. That is a heck of a lot easier than I had thought it would be. Glad it could help. Thanks for your contributions Chris. cheers, Chris _ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail
Re: Xwinclip doesn't work with Konsole
Hi! Just to check (I don't have kde available) this when you say copy/paste do you mean the edit menu copy and paste? No. Or do you mean highlight in Konsole and then use the middle button. Yes, this is what I meant. Best Regards Jozsef Kercso
Re: Xwinclip doesn't work with Konsole
Hi! Damn this is interesting. Just to check (I don't have kde available) this when you say copy/paste do you mean the edit menu copy and paste? No. good good. Or do you mean highlight in Konsole and then use the middle button. Yes, this is what I meant. What does the output of my xwinclip have in it? does it show something like : no xa_primary owner or maybe XToWindows XA_PRIMARY convertselect() failed It would mean that there is some dodgy non-standard activity going on or that the format can't be converted. Chris _ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail
RE: Xwinclip doesn't work with Konsole
-Original Message- From: Chris Twiner [SMTP:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 4:08 PM To: [EMAIL PROTECTED] Subject: Re: Xwinclip doesn't work with Konsole It would mean that there is some dodgy non-standard activity going on or that the format can't be converted. I don't know the details, but I've heard it said that some aspect of copy and paste is broken in Qt 2 when it comes to interoperating with clients using other widget sets, but is fixed in Qt 3. Maybe this is what you're up against? -- Peter Oliver Midrange Services Aquila Networks Services Ltd Confidentiality: This e-mail and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this e-mail in error, use of this information (including disclosure, copying or distribution) may be unlawful. Please notify [EMAIL PROTECTED] and delete the message immediately. Security: Internet e-mail is not a 100% secure communications medium. Viruses: This e-mail (and any attachments) has been checked (using Sophos Sweep 3.63 + patches) and found to be clean from any virus infection before leaving. Therefore neither Aquila Networks Services Ltd nor Midlands Electricity plc or any of their group undertakings (as defined by the Companies Act 1989) (together referred to as the Companies) accept legal responsibility for this message or liability for the consequences of any computer viruses which may have been transmitted by this e-mail. Monitoring: All electronic communications with the Companies may be monitored in accordance with the UK Regulation of Investigatory Powers Act, Lawful Business Practice Regulations, 2000. If you do not consent to such monitoring, you should contact the sender of this e-mail. Aquila Networks Services Limited, Registered office: Whittington Hall, Whittington, Worcester, WR5 2RB Registered in England and Wales number 3600545 This e-mail may be sent on behalf of any of the Companies.
Compile error with current CVS
Hi, An update to the current XFree86 CVS revealed some compile problems. xc/programs/Xserver/hw/xfree86/common/xf86.h now includes xf86Pci.h which is found in xc/programs/Xserver/hw/xfree86/os-support/bus. The patch adds the path to the include list, but I'm not sure if this is the correct fix. Maybe Alan put some light to this topic. bye ago Index: programs/Xserver/xkb/Imakefile === RCS file: /cvs/xc/programs/Xserver/xkb/Imakefile,v retrieving revision 3.19 diff -u -r3.19 Imakefile --- programs/Xserver/xkb/Imakefile 2002/11/20 04:49:02 3.19 +++ programs/Xserver/xkb/Imakefile 2002/11/20 20:19:18 @@ -26,7 +26,7 @@ XKB_DDXDEFS = XkbServerDefines #ifdef XFree86Version -XF86INCLUDES = -I$(XF86COMSRC) -I$(XF86OSSRC) +XF86INCLUDES = -I$(XF86COMSRC) -I$(XF86OSSRC) -I$(XF86OSSRC)/bus XF86_OBJS = xf86KillSrv.o xf86VT.o xf86Private.o #endif --
cygwin/xfree86 startup problem
I had a problem exactly like the one below except I did install everything in C drive in the proper directories. The only thing I can thin is that I downloaded the cygwin setup.exe to a zip and ran it and downloaded the files to the zip. Then I copied setup.exe and all the cygwin files directly to c:/cyginstall. Then I ran setup.exe again, this time choosing to install from local directory to c:/cygwin. Then I copied the zipped xfree binaries from my zip to c:/cygwin/tmp. Then I installed cygwin/xfree86 exactly according to the users guide, but when I doubleclicked on startxwin.bat, I got the error it could not find cygwin1.dll even though I could see it sitting there in c:/cygwin/bin. There were also messages in the terminal about not being able to locate Xwin, Xterm, and twm. Editing the path to explicitly include c:\ does nothing in my case. I am also using Windows 98SE. Phil Blood -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Harold L Hunt Sent: Friday, July 12, 2002 10:27 AM To: willichnicht habichnet; [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: Re: Bug in startxwin.bat after installing with setup.exe in win98SE First: Wrong mailing list. Send all Cygwin/XFree86 questions to [EMAIL PROTECTED] I have redirected this discussion to the correct mailing list, but be careful to remove [EMAIL PROTECTED] from the ``cc'' if you reply to this message. Second: Read the extensive comments in startxwin.bat that talk about CYGWIN_ROOT. Not specifying a drive causes the path to reference \cygwin on the current drive (such as c:\cygwin, or d:\cygwin). You have to run startxwin.bat from the same drive that Cygwin is installed on, or it won't get the correct drive letter. startxwin.bat has been using this path setting system for over a year (I think) and not a *single* person has reported a problem with it until today. You must be doing something on your system that other users thought better of (such as running startxwin.bat from the d drive, or from a network drive, when Cygwin is installed on the c drive). I have to disagree. Absence of complaints does not imply absence of errors. Many people become used to buggy software that only works if you don't deviate from the well-trodden path. That's no excuse to fail to fix a known problem. The comments you mention: REM REM The path in the CYGWIN_ROOT environment variable assignment assume REM that Cygwin is installed in a directory called 'cygwin' in the root REM directory of the current drive. You will only need to modify REM CYGWIN_ROOT if you have installed Cygwin in another directory. For REM example, if you installed Cygwin in \foo\bar\baz\cygwin, you will need REM to change \cygwin to \foo\bar\baz\cygwin. REM REM This batch file will almost always be run from the same drive (and REM directory) as the drive that contains Cygwin/XFree86, therefore you will REM not need to add a drive letter to CYGWIN_ROOT. For example, you do REM not need to change \cygwin to c:\cygwin if you are running this REM batch file from the C drive. REM don't explain what is wrong with adding a drive lettter, only that it almost always works because the drive letter already matches. You need the drive letter to make an absolute path. (By the way, . is redundant in a Windows path anyway.) Harold willichnicht habichnet [EMAIL PROTECTED] said: i installed cygwin/XFree86 with the recommended setup.exe in win98SE. starting X with startxwin.bat resulted in errormessages like cygwin1.dll not found etc. in startxwin.bat there was a line SET CYGWIN_ROOT=\cygwin which seemed to be incorrect because theres no drive given. after editing it to SET CYGWIN_ROOT=c:\cygwin (installation path) everything works fine. i installed it on a win2000 too without these problems but on an other win98-system same corrupted startxwin.bat was there. i have not tried an other installation directory or drive. this message is only for information; i need no other solution. Michael Heide
Re: ntsec patch 1: uid==gid, chmod, alloc_sd, is_grp_member
On Sun, Nov 17, 2002 at 10:42:55PM -0500, Pierre A. Humblet wrote: At 06:56 PM 11/15/2002 +0100, Corinna Vinschen wrote: On Fri, Nov 15, 2002 at 12:29:44PM -0500, Pierre A. Humblet wrote: Alternatively I could add it, but add a check for group sid is SYSTEM, and then skip the step. That would be very easy to do, and to remove later when ssh is ready. I like this best actually. Good idea! Me too. But that must go into both functions, get_attribute_from_acl() and alloc_sd(). OK for get_attribute_from_acl (which is where the /var/empty problem originates), but why for alloc_sd () ? The patch will do the right thing, whereas the current code can give rise to unexpected results. See below. Ok. Applied. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc.
Re: ntsec patch #4: passwd and group
On Sun, Nov 17, 2002 at 10:44:18PM -0500, Pierre A. Humblet wrote:
Corinna,
Almost same as last week, but with the diff and ChangeLog against current CVS.
I also made a small change.
[...]
Some questions and comments:
* security.h: [...]. Undeclare internal_getpwent.
^^^
You didn't.
* passwd.cc (getpwsid): Create.
(internal_getpwent): Suppress.
(read_etc_passwd): Make static. Rewrite the code for the completion
line. Set curr_lines to 0.
(parse_pwd): Change type to static int. Return 0 for short lines.
(add_pwd_line): Pay attention to the value of parse_pwd.
(search_for): Do not look for nor return the DEFAULT_UID.
* grp.cc (read_etc_group): Make static. Free gr_mem and set
curr_lines to 0. Always call add_pwd_line. Rewrite the code for the
completion line.
(parse_grp): If grp.gr_mem is empty, set it to null_ptr.
Never NULL gr_passwd.
(getgrgid32): Only return the default if ntsec is off and the gid is
ILLEGAL_GID.
* sec_helper.cc (cygsid::get_id): Use getpwsid and getgrsid;
(cygsid_getfrompw): Clean up last line.
(cygsid_getfromgr): Ditto.
(is_grp_member): Use getpwuid32 and getgrgid32.
* uinfo.cc (internal_getlogin): Set DEFAULT_GID at start.
Use getpwsid. Move the read of /etc/group after the second access
of /etc/passwd. Change some debug_printf.
-/* FIXME: should be static but this is called in uinfo_init outside this
- file */
-void
+static void
read_etc_group ()
Do I miss something? I don't see that in this patch.
@@ -150,76 +145,74 @@ read_etc_group ()
if (group_state != initializing)
{
group_state = initializing;
+ for (int i = 0; i curr_lines; i++)
+ if ((group_buf + i)-gr_mem != null_ptr)
+ free ((group_buf + i)-gr_mem);
+
+ curr_lines = 0;
if (gr.open (/etc/group))
{
char *line;
while ((line = gr.gets ()) != NULL)
- if (strlen (line))
- add_grp_line (line);
+add_grp_line (line);
group_state.set_last_modified (gr.get_fhandle (), gr.get_fname ());
- group_state = loaded;
gr.close ();
debug_printf (Read /etc/group, %d lines, curr_lines);
}
- else /* /etc/group doesn't exist -- create default one in memory */
- {
- char group_name [UNLEN + 1];
- DWORD group_name_len = UNLEN + 1;
- char domain_name [INTERNET_MAX_HOST_NAME_LENGTH + 1];
- DWORD domain_name_len = INTERNET_MAX_HOST_NAME_LENGTH + 1;
- SID_NAME_USE acType;
+
+ /* Complete /etc/group in memory if needed */
+ if (!getgrgid32 (myself-gid))
?!? How is that supposed to work? We're in group_state==initializing,
therefore in getgrgid32(), read_etc_group() is called. Isn't that
somewhat dangerous?
+ cygheap-user.groups.pgsid.string (strbuf);
+ if (!(gr = getgrsid (cygheap-user.groups.pgsid)))
+ {
+ if (!LookupAccountSidA (NULL, cygheap-user.groups.pgsid,
+ group_name, group_name_len,
+ domain_name, domain_name_len,
+ acType))
+ debug_printf (Failed to get primary group name. %E);
Didn't you propose to get rid of the LookupAccountSidA() calls?
- return allow_ntsec ? NULL : default_grp;
+ return (!allow_ntsec gid == ILLEGAL_GID)?default_grp:NULL;
I'd better like
return allow_ntsec || gid != ILLEGAL_GID ? NULL : default_grp;
extern C struct __group16 *
@@ -482,13 +474,9 @@ setgroups32 (int ngroups, const __gid32_
for (int gidy = 0; gidy gidx; gidy++)
if (grouplist[gidy] == grouplist[gidx])
goto found; /* Duplicate */
- for (int gidy = 0; (gr = internal_getgrent (gidy)); ++gidy)
- if (gr-gr_gid == (__gid32_t) grouplist[gidx])
- {
- if (gsids.addfromgr (gr))
- goto found;
- break;
- }
+ if ((gr = getgrgid32 (grouplist[gidx]))
Ahem, I thought we agreed that we don't call external functions from
inside Cygwin? Never mind, there are still some of them which we have
to eliminate, anyway.
