autoexpect appears to be missing
Hello, I appear to be missing the autoexpect command, yet I have latest full versions of: expect tcltk dejagnu (plus source) I searched for this problem in the Documentation page, list archives, FAQ, and manpage. Do I need to install something else? thanks, (please include me on any reply as I'm currently off the list) Scott -- 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 Release process
-Original Message- From: Christopher Faylor [mailto:[EMAIL PROTECTED]] Sent: Monday, January 27, 2003 6:08 PM To: [EMAIL PROTECTED] Subject: Re: cygwin Release process On Mon, Jan 27, 2003 at 04:54:25PM -0500, Scott Prive wrote: William, The ntsec problem by all accounts was a one-time switch that burned a lot of people. It seems like a great feature (not completely using it myself), and when I upgraded to it I had NO idea of the impending change. I should have known better than to perform blind upgrades. I've been using Cywin for maybe 3 years (?) now and that's the only major problem I ever had. It's STILL not resolved, and I do not have time to attempt correcting it since I have required level of functionality. I'd love to see a release process something like Redhat's or Debian's. But that's not enough -- someone has to maintain it. As a FORMER ;-) Debian user I can say there are some downsides to too much process... like a stable tag that's so old almost no one runs 100% from that branch. Ok, that's three people complaining about the ntsec change. Don't you see some faulty reasoning here? How long do you suppose CYGWIN=ntsec existed? It was around for years. Lots of people used it. There was no way to anticipate that it would cause problems for some people. I tried to use my experience to bring William around to your point of view. Because William is concerned about future upgrades, I presented some scenarios for guarding against upgrades. I presume you're using Cygwin in a development context -- where it's good if you have more eyeballs testing the current code. I'll go out on a limb and assume William used Cygwin in a production environment -- where the opposite is true. You stay on something you understand, and if you don't knowingly apply major upgrades. If we had made the decision to turn CYGWIN=ntsec on in the next major stable release, then all of horrendous problems that were suffered (which, AFAIK, are solved these days by saying CYGWIN=nontsec) would still have manifested. The change *was* announced. The functionality *was* previously tested. I don't disagree that the change was announced. In hindsight, I see it was. Arthur Dent got an announcement before his home was demolished for a bypass (apologies to those who don't get the HHGTTG reference). :-) To this day, it's still not on the Cygwin.com home page under What's New. I don't doubt it is mentioned under a particular Release announcement. For better or worse, people in the Windows space wouldn't look there anyways. They wouldn't proactively subscribe to a mailing list and look for problem reports. What many users would assume is a setup utility provides a mechanism for a packager to provide critical warnings that must be acknowledged. I can't offer that improvement as a patch -- the best I can do is stay out of the way by assuming responsibility for my own upgrade testing. But it would be nice :-) I'd like to end this thread with the following comment: The hours I lost on my ntsec upgrade (and my botched attempts to fix it which probably made it worse) are *insignificant* to the hours of productivity I GAINED by having a mostly-complete GNU environment under Windows. Cygwin saved me from a complete rewrite task, of scripts which always assumed Linux. I can't understate how much I gained here. You made the `almost impossible', possible. Thank you (everyone). This is not an argument for a stable release. On the contrary, it's an argument for quick corrective releases to fix the problem. You make a good point about the old, stable release. That's one of the drawbacks of such a plan. What about errata? Are errata going to be supplied for the stable release? Who is going to decide when to make a new release? Is there going to be a voting process? Who gets to vote? Who gets to arbitrate disputes? Debian has a structure for this kind of thing. So does Red Hat. I don't see any evidence that the Cygwin community has the type of committment required for this kind of activity and, I, certainly don't want to be worrying about stuff like this. It requires someone with both the desire and organizational skills to pull it together. The effort to do this can't be trivialized. Agreed. I will happily provide the disk space for this but I am not going to be changing the way I release the packages I provide. If someone wants to take my packages, decide if they are experimental or stable, and put them into a different release framework then more power to them. Seriously. I'll applaud and commend their efforts. 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/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe
RE: Cygwin Release process
William, The ntsec problem by all accounts was a one-time switch that burned a lot of people. It seems like a great feature (not completely using it myself), and when I upgraded to it I had NO idea of the impending change. I should have known better than to perform blind upgrades. I've been using Cywin for maybe 3 years (?) now and that's the only major problem I ever had. It's STILL not resolved, and I do not have time to attempt correcting it since I have required level of functionality. I'd love to see a release process something like Redhat's or Debian's. But that's not enough -- someone has to maintain it. As a FORMER ;-) Debian user I can say there are some downsides to too much process... like a stable tag that's so old almost no one runs 100% from that branch. Here are some suggestions for you William: 1) maintain an internal mirror. Mirroring is very very easy to set up, and you will save your company money by eliminating random bandwidth spikes every time someone upgrades. 2) Make it EASY for your users to use your mirror (see step 3) 3) In the setup.exe GUI, you can manually add in a custom mirror. Find out if you can preload your mirror into the GUI, and remove standard mirrors. You may need to rebuild setup.exe; I'm not certain. 4) Make it more difficult for users to NOT use your mirror. If you're evil. :-) 5) If you're running internal software/testsuites on top of Cygwin, you can build a table of version numbers you expect, and what you see on the system. You can automate the building of this table so it is not a hassle every time you approve upgrade snapshots. If you think some users will install from the net anyways -- this will save you some debugging/triage time. disclaimer: I do *not* know what the current capabilities of setup.exe are. You might find discussion along these lines in the archives. Hope this helps! Scott -Original Message- From: William A. Hoffman [mailto:[EMAIL PROTECTED]] Sent: Monday, January 27, 2003 3:24 PM To: Max Bowsher; [EMAIL PROTECTED] Subject: Re: Cygwin Release process Well, if I am the only person with this opinion, then you are right. I should stop complaining and burn a CD. However, I suspect that I am not alone in wanting a more stable cygwin.It will be hard to prove my case, as the folks that read this list and post to it, tend to be more developer oriented, and are more interested in not missing out on the latest features than having a stable platform. There must be some reason that RedHat, Debian and all the major linux distributions have releases. I belive that if this were setup, and download stats were created, it would be come the most common type of download for cygwin. -Bill At 08:07 PM 1/27/2003 +, Max Bowsher wrote: If this is not good enough for you, then *just burn a CD*. There is no need to force this artificial 'release' policy on the Cygwin project. 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/ -- 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: Lesstif compilation problem and setup.exe
-Original Message- From: Mark Manning [mailto:markem;ev1.net] Sent: Monday, November 11, 2002 12:20 PM To: [EMAIL PROTECTED] Subject: Lesstif compilation problem and setup.exe snipped out your primary problem which I can't help with.. If you need a screen dump - let me know. I can either try to capture the text itself (didn't work before but I can try again) or do a screen capture (found a nice on-line product). Didn't want to post the screen capture. See `man script`. Run this before doing your compile, and you'll begin a disk log of everything on your terminal. Even if the process dies at the same as the others, it should have a record up until that point. If it's a reasonable size for posting, the general concensus is to post as an attachment so it doesn't generate false hits in the list archive searches.. For your primary concern, hopefully someone else will step up with real answers. :-) You might also attempt compiling building a smaller package to triage what breaks your system (ie, try packaging bash and avoiding anything X-related). -- 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: gdb hangs on a 486
At this point, I think most (all?) Cygwin packages are configured like this. Whether or not that's true, it's not unwarranted. There's good reason to make use of the newer architectures' capabilities. At the risk of asking for Yet Another Feature ... and I'm thinking out loud more than anything else... it would be friendly for the setup utility to do a CPU check vs. the packages you selected. I know.. patches gratefully accepted (You wouldn't want a patch in C from me. trust me :-) The whole system was downloaded through setup within the past 20 days. Gdb came up in a windowed rather than command line version. After the hang the mouse was dead and the system needed rebooting. I normally can run for weeks without reboots. gdb -nw Under DJGPP I am running gdb 5.1.1, with no apparent difficulties. There the configuration says i386-pc-msdosdjgpp Sounds like you may want to get the source, reconfigure, and build your own version targeting i386 or i486. Larry Hall [EMAIL PROTECTED] RFK Partners, Inc. http://www.rfk.com 838 Washington Street (508) 893-9779 - RFK Office Holliston, MA 01746 (508) 893-9889 - FAX -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: gdb hangs on a 486