@@ -208,24 +190,17 @@ is_grp_member (__uid32_t uid, __gid32_t
}
/* Otherwise try getting info from examining passwd and group files. */
- for (int idx = 0; (pw = internal_getpwent (idx)); ++idx)
-if ((__uid32_t) pw-pw_uid == uid)
- {
- /* If gid == primary group of uid, return immediately. */
- if ((__gid32_t) pw-pw_gid == gid)
- return TRUE;
- /* Otherwise search for supplementary user list of this group. */
- for (idx = 0; (gr = internal_getgrent (idx)); ++idx)
- if ((__gid32_t) gr-gr_gid == gid)
- {
- if (gr-gr_mem)
Re: ntsec patch #4: passwd and group
On Wed, Nov 20, 2002 at 10:04:53AM -0500, Pierre A. Humblet wrote: Corinna Vinschen wrote: Ahem, I thought we agreed that we don't call external functions from inside Cygwin? Never mind, there are still some of them which we have to eliminate, anyway. I didn't know about that policy but it suits me fine. As we discussed, internal calls to passwd/group functions should never reread the files, so new entry points are called for. I was going to do that in a second step, it wasn't a goal when I started. The problem I'm talking about is beyond this patch. Just reiterating (so we know that we're talking about the same): We talked about the need to use a static buffer in the external getpwXXX and getgrXXX functions to allow apps to use the last return value (a pointer) even though the file is going to be reread which destroys all buffer space. If Cygwin calls these functions internally, the static buffers would be overwritten and the application itself gets false data. As you say, we need to substitute all internal calls to getpwXXX and getgrXXX funcs (except getXXsid, of course) by calling appropriate internal functions as internal_getpwuid(). Then we should be safe again. Probably it would make sense to rename getXXsid() funcs to internal_getXXsid now to have a consistent naming scheme? How do you want to proceed? Apply this patch and undeclare internal_getpwent, remove LookupAccountSidA(), apply your I'd better like and introduce internal lookup functions in a few days, or prepare a single all-encompassing patch in a few days? I patched your patch already slightly: - Undeclare internal_getpwent - I'd better like - Avoid a compiler warning by using %lu instead of %u in sprintf'ing uids and gids. - A few minor formatting issues. How would you like to remove LookupAccountSidA? Just remove it and debug_printf (Failed to get primary group name.) ? Rewriting the external funcs and creating and using the internal funcs is ok for another patch. Corinna
Re: ntsec patch #4: passwd and group
Corinna Vinschen wrote: The problem I'm talking about is beyond this patch. Just reiterating (so we know that we're talking about the same): We talked about the need to use a static buffer in the external getpwXXX and getgrXXX functions to allow apps to use the last return value (a pointer) even though the file is going to be reread which destroys all buffer space. If Cygwin calls these functions internally, the static buffers would be overwritten and the application itself gets false data. As you say, we need to substitute all internal calls to getpwXXX and getgrXXX funcs (except getXXsid, of course) by calling appropriate internal functions as internal_getpwuid(). Then we should be safe again. ...and we don't need to change anything to the external functions (so we know that we're talking about the same). See below. Probably it would make sense to rename getXXsid() funcs to internal_getXXsid now to have a consistent naming scheme? Yes. I patched your patch already slightly: Thanks. How would you like to remove LookupAccountSidA? Just remove it and debug_printf (Failed to get primary group name.) ? OK, the debug printf can be removed too, the result will be visible in the other debug_printf below. A more ambitious way is to see if the sid is local or not. If it is local we can safely call LookupAccountSidA. Rewriting the external funcs and creating and using the internal funcs is ok for another patch. Glad we talk! The external non multithread functions are already OK (i.e. POSIX), as long as we have the new internal functions. In other words, if the internal functions don't reread the files, then there is no need for new static buffers. The existing buffers can only be wiped out on the next external call, which is OK. The multithread issues are still outside the next patch. Pierre
Re: ntsec patch #4: passwd and group
On Wed, Nov 20, 2002 at 11:29:08AM -0500, Pierre A. Humblet wrote: How would you like to remove LookupAccountSidA? Just remove it and debug_printf (Failed to get primary group name.) ? OK, the debug printf can be removed too, the result will be visible in the other debug_printf below. A more ambitious way is to see if the sid is local or not. If it is local we can safely call LookupAccountSidA. I chose the less ambitious way for now. Rewriting the external funcs and creating and using the internal funcs is ok for another patch. Glad we talk! The external non multithread functions are already OK (i.e. POSIX), as long as we have the new internal functions. In other words, if the internal functions don't reread the files, then there is no need for new static buffers. The existing buffers can only be wiped out on the next external call, which is OK. You're correct, sure. Somehow I was already fixed on the static buffer track. Applied with changes. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc.
Re: bug report: poll() with listen sockets always gives POLLERR
On Tue, Nov 19, 2002 at 01:34:30PM +, Steven O'Brien wrote: Hi The current implementation of poll() does not behave correctly with listen sockets. It always gives a POLLERR revent when a connection request is received. I believe the error is in poll.cc lines 96-108: [...] Thanks for the report. I've checked in a patch. Please try the next developers snapshot. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Hi, /usr/lib/gcc-lib/i686-pc-mingw32/3.2/../../../../i686-pc-mingw32/bin/ld.exe --shared -Bdynamic -e _DllMainCRTStartup@12 -o hello.dll -export-all /usr/lib/gcc-lib/i686-pc-mingw32/3.2/../../../../i686-pc-mingw32/lib/dllcrt2 .o /usr/lib/gcc-lib/i686-pc-mingw32/3.2/crtbegin.o -L/usr/lib/gcc-lib/i686-pc-m ingw32/3.2 -L/usr/lib/gcc-lib/i686-pc-mingw32/3.2/../../../../i686-pc-mingw3 2/lib -L/usr/lib/gcc-lib/i686-pc-mingw32/3.2/../../.. /var/tmp/ccqxxSSp.o --out-implib libhello.dll.a -lmingw32 -lgcc -lmoldname -lmingwex -lmsvcrt -lmingw32 -luse r32 -lkernel32 -ladvapi32 -lshell32 -lmingw32 -lgcc -lmoldname -lmingwex -lm svcrt /usr/lib/gcc-lib/i686-pc-mingw32/3.2/crtend.o Using gcc-3.2-3. this is vastly different then the machine's I have. I have attached an untouched spec file fresh from setup.exe. Are there any differences with yours? Chris _ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus specs.cp.gz Description: GNU Zip compressed data -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: bug report: poll() with listen sockets always gives POLLERR
Corinna Vinschen wrote: On Tue, Nov 19, 2002 at 01:34:30PM +, Steven O'Brien wrote: Hi The current implementation of poll() does not behave correctly with listen sockets. It always gives a POLLERR revent when a connection request is received. I believe the error is in poll.cc lines 96-108: [...] Thanks for the report. I've checked in a patch. Please try the next developers snapshot. Corinna Thanks Corinna, but its still not quite right. poll() needs to set POLLIN when a connection request is received on a listen socket. The following patch completes the job: --- poll.cc 2002-11-20 08:57:03.0 + +++ poll.cc-fix 2002-11-20 09:01:21.0 + @@ -111,9 +111,12 @@ poll (struct pollfd *fds, unsigned int n Unfortunately, recvfrom() doesn't make much sense then. It returns WSAENOTCONN in that case. Since that's not actually an error, -we must not set POLLERR. */ +we must not set POLLERR, but POLLIN to +indicate the request. */ if (WSAGetLastError () != WSAENOTCONN) fds[i].revents |= POLLERR; + else + fds[i].revents |= POLLIN; break; case 0: /* Closed on the read side. */ fds[i].revents |= POLLHUP; Regards, Steven -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Conflict between libcygwin.a and GCC core libraries
Max Bowsher wrote: Actually, pipe and kill might not be. Don't know. You'd have to look and find out. Well, the undefined references to pipe and kill were solved by linking one of the forbidden libraries (-lcygwin, -lc, -lg...), so it seems there are at least dummy implementations, or do I misinterpret this? As it looks like now, my only alternatives are: Linking one of those forbidden libraries and getting a multiple definition of `atexit', or not linking them and getting that bunch of undefined references. What can I do here? I've been sitting on this problem for a week now, and I don't have any ideas what more to try. Markus -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
You've left out the actual error. I think that -lm is an alias of -lcygwin. Get rid of it. Neither Cygwin or MinGW has a seperate math library. Max. The error was in my previous post. But thanks, the -lm was the problem. I left it in since gnucap being an electrical simulator it didn't seem stupid to link with the math library. Is there any reason for -lm to be an alias of -lcygwin ? I looks rather confusing. Anyway, thanks again, it works. Denis. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
bash builtin pwd returns Windows style names
Hello, I observed that in previous releases of Cygwin (I think it was 1.3.12) the bash builtin pwd returned directory names in the form: /cygdrive/c/classes but if I try this now with the release 1.3.14 (and also with 1.3.15 and bash 2.05b) I get c:/classes I wonder if this has been intentionally changed, and if so, whether it is possible to configure it either way. Thanks in advance for comments on this, Norbert Richter -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Conflict between libcygwin.a and GCC core libraries
Subject: Re: Conflict between libcygwin.a and GCC core libraries From: Markus Gerwinski [EMAIL PROTECTED] Date: Wed, 20 Nov 2002 10:36:14 +0100 To: [EMAIL PROTECTED] Max Bowsher wrote: Actually, pipe and kill might not be. Don't know. You'd have to look and find out. Well, the undefined references to pipe and kill were solved by linking one of the forbidden libraries (-lcygwin, -lc, -lg...), so it seems there are at least dummy implementations, or do I misinterpret this? As it looks like now, my only alternatives are: Linking one of those forbidden libraries and getting a multiple definition of `atexit', or not linking them and getting that bunch of undefined references. What can I do here? I've been sitting on this problem for a week now, and I don't have any ideas what more to try. You are mixing the Cygwin and MinGW environments. This *will not work*. The thing you are trying to compile does not appear to have been ported to allow it to run on native Windows, and to run it on Cygwin would require a Cygwin-compiled gtk+, which is a whole new can of worms. You have 2 options, as far as I can see - give up, or devote lots of time to either porting GPA, or compiling a Cygwin gtk+ (and all its many dependencies.) Max. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Conflict between libcygwin.a and GCC core libraries
Max Bowsher wrote: You are mixing the Cygwin and MinGW environments. This *will not work*. The thing you are trying to compile does not appear to have been ported to allow it to run on native Windows, and to run it on Cygwin would require a Cygwin-compiled gtk+, which is a whole new can of worms. You have 2 options, as far as I can see - give up, or devote lots of time to either porting GPA, or compiling a Cygwin gtk+ (and all its many dependencies.) Okay, I'll try another way. Thanks alot for your help Markus -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Re: server refused our key
Subject: Re: Re: server refused our key From: Manfred Köhler [EMAIL PROTECTED] Date: Wed, 20 Nov 2002 10:24:33 +0100 To: Max Bowsher [EMAIL PROTECTED] I have setup sshd using privilege separation. ssh login works fine but sshd doesn't accept my public key, which works fine using ssh on other UNIX machines. To diagnose properly, we need to see the *server* debug log. O.K. file is attached I have also added debugging output generated by ssh command. Hopefully this will help to fix the problem. Ah. I see the problem. Your home directory is on a Windows share. Since Windows doesn't know your password, it can't access the server, so ssh can't get to your authorized_keys file. Hard to say how to solve this. I don't think you will be able to store your authorized_keys on a Windows share. Max. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Subject: Re: mno_cygwin gcc 3.2 From: Chris Twiner [EMAIL PROTECTED] Date: Wed, 20 Nov 2002 08:26:54 + To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Hi, this is vastly different then the machine's I have. I have attached an untouched spec file fresh from setup.exe. Are there any differences with yours? I'm away from my machine at the moment, but since you are having problems with gcc, may I suggest you upgrade to the latest version? Cygwin's gcc-3.2 has been bugfixed twice since its initial -1 release, current being gcc-3.2-3. Max. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Subject: Re: mno_cygwin gcc 3.2 From: Denis Dupeyron [EMAIL PROTECTED] Date: Wed, 20 Nov 2002 10:44:13 +0100 To: [EMAIL PROTECTED] You've left out the actual error. I think that -lm is an alias of -lcygwin. Get rid of it. Neither Cygwin or MinGW has a seperate math library. The error was in my previous post. But thanks, the -lm was the problem. I left it in since gnucap being an electrical simulator it didn't seem stupid to link with the math library. Is there any reason for -lm to be an alias of -lcygwin ? I looks rather confusing. It makes sense as far as Cygwin goes - there is no seperate math library, so just link to the library the math functions are in. It is a problem for -mno-cygwin mode, though. Ideally, -mno-cygwin mode would know not to use /usr/lib, but I believe gcc makes it difficult to override this. Max. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Hi, this is vastly different then the machine's I have. I have attached an untouched spec file fresh from setup.exe. Are there any differences with yours? I'm away from my machine at the moment, but since you are having problems with gcc, may I suggest you upgrade to the latest version? Cygwin's gcc-3.2 has been bugfixed twice since its initial -1 release, current being gcc-3.2-3. Max. I tried to do that on my home laptop but it didn't seem to make a difference. -dumpversion still showed 3.2 not 3.2-3 and the release number didn't seem to change. I'll try on another machine. Chris _ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Subject: Re: mno_cygwin gcc 3.2 From: Chris Twiner [EMAIL PROTECTED] Date: Wed, 20 Nov 2002 10:24:38 + To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Hi, this is vastly different then the machine's I have. I have attached an untouched spec file fresh from setup.exe. Are there any differences with yours? I'm away from my machine at the moment, but since you are having problems with gcc, may I suggest you upgrade to the latest version? Cygwin's gcc-3.2 has been bugfixed twice since its initial -1 release, current being gcc-3.2-3. Max. I tried to do that on my home laptop but it didn't seem to make a difference. -dumpversion still showed 3.2 not 3.2-3 and the release number didn't seem to change. I'll try on another machine. Chris $ cygcheck -c gcc will show you the version installed. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SSHD: error initializing windows sockets if I use .ssh/authorized_keys
Hi Cygwin-List! Just discoverd that the same problem exists with .ssh/authorized_keys2. Can someone reproduce the problem? Cheers, Tino - Tino Lange wrote: Hi again! I have some new problems with sshd on Windows 2000. (all latest cygwin packages from today) I try to login on my Win 2000 machine from my Linux machine. Works fine so far. (Thanks for your recent help!) But: if I use RSA authentication with authorized_keys to get on the cygwin machine it still works but I don't have network/sockets - the windows socket interface cannot be initialized?! If I remove authorized_keys and login via password it works again. Strange, or? Here's a copy and paste: tino@tinux:~ ssh tinont Tino@tinont's password: Last login: Sat Nov 16 18:12:09 2002 from tinux.tino-lange.de Fanfare!!! You are successfully logged in to this server!!! Tino@TINONT ~ $ ping tinux Ping tinux [192.168.1.1] mit 32 Bytes Daten: Antwort von 192.168.1.1: Bytes=32 Zeit10ms TTL=64 Antwort von 192.168.1.1: Bytes=32 Zeit10ms TTL=64 Ping-Statistik fr 192.168.1.1: Pakete: Gesendet = 4, Empfangen = 4, Verloren = 0 (0% Verlust), Ca. Zeitangaben in Millisek.: Minimum = 0ms, Maximum = 0ms, Mittelwert = 0ms Tino@TINONT ~ $ cat .ssh/identity.pub .ssh/authorized_keys Tino@TINONT ~ $ chmod 644 .ssh/authorized_keys Tino@TINONT ~ $ logout Connection to tinont closed. tino@tinux:~ ssh tinont Last login: Sat Nov 16 18:12:51 2002 from tinux.tino-lange.de Fanfare!!! You are successfully logged in to this server!!! Tino@TINONT ~ $ ping tinux Fehler bei der Initialisierung der Windows-Socketsschnittstelle. Fehlercode 0. Tino@TINONT ~ $ rm .ssh/authorized_keys Tino@TINONT ~ $ logout Connection to tinont closed. tino@tinux:~ ssh tinont Last login: Sat Nov 16 18:12:51 2002 from tinux.tino-lange.de Fanfare!!! You are successfully logged in to this server!!! Tino@TINONT ~ $ ping tinux Ping tinux [192.168.1.1] mit 32 Bytes Daten: Antwort von 192.168.1.1: Bytes=32 Zeit10ms TTL=64 Antwort von 192.168.1.1: Bytes=32 Zeit10ms TTL=64 Ping-Statistik fr 192.168.1.1: Pakete: Gesendet = 4, Empfangen = 4, Verloren = 0 (0% Verlust), Ca. Zeitangaben in Millisek.: Minimum = 0ms, Maximum = 0ms, Mittelwert = 0ms Tino@TINONT ~ $ Do you have an idea? Thanks! Tino -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SSHD: error initializing windows sockets if I use .ssh/authorized_keys
On Wed, Nov 20, 2002 at 11:35:10AM +0100, Tino Lange wrote: Hi Cygwin-List! Just discoverd that the same problem exists with .ssh/authorized_keys2. Can someone reproduce the problem? No. Btw., did you read http://cygwin.com/bugs.html? And *additionally* what is the output of getfacl $WINDIR/system32/wsock32.dll and getfacl $WINDIR/system32/winsock.dll Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: bug report: poll() with listen sockets always gives POLLERR
On Wed, Nov 20, 2002 at 09:07:30AM +, Steven O'Brien wrote: Corinna Vinschen wrote: On Tue, Nov 19, 2002 at 01:34:30PM +, Steven O'Brien wrote: Hi The current implementation of poll() does not behave correctly with listen sockets. It always gives a POLLERR revent when a connection request is received. I believe the error is in poll.cc lines 96-108: [...] Thanks for the report. I've checked in a patch. Please try the next developers snapshot. Corinna Thanks Corinna, but its still not quite right. poll() needs to set POLLIN Oops, sure. Thanks, I've applied the patch. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
$ cygcheck -c gcc will show you the version installed. got the right version on this computer and the spec file is still broken, it just adds dllcrt2.o not the path as yours does. Chris _ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: bash builtin pwd returns Windows style names
On Wed, Nov 20, 2002 at 10:54:16AM +0100, Richter Norbert wrote: Hello, I observed that in previous releases of Cygwin (I think it was 1.3.12) the bash builtin pwd returned directory names in the form: /cygdrive/c/classes but if I try this now with the release 1.3.14 (and also with 1.3.15 and bash 2.05b) I get c:/classes bash-2.05b$ uname -a CYGWIN_NT-5.1 CORINNA 1.3.15(0.63/3/2) 2002-11-07 13:57 i686 unknown bash-2.05b$ cd d: bash-2.05b$ pwd /cygdrive/d Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SSHD: error initializing windows sockets if I use .ssh/authorized_keys
Corinna Vinschen wrote: Hi Corinna! No. Btw., did you read http://cygwin.com/bugs.html? You think my mail didn't contain enough information or wasn't formed well? a) The subject line describes the problem very good b) In the body there's a huge shell section captured, exactly showing the problem c) I told you the relevant packages (latest version from the cygwin release) d) the only multiple exclamation marks were from the sshd login itself in the quoted section... e) I didn't assume that the bug is trivial f) I didnt tell you personal stories - the mail was focused on the problem itself The only point that is left is the cygcheck.out - did you mean that? OK - why didn't I include a cygcheck.out? a) maybe the information was already enough for you to locate the problem b) [EMAIL PROTECTED] is a huge list - I don't want to spam lots of people with this information - if someone needs it I can send it to him personally I don't like to recieve lot's of cygcheck.out a day as a public reader of this list myself. c) cygcheck.out can contain *lots* of personal information. I don't want Microsoft to spy on my computer why should I broadcast such information by myself actively on a mailinglist that's very huge, get's archived and so on? I have no Problem sending this file to you, Corinna, or some of your collegues - if you need such detailed information because all the other information was not enough. Just ask for it if you feel that you can maybe help and that you are the appropiate maintainer for my problem. You just got it with separate mail. And *additionally* what is the output of getfacl $WINDIR/system32/wsock32.dll and getfacl $WINDIR/system32/winsock.dll Of course - it's on its way. Thanks and best regards Tino -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Remote mounts not seen in remote bash shell
are you using pubkey or password authentication ? if your using pubkey then try password. -Original Message- From: Jack Twilley [mailto:[EMAIL PROTECTED]] Sent: 19 November 2002 20:05 To: [EMAIL PROTECTED] Subject: Remote mounts not seen in remote bash shell I have a Win2kAS box with the most recent Cygwin package installed. I am logged into the box with my own userid which has Administrator privileges. When I am running a local bash shell, that is on the console of the box, I can see remote drives with mount: $ mount C:\cygwin\bin on /usr/bin type system (binmode) C:\cygwin\lib on /usr/lib type system (binmode) C:\cygwin on / type system (binmode) c: on /cygdrive/c type user (binmode,noumount) h: on /cygdrive/h type user (binmode,noumount) j: on /cygdrive/j type user (binmode,noumount) The drives are also accessible via their mount locations -- //FILESERVER1/Home/jtwilley and //chez/jtwilley, respectively. When I am running a remote bash shell, that is via ssh'ing into the box from another computer, I cannot see remote drives with mount. The H: and J: drives are not visible, and they cannot be accessed via /cygdrive/j or //chez/jtwilley from the remote bash shell. How do I fix this? The whole reason I installed Cygwin was so I could ssh into the box, run a script stored on a remote machine and move files from one machine to another. This is not working as well as I'd hoped. I searched all over via Google and the FAQ and the User Guide and found nothing helpful. Jack. -- Jack Twilley // Tier 2 Support Engineer // Brightmail Inc. jtwilley at brightmail dot com -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SSHD: error initializing windows sockets if I use .ssh/authorized_keys
On Wed, Nov 20, 2002 at 12:12:51PM +0100, Tino Lange wrote: The only point that is left is the cygcheck.out - did you mean that? Yes. OK - why didn't I include a cygcheck.out? a) maybe the information was already enough for you to locate the problem No. b) [EMAIL PROTECTED] is a huge list - I don't want to spam lots of people with this information - if someone needs it I can send it to him personally I don't like to recieve lot's of cygcheck.out a day as a public reader of this list myself. It's one (better: the) purpose of the cygwin ML to send bug reports here and to receive help if somebody actually knows a solution. c) cygcheck.out can contain *lots* of personal information. I don't want Microsoft to spy on my computer why should I broadcast such information by myself actively on a mailinglist that's very huge, get's archived and so on? Then obfuscate stuff in the cygcheck output which you take for private. I have no Problem sending this file to you, Corinna, or some of your Perhaps, but *I* have problems with getting PM about Cygwin stuff. Did you read the section Shouldn't I just send email to straight to a cygwin developer or package maintainer?, especially the Reply-To: passage? You just got it with separate mail. No, I didn't so far and I don't want to get PM further on. This is the reason I'm setting the Reply-To: to the ML and not using my real mail address. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: gcc-3.2-3: bootstrap build fails (HAVE_DECL_GETOPT not in config.h?)
Christopher Faylor wrote: On Wed, Nov 20, 2002 at 10:14:59AM +1100, Nigel Stewart Fiona Smith wrote: Michael, I have encountered the same problem on my setup. Regards, I have not encountered the same problem in my setup. HTH, cgf I have the same problem and i suspect that this is due to the definition of getopt in unistd.h. IMHO this should be only in getopt.h or unistd.h should define HAVE_DECL_GETOPT too. As a workaround i have removed the declaration of getopt in gcc's getopt.h. Thomas -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
how to configure rsh daemon
I need to run rsh from a non interactive script . I known i have to configure the .rhosts file but i d'ont known the format,where to put it and how to associate the local user whit the remote user. Can someone tell me where to find some documentation on the web ? Thanks in advance. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
FW: qt with other compiler on windows
Hi all, this thread I have found on [EMAIL PROTECTED]: I've experimented with using cygwin as my development environment... My initial goal was to compile using gcc, as we are using gcc in the MacOSX world. I ended up grabbing the qmake and moc from the KDE distribution to accomplish this. I have not proceeded past compilation, because I have not had the time to create the glue libraries between Qt's .lib files and the .a format of the GNU tools. Also, some Windows concepts are not available on cygwin, such as COM/ActiveX. Can anyone tell me if this is still true ? So, the use of cygwin/gcc as a complete environment may never be possible for your application. Ralf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Is GCC buggy ?
CONTENTS:
*
SYMPTOMS
SOURCE of the test program
COMPILATION + EXECUTION
STRACE
Execution de cygcheck -s -v -r
-
SYMPTOMS:
when I compile a source defining a function and calling it,
the program coredumps.
I already compiled on this machine using 2.95.
This happens since the update gcc 3.2
I fully reinstall cygwin (today: 11/20/2002 at 14h:00( PARIS/FRANCE)
packages from programming.ccp14.uk), the same always happens.
What I compiled, rox file manager for instance, is always functionnal.
-
SOURCE of the test program
jrc@NADIAV2 ~/tmp $cat f.c
#include stdio.h
int p(char *s);
int main(int argc, char *argv[], char *envp[]) {
int r= 0 ;
printf(DEBUT\n) ;
r=p(azerty) ;
printf(FIN\n) ;
exit(0) ;
}
int p(char *s){
fprintf(stdout,dans p:%d %s\n,s,s) ;
return(2);
}
-
COMPILATION + EXECUTION
+SIMPLE
===
jrc@NADIAV2 ~/tmp $gcc f.c -o f f
DEBUT
Segmentation fault (core dumped)
+DETAIL
===
-- note message on nonexistant include directory, after full cygwin install.
jrc@NADIAV2 ~/tmp $gcc -v f.c -o f f
Reading specs from /bin/../lib/gcc-lib/i686-pc-cygwin/3.2/specs
Configured with: /netrel/src/gcc-3.2-1/configure --enable-
languages=c,c++,f77,java --enable-libgcj --enable-threads=posix --with-system-
zlib --enable-nls --without-included-gettext --enable-interpreter --disable-
sjlj-exceptions --disable-version-specific-runtime-libs --enable-shared --
build=i686-pc-linux --host=i686-pc-cygwin --target=i686-pc-cygwin --enable-
haifa --prefix=/usr --exec-prefix=/usr --sysconfdir=/etc --libdir=/usr/lib --
includedir=/nonexistent/include --libexecdir=/usr/sbin
Thread model: posix
gcc version 3.2 20020818 (prerelease)
/bin/../lib/gcc-lib/i686-pc-cygwin/3.2/cc1.exe -lang-c -v -
iprefix /bin/../lib/gcc-lib/i686-pc-cygwin/3.2/ -D__GNUC__=3 -
D__GNUC_MINOR__=2 -D__GNUC_PATCHLEVEL__=0 -D__GXX_ABI_VERSION=102 -D_X86_=1 -
D_X86_=1 -Asystem=winnt -D__NO_INLINE__ -D__STDC_HOSTED__=1 -Acpu=i386 -
Amachine=i386 -Di386 -D__i386 -D__i386__ -D__tune_i686__ -D__tune_pentiumpro__ -
D__tune_pentium2__ -D__tune_pentium3__ -D__stdcall=__attribute__
((__stdcall__)) -D__fastcall=__attribute__((__fastcall__)) -
D__cdecl=__attribute__((__cdecl__)) -D_stdcall=__attribute__((__stdcall__)) -
D_fastcall=__attribute__((__fastcall__)) -D_cdecl=__attribute__((__cdecl__)) -
D__declspec(x)=__attribute__((x)) -D__i386__ -D__i386 -D__CYGWIN32__ -
D__CYGWIN__ -Dunix -D__unix__ -D__unix -isystem /usr/lib/gcc-lib/i686-pc-
cygwin/3.2/../../../../include/w32api -isystem /usr/lib/gcc-lib/i686-pc-
cygwin/3.2/../../../../i686-pc-cygwin/lib/../../include/w32api f.c -quiet -
dumpbase f.c -version -o /cygdrive/c/DOCUME~1/jrc/LOCALS~1/Temp/ccqSNVlP.s
GNU CPP version 3.2 20020818 (prerelease) (cpplib) (80386, BSD syntax)
GNU C version 3.2 20020818 (prerelease) (i686-pc-cygwin)
compiled by GNU C version 3.2 20020815 (prerelease).
ignoring nonexistent directory /i686-pc-cygwin/include
ignoring nonexistent directory /usr/i686-pc-cygwin/include
ignoring duplicate directory /usr/i686-pc-cygwin/lib/../../include/w32api
ignoring duplicate directory /usr/lib/gcc-lib/i686-pc-cygwin/3.2/include
#include ... search starts here:
#include ... search starts here:
/usr/include/w32api
/lib/gcc-lib/i686-pc-cygwin/3.2/include
/usr/local/include
/usr/include
End of search list.
/usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../../i686-pc-cygwin/bin/as.exe --
traditional-format -
o /cygdrive/c/DOCUME~1/jrc/LOCALS~1/Temp/ccQ9MQmJ.o /cygdrive/c/DOCUME~1/jrc/LOC
ALS~1/Temp/ccqSNVlP.s
/bin/../lib/gcc-lib/i686-pc-cygwin/3.2/collect2.exe -Bdynamic --dll-search-
prefix=cyg -o f.exe /bin/../lib/gcc-lib/i686-pc-
cygwin/3.2/../../../crt0.o /bin/../lib/gcc-lib/i686-pc-cygwin/3.2/crtbegin.o -
L/bin/../lib/gcc-lib/i686-pc-cygwin/3.2 -L/bin/../lib/gcc-lib -L/usr/lib/gcc-
lib/i686-pc-cygwin/3.2 -L/bin/../lib/gcc-lib/i686-pc-cygwin/3.2/../../.. -
L/usr/lib/gcc-lib/i686-pc-
cygwin/3.2/../../.. /cygdrive/c/DOCUME~1/jrc/LOCALS~1/Temp/ccQ9MQmJ.o -lgcc -
lcygwin -luser32 -lkernel32 -ladvapi32 -lshell32 -lgcc /bin/../lib/gcc-lib/i686-
pc-cygwin/3.2/crtend.o
DEBUT
Segmentation fault (core dumped)
-
STRACE
jrc@NADIAV2 ~/tmp $strace -o strace.out f
14322 [main] f 3300 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION
17994 [main] f 3300 open_stackdumpfile: Dumping stack trace to f.exe.stackdump
STRACE.OUT
**
Program name: c:\home\jrc\cygwin\tmp\f.exe (3300)
App version: 1003.15, api: 0.63
DLL version: 1003.15, api: 0.63
DLL build:2002-11-06 22:41
OS version: Windows NT-5.1
Heap size:402653184
Date/Time:2002-11-20 15:33:23
RE: sshd: server refused our key
Please keep all replies on the cygwin mailing list. This helps you because more people are able to diagnose your problem. It helps others because they are able to search the mailing list archives for useful information. 1. In your reply below, you list the following for your home directory: drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . Try: chmod 750 ~ chgrp SYSTEM ~ 2. And for your ~/.ssh directory: drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh Try: chmod 700 ~/.ssh Please change both ~ and ~/.ssh. It is not enough to only change one. If these changes do not fix your problem, then please include the output of the 'mount' command in your reply. -Original Message- From: Manfred Köhler [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 5:58 AM To: Harig, Mark A. Subject: Re: RE: sshd: server refused our key Harig, Mark A. [EMAIL PROTECTED] schrieb am 19.11.02 17:11:23: As requested at http://cygwin.com/bugs.html: o Please describe how to reproduce the problem, including a test case, if possible. In your case, please list the commands that you ran to set up sshd on your Cygwin machine. ssh-host-config -y ...set cygwin to defaults chmod 755 /var/empty mkgroup -l -d domain/etc/group mkpasswd -l -d domain/etc/passwd o Please include at least the version number of the Cygwin release you are using along with the operating system name and its version number, for example, cygwin v1.3.13 under NT 4.0. Please refer attached file. o Most of the information about your Cygwin environment is listed by running 'cygcheck -s -v -r cygcheck.txt'. Please include cygcheck.txt *AS AN ATTACHMENT* to your report. It is important that you include it as an attachment so that searches of the mailing-list archives give fewer false matches. Please refer attached file Some things to check (that is, email back to this mailing list): The permissions and ownership of: - your home directory drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . - your home/.ssh directory drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh - your home/.ssh files seen on UNIX: -rw-r--r--1 mk group 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk group 546 Nov 20 08:48 authorized_keys2 -rw---1 mk group 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk group 218 Nov 19 13:44 id_rsa.pub -rw---1 mk group 523 Nov 19 13:44 identity -rw-r--r--1 mk group 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk group1442 Nov 20 11:50 known_hosts -rw---1 mk group 512 Nov 20 11:50 random_seed seen inside ssh session: -rw-r--r--1 mk Domain U 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk Domain U 546 Nov 20 08:48 authorized_keys2 -rw-r--r--1 mk Domain U 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk Domain U 218 Nov 19 13:44 id_rsa.pub -rw-r--r--1 mk Domain U 523 Nov 19 13:44 identity -rw-r--r--1 mk Domain U 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk Domain U 1442 Nov 20 11:50 known_hosts -rw-r--r--1 mk Domain U 512 Nov 20 11:52 random_seed I do not know why private files become readable for everyone. There is also a samba server running to mount files on Windows. Hopefully these informations will help you to solve the problem. Cygwin Win95/NT Configuration Diagnostics Current System Time: Tue Nov 05 17:34:00 2002 Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 3 Path: C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin c:\WINNT\system32 c:\WINNT c:\WINNT\System32\Wbem c:\WINNT\system32\nls c:\WINNT\system32\nls\English c:\Programme\Gemeinsame Dateien\Adaptec Shared\System c:\Programme\Rational\ClearCase\bin c:\Programme\TclTk\bin Z . Y . X . C:\cygwin\usr\X11R6\bin SysDir: C:\WINNT\System32 WinDir: C:\WINNT HOME = `C:\cygwin\home\weberr' MAKE_MODE = `unix' PWD = `/home/weberr' USER = `weberr' ALLUSERSPROFILE = `C:\Dokumente und Einstellungen\All Users' APPDATA = `C:\Dokumente und Einstellungen\weberr\Anwendungsdaten' CLEARCASE_PRIMARY_GROUP = `pars' COMMONPROGRAMFILES = `C:\Programme\Gemeinsame Dateien' COMPUTERNAME = `XTERM1192' COMSPEC = `C:\WINNT\system32\cmd.exe' HOMEDRIVE = `C:' HOMEPATH = `\' LOGONSERVER = `\\XTERM1192' MANPATH = `:/usr/ssl/man' NUMBER_OF_PROCESSORS = `1' NWLANGUAGE = `DEUTSCH' NWUSERNAME = `weberr' OLDPWD = `/usr/bin' OS2LIBPATH = `C:\WINNT\system32\os2\dll;' OS = `Windows_NT' PATHEXT = `%PATHEXT%;.tcl' PROCESSOR_ARCHITECTURE = `x86' PROCESSOR_IDENTIFIER = `x86
Re: Is GCC buggy ?
I had no problems running the code on NT4 SP6, although had the following Package's version were different. _update-info-dir 0078-1 bash 2.05b-5 binutils 20020706-2 cygwin 1.3.14-1 openssh 3.4.pl-5 sed 3.02-1 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: sshd: server refused our key
On Wed, Nov 20, 2002 at 10:15:24AM -0500, Harig, Mark A. wrote: Please keep all replies on the cygwin mailing list. This helps you because more people are able to diagnose your problem. It helps others because they are able to search the mailing list archives for useful information. 1. In your reply below, you list the following for your home directory: drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . Try: chmod 750 ~ chgrp SYSTEM ~ 2. And for your ~/.ssh directory: drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh Try: chmod 700 ~/.ssh Please change both ~ and ~/.ssh. It is not enough to only change one. I don't like the idea to change these permissions. This is actually unlikely to help. Especially what's the reasoning to change group membership to SYSTEM? The below permissions The permissions and ownership of: - your home directory drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . - your home/.ssh directory drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh - your home/.ssh files seen on UNIX: -rw-r--r--1 mk group 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk group 546 Nov 20 08:48 authorized_keys2 -rw---1 mk group 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk group 218 Nov 19 13:44 id_rsa.pub -rw---1 mk group 523 Nov 19 13:44 identity -rw-r--r--1 mk group 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk group1442 Nov 20 11:50 known_hosts -rw---1 mk group 512 Nov 20 11:50 random_seed are perfectly fine. What irritates me is the seen on UNIX and seen inside ssh session. What does that mean? From the cygcheck output I would think the home dir is on the local NTFS drive C:. So how can you see anything from UNIX? You know that the permission translation between UNIX and NT via Samba doesn't work flawlessly, don't you? Could you please enlighten us what the above wording is trying to say? And especially interesting are the permissions on these files seen in a *local* NT session on that very machine you're trying to connect via ssh. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Is GCC buggy ? : solved !
replacing gcc 3.2-1 by gcc 3.2-3 (found on sunsite chlab) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
ok finally cracked it:
g++ -mno-cygwin -enable-threads=no -nostdlib -shared -v -e _DllMain@12 -o
xwinclip_hook.dll xwinclip_hook.o -lkernel32 -luser32
not that the :-enable-threads=no seems to matter a damn.
I don't use any c++ but it still asks for and unresolved symbol:
__gxx_personality_v0
so I added a :
extern C {long __gxx_personality_v0;}
And that seems to keep everything happy.
Not pretty but it works.
In answer to the mingw stuff I added the paths to $LIBRARY_PATH and then it
caused pthread issues which I couldn't seem to resolve hence the -nostdlib.
Thanks,
Chris
_
Tired of spam? Get advanced junk mail protection with MSN 8.
http://join.msn.com/?page=features/junkmail
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Is GCC buggy ?
Jean-Raymond, For practical purposes, no, GCC is not buggy. Which is not to say it is bug-free. No large software program is literally bug-free. Nonetheless, I compiled your program and it did what one would expect. No fireworks: DEBUT dans p:4198551 azerty FIN % gcc --version gcc (GCC) 3.2 20020927 (prerelease) % cygcheck -c gcc Cygwin Package Information Package Version gcc 3.2-3 Randall Schulz Mountain View, CA USA P.S. In the future, please place your cygcheck output in an attachment. I helps control clutter in the mailing list archives and reduces false hits when people perform package-, library- or program-specific searches of the archives. Likewise for cygcheck output. In this case, compression is usually called for (use bzip2) and even without it, putting the content in an attachment protects against line wrapping. At 06:55 2002-11-20, CHAUVIERE Jean-Raymond wrote: When I compile a source defining a function and calling it, the program coredumps. I already compiled on this machine using 2.95. This happens since the update gcc 3.2 I fully reinstall cygwin (today: 11/20/2002 at 14h:00( PARIS/FRANCE) packages from programming.ccp14.uk), the same always happens. What I compiled, rox file manager for instance, is always functional. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: sshd: server refused our key
Mark, I think, we can't solve the problem because my home directory is a windows share. Every windows share becomes permission 755. = ssh daemon rejects files in .ssh folder which should be strictly private. = I'm prompted to enter my password. Please refer: http://sources.redhat.com/ml/cygwin/2001-10/msg01011.html Harig, Mark A. wrote: . -Original Message- From: Manfred Köhler [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 5:58 AM To: Harig, Mark A. Subject: Re: RE: sshd: server refused our key Harig, Mark A. [EMAIL PROTECTED] schrieb am 19.11.02 17:11:23: As requested at http://cygwin.com/bugs.html: o Please describe how to reproduce the problem, including a test case, if possible. In your case, please list the commands that you ran to set up sshd on your Cygwin machine. ssh-host-config -y ...set cygwin to defaults chmod 755 /var/empty mkgroup -l -d domain/etc/group mkpasswd -l -d domain/etc/passwd o Please include at least the version number of the Cygwin release you are using along with the operating system name and its version number, for example, cygwin v1.3.13 under NT 4.0. Please refer attached file. o Most of the information about your Cygwin environment is listed by running 'cygcheck -s -v -r cygcheck.txt'. Please include cygcheck.txt *AS AN ATTACHMENT* to your report. It is important that you include it as an attachment so that searches of the mailing-list archives give fewer false matches. Please refer attached file Some things to check (that is, email back to this mailing list): The permissions and ownership of: - your home directory drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . - your home/.ssh directory drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh - your home/.ssh files seen on UNIX: -rw-r--r--1 mk group 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk group 546 Nov 20 08:48 authorized_keys2 -rw---1 mk group 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk group 218 Nov 19 13:44 id_rsa.pub -rw---1 mk group 523 Nov 19 13:44 identity -rw-r--r--1 mk group 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk group1442 Nov 20 11:50 known_hosts -rw---1 mk group 512 Nov 20 11:50 random_seed seen inside ssh session: -rw-r--r--1 mk Domain U 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk Domain U 546 Nov 20 08:48 authorized_keys2 -rw-r--r--1 mk Domain U 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk Domain U 218 Nov 19 13:44 id_rsa.pub -rw-r--r--1 mk Domain U 523 Nov 19 13:44 identity -rw-r--r--1 mk Domain U 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk Domain U 1442 Nov 20 11:50 known_hosts -rw-r--r--1 mk Domain U 512 Nov 20 11:52 random_seed I do not know why private files become readable for everyone. There is also a samba server running to mount files on Windows. Hopefully these informations will help you to solve the problem. Cygwin Win95/NT Configuration Diagnostics Current System Time: Tue Nov 05 17:34:00 2002 Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 3 Path: C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin c:\WINNT\system32 c:\WINNT c:\WINNT\System32\Wbem c:\WINNT\system32\nls c:\WINNT\system32\nls\English c:\Programme\Gemeinsame Dateien\Adaptec Shared\System c:\Programme\Rational\ClearCase\bin c:\Programme\TclTk\bin Z . Y . X . C:\cygwin\usr\X11R6\bin SysDir: C:\WINNT\System32 WinDir: C:\WINNT HOME = `C:\cygwin\home\weberr' MAKE_MODE = `unix' PWD = `/home/weberr' USER = `weberr' ALLUSERSPROFILE = `C:\Dokumente und Einstellungen\All Users' APPDATA = `C:\Dokumente und Einstellungen\weberr\Anwendungsdaten' CLEARCASE_PRIMARY_GROUP = `pars' COMMONPROGRAMFILES = `C:\Programme\Gemeinsame Dateien' COMPUTERNAME = `XTERM1192' COMSPEC = `C:\WINNT\system32\cmd.exe' HOMEDRIVE = `C:' HOMEPATH = `\' LOGONSERVER = `\\XTERM1192' MANPATH = `:/usr/ssl/man' NUMBER_OF_PROCESSORS = `1' NWLANGUAGE = `DEUTSCH' NWUSERNAME = `weberr' OLDPWD = `/usr/bin' OS2LIBPATH = `C:\WINNT\system32\os2\dll;' OS = `Windows_NT' PATHEXT = `%PATHEXT%;.tcl' PROCESSOR_ARCHITECTURE = `x86' PROCESSOR_IDENTIFIER = `x86 Family 15 Model 2 Stepping 4, GenuineIntel' PROCESSOR_LEVEL = `15' PROCESSOR_REVISION = `0204' PROGRAMFILES = `C:\Programme' PROMPT = `$P$G' PS1 = `\[\033]0;\w\007 \033[32m\]\u@\h \[\033[33m\w\033[0m\] $ ' SHLVL = `1' SYSTEMDRIVE = `C:' SYSTEMROOT = `C:\WINNT' TEMP = `c:\DOKUME~1\weberr\LOKALE~1\Temp' TERM = `cygwin' TMP = `c:\DOKUME~1\weberr\LOKALE~1\Temp' USERDOMAIN = `XTERM1192' USERNAME = `weberr' USERPROFILE = `C:\Dokumente und Einstellungen\weberr' WINDIR = `C:\WINNT'
Re: sshd: server refused our key
Mark, I think we can't solve the problem, because my home is a windows share. Every windows share is mounted with permission 755! Therefore sshd rejects the files ins ~/.ssh which should be private. So I'm prompted for password. Please refer http://sources.redhat.com/ml/cygwin/2001-10/msg01011.html Manfred Harig, Mark A. wrote: Please keep all replies on the cygwin mailing list. This helps you because more people are able to diagnose your problem. It helps others because they are able to search the mailing list archives for useful information. 1. In your reply below, you list the following for your home directory: drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . Try: chmod 750 ~ chgrp SYSTEM ~ 2. And for your ~/.ssh directory: drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh Try: chmod 700 ~/.ssh Please change both ~ and ~/.ssh. It is not enough to only change one. If these changes do not fix your problem, then please include the output of the 'mount' command in your reply. -Original Message- From: Manfred Köhler [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 5:58 AM To: Harig, Mark A. Subject: Re: RE: sshd: server refused our key Harig, Mark A. [EMAIL PROTECTED] schrieb am 19.11.02 17:11:23: As requested at http://cygwin.com/bugs.html: o Please describe how to reproduce the problem, including a test case, if possible. In your case, please list the commands that you ran to set up sshd on your Cygwin machine. ssh-host-config -y ...set cygwin to defaults chmod 755 /var/empty mkgroup -l -d domain/etc/group mkpasswd -l -d domain/etc/passwd o Please include at least the version number of the Cygwin release you are using along with the operating system name and its version number, for example, cygwin v1.3.13 under NT 4.0. Please refer attached file. o Most of the information about your Cygwin environment is listed by running 'cygcheck -s -v -r cygcheck.txt'. Please include cygcheck.txt *AS AN ATTACHMENT* to your report. It is important that you include it as an attachment so that searches of the mailing-list archives give fewer false matches. Please refer attached file Some things to check (that is, email back to this mailing list): The permissions and ownership of: - your home directory drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . - your home/.ssh directory drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh - your home/.ssh files seen on UNIX: -rw-r--r--1 mk group 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk group 546 Nov 20 08:48 authorized_keys2 -rw---1 mk group 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk group 218 Nov 19 13:44 id_rsa.pub -rw---1 mk group 523 Nov 19 13:44 identity -rw-r--r--1 mk group 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk group1442 Nov 20 11:50 known_hosts -rw---1 mk group 512 Nov 20 11:50 random_seed seen inside ssh session: -rw-r--r--1 mk Domain U 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk Domain U 546 Nov 20 08:48 authorized_keys2 -rw-r--r--1 mk Domain U 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk Domain U 218 Nov 19 13:44 id_rsa.pub -rw-r--r--1 mk Domain U 523 Nov 19 13:44 identity -rw-r--r--1 mk Domain U 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk Domain U 1442 Nov 20 11:50 known_hosts -rw-r--r--1 mk Domain U 512 Nov 20 11:52 random_seed I do not know why private files become readable for everyone. There is also a samba server running to mount files on Windows. Hopefully these informations will help you to solve the problem. Cygwin Win95/NT Configuration Diagnostics Current System Time: Tue Nov 05 17:34:00 2002 Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 3 Path: C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin c:\WINNT\system32 c:\WINNT c:\WINNT\System32\Wbem c:\WINNT\system32\nls c:\WINNT\system32\nls\English c:\Programme\Gemeinsame Dateien\Adaptec Shared\System c:\Programme\Rational\ClearCase\bin c:\Programme\TclTk\bin Z . Y . X . C:\cygwin\usr\X11R6\bin SysDir: C:\WINNT\System32 WinDir: C:\WINNT HOME = `C:\cygwin\home\weberr' MAKE_MODE = `unix' PWD = `/home/weberr' USER = `weberr' ALLUSERSPROFILE = `C:\Dokumente und Einstellungen\All Users' APPDATA = `C:\Dokumente und Einstellungen\weberr\Anwendungsdaten' CLEARCASE_PRIMARY_GROUP = `pars' COMMONPROGRAMFILES = `C:\Programme\Gemeinsame Dateien' COMPUTERNAME = `XTERM1192' COMSPEC = `C:\WINNT\system32\cmd.exe' HOMEDRIVE = `C:' HOMEPATH = `\' LOGONSERVER = `\\XTERM1192' MANPATH = `:/usr/ssl/man' NUMBER_OF_PROCESSORS
Setup's Choose a download source screen...
When I run the Cygwin setup program, the second screen titled Choose a
download source has 3 radio buttons. More often than not I select the one
labeled Install from Local Directory. Unfortunately, setup does not
remember that selection the next time I run it, always defaulting to
Install from Internet.
I suppose I could patch the file source.cc like so:
--- source.cc.orig 2002-11-20 11:20:08.0 -0500
+++ source.cc 2002-11-20 11:21:13.0 -0500
@@ -79,7 +79,7 @@
SourcePage::OnActivate ()
{
-source = IDC_SOURCE_NETINST;
+source = IDC_SOURCE_CWD;
load_dialog (GetHWND ());
// Check to see if any radio buttons are selected. If not, select a
default.
But it seems a better solution would be to have setup save whatever I last
picked in setup.ini or in the Windows Registry.
How hard would it be to implement this feature? I suppose given enough
guidance I could implement it myself and offer up a patch.
_
Add photos to your e-mail with MSN 8. Get 2 months FREE*.
http://join.msn.com/?page=features/featuredemail
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
On Wed, Nov 20, 2002 at 10:44:13AM +0100, Denis Dupeyron wrote: You've left out the actual error. I think that -lm is an alias of -lcygwin. Get rid of it. Neither Cygwin or MinGW has a seperate math library. Max. The error was in my previous post. But thanks, the -lm was the problem. I left it in since gnucap being an electrical simulator it didn't seem stupid to link with the math library. Is there any reason for -lm to be an alias of -lcygwin ? I looks rather confusing. Reason? Nah, it's all just random guesswork. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
On Wed, Nov 20, 2002 at 10:24:03AM +, Max Bowsher wrote: It makes sense as far as Cygwin goes - there is no seperate math library, so just link to the library the math functions are in. libm.a isn't a simple alias for libcygwin.a. It contains only the math functions from libcygwin.a. Ideally, -mno-cygwin mode would know not to use /usr/lib, but I believe gcc makes it difficult to override this. It's ld that makes it difficult, actually. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: bash builtin pwd returns Windows style names
On Wed, Nov 20, 2002 at 12:08:09PM +0100, Corinna Vinschen wrote: On Wed, Nov 20, 2002 at 10:54:16AM +0100, Richter Norbert wrote: I observed that in previous releases of Cygwin (I think it was 1.3.12) the bash builtin pwd returned directory names in the form: /cygdrive/c/classes but if I try this now with the release 1.3.14 (and also with 1.3.15 and bash 2.05b) I get c:/classes bash-2.05b$ uname -a CYGWIN_NT-5.1 CORINNA 1.3.15(0.63/3/2) 2002-11-07 13:57 i686 unknown bash-2.05b$ cd d: bash-2.05b$ pwd /cygdrive/d Try doing a cd c:/tmp or something similar. Bash will dutifully report that you are in the c:/tmp directory at that point. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Possible bug in non blocking read from /dev/tty
I'm converting from a program linux to Cygwin. This program uses non
blocking read from the terminal by setting the VMIN and VTIME in the
`termios' structure and doing a non CANON read, like this:
ttyarg.c_cc [VMIN] = 0 ;
ttyarg.c_cc [VTIME] = 1 ;
ttyarg.c_lflag = 0 ;
The read does BLOCK (wait for a character to be typed) and does not
return after 1/10 second as it supposed (It works OK on linux).
I added a call to `stty -a' before and after I change the terminal
attributes. here is the output:
Before:
speed 38400 baud; rows 49; columns 105; line = 0;
intr = ^C; quit = ^\; erase = ^H; kill = ^U; eof = ^D; eol = undef; eol2 = undef;
start = ^Q;
stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W; lnext = ^V; flush = ^O; min = 1; time =
2;
-parenb -parodd cs8 -hupcl -cstopb cread -clocal -crtscts
-ignbrk brkint ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff -iuclc
-ixany imaxbel
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
isig icanon iexten echo echoe echok -echonl -noflsh -tostop echoctl echoke
After:
speed 38400 baud; rows 49; columns 105; line = 0;
intr = ^C; quit = ^\; erase = ^H; kill = ^U; eof = ^D; eol = undef; eol2 = undef;
start = ^Q;
stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W; lnext = ^V; flush = ^O; min = 0; time =
1;
-parenb -parodd cs8 -hupcl -cstopb cread -clocal -crtscts
ignbrk -brkint ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl -ixon -ixoff -iuclc
-ixany imaxbel
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
-isig -icanon -iexten -echo -echoe -echok -echonl -noflsh -tostop -echoctl -echoke
As you can see the VMIN VTIME values were set, and the icanon was
changed to -icanon.
My system is P4, 1.7GH, running M$Windows XP. Output of uname -a :
CYGWIN_NT-5.1 P4-1700-XP 1.3.15(0.63/3/2) 2002-11-07 13:57 i686 unknown
Setting the CYGWIN environment variable to tty/notty have no effect.
A full test program with debug messages is below.
Ehud.
// test_read_term.c program
#include stdio.h
#include stdlib.h
#include unistd.h
#include termios.h
int main ( int argc, char *argv[] )
{
FILE *devttyr = NULL ; /* stream for terminal I/O */
struct termios ttyarg, intrm ; /* arguments for termios */
char buf [ 50 ] , *lni ;
int ftty, n , rdn , length ; /* devtty file number, temp n, rdn */
length = atoi ( argv [ 1 ] ) ; /* real length to read */
lni = buf ; /* pointer to input */
if ( ( devttyr = fopen ( /dev/tty , r+ ) ) == NULL )/* open (terminal ?)