Agreed. CPU specific-packages for obsolete platforms are not needed, and my remark was not intended to suggest something far less than this. What I meant was, in SETUP.EXE provide some warning to the end user that the packages they have selected will not run on their CPU. Allow them to continue if they acknowledge the warning. Imagine waiting for install to complete (probably on 56K), and then realize it's i586+ only. Having a check in the installer means you only wasted a ~250Kb download. -Scott -Original Message- From: [EMAIL PROTECTED] [mailto:lhall;pop.ma.ultranet.com] Sent: Thursday, October 31, 2002 12:32 PM To: Scott Prive; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: gdb hangs on a 486 Hm, an interesting thought. This would require packages to provide some information, probably in their setup.hint, to indicate their configuration target. Could work. But unless there are packages that are configured specifically for other than the default i686, I don't think it would be a feature that would get much use. But I'm willing to be proven wrong on this. :-) Larry Original Message: - From: Scott Prive [EMAIL PROTECTED] Date: Thu, 31 Oct 2002 10:06:22 -0500 To: [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: RE: gdb hangs on a 486 At this point, I think most (all?) Cygwin packages are configured like this. Whether or not that's true, it's not unwarranted. There's good reason to make use of the newer architectures' capabilities. At the risk of asking for Yet Another Feature ... and I'm thinking out loud more than anything else... it would be friendly for the setup utility to do a CPU check vs. the packages you selected. I know.. patches gratefully accepted (You wouldn't want a patch in C from me. trust me :-) The whole system was downloaded through setup within the past 20 days. Gdb came up in a windowed rather than command line version. After the hang the mouse was dead and the system needed rebooting. I normally can run for weeks without reboots. gdb -nw Under DJGPP I am running gdb 5.1.1, with no apparent difficulties. There the configuration says i386-pc-msdosdjgpp Sounds like you may want to get the source, reconfigure, and build your own version targeting i386 or i486. Larry Hall [EMAIL PROTECTED] RFK Partners, Inc. http://www.rfk.com 838 Washington Street (508) 893-9779 - RFK Office Holliston, MA 01746 (508) 893-9889 - FAX -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ mail2web - Check your email from the web at http://mail2web.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: gdb hangs on a 486
Thanks for the clarification Scott. Larry NP. When someone contributes a patch, I'll be sure to transfer those thanks since I don't deserve them (suggestions are *always* free..) ;-) Original Message: - From: Scott Prive [EMAIL PROTECTED] Date: Thu, 31 Oct 2002 13:30:41 -0500 To: [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: RE: gdb hangs on a 486 Agreed. CPU specific-packages for obsolete platforms are not needed, and my remark was not intended to suggest something far less than this. What I meant was, in SETUP.EXE provide some warning to the end user that the packages they have selected will not run on their CPU. Allow them to continue if they acknowledge the warning. Imagine waiting for install to complete (probably on 56K), and then realize it's i586+ only. Having a check in the installer means you only wasted a ~250Kb download. -Scott -Original Message- From: [EMAIL PROTECTED] [mailto:lhall;pop.ma.ultranet.com] Sent: Thursday, October 31, 2002 12:32 PM To: Scott Prive; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: gdb hangs on a 486 Hm, an interesting thought. This would require packages to provide some information, probably in their setup.hint, to indicate their configuration target. Could work. But unless there are packages that are configured specifically for other than the default i686, I don't think it would be a feature that would get much use. But I'm willing to be proven wrong on this. :-) Larry Original Message: - From: Scott Prive [EMAIL PROTECTED] Date: Thu, 31 Oct 2002 10:06:22 -0500 To: [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: RE: gdb hangs on a 486 At this point, I think most (all?) Cygwin packages are configured like this. Whether or not that's true, it's not unwarranted. There's good reason to make use of the newer architectures' capabilities. At the risk of asking for Yet Another Feature ... and I'm thinking out loud more than anything else... it would be friendly for the setup utility to do a CPU check vs. the packages you selected. I know.. patches gratefully accepted (You wouldn't want a patch in C from me. trust me :-) The whole system was downloaded through setup within the past 20 days. Gdb came up in a windowed rather than command line version. After the hang the mouse was dead and the system needed rebooting. I normally can run for weeks without reboots. gdb -nw Under DJGPP I am running gdb 5.1.1, with no apparent difficulties. There the configuration says i386-pc-msdosdjgpp Sounds like you may want to get the source, reconfigure, and build your own version targeting i386 or i486. Larry Hall [EMAIL PROTECTED] RFK Partners, Inc. http://www.rfk.com 838 Washington Street (508) 893-9779 - RFK Office Holliston, MA 01746 (508) 893-9889 - FAX -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ mail2web - Check your email from the web at http://mail2web.com/ . mail2web - Check your email from the web at http://mail2web.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: gdb hangs on a 486 (dead msg thread)
Yes, I read your email.. when it arrived. If there were no latency in mail, I would *even* have seen it before replying to Larry. I suspect redhat.com email routes more quickly on the inside :-) -Original Message- From: Christopher Faylor [mailto:cgf;redhat.com] Sent: Thursday, October 31, 2002 2:56 PM To: [EMAIL PROTECTED] Subject: Re: gdb hangs on a 486 On Thu, Oct 31, 2002 at 02:46:58PM -0500, Scott Prive wrote: Thanks for the clarification Scott. NP. When someone contributes a patch, I'll be sure to transfer those thanks since I don't deserve them (suggestions are *always* free..) ;-) Is anyone reading my previous message? This SHOULD NOT BE AN ISSUE. There is no need for a patch. 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/ -- 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: Looking for cygpcre.dll
This (and other) dll's get installed by programs that need it. You do not need to worry about picking dll's -- it's all automatic by setup.exe If I understand you, you just want the dll, and you're not terribly curious to know what package it comes from. So, one answer is run SETUP, select something, and verify you have the file. If you need to know exactly which package provides this, I cannot answer you. -Scott -Original Message- From: jblazi [mailto:jblazi;gmx.de] Sent: Thursday, October 31, 2002 6:08 PM To: [EMAIL PROTECTED] Subject: Looking for cygpcre.dll I need cygpcre.dll. I found it with search on the cygwin site but I still do not know how to get it (via setup). Can anybody help me? TIA, -- Janos Blazi -- 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: REQUEST FOR AN ASSISTANCE (can we get an ALL CAPS spam filter?)
Any chance the list maintainer could add a filter for all-caps emails? -- 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: domain users problem
No one here will know about your non-Cygwin tools. I'm sure you can get this all to work, but you'll need to solve the problem yourself. Here are some suggestions for your debugging; they MAY or MAY NOT work but could provide you with clues. Also, these are generalizations and I may not be 100% accurate. Hopefully this help you some.. From a shell -- Cygwin OR MS Windows CDM.EXE, you can switch to the other environment using commands. --- ex: (from CMD.EXE) Microsoft Windows 2000 [Version 5.00.2195] (C) Copyright 1985-2000 Microsoft Corp. C:\WINNTsh $ pwd /cygdrive/c/WINNT $ cmd Microsoft Windows 2000 [Version 5.00.2195] (C) Copyright 1985-2000 Microsoft Corp. C:\sh -c 'ls -l *sh' sh -c 'ls -l *sh' -r-xr-xr-x1 Administ None 1083 Oct 7 14:59 watch.sh If you want to mix commands or run Cygwin commands from a C: prompt, you need to make sure your Windows Environment variable PATH includes the path for Cygwin's bin directory. Must be defined as Windows-style path like so: C:\Cygwin\bin If you can then run Cygwin commands from a C: prompt, be aware that Cygwin expects UNIX style pathing /tmp not c:\temp. Highly reccomended you search the list archives and documentation regarding cygpath. I'd try to determine why you get long timeouts before your system gives up with command not found. That's a clue you have a pathing issue. You might want to create testing shell or BAT scripts throughout your command chain, and have them echo to the screen what they get as parameters (I'm assuming something in your setup calls another command which can't be found). Personally I'd suggest you consider staying inside the Cygwin environment and use Cygwin's sshd server to remotely execute commands. Setting up the sshd server is not always trivial due to NT permissions, etc. but when it's working, it's REALLY nice. Good luck! -Scott -Original Message- From: Steven Weiss [mailto:sweiss;iafrica.com] Sent: Sunday, October 27, 2002 9:35 AM To: [EMAIL PROTECTED] Subject: domain users problem Cygwin 1.3.10 Ataman 3.1 W2000 Hi I am having a performance problem when executing cygwin commands i.e. ls when running the command through Ataman rexec. Example: Domain user from client PC - rexec SERVER cmd Ataman validates the user as the domain users and presents the user with a c: prompt. The user types 'ls' and it takes about 30 seconds for a directory listing to come back. Also when typing 'bash', the command just hangs in a sort of interative mode and any additional commands typed return a ': command not found' error. If the user is not a domain user and is local to SERVER, then I do not have any problems. Also if the user logs directly in at the Console I do not have any problems. I have another server configured exactly the same way and am not getting this problem and I cannot see anything different. Regards Steven Weiss At command prompt USER-A types ls I'm using a W2000 rexec from a client to to execute a b -- 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: == operand not found