for read */
{
fprintf ( stderr , Could not open /dev/tty ) ;
exit ( 1 ) ;
}
ftty = fileno ( devttyr ) ; /* file number of devtty */
tcgetattr( ftty , intrm ) ;/* get stdin arguments (SAVE) */
ttyarg = intrm ;/* saved original configuration */
ttyarg.c_cc [VMIN] = 0 ;/* min of 0 chars */
ttyarg.c_cc [VTIME] = 1 ; /* wait max vtime_val/10 sec (def 15/10) */
ttyarg.c_iflag = IGNBRK | IGNPAR | ICRNL | IMAXBEL ;/* set these modes only on
input */
ttyarg.c_lflag = 0 ;/* clear lflag of all processing */
system ( stty -a ) ;
n = tcsetattr( ftty , TCSANOW , ttyarg ) ; /* set changed tty arguments */
fprintf ( stderr , result of tcsetattr is %d\n , n ) ;
system ( stty -a ) ;
lni [ length + 1 ] = 0 ;/* last pos must be Z byte */
while ( length 0 )/* read until buffer full */
{
fprintf ( stderr , before read --\n ) ;
rdn = length ; /* min (length, MAX_BEL) */
n = read ( ftty , lni , rdn ) ; /* read devtty, max n chars, up to 1/2 sec
wait */
fprintf ( stderr , AFTER read n=%d -\n , n ) ;
switch ( n )/* read devtty, max n chars, up to 1/2 sec
wait */
{
case -1:/* error */
fprintf ( stderr , Error occurred \n ) ;
length = 0 ;/* exit while loop */
case 0:/* no input (do nothing) */
break; /* all ok */
default:/* input read, check for terminating NL */
lni += n ; /* next address */
*lni = 0 ; /* must (we may stop reading) */
length -= n ; /* length left to read */
if ( *(lni-1) == '\n' ) /* if last char is NL */
length = 0 ;/* dont read any more */
}
}
tcsetattr( ftty , TCSANOW , intrm ) ; /* reset devtty to saved value */
return ( 0 ) ;
}
/*=== end of test_read_term.c =
--
Ehud Karni Tel: +972-3-7966-561 /\
Mivtach - Simon Fax: +972-3-7966-667 \ / ASCII Ribbon
Re: SSHD: error initializing windows sockets if I use .ssh/authorized_keys
On Wed, Nov 20, 2002 at 12:51:29PM +0100, Corinna Vinschen wrote: No, I didn't so far and I don't want to get PM further on. This is the reason I'm setting the Reply-To: to the ML and not using my real mail address. This is all very confusing. How about if I drop you a personal note to discuss it? cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SSHD: error initializing windows sockets if I use .ssh/authorized_keys
On Wed, Nov 20, 2002 at 11:49:02AM -0500, Chris Faylor wrote: On Wed, Nov 20, 2002 at 12:51:29PM +0100, Corinna Vinschen wrote: No, I didn't so far and I don't want to get PM further on. This is the reason I'm setting the Reply-To: to the ML and not using my real mail address. This is all very confusing. How about if I drop you a personal note to discuss it? cgf *kick* Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: bash builtin pwd returns Windows style names
On Wed, Nov 20, 2002 at 11:42:53AM -0500, Christopher Faylor wrote: Try doing a cd c:/tmp or something similar. Bash will dutifully report that you are in the c:/tmp directory at that point. Ok. That's a bash problem. It doesn't recognize c: as a drive but as the beginning of a relative path. This is weird since bash has special handling for drive letters under Cygwin in lib/sh/pathcanon.c, function sh_canonpath(). For some reason it just doesn't work... Chet? Do you have an idea how to solve that? Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: bash builtin pwd returns Windows style names
Try doing a cd c:/tmp or something similar. Bash will dutifully report that you are in the c:/tmp directory at that point. Ok. That's a bash problem. It doesn't recognize c: as a drive but as the beginning of a relative path. This is weird since bash has special handling for drive letters under Cygwin in lib/sh/pathcanon.c, function sh_canonpath(). For some reason it just doesn't work... Chet? Do you have an idea how to solve that? Start looking at it with a debugger. The bash function absolute_pathname() should report that c:/tmp is absolute. If it succeeds, set a breakpoint in change_to_directory() and see what it's doing. I'm not on the cygwin mailing list, and can't provide any debugging assistance right now, so contact me directly if you have more questions. Chet -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ( ``Discere est Dolere'' -- chet ) Chet Ramey, ITS, CWRU[EMAIL PROTECTED]http://cnswww.cns.cwru.edu/~chet/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
bind
Just a quicky to ask if anyone is working on a supported port of bind. i found a port of 8.1.2 which gave me dig which was all i realy need it for, but it would be nice to see it as a supported package. (my C programming skills are past hello world but nowhere near porting bind, otherwise i'd give it a go.) --- Vince Hoffman NT Administrator Circle.Com (London) P: 020 7959 7576 M: 0777 5822213 E: [EMAIL PROTECTED] Icq: 50905590 PGP ID: 0xC859C861 --- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Shankar Unni wrote: He has a valid question: why is -lm an alias to -lcygwin Never mind - I saw the followup a few posts later. Still, it would be nice to be able to leave -lm in the link list even if linking -mno-cygwin, for the sake of Makefile cleanness.. -- Shankar. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: sshd: server refused our key
You might try reading a recent thread of messages in the mailing list archive with the subject line: Is RSA authentication on SSH still broken? It started back on 5 November 2002. -Original Message- From: Manfred Köhler [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 11:34 AM To: Harig, Mark A. Cc: [EMAIL PROTECTED] Subject: Re: sshd: server refused our key Mark, I think we can't solve the problem, because my home is a windows share. Every windows share is mounted with permission 755! Therefore sshd rejects the files ins ~/.ssh which should be private. So I'm prompted for password. Please refer http://sources.redhat.com/ml/cygwin/2001-10/msg01011.html Manfred Harig, Mark A. wrote: Please keep all replies on the cygwin mailing list. This helps you because more people are able to diagnose your problem. It helps others because they are able to search the mailing list archives for useful information. 1. In your reply below, you list the following for your home directory: drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . Try: chmod 750 ~ chgrp SYSTEM ~ 2. And for your ~/.ssh directory: drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh Try: chmod 700 ~/.ssh Please change both ~ and ~/.ssh. It is not enough to only change one. If these changes do not fix your problem, then please include the output of the 'mount' command in your reply. -Original Message- From: Manfred Köhler [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 5:58 AM To: Harig, Mark A. Subject: Re: RE: sshd: server refused our key Harig, Mark A. [EMAIL PROTECTED] schrieb am 19.11.02 17:11:23: As requested at http://cygwin.com/bugs.html: o Please describe how to reproduce the problem, including a test case, if possible. In your case, please list the commands that you ran to set up sshd on your Cygwin machine. ssh-host-config -y ...set cygwin to defaults chmod 755 /var/empty mkgroup -l -d domain/etc/group mkpasswd -l -d domain/etc/passwd o Please include at least the version number of the Cygwin release you are using along with the operating system name and its version number, for example, cygwin v1.3.13 under NT 4.0. Please refer attached file. o Most of the information about your Cygwin environment is listed by running 'cygcheck -s -v -r cygcheck.txt'. Please include cygcheck.txt *AS AN ATTACHMENT* to your report. It is important that you include it as an attachment so that searches of the mailing-list archives give fewer false matches. Please refer attached file Some things to check (that is, email back to this mailing list): The permissions and ownership of: - your home directory drwxr-xr-x 138 mk group 24576 Nov 20 11:48 . - your home/.ssh directory drwxr-xr-x2 mk group 4096 Nov 19 13:44 .ssh - your home/.ssh files seen on UNIX: -rw-r--r--1 mk group 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk group 546 Nov 20 08:48 authorized_keys2 -rw---1 mk group 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk group 218 Nov 19 13:44 id_rsa.pub -rw---1 mk group 523 Nov 19 13:44 identity -rw-r--r--1 mk group 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk group1442 Nov 20 11:50 known_hosts -rw---1 mk group 512 Nov 20 11:50 random_seed seen inside ssh session: -rw-r--r--1 mk Domain U 545 Nov 20 08:48 authorized_keys -rw-r--r--1 mk Domain U 546 Nov 20 08:48 authorized_keys2 -rw-r--r--1 mk Domain U 887 Nov 19 13:44 id_rsa -rw-r--r--1 mk Domain U 218 Nov 19 13:44 id_rsa.pub -rw-r--r--1 mk Domain U 523 Nov 19 13:44 identity -rw-r--r--1 mk Domain U 327 Nov 19 13:44 identity.pub -rw-r--r--1 mk Domain U 1442 Nov 20 11:50 known_hosts -rw-r--r--1 mk Domain U 512 Nov 20 11:52 random_seed I do not know why private files become readable for everyone. There is also a samba server running to mount files on Windows. Hopefully these informations will help you to solve the problem. - --- Cygwin Win95/NT Configuration Diagnostics Current System Time: Tue Nov 05 17:34:00 2002 Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 3 Path:C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin c:\WINNT\system32 c:\WINNT c:\WINNT\System32\Wbem c:\WINNT\system32\nls c:\WINNT\system32\nls\English c:\Programme\Gemeinsame Dateien\Adaptec
RE: [ANNOUNCEMENT] Updated sysvinit packake available for download
Outstanding! Sergey, are you planning on moving your chkconfig and xinetd ports over soon? Updated sysvinit package available for download. What's new: postinstall script modified to create /var/log and /var/run dirs and empty utmp/wtmp files (if not exist). Removed some manual pages -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: [ANNOUNCEMENT] Updated sysvinit packake available for download
Chkconfig is ready to release, xinetd requires upcoming 1.3.16 cygwin1.dll -Original Message- From: Jim Drash [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 1:31 PM To: [EMAIL PROTECTED] Subject: RE: [ANNOUNCEMENT] Updated sysvinit packake available for download Outstanding! Sergey, are you planning on moving your chkconfig and xinetd ports over soon? Updated sysvinit package available for download. What's new: postinstall script modified to create /var/log and /var/run dirs and empty utmp/wtmp files (if not exist). Removed some manual pages -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: .rhosts on W2K w/o ntsec
Also, the directories created by Cygwin with ntsec do have inheritance turned on. In fact that inheritance determines the ACL of files created by Cygwin when ntsec is off, and also the ACL created by most Windows applications. Incidentally you can display these stupid permissions with getfacl and change them with setfacl, so you could add Administrators if needed. Hmmm it seems as if you mis-interpreted (is this a word?) my problem: The permissions set by Cygwin with ntsec are absolutely OK. I'm having problems with permissions set by *native* Windows programs when they create files in my Cygwin home directory I just did some tests with CYGWIN=ntsec and it seems as if it's better than it used to be a year ago or so. The only thing that doesn't work is typing something like cmd /c xxx.doc to start the according application automatically if the according file is not executable but I can write a little script that looks into /proc/registry and figures out how to open a file of a given type. I'll give it a shot, convert all my files to NT security and see how it goes. Thanks again. Cheers, --Christian -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
On Wed, Nov 20, 2002 at 09:24:35AM -0800, Shankar Unni wrote: Christopher Faylor wrote: On Wed, Nov 20, 2002 at 10:44:13AM +0100, Denis Dupeyron wrote: Is there any reason for -lm to be an alias of -lcygwin ? I looks rather confusing. Reason? Nah, it's all just random guesswork. Now, now. :-) He has a valid question: why is -lm an alias to -lcygwin, instead of simply being stubbed out with an empty archive? As I stated in other email, -lm is not an alias for -lcygwin. Changing things just to accommodate people who want to use -lm -mno-cygwin is not a viable option. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: CYGWIN=codepage:oem in User's Guide
On Tue, 19 Nov 2002, Joshua Daniel Franklin wrote: After researching a bit on the Internet, I think I understand what the codepage is and how it works in the CYGWIN environment variable. Please read this snippit of text meant eventually for http://cygwin.com/cygwin-ug-net/using-cygwinenv.html and give me any feedback about any errors: icodepage:[ansi|oem]/i - Windows console applications use one of two character sets for drawing characters, ansi or oem. The first is the default, called ansi since the Windows 1.0 set was the ANSI Latin1 (ISO 8859-1) standard, though the character sets have since diverged from any standard. The second character sets are the older, DOS-based character set, called oem since originally they were encoded on the firmware of IBM PCs by OEMs. If you find that some characters (especially non-US or 'graphical' ones) do not display correctly in Cygwin, you can manually set which codepage you want to use. How about the following (fixes small typos and adds the upcoming codepage:none option): icodepage:[ansi|oem|none]/i - Windows console applications can use different character sets (codepages) for drawing characters. The first setting, called ansi is the default. This character set contains various forms of latin characters used in European languages. The name originates from the ANSI Latin1 (ISO 8859-1) standard, used in Windows 1.0, though the character sets have since diverged from any standard. The second setting selects an older, DOS-based character set, containing various line drawing and special characters. It's called oem since it was originally encoded in the firmware of the IBM PCs by the original equipment manufacturers (OEMs). The third setting allows using non-cygwin tools, such as chcp, to select a particular codepage. If you find that some characters (especially non-US or 'graphical' ones) do not display correctly in Cygwin, you can use this option to select an appropriate codepage. -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! Water molecules expand as they grow warmer (C) Popular Science, Oct'02, p.51 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: .rhosts on W2K w/o ntsec
Christian, Use cygstart docFile instead of cmd /c docFile Cygstart doesn't require the execute bit on the document file as cmd /c does. I don't know how cygstart works internally, but I imagine it looks up the association and launches the appropriate application without regard for the execute bits on the document file being opened. Randall Schulz Mountain View, CA USA At 10:42 2002-11-20, Christian Mueller wrote: ... I just did some tests with CYGWIN=ntsec and it seems as if it's better than it used to be a year ago or so. The only thing that doesn't work is typing something like cmd /c xxx.doc to start the according application automatically if the according file is not executable but I can write a little script that looks into /proc/registry and figures out how to open a file of a given type. I'll give it a shot, convert all my files to NT security and see how it goes. Thanks again. Cheers, --Christian -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Errors when building setup.exe
I downloaded the source for the Cygwin setup program (via setup), ran configure, and then make. All went smoothly until this: Making all in libgetopt++ make[2]: Entering directory `/usr/src/setup-0/libgetopt++' source='src/GetOption.cc' object='src/GetOption.lo' libtool=yes \ depfile='src/.deps/GetOption.Plo' tmpdepfile='src/.deps/GetOption.TPlo' \ depmode=gcc3 /bin/bash ./cfgaux/depcomp \ /bin/bash ./libtool --mode=compile g++ -DHAVE_CONFIG_H -I. -I. -I./include -I./include-g -O2 -c -o src/GetOption.lo `test -f 'src/GetOption.cc' || echo './'`src/GetOption.cc mkdir src/.libs g++ -DHAVE_CONFIG_H -I. -I. -I./include -I./include -g -O2 -c src/GetOption.cc -MT src/GetOption.lo -MD -MP -MF src/.deps/GetOption.TPlo -DPIC -o src/.libs/GetOption.o In file included from include/getopt++/GetOption.h:19, from src/GetOption.cc:19: include/getopt++/OptionSet.h:31: `ostream' was not declared in this scope include/getopt++/OptionSet.h:31: parse error before `)' token In file included from src/GetOption.cc:20: include/getopt++/Option.h:33: ISO C++ forbids declaration of `string' with no type include/getopt++/Option.h:33: `string' declared as a `virtual' field include/getopt++/Option.h:33: parse error before `const' include/getopt++/Option.h:35: ISO C++ forbids declaration of `string' with no type include/getopt++/Option.h:35: `string' declared as a `virtual' field include/getopt++/Option.h:35: declaration of `int Option::string' include/getopt++/Option.h:33: conflicts with previous declaration `int Option::string' include/getopt++/Option.h:35: parse error before `const' include/getopt++/Option.h:35: duplicate member `Option::string' make[2]: *** [src/GetOption.lo] Error 1 make[2]: Leaving directory `/usr/src/setup-0/libgetopt++' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/usr/src/setup-0' make: *** [all] Error 2 Shouldn't this package build cleanly out of the box? Am I missing something? _ STOP MORE SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: gcc-3.2-3: bootstrap build fails (HAVE_DECL_GETOPT not in config.h?)
I did a google search to try to find specs for unistd.h. I found this link: http://www.opengroup.org/onlinepubs/007908799/xsh/unistd.h.html Not sure how authoritative it is, but it seems to say unistd.h should either directly or indirectly declare getopt(). I think getopt.h in gcc is an implementation detail of gcc. There is a getopt.h in /usr/include but I think that's an implementation detail of unistd.h. Also, noticed that the getopt.h in /usr/include is different from the getopt.h in gcc (?). Mike -Original Message- I have the same problem and i suspect that this is due to the definition of getopt in unistd.h. IMHO this should be only in getopt.h or unistd.h should define HAVE_DECL_GETOPT too. As a workaround i have removed the declaration of getopt in gcc's getopt.h. Thomas -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
ATTENZIONE...
http://ilcoyote.iscool.net da oggi puoi vedere in diretta il primo sexy-bar stile coyote ugly guarda le lap dance delle nostre bariste in diretta su internet!!! http://ilcoyote.iscool.net per non ricevere altri messaggi: [EMAIL PROTECTED] -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
compile with -fomit-frame-pointer and gcc
I have the following to report, perhaps it is a bug in gcc for Cygwin. I wrote a C subroutine that allocated 6 integers on the stack each time it was called. Under Cygwin gcc 2.95x this worked fine. However, when compiled with the current gcc 3.2 it fails if compiled using the -fomit-frame-pointer optimization (default from configure). Without this optimization turned on all is well. Declaring the array static cleared up this problem and the program can be compiled and run sucessfully using -fomit-frame-pointer. Cheers, Charles Werner -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: .rhosts on W2K w/o ntsec
On 20-11-2002 19:51, Randall R Schulz wrote: I don't know how cygstart works internally, but I imagine it looks up the association and launches the appropriate application without regard for the execute bits on the document file being opened. Luckily, cygstart doesn't need to do such things. ;-) It just passes the filename to the ShellExecute API function, which doesn't seem to care about the execute bits. - Michael -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Remote mounts not seen in remote bash shell
Vince == Vince Hoffman [EMAIL PROTECTED] writes: Vince are you using pubkey or password authentication ? if your using Vince pubkey then try password. I am able to successfully log in to the machine with RSA keys. The problem was eventually solved by running the following command from a remote bash shell: net use k: chez\\jtwilley myrealpassword /persistent:yes And yes, I had to use my real password. This enabled me to use /cygdrive/j as well as /cygdrive/k and //chez/jtwilley. Goodness. I can kinda sorta see why, but it's really ugly and annoying. I have another problem which is only peripherally related to cygwin so I'll put it in another message. Thanks! Jack. -- Jack Twilley // Tier 2 Support Engineer // Brightmail Inc. jtwilley at brightmail dot com -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: .rhosts on W2K w/o ntsec
Also, the directories created by Cygwin with ntsec do have inheritance turned on. In fact that inheritance determines the ACL of files created by Cygwin when ntsec is off, and also the ACL created by most Windows applications. Incidentally you can display these stupid permissions with getfacl and change them with setfacl, so you could add Administrators if needed. Hmmm it seems as if you mis-interpreted (is this a word?) my problem: The permissions set by Cygwin with ntsec are absolutely OK. I'm having problems with permissions set by *native* Windows programs when they create files in my Cygwin home directory I just did some tests with CYGWIN=ntsec and it seems as if it's better than it used to be a year ago or so. The only thing that doesn't work is typing something like cmd /c xxx.doc to start the according application automatically if the according file is not executable but I can write a little script that looks into /proc/registry and figures out how to open a file of a given type. I'll give it a shot, convert all my files to NT security and see how it goes. Thanks again. Cheers, --Christian OK -- I tried it, converted all my files to ntsec and played around with some programs. And, I'll go right back, remove all NT-level permissions and turn off ntsec. Files saved by Windows applications such as JBuilder (actually Java but a Windows JRE) end up as follows: -r-xr-xr-x 1 root none 59486 Nov 20 20:46 MainFrame.java* Without ntsec, the file looks as it should: -rw-r--r-- 1 chris users 59486 Nov 20 20:46 MainFrame.java Furthermore, backing up the Cygwin directory to a CD with a general purpose CD writer program such as Nero doesn't work anymore because Nero can't read some of the files. Windows is not Unix and Windows programs typically don't care about file permissions (especially things like should a file be executable), thus files created by Windows programs will always end up with incorrect permissions in Cygwin with ntsec turned on. Back to the original question: Would it be possible to either (optionally) turn off the owner check in ruserok() or to respect the UID/GID values in the extended attributes if CYGWIN=ntea nontsec plus something like ntea_uid? Cheers, --Christian -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Errors when building setup.exe
Thomas Chadwick [EMAIL PROTECTED] wrote: I downloaded the source for the Cygwin setup program (via setup), ran configure, and then make. All went smoothly until this: [errors snipped] Shouldn't this package build cleanly out of the box? It did, when it was released. Setup was only recently ported to compile with gcc3. Get the latest source from cvs. That should compile cleanly. http://sources.redhat.com/cygwin-apps/setup.html Max. PS to Robert Collins: Perhaps the setup source tarball should be removed from the mirrors? Its so old as to be impractical to use now. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
Chris Twiner [EMAIL PROTECTED] wrote:
ok finally cracked it:
g++ -mno-cygwin -enable-threads=no -nostdlib -shared -v -e
_DllMain@12 -o xwinclip_hook.dll xwinclip_hook.o -lkernel32 -luser32
not that the :-enable-threads=no seems to matter a damn.
I don't use any c++ but it still asks for and unresolved symbol:
__gxx_personality_v0
so I added a :
extern C {long __gxx_personality_v0;}
And that seems to keep everything happy.
Not pretty but it works.
For how long? This is an ugly hack.
Are you really trying to compile xwinclip for MinGW? Why?
Max.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Setup's Choose a download source screen...
Thomas Chadwick [EMAIL PROTECTED] wrote:
When I run the Cygwin setup program, the second screen titled Choose
a download source has 3 radio buttons. More often than not I select
the one labeled Install from Local Directory. Unfortunately, setup
does not remember that selection the next time I run it, always
defaulting to Install from Internet.
I suppose I could patch the file source.cc like so:
--- source.cc.orig 2002-11-20 11:20:08.0 -0500
+++ source.cc 2002-11-20 11:21:13.0 -0500
@@ -79,7 +79,7 @@
SourcePage::OnActivate ()
{
-source = IDC_SOURCE_NETINST;
+source = IDC_SOURCE_CWD;
load_dialog (GetHWND ());
// Check to see if any radio buttons are selected. If not, select a
default.
But it seems a better solution would be to have setup save whatever I
last picked in setup.ini or in the Windows Registry.
How hard would it be to implement this feature? I suppose given
enough guidance I could implement it myself and offer up a patch.
The principal difficulty is that there is no good framework in place for
persisting setup's configuration.
Its probably time to discuss how to deal with this, as it will undoubtably
become neccessary in the short term.
For the time being, it may help that current cvs version of setup has
command line options which can set the default of this page.
Move this thread to [EMAIL PROTECTED] All setup.exe discussion happens
there.
Max.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: mno_cygwin gcc 3.2
For how long? This is an ugly hack. Indeed it is. The 2.95 release is ok though, 3.2 forces this hack. Are you really trying to compile xwinclip for MinGW? Why? Not entirely, a version I have written that resolves claiming ownership of the primary selection. It uses a windows hook and as such must not use cygwin1.dll (which by default is not in the path). Chris _ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Text Problems with Cygwin 1.3.14 and TinyFugue WAS Re: Text problems with Cygwin 1.1.14 and TinyFugue
On Friday 15 November 2002 08:46 am, you wrote: Does anyone still have snapshots of the cygwin dll from betweem 1.3.13-2 and 1.3.14. Note, in the original subject and and messages I had somehow thought the current versions of Cygwin were 1.1.x, not 1.3.x. :) I am currently running 1.3.13 with no problems. It is 1.3.14 and above that show this problem. Curiouser and curiouser. With 1.3.15 and the 11/20 snapshot I can run tf from within strace. That is, I can enter text into it and even connect to worlds. That is not to say that I can *use* it, the output doesn't work quite correctly. HTH. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
1.3.15 - with long command line, command hangs + thrashes. nt4 (?bash/tcsh)
A while ago I posted a weird bug; it's something I can work around but I figured that I should write up a better description. I don't see this bug on every machine I use, which makes it especially puzzling. On NT4, I see the following: I find that once a command line exceeds a certain number of characters, and I call an external program, things get stuck. I don't observe the same thing on my w2k laptop. For example: $ LONGARG= 1 $ echo $LONGARG works fine, which seems to use a built-in echo command, but $ /usr/bin/echo $LONGARG hangs. In the process viewer, it keeps using more and more memory. Eventually the system falls down. If I shorten LONGARG, it doesn't hang. When I strace the hanging process all I get is $ strace -p 378 Attached to pid 378 (windows pid 378) 13 13 [sig] bash 378! wait_sig: looping I note the same situation in bash and tcsh. cygcheck.out.bz2 cygcheck.out.bz2 Description: Binary data -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Errors when building setup.exe
On Thu, 2002-11-21 at 07:19, Max Bowsher wrote: PS to Robert Collins: Perhaps the setup source tarball should be removed from the mirrors? Its so old as to be impractical to use now. Can't do that - it's the source code that must be offered (GPL :}). Rob Followup to cygwin-apps please. -- --- GPG key available at: http://users.bigpond.net.au/robertc/keys.txt. --- signature.asc Description: This is a digitally signed message part
rxvt hangs upon resize...
Hello, I have been experiencing very frequent rxvt hangs when resizing windows ever since I installed 1.3.15 (or close enough as to make no difference, although the issue seems to be related). Upon resizing the main window, the rxvt child windows seem to take an indordinate amount of time handling the WM_RESIZE events (I am running cygwin under Windows XP SP1, and windows resize immediatly upon dragging an edge), hogging my CPU to the extent where it is actually preferrable to reboot the machine. I cannot do anything at all with the machine when rxvt flies off the handle - cannot close the window, invoke the Task Manager with Ctrl-Alt-Del, or even ssh into it an issue a kill command. And I am sure that rxvt is causing this, since I actually had Task Manager up and running at one point and (after a good 30s of CPU torture) I saw an rxvt instance with over 90% CPU load. (Debugging this with the likes of Spy++ is also impossible, since rxvt seems to never let go of the CPU) A temporary fix for extended coding sessions is to just launch X11 (where I have no problems whatsoever running rxvt under Windowmaker), but shuffling two sets of virtual desktops is awful. Is this a known problem? I seem to recall seeing something on it on the mailing list archives before, but I cannot find it now. Rui Carmo -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: CYGWIN=codepage:oem in User's Guide
--- Igor Pechtchanski [EMAIL PROTECTED] wrote: How about the following (fixes small typos and adds the upcoming codepage:none option): This looks good though I'd like to wait until none is actually a valid option for people reading the User's Guide. Thanks for the feedback. __ Do you Yahoo!? Yahoo! Web Hosting - Let the expert host your site http://webhosting.yahoo.com -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygwin=codepage:oem in User's Guide
On Wed, Nov 20, 2002 at 01:56:20PM -0800, Joshua Daniel Franklin wrote: --- Igor Pechtchanski [EMAIL PROTECTED] wrote: How about the following (fixes small typos and adds the upcoming codepage:none option): This looks good though I'd like to wait until none is actually a valid option for people reading the User's Guide. Any chance of that happening for 1.3.16? Can we get the patch in soon? Thanks for the feedback. Ditto. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Another less/rxvt prob: Not interpreting backspace?
Hi, All, :) Thank You! to Sergei and Matthijs who graciously gave suggestions and commiseration about my problem with 'less' not determining the terminal size correctly. Through no fault of my own, the problem seems to have ameliorated itself: running 'less' in a new rxvt instance no longer exhibits the quirky behavior of trying to scroll only 24 lines in a 25-line terminal. (Note that rxvt is still opening a window that is one row greater than what I ask for. I ask for '-geometry 80x24' and get 80x25, but that's a niggle.) However, 'less' now seems to be struggling to interpret keystrokes. Specifically backspace: when editing a search pattern, instead of erasing the previous character, 'less' inserts ^?. Ctrl-backspace seems to perform the function of backspace, as does Ctrl-h. Once again, this behavior didn't crop up until my recent update from 1.3.12 or so to 1.3.15 on Friday. Here is what I believe to be the relevant info about my system. If this isn't clear or complete enough, please let me know! Installation info: cygwin: 1.3.15-2 rxvt: 2.7.2-14 bash: 2.05b-7 less: 374-1 termcap: 20020930-1 terminfo: 5.2-3 Win2K SP2 rxvt invocation: C:\WINNT\system32\cmd.exe /c start C:\cygwin\bin\rxvt.exe -tn rxvt -fg grey -bg black -fn Fixedsys -sr -sl 5000 -e /bin/bash --login -i Relevant environment variables: TERM=xterm LESS=iFMSX# 8 CYGWIN=tty nontsec Any ideas on how I can investigate this problem further? ---Jason -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: emacs 100% cpu usage bug
Please check out the project web page for links to available information and ports: http://cygwin.com/ . If you don't see what you need there, then the cygwin mailing list is the best place to make observations or get questions answered. Information on the mailing list is available at the project web page. For your convenience, I've reset the Reply-To: address to point to the cygwin mailing list. I've also Cc'ed this reply there. On Wed, Nov 20, 2002 at 07:06:56PM -0500, Zdzislaw Meglicki wrote: Dear Christopher, I have downloaded and installed the latest, I presume, version of Cygwin, with some additional updates downloaded only today from a NASA Cygwin mirror, and... I am still plagued by the emacs 100% CPU usage bug, i.e., when emacs is invoked in the X11 environment, it spins and doesn't come up. When it is invoked in a no-X11 environment, it works fine. I have perused the mailing lists, both cygwin-xfree and cygwin, and found that the latest snapshot, I guess of cygwin1.dll, fixes the problem. So my question is how I can download this latest snapshot. I haven't seen it on Cygwin mirrors at the ANL and NASA. But they seem to have various other fixes, e.g., for gcc. Best regards, -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Another less/rxvt prob: Not interpreting backspace?