This one is in the FAQ on the Cygwin website. I'll give you hint: ;-) $ /bin/bash testme.sh 1 Hello World Scott -Original Message- From: Nitin Gupta [mailto:gupta;equator.com] Sent: Wednesday, October 23, 2002 6:22 PM To: [EMAIL PROTECTED] Subject: == operand not found Hi, following script runs fine on linux, but not on cygwin. Please let me know equivalent of == on cygwin. Thanks, Nitin #!/bin/sh if [ $1 == 1 ]; then echo Hello World fi -- 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: == operand not found
-Original Message- From: Randall R Schulz [mailto:rrschulz;cris.com] Sent: Wednesday, October 23, 2002 7:30 PM To: [EMAIL PROTECTED] Subject: Re: == operand not found Nitin, You're most likely accustomed on your Linux system to /bin/sh being BASH. On Cygwinm /bin/sh is ASH, and it is far more minimal in its implementation of the POSIX shell standard, This makes me ask a few questions.. 1) Why is ash the default? At least on UNIX systems that use true sh -- usually just /bin/bash in /bin/sh compatibility mode -- I can understand THAT because plain sh is, well... traditional. :-) Bash2 seems closer to most expectations; ash doesn't seem to add any value. 2) How would a user know they are defaulting to ash? a) The first place I would look is /etc/password for my default, which clearly states /bin/bash (at least for me it does). b) Next I would ls -l on /bin/whatever to see if it is a symbolic link to something else. Even on NTFS, /bin/sh or /usr/bin/sh do not appear to be links. -Scott and does not provide == as an equivalent for = in the test (a.k.a. [) built-in. Randall Schulz Mountain View, CA USA At 15:22 2002-10-23, Nitin Gupta wrote: Hi, following script runs fine on linux, but not on cygwin. Please let me know equivalent of == on cygwin. Thanks, Nitin #!/bin/sh if [ $1 == 1 ]; then echo Hello World fi -- 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: cygwinl.dll
-Original Message- From: David Starks-Browning [mailto:starksb;ebi.ac.uk] Sent: Monday, October 21, 2002 8:18 AM To: nemrut cesetevi Cc: [EMAIL PROTECTED] Subject: cygwinl.dll On Monday 21 Oct 02, nemrut cesetevi writes: i installed cygwin-b20 full setup.but i dont find cygwinl.dll file Where did you get cygwin-b20 full setup? anywhere .where can i donwload full cygwin ? dont say cygwin.com plz :/ Why not? Is this a joke? This is more of a reply to nemrut, or just me talking out loud on a Monday morning... :-) It *might* be that nemrut has Cygwin B20 on CD-ROM, and has bad dialup internet service, where it will cost him money to download the latest from cygwin.com. That's a valid reason to [want to] stay on an older version.. If this IS the case, nemrut should ask the question, `where could he find CD-ROMs with a more recent Cygwin?`. Unfortunately I do not know the answer. I checked cheapbytes.com but they don't offer Cygwin discs. But perhaps this post will prompt nemrut to offer more information... -Scott David -- 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: Latest update problems (ntsec and file permissions, I think)(tagged releases comment)
-Original Message- From: William A. Hoffman [mailto:billlist;nycap.rr.com] Sent: Monday, October 21, 2002 10:12 AM To: [EMAIL PROTECTED] Subject: Latest update problems (ntsec and file permissions, I think) Once I did that, things were better. Is this a problem that is going to be fixes, or are people supposed to know that they have to run mkpasswd -du after running setup? BTW, I also tried setting CYGWIN to nontsec, but this had no effect on the problem. If a new user were to try cygwin right now, I think they would be a bit confused. -Bill Personally, I watch the list and am more careful now about upgrading a box I use in production. This can be intimidating to new users, especially when the norm is to expect a Setup program to present you with Release Notes. Cygwin reminds me of running the testing tree of Debian: something is updated every day, and by participating in this tree you assume a higher level of risk than usual Windows software. I think what is sorely needed is a tagged release, or bundled packages. The way this could work is you are locked into a version set, and can add and remove packages without always getting the latest code. The benefit here would be a set of known issues for the release, and a proper set of Release Notes could be made for each release. I get the impression patches for releases and show release notes would be accepted (things I couldn't offer). -Scott -- 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 -c
Chris F.: Please do write that book! I'm negotiating as fast as I can. :-) cgf Would it help if everyone politely asked the potential publisher when they'll ever have a Cygwin book? ;-) -- 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 feature request - setup.exe release notes
Feature request: For SOME packages, release Notes that can be pulled up from the installer. Many people run Cygwin and are not members of this list. Ideally, people would read the mailing list before installing, and again before upgrading. This is not the reality and human nature is difficult to change. So in the spirit of this, a nice feature would there were a pop-up warning or note for risky upgrades, or first time installs. This would not need to be done for most packages, just those that may change expected or assumed behavior. For example: I saw the post about the security defaults changing. If you're on this list, you would know this, but otherwise you are in for a big surprise in cvs file attributes, etc. There might be other examples but this is what comes to me now. In short: if it's worth warning users on the list about, the software might do the same. Is this a bad idea? (And no, I can't send patches for this :-) -Scott -- 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/
Perl system() and Cygwin
Hello,
I'm experiencing some non-intuitive behavior when -- under Cygwin Perl -- calling
system commands. The FAQ alludes to differences between Cygwin and systems like
Linux.. this might be another one of these differences, or it might be dumb-user-error.
I have a solution I can live with. I just don't like not knowing what breaks my other
attempts.
I want to do the Perl equivalent of
net use Q: e04-3000\\foo foo /user:foo
This always works in bash, running 'net' and everything else is an argument. This
works under Cygwin /bin/bash just fine.
In Perl, you can call commands like so:
system(net use Q: e04-3000\\foo foo /user:foo);
This fails, with the net command returning Error 67 (resource not found).
Unfortunately there's no way it tells me what it THINKS it saw.
This also fails:
system(net, use Q: e04-3000\\foo foo /user:foo);
*** This WORKS:
system(net, use, Q:, e04-3000\\foo, foo, /user:foo);
But it's *evil*.
Since the above works, I try less evil-looking code:
$cmd = net use Q: e04-3000\\foo foo /user:foo
$mountcode = system (split(' ',$cmd));
to simulate the list that worked. This also fails.
I've been executing code like this (all one string) all along under Linux Perl. I'm
assuming Perl's system() on Cygwin executes CMD.EXE as a subshell (this true?) and,
well, this is one of those differences
Any Thoughts? Thanks.
-Scott
--
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: RESOLVED: Perl system() and Cygwin
-Original Message-
From: Igor Pechtchanski [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, October 15, 2002 4:00 PM
To: Scott Prive
Cc: [EMAIL PROTECTED]
Subject: Re: Perl system() and Cygwin
On Tue, 15 Oct 2002, Scott Prive wrote:
Hello,
I'm experiencing some non-intuitive behavior when -- under
Cygwin Perl
-- calling system commands. The FAQ alludes to differences between
Cygwin and systems like Linux.. this might be another one of these
differences, or it might be dumb-user-error.
I have a solution I can live with. I just don't like not
knowing what
breaks my other attempts.
I want to do the Perl equivalent of
net use Q: e04-3000\\foo foo /user:foo
This always works in bash, running 'net' and everything else is an
argument. This works under Cygwin /bin/bash just fine.
In Perl, you can call commands like so:
system(net use Q: e04-3000\\foo foo /user:foo);
This fails, with the net command returning Error 67 (resource not
found). Unfortunately there's no way it tells me what it
THINKS it saw.
This also fails:
system(net, use Q: e04-3000\\foo foo /user:foo);
*** This WORKS:
system(net, use, Q:, e04-3000\\foo, foo, /user:foo);
But it's *evil*.
Since the above works, I try less evil-looking code:
$cmd = net use Q: e04-3000\\foo foo /user:foo
$mountcode = system (split(' ',$cmd));
to simulate the list that worked. This also fails.
I've been executing code like this (all one string) all along under
Linux Perl. I'm assuming Perl's system() on Cygwin
executes CMD.EXE as
a subshell (this true?) and, well, this is one of those
differences
Any Thoughts? Thanks.
-Scott
Scott,
Any perl will behave like this, even the Linux one. Your
problem is the
backslashes. Since you provide the command all in one chunk,
a shell is
invoked to parse the arguments, and it obviously sees fewer
backslashes
than you provided (because some are used as escapes in perl
itself). So,
if you want to use the single-parameter system() call, you
have to escape
each backslash twice (i.e., double the number of backslashes).
I don't really know why the split doesn't work, except to
guess that it
might be interpreting the result in a scalar context... Try
assigning it
to a list first.
Igor
You're right. Wrote a small shell script that echo'd out $* $@ so I could verify the
translation loss. I was thinking this worked in Linux because I often put commands
in a string, which provide some protection... but I never send to the shell so
double-escaping seemed like a non-issue.
So, I have a double-escaped UNC (ugg-lee! :-) to mount CIFS.
DB36 system(net use Q: e04-3000foo foo /user:foo);
The command completed successfully.
Thanks,
Scott
--
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: paths like //usr/local
FYI: I actually *had* hostnames like var, tmp and so on at one time... :-) Cygwin maps UNIX/POSIX behavior on top of NT, but NT was designed to be compatible with DOS's *broken* conventions, so NT is half-broken. Any kind of parsing for mounts like you suggest would probably incur a performance hit, generate limitations, and cause breakage everywhere. The escaping can be annoying but it's livable IMO. I suggest not to use //, but to use the proper path convention -- and escape accordingly. This is more keystrokes and more to remember, but it seems bugfree and 100% consistent. I shudder at the thought of `rm -rf /path` going to an interpreted mount point. Example: Desired path: \\server\share Bash: server\\share Perl - going THROUGH bash, using system(): servershare Slightly off topic, but standard Bash *does* support hostname completion. You have to configure it -- see standard BASH2 documentation online. I have no idea how well it works with Cygwin and with UNC pathing, but it works on UNIX with automount and ssh. If nothing, it could be extended by someone for UNC. -Scott -Original Message- From: Jan Nieuwenhuizen [mailto:[EMAIL PROTECTED]] Sent: Tuesday, October 15, 2002 4:01 PM To: [EMAIL PROTECTED] Subject: Re: paths like //usr/local Christopher Faylor [EMAIL PROTECTED] writes: cygwin allows the user to specify paths like: c:\foo\bar and c:/foo/bar. Similarly, it allows //foo/bar and \\foo\bar . If that doesn't satisfy you then you can go back to the Because we're mean argument. I've been hurt by this too, and it makes me think. It would be even more satisfactory if some configurable list of 'hosts' would map to //localhost/. Hosts with names such as \\bin, \\etc, \\tmp, \\usr or \\var come to mind. Now if such a thing could be implemented without some horrible kludge, would that be nice? Jan. -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Cygwin can't write to CIFS... but cmd.exe *can*
If one is authenticated against a remote CIFS share (Linux), should there be a difference in permissions between Cygwin and CMD.EXE? Example: After authentication/mount (via net use), I try `echo foo myfile.txt` In the Cygwin shell, this fails and I get a 0-byte file. If I start CMD.EXE from the same shell, it works in the CMD processor. Could someone please explain the inconsistency? If not, suggestions for workaround ARE welcome (at a higher level I'm working off a Perl script but I reproduced the above in bash using just 'echo'). To eliminate sshd issues, I am running things locally (using WinVNC to export my display). -Scott Cygwin Win95/NT Configuration Diagnostics Current System Time: Mon Oct 14 10:08:54 2002 Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 2 Path: C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin c:\WINNT\system32 c:\WINNT c:\WINNT\System32\Wbem SysDir: C:\WINNT\System32 WinDir: C:\WINNT CYGWIN = `tty ntsec' HOME = `C:\cygwin\home\Administrator' MAKE_MODE = `unix' PWD = `/home/Administrator' USER = `Administrator' ALLUSERSPROFILE = `C:\Documents and Settings\All Users' APPDATA = `C:\Documents and Settings\Administrator.QA\Application Data' COMMONPROGRAMFILES = `C:\Program Files\Common Files' COMPUTERNAME = `QA2000TEST' COMSPEC = `C:\WINNT\system32\cmd.exe' CVSREAD = `y' CVSROOT = `:pserver:[EMAIL PROTECTED]:/var/cvsroot_generic' HOMEDRIVE = `C:' HOMEPATH = `\' LOGONSERVER = `\\QACONTROL' MANPATH = `:/usr/ssl/man' NUMBER_OF_PROCESSORS = `1' OLDPWD = `/usr/bin' OS2LIBPATH = `C:\WINNT\system32\os2\dll;' OS = `Windows_NT' PATHEXT = `.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH' PROCESSOR_ARCHITECTURE = `x86' PROCESSOR_IDENTIFIER = `x86 Family 6 Model 8 Stepping 6, GenuineIntel' PROCESSOR_LEVEL = `6' PROCESSOR_REVISION = `0806' PROGRAMFILES = `C:\Program Files' PROMPT = `$P$G' PS1 = `\[\033]0;\w\007 \033[32m\]\u@\h \[\033[33m\w\033[0m\] $ ' SHLVL = `1' SINC = `/usr/pool/stest/linux/bin/include' SYSTEMDRIVE = `C:' SYSTEMROOT = `C:\WINNT' TEMP = `c:\DOCUME~1\ADMINI~1.QA\LOCALS~1\Temp' TERM = `cygwin' TMP = `c:\DOCUME~1\ADMINI~1.QA\LOCALS~1\Temp' USERDNSDOMAIN = `qa.storigen.com' USERDOMAIN = `QA' USERNAME = `Administrator' USERPROFILE = `C:\Documents and Settings\Administrator.QA' WINDIR = `C:\WINNT' _ = `/usr/bin/cygcheck.exe' Use `-r' to scan registry a: fd N/AN/A c: hd NTFS 14645Mb 16% CP CS UN PA FC d: cd N/AN/A w: net NTFS2022Mb 64% CP CSPAfoo C:\cygwin / system binmode C:\cygwin/bin /usr/bin system binmode C:\cygwin/lib /usr/lib system binmode . /cygdrive userbinmode,cygdrive Found: C:\cygwin\bin\bash.exe Found: C:\cygwin\bin\cat.exe Found: C:\cygwin\bin\cpp.exe Found: C:\cygwin\bin\find.exe Found: C:\cygwin\bin\gcc.exe Not Found: gdb Found: C:\cygwin\bin\ld.exe Found: C:\cygwin\bin\ls.exe Found: C:\cygwin\bin\make.exe Found: C:\cygwin\bin\sh.exe 58k 2002/05/07 C:\cygwin\bin\cygbz2-1.dll - os=4.0 img=1.0 sys=4.0 cygbz2-1.dll v0.0 ts=2002/5/7 2:33 625k 2002/08/09 C:\cygwin\bin\cygcrypto.dll - os=4.0 img=1.0 sys=4.0 cygcrypto.dll v0.0 ts=2002/8/9 16:20 452k 2002/07/17 C:\cygwin\bin\cygcurl-2.dll - os=4.0 img=1.0 sys=4.0 cygcurl-2.dll v0.0 ts=2002/7/17 10:50 45k 2001/04/25 C:\cygwin\bin\cygform5.dll - os=4.0 img=1.0 sys=4.0 cygform5.dll v0.0 ts=2001/4/25 1:28 35k 2002/01/09 C:\cygwin\bin\cygform6.dll - os=4.0 img=1.0 sys=4.0 cygform6.dll v0.0 ts=2002/1/9 1:03 19k 2002/02/20 C:\cygwin\bin\cyggdbm.dll - os=4.0 img=1.0 sys=4.0 cyggdbm.dll v0.0 ts=2002/2/19 22:05 17k 2001/06/28 C:\cygwin\bin\cyghistory4.dll - os=4.0 img=1.0 sys=4.0 cyghistory4.dll v0.0 ts=2001/1/6 23:34 20k 2002/10/10 C:\cygwin\bin\cyghistory5.dll - os=4.0 img=1.0 sys=4.0 cyghistory5.dll v0.0 ts=2002/10/10 13:28 929k 2002/06/24 C:\cygwin\bin\cygiconv-2.dll - os=4.0 img=1.0 sys=4.0 cygiconv-2.dll v0.0 ts=2002/6/24 14:24 22k 2001/12/13 C:\cygwin\bin\cygintl-1.dll - os=4.0 img=1.0 sys=4.0 cygintl-1.dll v0.0 ts=2001/12/13 4:28 28k 2002/09/20 C:\cygwin\bin\cygintl-2.dll - os=4.0 img=1.0 sys=4.0 cygintl-2.dll v0.0 ts=2002/9/19 23:13 21k 2001/06/20 C:\cygwin\bin\cygintl.dll - os=4.0 img=1.0 sys=4.0 cygintl.dll v0.0 ts=2001/6/20 13:09 81k 2000/12/05 C:\cygwin\bin\cygitcl30.dll - os=4.0 img=1.0 sys=4.0 cygitcl30.dll v0.0 ts=2000/11/25 20:43 35k 2000/12/05 C:\cygwin\bin\cygitk30.dll - os=4.0 img=1.0 sys=4.0 cygitk30.dll v0.0 ts=2000/11/25 20:43 26k 2001/04/25 C:\cygwin\bin\cygmenu5.dll - os=4.0 img=1.0 sys=4.0
RE: Cygwin can't write to CIFS... but cmd.exe *can* (more)
...of course, when I do this (in either example), I have cd'd to the CIFS share (/cygdrive/w/ in both cases) Also, the share is authenticated as a test account other than who I am in the shell (shell user=Administrator; CIFS authenticated as user 'foo'). I'm wondering if this has anything to do with my problem, but one would expect to be able to authenticate CIFS shares as other users (I even tried mapping the drive under plain Explorer). What puzzles me is if I start cmd.exe as a subprocess of bash, the writes succeed. -Scott Example: After authentication/mount (via net use), I try `echo foo myfile.txt` -- 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: RESOLVED: Cygwin can't write to CIFS... but cmd.exe *can* (more)
YES thank you sir! :-D that helped me find this: http://sources.redhat.com/ml/cygwin/2000-12/msg00877.html For the archive: Add to NT Environment nontsec=smb (no quotes, append to your existing CYGWIN value). I did search the list before posting, but my search (CIFS, Permission) got ZERO hits. Surprisingly, even the refined search (smb nontsec) got only a few hits... all talking about this problem BEFORE the feature code was added (discussion was still taking place as to what the variable should be, including ilikepie=yes :-) Interestingly, some folks had suggested that the Cygwin default mirror CMD.EXE behavior, as it would be less confusing (I agree). I didn't see a reply to that suggestion. This would be a great addition to the FAQ IMO, even if it only references a search to the archives, because you need to know the answer to get a hit on this search... -Scott -Original Message- From: Donald MacVicar [mailto:[EMAIL PROTECTED]] Sent: Monday, October 14, 2002 11:12 AM To: [EMAIL PROTECTED] Subject: Re: Cygwin can't write to CIFS... but cmd.exe *can* (more) There is loads of stuff in the list archives on this. If you have local user accounts on the Win machine (rather than domain accounts) and access using shares using the unix account then the UID/GID are different, you can set CYGWIN too some value - I think it is smbnontsec, someone can correct me if I cam wrong. Using a different user for authentication on the samba share that the current user will work fine under win but because of the UID changes will not work under cygwin unless the smbnontsec is added to the CYGWIN enviroment variable. A search of the archives will give you plenty more on this. Donald. Igor Pechtchanski wrote: Scott, I've had some trouble with file permissions on samba shares under Win2k. Not anything as severe as yours, but the files created on a share didn't inherit the world read permissions of the directories (and those couldn't be set). I wonder if these are related? Igor On Mon, 14 Oct 2002, Scott Prive wrote: ...of course, when I do this (in either example), I have cd'd to the CIFS share (/cygdrive/w/ in both cases) Also, the share is authenticated as a test account other than who I am in the shell (shell user=Administrator; CIFS authenticated as user 'foo'). I'm wondering if this has anything to do with my problem, but one would expect to be able to authenticate CIFS shares as other users (I even tried mapping the drive under plain Explorer). What puzzles me is if I start cmd.exe as a subprocess of bash, the writes succeed. -Scott Example: After authentication/mount (via net use), I try `echo foo myfile.txt` -- 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: RESOLVED: cygwin can't write to CIFS... but cmd.exe *can* (more)
nontsec=smb probably just turns off ntsec entirely. The CYGWIN environment variable parser is not that sophisticated. oops. I don't suppose I would even notice ntsec being off notice given my limited needs. You are probably right, and this is likely a `side effect/parser issue'... I've since moved my environment to the documented variable nosmbntsec 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/ -- 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 can't write to CIFS... but cmd.exe *can* (more)