Set env var SHELL to bash before starting rxvt. -Original Message- From: Jason Tiller [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 6:59 PM To: [EMAIL PROTECTED] Subject: Another less/rxvt prob: Not interpreting backspace? Hi, All, :) Thank You! to Sergei and Matthijs who graciously gave suggestions and commiseration about my problem with 'less' not determining the terminal size correctly. Through no fault of my own, the problem seems to have ameliorated itself: running 'less' in a new rxvt instance no longer exhibits the quirky behavior of trying to scroll only 24 lines in a 25-line terminal. (Note that rxvt is still opening a window that is one row greater than what I ask for. I ask for '-geometry 80x24' and get 80x25, but that's a niggle.) However, 'less' now seems to be struggling to interpret keystrokes. Specifically backspace: when editing a search pattern, instead of erasing the previous character, 'less' inserts ^?. Ctrl-backspace seems to perform the function of backspace, as does Ctrl-h. Once again, this behavior didn't crop up until my recent update from 1.3.12 or so to 1.3.15 on Friday. Here is what I believe to be the relevant info about my system. If this isn't clear or complete enough, please let me know! Installation info: cygwin: 1.3.15-2 rxvt: 2.7.2-14 bash: 2.05b-7 less: 374-1 termcap: 20020930-1 terminfo: 5.2-3 Win2K SP2 rxvt invocation: C:\WINNT\system32\cmd.exe /c start C:\cygwin\bin\rxvt.exe -tn rxvt -fg grey -bg black -fn Fixedsys -sr -sl 5000 -e /bin/bash --login -i Relevant environment variables: TERM=xterm LESS=iFMSX# 8 CYGWIN=tty nontsec Any ideas on how I can investigate this problem further? ---Jason -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
file missing when trying to use gcc
hi, i'm having problems with gcc. every time i try to compile a program i get this error: liquid@LIQUID ~ $ gcc -o x x.c gcc: installation problem, cannot exec `as': No such file or directory liquid@LIQUID ~ $ i've reinstalled it but the problem continues. i think there's a missing package or something like. please send me the solution to this problem or the name of the package who contains the `as' program. thank you. -Allan liquid. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
1.3.15 (and earlier): cdrecord fifo bug
hello, using cygwin 1.3.15-2 (and earlier versions, i will elaborate later) to compile cdrecord 1.11a39 (and earlier) makes it produce a buffer underrun immediately after writing has begun. fifo will go down from 100% straight to 0% in about 1-2 seconds. jörg schilling thinks it might be a scheduling bug in cygwin. here is howto reproduce this error: grab any of the binaries from http://www.geoshock.com/cdrtools/ (or compile your own of course) and burn a cd, using this command f.e.: mkisofs -J -R * | cdrecord -v -speed=24 - you can add -driveropts=burnfree if your writer supports it, that will save the cd-r, but fifo will stay at 0% during the whole time. it will happen every time, i tested it on 2 computers with two different writers. now comes the interesting part: i found some older cygwin binaries from 1.1.18 (cygwin1.dll, /usr and some binaries), mixed them with the 1.3.15-2, compiled cdrecord 1.11a39 with it and fifo works flawlessy! not a single buffer underrun up to now. that means the bug was introduced sometime between 1.1.18 and now (i never used cygwin before). thomas -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: file missing when trying to use gcc
liquid [EMAIL PROTECTED] wrote: hi, i'm having problems with gcc. every time i try to compile a program i get this error: liquid@LIQUID ~ $ gcc -o x x.c gcc: installation problem, cannot exec `as': No such file or directory liquid@LIQUID ~ $ i've reinstalled it but the problem continues. i think there's a missing package or something like. please send me the solution to this problem or the name of the package who contains the `as' program. thank you. http://cygwin.com/packages/ Search for /as.exe Setup's dependency support should have prevented this, though. Did you make a deliberate effort to minimize the packages you were installing? Max. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: file missing when trying to use gcc
Allan, We'll go you one better and show you how to find this out for yourself: http://cygwin.com/packages/ Here's a hint. To find binary executables include the .exe suffix. This is especially helpful if the name for which you're searching is short and might well appear as a substring in other file names included in some Cygwin package. Also note that the server that handles these look-ups is often rather slow or burdened, so be patient when you query. Good luck. Randall Schulz Mountain View, CA USA P.S. I wish I could help you with your malfunctioning shift key, but you'll have to fix that all on your own. At 16:16 2002-11-20, liquid wrote: hi, i'm having problems with gcc. every time i try to compile a program i get this error: liquid@LIQUID ~ $ gcc -o x x.c gcc: installation problem, cannot exec `as': No such file or directory liquid@LIQUID ~ $ i've reinstalled it but the problem continues. i think there's a missing package or something like. please send me the solution to this problem or the name of the package who contains the `as' program. thank you. -Allan liquid. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Another less/rxvt prob: Not interpreting backspace?
Yo, Again, Sergei! :) On Wed, 20 Nov 2002, Sergei Okhapkin wrote: Set env var SHELL to bash before starting rxvt. Thanks for taking a stab at this. :) I did as you suggested, but this didn't seem to change less's behavior. Obviously I'm ignorant, so I'm wondering why you thought this might fix me up... ---Jason -Original Message- From: Jason Tiller [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 6:59 PM To: [EMAIL PROTECTED] Subject: Another less/rxvt prob: Not interpreting backspace? Hi, All, :) Thank You! to Sergei and Matthijs who graciously gave suggestions and commiseration about my problem with 'less' not determining the terminal size correctly. Through no fault of my own, the problem seems to have ameliorated itself: running 'less' in a new rxvt instance no longer exhibits the quirky behavior of trying to scroll only 24 lines in a 25-line terminal. (Note that rxvt is still opening a window that is one row greater than what I ask for. I ask for '-geometry 80x24' and get 80x25, but that's a niggle.) However, 'less' now seems to be struggling to interpret keystrokes. Specifically backspace: when editing a search pattern, instead of erasing the previous character, 'less' inserts ^?. Ctrl-backspace seems to perform the function of backspace, as does Ctrl-h. Once again, this behavior didn't crop up until my recent update from 1.3.12 or so to 1.3.15 on Friday. Here is what I believe to be the relevant info about my system. If this isn't clear or complete enough, please let me know! Installation info: cygwin: 1.3.15-2 rxvt: 2.7.2-14 bash: 2.05b-7 less: 374-1 termcap: 20020930-1 terminfo: 5.2-3 Win2K SP2 rxvt invocation: C:\WINNT\system32\cmd.exe /c start C:\cygwin\bin\rxvt.exe -tn rxvt -fg grey -bg black -fn Fixedsys -sr -sl 5000 -e /bin/bash --login -i Relevant environment variables: TERM=xterm LESS=iFMSX# 8 CYGWIN=tty nontsec Any ideas on how I can investigate this problem further? ---Jason -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Cygwin problem...
If I installed Cygwin on a computer, burned it to another computer should it work? It doesn't for me. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Another less/rxvt prob: Not interpreting backspace?
Because I have almost the same backspace problem when /bin/sh is running inside Windows version of rxvt. I have no problems with /bin/bash or when rxvt is running in X-Window mode. -Original Message- From: Jason Tiller [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 7:28 PM To: [EMAIL PROTECTED] Subject: RE: Another less/rxvt prob: Not interpreting backspace? Yo, Again, Sergei! :) On Wed, 20 Nov 2002, Sergei Okhapkin wrote: Set env var SHELL to bash before starting rxvt. Thanks for taking a stab at this. :) I did as you suggested, but this didn't seem to change less's behavior. Obviously I'm ignorant, so I'm wondering why you thought this might fix me up... ---Jason -Original Message- From: Jason Tiller [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 6:59 PM To: [EMAIL PROTECTED] Subject: Another less/rxvt prob: Not interpreting backspace? Hi, All, :) Thank You! to Sergei and Matthijs who graciously gave suggestions and commiseration about my problem with 'less' not determining the terminal size correctly. Through no fault of my own, the problem seems to have ameliorated itself: running 'less' in a new rxvt instance no longer exhibits the quirky behavior of trying to scroll only 24 lines in a 25-line terminal. (Note that rxvt is still opening a window that is one row greater than what I ask for. I ask for '-geometry 80x24' and get 80x25, but that's a niggle.) However, 'less' now seems to be struggling to interpret keystrokes. Specifically backspace: when editing a search pattern, instead of erasing the previous character, 'less' inserts ^?. Ctrl-backspace seems to perform the function of backspace, as does Ctrl-h. Once again, this behavior didn't crop up until my recent update from 1.3.12 or so to 1.3.15 on Friday. Here is what I believe to be the relevant info about my system. If this isn't clear or complete enough, please let me know! Installation info: cygwin: 1.3.15-2 rxvt: 2.7.2-14 bash: 2.05b-7 less: 374-1 termcap: 20020930-1 terminfo: 5.2-3 Win2K SP2 rxvt invocation: C:\WINNT\system32\cmd.exe /c start C:\cygwin\bin\rxvt.exe -tn rxvt -fg grey -bg black -fn Fixedsys -sr -sl 5000 -e /bin/bash --login -i Relevant environment variables: TERM=xterm LESS=iFMSX# 8 CYGWIN=tty nontsec Any ideas on how I can investigate this problem further? ---Jason -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Another less/rxvt prob: Not interpreting backspace?
Hi, Sergei, :) On Wed, 20 Nov 2002, Sergei Okhapkin wrote: Because I have almost the same backspace problem when /bin/sh is running inside Windows version of rxvt. I have no problems with /bin/bash or when rxvt is running in X-Window mode. That seems reasonable. Did it work for you? I did a little scrounging via Google and saw a message that referred to something *very* similar - instead of backspace producing '^?', it produced '^H'. The suggested solution was: stty erase ^H I modifed that to fit my situation and tried: stty erase ^? This seems to have solved my problem because now 'less' properly interprets backspace. The post from which I (gratefully) stole this information (http://www.cygwin.com/ml/cygwin/1999-07/msg00203.html) ended with this caveat: This might have unexpected side effects in other programs, but at least not in bash which puts the tty into raw mode (by way of readline). Anybody know if I should expect untoward consequences by putting 'stty erase ^?' in my .bash_profile? Thanks! ---Jason -Original Message- From: Jason Tiller [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 7:28 PM To: [EMAIL PROTECTED] Subject: RE: Another less/rxvt prob: Not interpreting backspace? Yo, Again, Sergei! :) On Wed, 20 Nov 2002, Sergei Okhapkin wrote: Set env var SHELL to bash before starting rxvt. Thanks for taking a stab at this. :) I did as you suggested, but this didn't seem to change less's behavior. Obviously I'm ignorant, so I'm wondering why you thought this might fix me up... ---Jason -Original Message- From: Jason Tiller [mailto:[EMAIL PROTECTED]] Sent: Wednesday, November 20, 2002 6:59 PM To: [EMAIL PROTECTED] Subject: Another less/rxvt prob: Not interpreting backspace? Hi, All, :) Thank You! to Sergei and Matthijs who graciously gave suggestions and commiseration about my problem with 'less' not determining the terminal size correctly. Through no fault of my own, the problem seems to have ameliorated itself: running 'less' in a new rxvt instance no longer exhibits the quirky behavior of trying to scroll only 24 lines in a 25-line terminal. (Note that rxvt is still opening a window that is one row greater than what I ask for. I ask for '-geometry 80x24' and get 80x25, but that's a niggle.) However, 'less' now seems to be struggling to interpret keystrokes. Specifically backspace: when editing a search pattern, instead of erasing the previous character, 'less' inserts ^?. Ctrl-backspace seems to perform the function of backspace, as does Ctrl-h. Once again, this behavior didn't crop up until my recent update from 1.3.12 or so to 1.3.15 on Friday. Here is what I believe to be the relevant info about my system. If this isn't clear or complete enough, please let me know! Installation info: cygwin: 1.3.15-2 rxvt: 2.7.2-14 bash: 2.05b-7 less: 374-1 termcap: 20020930-1 terminfo: 5.2-3 Win2K SP2 rxvt invocation: C:\WINNT\system32\cmd.exe /c start C:\cygwin\bin\rxvt.exe -tn rxvt -fg grey -bg black -fn Fixedsys -sr -sl 5000 -e /bin/bash --login -i Relevant environment variables: TERM=xterm LESS=iFMSX# 8 CYGWIN=tty nontsec Any ideas on how I can investigate this problem further? ---Jason -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygwin problem...
On Wed, Nov 20, 2002 at 07:32:55PM -0500, [EMAIL PROTECTED] wrote: If I installed Cygwin on a computer, burned it to another computer should it work? It doesn't for me. http://cygwin.com/bugs.html cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