Igor, I suspect the resolution to my problem will be different from yours. For me, I just wanted shares mounted under Cygwin to behave the same as if I mounted it under Command Prompt. I got this result by disabling ntsec for CIFS shares (nosmbntsec). I couldn't provide an answer for your inheriting-permissions issue, sorry. -Scott (Eagerly awaiting someone to write an O'Reily book on Cygwin :-) -Original Message- From: Igor Pechtchanski [mailto:[EMAIL PROTECTED]] Sent: Monday, October 14, 2002 10:54 AM To: Scott Prive Cc: [EMAIL PROTECTED] Subject: RE: Cygwin can't write to CIFS... but cmd.exe *can* (more) Scott, I've had some trouble with file permissions on samba shares under Win2k. Not anything as severe as yours, but the files created on a share didn't inherit the world read permissions of the directories (and those couldn't be set). I wonder if these are related? Igor On Mon, 14 Oct 2002, Scott Prive wrote: ...of course, when I do this (in either example), I have cd'd to the CIFS share (/cygdrive/w/ in both cases) Also, the share is authenticated as a test account other than who I am in the shell (shell user=Administrator; CIFS authenticated as user 'foo'). I'm wondering if this has anything to do with my problem, but one would expect to be able to authenticate CIFS shares as other users (I even tried mapping the drive under plain Explorer). What puzzles me is if I start cmd.exe as a subprocess of bash, the writes succeed. -Scott Example: After authentication/mount (via net use), I try `echo foo myfile.txt` -- 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/
Oreilly book on Cygwin
I sent my request to [EMAIL PROTECTED] Hopefully they sense enough demand to investigate this topic... -Original Message- From: Larry Hall (RFK Partners, Inc) [mailto:[EMAIL PROTECTED]] Sent: Monday, October 14, 2002 3:24 PM To: Scott Prive; [EMAIL PROTECTED] Subject: RE: Cygwin can't write to CIFS... but cmd.exe *can* (more) At 03:01 PM 10/14/2002, Scott Prive wrote: (Eagerly awaiting someone to write an O'Reily book on Cygwin :-) Sounds good to me. My guess is that someone here would be interested in doing so if O'Reilly expressed interest. ;-) Larry Hall [EMAIL PROTECTED] RFK Partners, Inc. http://www.rfk.com 838 Washington Street (508) 893-9779 - RFK Office Holliston, MA 01746 (508) 893-9889 - FAX -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
weird Cygwin+CIFS issues (not symlinks...): filename contains garbage mounting under ssh
This question deals with weird, undeletable files under CIFS mounts, and an issue creating drive mounts under ssh. They may be related. Problem #1 - garbage-chars in filename Drive W: mounted using MS' net use command. I am operating under a remote ssh login. I am using Cygwin on a Linux CIFS share and get denied message when I DO have correct permissions. This error leaves weird filenames in the dir. $ echo foo: HJHKH ()) /cygdrive/w/bashwrite -bash: /cygdrive/w/bashwrite: Permission denied copies don't work either: $ cp /cygdrive/c/expectedfile /cygdrive/w/newwrite cp: cannot create regular file `/cygdrive/w/newwrite': Permission denied $ ls -l /cygdrive/w ls: /cygdrive/w/bashwrite???X: No such file or directory ls: /cygdrive/w/newwriteX: No such file or directory total 1 -rwxr--r--1 2000 10016 Oct 13 17:24 expectedfile OK. I'll try it in CMD.EXE (under Cygwin) Administrator@QA2000TEST ~ $ cmd Microsoft Windows 2000 [Version 5.00.2195] (C) Copyright 1985-2000 Microsoft Corp. C:\cygwin\home\Administratorw: w: W:\ls ls bashwrite???X expectedfile newwriteX W:\echo hello worldhello_world echo hello worldhello_world W:\dir dir Volume in drive W is foo Volume Serial Number is 108D-0104 Directory of W:\ 10/13/2002 07:04p DIR . 09/24/2002 04:17p DIR .. 10/13/2002 05:24p 6 expectedfile 10/13/2002 06:59p 0 bashwrite???X 10/13/2002 07:02p 0 newwriteX 10/13/2002 07:04p 15 hello_world 4 File(s) 21 bytes 2 Dir(s) 767,623,168 bytes free W:\more hello_world more hello_world q ? Administrator@QA2000TEST ~ $ Because I could create the file using cmd.exe, this points to my Cygwin setup... but it might be more complicated than that. Problem #2: Drive shares created under ssh session, do not work under graphical console of Windows 2000. Example: I ssh into system as Administrator, net use to mount a CIFS drive. In the graphical desktop (also logged in as Administrator), the drive letter is shown but it has a red X and cannot be used. If I use Explorer to mount a new drive letter to the same UNC path, that mount behaves properly. This feels like a permissions issue, and I'm wondering if my earlier ssh server issues (which *seems* resolved) are the cause, or if it's a CIFS issue I only see with Cygwin (see log above, doesn't happen in CMD.EXE) ANY insight would be most appreciated! -Scott -- 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: Executing a script that needs DOS path
Have you looked for cygpath in the archives (also `man cygpath`). cygpath will return a converted path. Be sure to use single quotes OR properly escape your input string or you will not get the expected result. I haven't used this myself, which is why I avoid a direct answer, but I've seen the question enough on the list and a cursory glance at the manpage says you want this... -Scott -Original Message- From: Ivan Dobrianov [mailto:[EMAIL PROTECTED]] Sent: Thursday, October 10, 2002 2:37 PM To: [EMAIL PROTECTED] Subject: Executing a script that needs DOS path Sorry if this has been answered a hundred times, but could not find anything the FAQ, doc, or archives. THE PROBLEM: o Say I have some intrepreter xxx.exe, that expects to get started like this xxx c:\home\my_script.xxx arg_1 arg_2 ... o I want to automate this process the usual way, by adding this to the begining of the script: #!/c/bin/xxx ... hoping to be able to say: my_script.xxx arg_1 arg_2 o *** This fails, because cygwin [or bash] passes the unix path /c/home/my_script.xxx to xxx.exe which it cannot interpret. Is there a solution to this OTHER THAN using a proxy shell that would do the unix-to-dos translation? The reason I don't like this solution is that the shell will glob and eat quotes, making it very hard to process filenames with spaces. Thanks for any hints! -- 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/
ssh service staring problem bad owner /var/empty but not fixed
Hello, I understand the problem I am about to ask is not uncommon, and I have made considerable effort to look for the answers in the archive... On an up-to-date (today) Cygwin install, sshd refuses to start (the MS Management console gives a useless error). On other systems, I have installed Cygwin sshd and it worked fine (I have not done this recently though and I understand there have been changes to ssd of sorts). When I first attempted this install some weeks back, I followed the guide at http://tech.erdelynet.com/cygwin-sshd.html I didn't actually run the permissions script as the author had just days before, pulled down the script. I'm not sure if the other steps on this page complicate my problem, so I'll mention it. The first thing I check is /var/log/sshd.log, and it's bad owner or mode for /var/empty. OK, it's some sort of NT permissions issue. A Google search tells me /var/empty should be chmod 700 or 755 (it's 755). grep /etc/passwd ssh shows ssh account is 1000:513, sshd privsep, home of /var/empty and shell of /bin/false I've also tried chowning the directory as SYSTEM:SYSTEM (or 18:18). I did notice in the MMC Groups panel, there is no VISIBLE group for sshd, but there is a sshd user. My Google searches tell me there should be a group, so I attempt to add the group sshd and make sshd user a member. I get the error: while attempting to create the group sshd on computer QA2000TEST: The account already exists. I get this error if I attempt to create the group sshd with or with-out the member sshd. I've Reinstalled openssh, and even selected Unininstall followed by Install in case there was a difference. The version of openssh I have is 3.4p1-5 I appreciate any help. I hope I have checked all of the obvious gotchas so I don't waste anyone's time. Thanks. -Scott -- 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: ssh service staring problem bad owner /var/empty but not fixed
-Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 12:02 PM To: cygml Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hi, I had that when i first installed it... i take it the permissions on files like /etc/sshd* /etc/ssh_host* are exclusive to the SYSTEM account (if your running a shared-style server) and the service is running as SYSTEM. Let's see...: $ ls -l /etc/ssh* -rw-r--r--1 Administ None 1049 Sep 5 15:59 /etc/ssh_config -rw-r--r--1 Administ None 668 Sep 5 15:19 /etc/ssh_host_dsa_key -rw-r--r--1 Administ None 614 Sep 5 15:19 /etc/ssh_host_dsa_key.pub -rw-r--r--1 Administ None 539 Sep 5 15:19 /etc/ssh_host_key -rw-r--r--1 Administ None 343 Sep 5 15:19 /etc/ssh_host_key.pub -rw-r--r--1 Administ None 883 Sep 5 15:19 /etc/ssh_host_rsa_key -rw-r--r--1 Administ None 234 Sep 5 15:19 /etc/ssh_host_rsa_key.pub -rw-r--r--1 Administ None 2041 Sep 5 15:59 /etc/sshd_config Is Administrator here perfectly synonymous with SYSTEM? Also, I'm not sure what you mean by shared style server, how to verify if that is my case, or how this would affect things. The service in MMC shows it logs on as Local System Account, interact with desktop NOT checked. Should this instead be running as sshd user or Administrator? I personally prefer to get things running the right way and not blow holes through local security. That said, this is a test lab system and I'd go the hack way to Make It Work... if I knew what to do next. I got around that problem my making the system user the owner of /var/empty with exclusive rwx permissions and group/other with none. if youre not running the svc as SYSTEM just adjust the owner to your user. I've already `chmod 700 /var/empty`. Not sure what you mean about ownership of the service. I'm not sure this was the correct thing to do, but I tried setting CYGWIN sshd to log on as Administrator, set the password, and now it returns Error 1069: Logon failure (the password IS correct). Have you had problems with ssh when logging in at all? I can't even get the service to START. my sshd has for some reason been denying access to anyone that trys to login to my CYGWIN server with a permission/access denied message. nothing in sshd.log but event-log shows a badpw error (very weird). i know the password is correct bacause im using terminal services to login to the server right now... hope the first bit helps, sorry to bore you with the latter :) No problem. :-D I've been reading everything I can on the subject. There might be enough demand for a Cygwin book; I'd buy one in a heartbeat. With problems like this you get the complexity UNIX is known for, with NT's lack of decent error reporting. When you're DONE, of course, you get powerful UNIX tools, with Win2K's good points (good points? A free PC in every box of MS Outlook) :-) I'm still stuck, if anyone else has ideas. Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Cygwin [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 4:12 PM Subject: ssh service staring problem bad owner /var/empty but not fixed Hello, I understand the problem I am about to ask is not uncommon, and I have made considerable effort to look for the answers in the archive... On an up-to-date (today) Cygwin install, sshd refuses to start (the MS Management console gives a useless error). On other systems, I have installed Cygwin sshd and it worked fine (I have not done this recently though and I understand there have been changes to ssd of sorts). When I first attempted this install some weeks back, I followed the guide at http://tech.erdelynet.com/cygwin-sshd.html I didn't actually run the permissions script as the author had just days before, pulled down the script. I'm not sure if the other steps on this page complicate my problem, so I'll mention it. The first thing I check is /var/log/sshd.log, and it's bad owner or mode for /var/empty. OK, it's some sort of NT permissions issue. A Google search tells me /var/empty should be chmod 700 or 755 (it's 755). grep /etc/passwd ssh shows ssh account is 1000:513, sshd privsep, home of /var/empty and shell of /bin/false I've also tried chowning the directory as SYSTEM:SYSTEM (or 18:18). I did notice in the MMC Groups panel, there is no VISIBLE group for sshd, but there is a sshd user. My Google searches tell me there should be a group, so I attempt to add the group sshd and make sshd user a member. I get the error: while attempting to create the group sshd on computer QA2000TEST: The account already exists. I get this error if I attempt to create the group sshd with or with-out the member sshd. I've Reinstalled openssh, and even selected Unininstall followed by Install
RE: ssh service staring problem bad owner /var/empty but not fixed
Looks like our problems are somewhat related. I wonder if anyone else has ideas... -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 1:44 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hey, What i meant by shared-server is that more than one person (other than you) would be accessing the server. So if it is a shared environment you might want to tighten security. In general you should run things like crond,sshd etc. as the SYSTEM user as Administrator doesnt have the required run as service tokens and others needed for a run-as-user service unless youve added them in [domain|local] security policy(s) thingys in Administrative tools. I dont know whats going on. I just had to stop sshd so i could so i could get rid of an ssh process that wouldnt go away, went away when the service stopped but now i cant restart it. I get these errors in the eventlog... Event Type: Error Event Source: sshd Event Category: None Event ID: 0 Date: 09/10/2002 Time: 17:57:14 User: NT AUTHORITY\SYSTEM Computer: W3 Description: The description for Event ID ( 0 ) in Source ( sshd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. The following information is part of the event: sshd : Win32 Process Id = 0xCA8 : Cygwin Process Id = 0xCA8 : starting service `sshd' failed: execv: 1, Operation not permitted. YES! I get exactly this message in Event Viewer, except execv=255 error=255 Event Type: Error Event Source: sshd Event Category: None Event ID: 0 Date: 09/10/2002 Time: 17:57:13 User: NT AUTHORITY\SYSTEM Computer: W3 Description: The description for Event ID ( 0 ) in Source ( sshd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. The following information is part of the event: sshd : Win32 Process Id = 0x950 : Cygwin Process Id = 0x950 : starting service `l' failed: redirect_fd: open (1, /var/log/sshd.log): 22, Invalid argument. I don't get this one exactly. The second error I get is line-for-line identical with the first event, minus the bit about execv=255 (not a different error number... just not there at all). are you getting anything similar? Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 6:23 PM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 12:02 PM To: cygml Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hi, I had that when i first installed it... i take it the permissions on files like /etc/sshd* /etc/ssh_host* are exclusive to the SYSTEM account (if your running a shared-style server) and the service is running as SYSTEM. Let's see...: $ ls -l /etc/ssh* -rw-r--r--1 Administ None 1049 Sep 5 15:59 /etc/ssh_config -rw-r--r--1 Administ None 668 Sep 5 15:19 /etc/ssh_host_dsa_key -rw-r--r--1 Administ None 614 Sep 5 15:19 /etc/ssh_host_dsa_key.pub -rw-r--r--1 Administ None 539 Sep 5 15:19 /etc/ssh_host_key -rw-r--r--1 Administ None 343 Sep 5 15:19 /etc/ssh_host_key.pub -rw-r--r--1 Administ None 883 Sep 5 15:19 /etc/ssh_host_rsa_key -rw-r--r--1 Administ None 234 Sep 5 15:19 /etc/ssh_host_rsa_key.pub -rw-r--r--1 Administ None 2041 Sep 5 15:59 /etc/sshd_config Is Administrator here perfectly synonymous with SYSTEM? Also, I'm not sure what you mean by shared style server, how to verify if that is my case, or how this would affect things. The service in MMC shows it logs on as Local System Account, interact with desktop NOT checked. Should this instead be running as sshd user or Administrator? I personally prefer to get things running the right way and not blow holes through local security. That said, this is a test lab system and I'd go the hack way to Make It Work... if I knew what to do next. I got around that problem my making the system user the owner of /var/empty with exclusive rwx permissions and group/other with none. if youre not running the svc as SYSTEM just adjust the owner to your user. I've already `chmod 700 /var/empty`. Not sure what you mean about ownership of the service. I'm not sure this was the correct thing to do, but I tried setting CYGWIN sshd to log on as Administrator, set the password, and now it returns Error 1069: Logon failure (the password IS correct). Have you had problems with ssh when
RE: ssh service staring problem bad owner /var/empty but not fixed
I can say this works fine on one system, which I installed a while back. Then I got it working on a SECOND system, which worked fine UNTIL I updated Cygwin. Then it broke. I sent an email to this list but never got a reply. Then I tried a THIRD system, and even a fresh install did not work. That first system which still works, I refuse to update Cygwin until I understand what broke everything. I've come to the conclusion that something changed in the packages, but obviously it's working on SOME people's systems, right? I see a lot of related questions in the recent archives, and suggestions (which I followed). Then again, I missed seeing any replies that said thanks, that fixed it... so it's possible those suggestions did not work for them either. -Scott -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 2:08 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Its pretty funky that this has started happening OOTB (out of the blue). have you had a working sshd? ... i forget. have you installed new soft,libs recently... have you downloaded new net-release packages as well? All ive done is install mysql-3.23.52 on cygwin-1.3.12-2, cant see that making a difference. Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 7:03 PM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed Looks like our problems are somewhat related. I wonder if anyone else has ideas... -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 1:44 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hey, What i meant by shared-server is that more than one person (other than you) would be accessing the server. So if it is a shared environment you might want to tighten security. In general you should run things like crond,sshd etc. as the SYSTEM user as Administrator doesnt have the required run as service tokens and others needed for a run-as-user service unless youve added them in [domain|local] security policy(s) thingys in Administrative tools. I dont know whats going on. I just had to stop sshd so i could so i could get rid of an ssh process that wouldnt go away, went away when the service stopped but now i cant restart it. I get these errors in the eventlog... Event Type: Error Event Source: sshd Event Category: None Event ID: 0 Date: 09/10/2002 Time: 17:57:14 User: NT AUTHORITY\SYSTEM Computer: W3 Description: The description for Event ID ( 0 ) in Source ( sshd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. The following information is part of the event: sshd : Win32 Process Id = 0xCA8 : Cygwin Process Id = 0xCA8 : starting service `sshd' failed: execv: 1, Operation not permitted. YES! I get exactly this message in Event Viewer, except execv=255 error=255 Event Type: Error Event Source: sshd Event Category: None Event ID: 0 Date: 09/10/2002 Time: 17:57:13 User: NT AUTHORITY\SYSTEM Computer: W3 Description: The description for Event ID ( 0 ) in Source ( sshd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. The following information is part of the event: sshd : Win32 Process Id = 0x950 : Cygwin Process Id = 0x950 : starting service `l' failed: redirect_fd: open (1, /var/log/sshd.log): 22, Invalid argument. I don't get this one exactly. The second error I get is line-for-line identical with the first event, minus the bit about execv=255 (not a different error number... just not there at all). are you getting anything similar? Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 6:23 PM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 12:02 PM To: cygml Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hi, I had that when i first installed it... i take it the permissions on files like /etc/sshd* /etc/ssh_host* are exclusive to the SYSTEM account (if your running a shared-style server) and the service is running as SYSTEM. Let's see...: $ ls -l /etc/ssh* -rw-r--r--1 Administ None 1049 Sep 5 15:59 /etc/ssh_config -rw-r--r--1 Administ
RE: ssh service staring problem bad owner /var/empty but not fixed
Attempting to run the sshd server as Administrator was purely an act of desperation. All along until them I've left it at default Local System. On my sshd-working system, all of those files belong to None. On the other, broken-sshd system, all of those files belong to Administrator:None. HOWEVER if I do a `chmod SYSTEM /etc/ssh*`, the command does NOT change ownership. It just returns to the prompt w/o error (echo #? shows 0). I think this is where the problem may lie, but if the command wont change owner then I'm blocked. Any ideas? Thanks. Scott -Original Message- From: Marius Seritan [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 2:11 PM To: Scott Prive; Elfyn; cygml Subject: Re: ssh service staring problem bad owner /var/empty but not fixed I am not sure if I understand all the details of your setup but here are some comments. Unless you typed an user name and password in the sshd service setup box you are running sshd as SYSTEM. SYSTEM is totally different from Administrator, the 2 accounts different sids, different privileges, network access capabilities. You need to have /etc/ssh*, /var/empty and /var/log/sshd.log belong to SYSTEM (chown SYSTEM ...) I hope this helps. Marius - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 11:03 AM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed Looks like our problems are somewhat related. I wonder if anyone else has ideas... -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 1:44 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hey, What i meant by shared-server is that more than one person (other than you) would be accessing the server. So if it is a shared environment you might want to tighten security. In general you should run things like crond,sshd etc. as the SYSTEM user as Administrator doesnt have the required run as service tokens and others needed for a run-as-user service unless youve added them in [domain|local] security policy(s) thingys in Administrative tools. I dont know whats going on. I just had to stop sshd so i could so i could get rid of an ssh process that wouldnt go away, went away when the service stopped but now i cant restart it. I get these errors in the eventlog... Event Type: Error Event Source: sshd Event Category: None Event ID: 0 Date: 09/10/2002 Time: 17:57:14 User: NT AUTHORITY\SYSTEM Computer: W3 Description: The description for Event ID ( 0 ) in Source ( sshd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. The following information is part of the event: sshd : Win32 Process Id = 0xCA8 : Cygwin Process Id = 0xCA8 : starting service `sshd' failed: execv: 1, Operation not permitted. YES! I get exactly this message in Event Viewer, except execv=255 error=255 Event Type: Error Event Source: sshd Event Category: None Event ID: 0 Date: 09/10/2002 Time: 17:57:13 User: NT AUTHORITY\SYSTEM Computer: W3 Description: The description for Event ID ( 0 ) in Source ( sshd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. The following information is part of the event: sshd : Win32 Process Id = 0x950 : Cygwin Process Id = 0x950 : starting service `l' failed: redirect_fd: open (1, /var/log/sshd.log): 22, Invalid argument. I don't get this one exactly. The second error I get is line-for-line identical with the first event, minus the bit about execv=255 (not a different error number... just not there at all). are you getting anything similar? Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 6:23 PM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 12:02 PM To: cygml Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hi, I had that when i first installed it... i take it the permissions on files like /etc/sshd* /etc/ssh_host* are exclusive to the SYSTEM account (if your running a shared-style server) and the service is running as SYSTEM. Let's see...: $ ls -l /etc/ssh* -rw-r--r--1 Administ None 1049 Sep 5 15:59 /etc/ssh_config -rw-r--r--1 Administ None
RE: ALMOST RESOLVED: ssh service staring problem bad owner /var/empty but not fixed (now password sync issue)
OK, done. qacontrol is the system where sshd works... qa2000test is the system where sshd fails to start. I ran cygcheck and diffed my results: the broken system lacked the CYGWIN=tty sec variable, which I added in the WIN2K GUI, restarted all my shells and verified the variable was being used. NOW I could properly chown the files! Getting closer! :-) After verifying /var/log/sshd.log, /etc/ssh* and /var/empty/ were all owned by SYSTEM:SYSTEM. However the sshd service still will not start... but at least the log error is hinting at corrective action (a good thing for people like me :), and my /etc/ssh* files are too open. Not wanting to set blanket permissions on /etc/ssh*, I fixed the permissions one-at-a-time, and attempted to start sshd. I encountered a misleading error message: if /var/empty is chmod 777, you can get bogus log messages like permissions 0777 for /etc/ssh_host_key' are too open. I fixed that problem, but continued to get the error until I did chmod 755 on /var/empty/. It might be possible for more error checking here. Well, NOW I can start the server, I get NO error messages... but the Administrator password is rejected. Fine: it's not talking to NT's password management. used a local shell to reset the Administrator password. I realize this breaks password sync and I do want to fix it.. but at least I have a workaround. If anyone knows what's misconfigured by that description, suggestion would be most welcome! :-D Thanks for the cygcheck suggestion. Did you still want me to mail these to you (for your debugging?) -Scott -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 2:21 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hi, Can you do a cygcheck on all of your cygwin machines so we can compare what exactly has changed `cygcheck -s -s -r'... it has to be a change in package. ill go through latest changes to see what has be upgraded in packages released in the last couple of weeks. I got people pis*ed because of this and need to try and get it sorted as im sure you do... Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 7:13 PM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed I can say this works fine on one system, which I installed a while back. Then I got it working on a SECOND system, which worked fine UNTIL I updated Cygwin. Then it broke. I sent an email to this list but never got a reply. Then I tried a THIRD system, and even a fresh install did not work. That first system which still works, I refuse to update Cygwin until I understand what broke everything. I've come to the conclusion that something changed in the packages, but obviously it's working on SOME people's systems, right? I see a lot of related questions in the recent archives, and suggestions (which I followed). Then again, I missed seeing any replies that said thanks, that fixed it... so it's possible those suggestions did not work for them either. -Scott -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 2:08 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Its pretty funky that this has started happening OOTB (out of the blue). have you had a working sshd? ... i forget. have you installed new soft,libs recently... have you downloaded new net-release packages as well? All ive done is install mysql-3.23.52 on cygwin-1.3.12-2, cant see that making a difference. Elfyn - Original Message - From: Scott Prive [EMAIL PROTECTED] To: Elfyn [EMAIL PROTECTED]; cygml [EMAIL PROTECTED] Sent: Wednesday, October 09, 2002 7:03 PM Subject: RE: ssh service staring problem bad owner /var/empty but not fixed Looks like our problems are somewhat related. I wonder if anyone else has ideas... -Original Message- From: Elfyn [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 1:44 PM To: cygml; Scott Prive Subject: Re: ssh service staring problem bad owner /var/empty but not fixed Hey, What i meant by shared-server is that more than one person (other than you) would be accessing the server. So if it is a shared environment you might want to tighten security. In general you should run things like crond,sshd etc. as the SYSTEM user as Administrator doesnt have the required run as service tokens and others needed for a run-as-user service unless youve added them in [domain|local] security policy(s) thingys in Administrative tools. I dont know whats going on. I just had to stop sshd so i could so i could get rid
port of watch?
Is anyone aware of a port of watch to Cygwin? For the curious, watch is a standard (?) tool on Linux which re-executes a command at 2 second intervals (configurable). ex: watch ls -l This is more convenient than looping code on the shell. On Linux, this is part of procps: [root@redhat root]# rpm --whatprovides -q `which watch` procps-2.0.7-21 Thanks, Scott -- 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: port of watch?
Thanks -- good work Igor! I've made a sticky note to add procps to my next round of lab updates, and this script will hold me over until I do.. Cheers, Scott -Original Message- From: Igor Pechtchanski [mailto:[EMAIL PROTECTED]] Sent: Monday, October 07, 2002 1:08 PM To: Scott Prive Cc: [EMAIL PROTECTED] Subject: Re: port of watch? On Mon, 7 Oct 2002, Scott Prive wrote: Is anyone aware of a port of watch to Cygwin? For the curious, watch is a standard (?) tool on Linux which re-executes a command at 2 second intervals (configurable). ex: watch ls -l This is more convenient than looping code on the shell. On Linux, this is part of procps: [root@redhat root]# rpm --whatprovides -q `which watch` procps-2.0.7-21 Thanks, Scott Scott, There is a procps package available for Cygwin... If you don't want to install the whole package, use the attached script (which took about 15 minutes to write) for similar functionality (minus difference highlighting). Igor -- 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: dumb escaping question when using Cygwin + NT commands
-Original Message-
From: Randall R Schulz [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, September 18, 2002 6:30 PM
To: [EMAIL PROTECTED]
Subject: Re: dumb escaping question when using Cygwin + NT commands
Scott,
At 15:15 2002-09-18, Scott Prive wrote:
Hello,
I get this odd problem when calling NT commands from Cygwin. I am
single-quoting the data, but the way I'm doing things
(probably wrong...)
does not like passing $1 function arguments to NT commands.
If I hardcode
the arguments internally, everything works.
The two example functions below are intended to behave identical.
#!/bin sh
mount_drive () {
# Syntax: net 'use' '*' '\\redhat\foo' 'foo' '/user:foo'
net 'use' 'F:' '\\redhat\foo' 'foo' '/user:foo'
echo The command returned $?
return $?;
}
Note that the status ($?) you're returning from the
mount_drive shell
procedure is that of the echo command, not that printed
_by_ the echo
command.
The only arguments in this example for which quoting changes the net
argument passed to the underlying command is the one that
includes redhat
and the asterisk. The others contain no special characters requiring
quoting or escaping to inhibit special interpretation.
mount_drive2 () {
net '$1' '$2' '$3' '$4' '$5'
echo we saw in mount_drive2: '$1' '$2' '$3' '$4' '$5'
echo The command returned $?
return $?;
}
The same $? issue exists here, of course.
You need to be aware of the difference between 'single
quotes' and double
quotes. Variable expansion is inhibited in single-quoted
arguments, but
not in double-quoted ones. Furthermore, double quoted
arguments protect
single quotes, making the non-special. So you've probably
confused yourself
into thinking that in this example the net command saw the
arguments you
passed to the mount_drive2 procedure. It did not. It saw
arguments each
consisting of a dollar sign followed by a digit. Then you
echoed a single
argument composed of some fixed text, some single quote marks
and some
expanded positional parameters.
Doh!
Thanks. A good nights sleep and coffee got me thinking about this on the way to work,
and then I read your post.
I misled myself because the ECHO command worked. A debugging habit from Perl is I
would print out my variables. Since the echo worked, I never questioned what I was
doing with quotes.
I assumed quotes controlled how data gets sent to commands, but apparently that's an
oversimplification: quotes protect data being sent to a NEW PROCESS.. and builtins
like echo are NOT a new process (`type echo). This explains why the echo command
understood what the heck was inside '$2', but the echo command did not.
Of course you know this; I'm just filling in the blanks for the benefit of mailing
list and Google searches. For all of last night, I actually believed the problem was
due to mixing NT commands and Cygwin.
Thanks again.
#
mount_drive
mount_drive2 'use' 'G:' '\\redhat\foo' 'foo' '/user:foo'
# END SCRIPT
the output I get from mount_drive2 is standard usage info,
indicating I
passed arguments incorrectly. However the debug echo *looks* correct.
Someone please point out my mistake, else I'm doomed to some
ugly hackish
workarounds ;-)
Thanks,
Scott
Randall Schulz
Mountain View, CA USA
--
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/
dumb escaping question when using Cygwin + NT commands
Hello,
I get this odd problem when calling NT commands from Cygwin. I am single-quoting the
data, but the way I'm doing things (probably wrong...) does not like passing $1
function arguments to NT commands. If I hardcode the arguments internally, everything
works.
The two example functions below are intended to behave identical.
#!/bin sh
mount_drive () {
# Syntax: net 'use' '*' '\\redhat\foo' 'foo' '/user:foo'
net 'use' 'F:' '\\redhat\foo' 'foo' '/user:foo'
echo The command returned $?
return $?;
}
mount_drive2 () {
net '$1' '$2' '$3' '$4' '$5'
echo we saw in mount_drive2: '$1' '$2' '$3' '$4' '$5'
echo The command returned $?
return $?;
}
#
mount_drive
mount_drive2 'use' 'G:' '\\redhat\foo' 'foo' '/user:foo'
# END SCRIPT
the output I get from mount_drive2 is standard usage info, indicating I passed
arguments incorrectly. However the debug echo *looks* correct.
Someone please point out my mistake, else I'm doomed to some ugly hackish workarounds
;-)
Thanks,
Scott
--
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/
Better solution for calling 'net use' from Cygwin Perl?
Hello, I had a .sh script that called Microsoft's net use command to mount drives under Win2k, and this worked: net use 't:' '\\myserver\share' However I needed to do this in Cygwin Perl so I could leverage an existing set of Perl libraries I have. For the life of me I could correctly execute the net use command from Perl, because the characters would get escaped or not interpreted correctly. The Perl workaround a coworker devised was: my $cmd = net use 't:' '\\; $cmd .= \\storigen1u21\\sfstest'; print $cmd; system $cmd; and this DOES work. This would be the end of story, except we want to understand what caused the problem in the first place. I suspect it was the brain dead Microsoft command processor getting involved with this system call. Anyone care to elaborate? Thanks, Scott -- 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: F-Secure anti-virus causes system crash when scanning the C:\cygwin directory.
Brad, a workaround: Configure your antivirus program to exclude scanning the Cygwin directory. This is a very common question.. might be in FAQ?, I'm not sure. Outdated AV software will accuse Cygwin of being a virus... I'm not so sure I would want (or trust) antivirus software that *bluescreens*... especially on a production box if that is the case. Cheers, Scott -Original Message- From: Brad Morgan [mailto:[EMAIL PROTECTED]] Sent: Thursday, May 09, 2002 2:13 PM To: [EMAIL PROTECTED] Subject: RE: F-Secure anti-virus causes system crash when scanning the C:\cygwin directory. I wanted to see if this audience had any cygwin specific knowledge that might be helpful in either solving or working around this problem. -- 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: MS Logo compliance - w2k / xp
I'm being asked if Cygwin is or will be 'MS Logo' compliant.. [snip] - You already have some responses to this, but here's one, from a different slant: *Why* are you being asked this? (Don't answer me! :-) I don't want to know, since I can't help). Questions sometimes can be leading; this question may indicate a more subtle issue, which is your real concern. For example, this question could be from a manager who isn't very technical, but could have been exposed to some of Microsoft's anti-free software FUD in the mainstream press. So he/she has (unfounded) but legitimate concerns about using free software. If this were the case, I'd consult some advocacy how-to's. Of course this might not be the issue either. Good luck on this one. -Scott -- 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[2]: setup.exe and mirror list question
OK, thanks. I didn't know if there was a way to do a safe search and replace within the binary file. I'll recompile then. It's a little more work because the mirror is on Linux, and I've never investigated cross-compiling from Linux (and someone said you can't run Cygwin under WINE anymore). Anyhow, Thanks, you've been a big help! -Scott -Original Message- From: Pavel Tsekov [mailto:[EMAIL PROTECTED]] Sent: Friday, April 05, 2002 3:11 AM To: Scott Prive Cc: [EMAIL PROTECTED] Subject: Re[2]: setup.exe and mirror list question Hello Scott, Thursday, April 04, 2002, 6:18:45 PM, you wrote: SP Thanks Paul... that's exactly what I needed to know. SP A follow-up question, if I may: SP Does Cygwin or Linux have such a resource editing tool? Nope - it doesn't have a tool to edit the ready to use compiled resource script. Still as I mentioned in my previous mail you can get the source code for setup.exe and edit the resource script (a text file) manually than recompile. This soulution seems more scriptable to me. SP I imagine there is one, and I would need to learn it, but I don't know what it would be called. A shell tool would empower me to automate the resource editing after each new setup.exe is SP downloaded (yipee!). SP Thanks, SP Scott SP -Original Message- SP From: Pavel Tsekov [mailto:[EMAIL PROTECTED]] SP Sent: Wednesday, April 03, 2002 2:36 AM SP To: Scott Prive SP Cc: Cygwin SP Subject: Re: setup.exe and mirror list question SP Hello Scott, SP Tuesday, April 02, 2002, 6:44:47 PM, you wrote: SP Robert (or anyone), SP Regarding the setup.exe Choose a Download Site page: SP What I am doing right now is running an internal Cygwin mirror, updated every 24h via rsync. When a user runs setup.exe from my mirror, I'd like my internal mirror to be the ONLY available mirror SP displayed in the install screen. SP I checked the FAQ discussion lists and could not see how this would be done. Is this possible? SP The setup program fetches the mirror list from a hardcoded URL which SP is found in the resource section of the executable. So you can use a SP resource editor to change this URL to something apropriate for you... SP Another way to achieve this is to get the setup.exe sources and change SP the URL in the resource (.rc) script file and compile your own SP version. Still I think both choices will make your version of SP setup.exe unsupported by this list ... or maybe I'm wrong :)) -- 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.exe and mirror list question
Thanks Paul... that's exactly what I needed to know. A follow-up question, if I may: Does Cygwin or Linux have such a resource editing tool? I imagine there is one, and I would need to learn it, but I don't know what it would be called. A shell tool would empower me to automate the resource editing after each new setup.exe is downloaded (yipee!). Thanks, Scott -Original Message- From: Pavel Tsekov [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 03, 2002 2:36 AM To: Scott Prive Cc: Cygwin Subject: Re: setup.exe and mirror list question Hello Scott, Tuesday, April 02, 2002, 6:44:47 PM, you wrote: SP Robert (or anyone), SP Regarding the setup.exe Choose a Download Site page: SP What I am doing right now is running an internal Cygwin mirror, updated every 24h via rsync. When a user runs setup.exe from my mirror, I'd like my internal mirror to be the ONLY available mirror SP displayed in the install screen. SP I checked the FAQ discussion lists and could not see how this would be done. Is this possible? The setup program fetches the mirror list from a hardcoded URL which is found in the resource section of the executable. So you can use a resource editor to change this URL to something apropriate for you... Another way to achieve this is to get the setup.exe sources and change the URL in the resource (.rc) script file and compile your own version. Still I think both choices will make your version of setup.exe unsupported by this list ... or maybe I'm wrong :)) -- 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/
setup.exe and mirror list question
Robert (or anyone), Regarding the setup.exe Choose a Download Site page: What I am doing right now is running an internal Cygwin mirror, updated every 24h via rsync. When a user runs setup.exe from my mirror, I'd like my internal mirror to be the ONLY available mirror displayed in the install screen. I checked the FAQ discussion lists and could not see how this would be done. Is this possible? thanks, Scott Prive -- 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: trouble running commands non-interactively (e.g. via ssh, cron, at)
No idea since I don't administer NT users, but I assume you've tried to isolate this by scripting other executables. So is this a 1) system-wide NT / ssh issue, 2) a possible bug in your script, or 3) is something special about these particular .exe files on your server? If those NT exe's have proper $? return values on success/fail, I don't see how this could be any different than a cygwin .exe. -Scott -Original Message- From: Jonathan C. Detert [mailto:[EMAIL PROTECTED]] Sent: Thursday, February 14, 2002 11:00 AM To: [EMAIL PROTECTED] Subject: trouble running commands non-interactively (e.g. via ssh, cron, at) Hello, I've got DLL version 1.3.6 installed on NT 4.0 SP6. I'm trying to create/delete user accounts in the domain that the cygwin box is in. If I (interactively) login into the cygwin box via ssh, I can successfully run the NT Resource Kit addusers command and the Ms. Exchange admin.exe command to create/delete accounts. However, If I try to run either of those commands non-interactively via ssh, they don't work. The commands return with no effect. At first, I thought the problem was something peculiar about the addusers and admin commands. So, I wrote perl programs to manage the accounts. However, the same problem applies to the perl programs - they work fine when run from an interactive login-shell, but not when run non-interactively from ssh. The same problem occurrs when I try to schedule programs via cron or at - the scheduled programs fail (i don't even know if they run - there's no log anywhere that I can find; I just know they don't do what they were intended to do). Any ideas? AtDhVaAnNkCsE -- Happy Landings, Jon Detert Unix System Administrator, Milwaukee School of Engineering 1025 N. Broadway, Milwaukee, Wisconsin 53202 -- 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/
Did anyone ever have success porting sgtty.h?
All, In older postings, some folks had problems due to sgtty.h not being = supported; instead they should use termios. Are there any utility functions for converting a sgtty dependent = function over to termios? From the archives, this appears to be a common = question. Assuming someone provides text for the FAQ maintainer, could = we get this question added to the FAQ? Or better yet, if someone who has = had success converting, went and documented their effort. Reading between the lines on the prior postings, I gather there are = reasons no one wants to port it to Cygwin -- reasons other than the = header being obsolete. If there is a bloody history, I'm curious. :-) Scott Prive PS - If anyone is curious, the application I am trying to compile is a C = Tcl app called Brewers Little Helper (formerly BrewNIX). The author = is, unfortunately, unreachable and the code is unmaintained (a pity = since the brew mailing lists pine for a OS portable brew calculator). = -- 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/
