RE: Concept question.
How big are the projects and what are the build/release process for the shared files? If projects are relatively small and the sharing is fairly add-hoc then I'd go for a single project. However if you've got a separate release process for the shared files then you could go for separate repositories. Rather than checking out and building the shared files as part of the project, pull in a jar (or lib or whatever) from a release directory somewhere. In fact, I'd be tempted to have a separate build and release of the shared files anyway so you can try and define some clear interfaces between the projects and not end up with a tangle. Matt -Original Message- From: Delos Nash [mailto:[EMAIL PROTECTED]] Sent: 21 June 2001 22:47 To: '[EMAIL PROTECTED]' Subject: Concept question. I have 4 programmers working on 2 different projects. All of them are using cvs off the server. I set up cvs originally as 2 different projects that each team could checkout. Now they tell me that they have Shared files that each project needs. How would you set it up? I thought I could set it up as one project with three different modules so they could checkout what they needed. Or Setup a third project with the shared files or is there another idea? -- If you have received this e-mail in error or wish to read our e-mail disclaimer statement and monitoring policy, please refer to http://www.drkw.com/disc/email/ or contact the sender. -- ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Concept question
The lead programmer today told me that he thinks he will have 25 different projects that will need hundreds of shared files between them. So my first ideals have gone out the window and I'm going to have to re-think what I'm going to do. Del Nash From: Helliwell, Matthew [EMAIL PROTECTED] To: '[EMAIL PROTECTED]' [EMAIL PROTECTED] Subject: RE: Concept question. Date: Fri, 22 Jun 2001 08:45:16 +0100 How big are the projects and what are the build/release process for the shared files? If projects are relatively small and the sharing is fairly add-hoc then I'd go for a single project. However if you've got a separate release process for the shared files then you could go for separate repositories. Rather than checking out and building the shared files as part of the project, pull in a jar (or lib or whatever) from a release directory somewhere. In fact, I'd be tempted to have a separate build and release of the shared files anyway so you can try and define some clear interfaces between the projects and not end up with a tangle. Matt -Original Message- From: Delos Nash [mailto:[EMAIL PROTECTED]] Sent: 21 June 2001 22:47 To: '[EMAIL PROTECTED]' Subject: Concept question. I have 4 programmers working on 2 different projects. All of them are using cvs off the server. I set up cvs originally as 2 different projects that each team could checkout. Now they tell me that they have Shared files that each project needs. How would you set it up? I thought I could set it up as one project with three different modules so they could checkout what they needed. Or Setup a third project with the shared files or is there another idea? -- If you have received this e-mail in error or wish to read our e-mail disclaimer statement and monitoring policy, please refer to http://www.drkw.com/disc/email/ or contact the sender. -- ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS file locking question
I found this email address and wanted to see if you could help me out on a question about how CVS works. First a little background: We are about to start some development with WebObjects v5.0 under Mac OS X which embeds a UI layer to CVS into its own IDE. We normally use NT servers here so the introduction of Macs has caused us to scramble to fit the different boxes/networking together. We are thinking of having the main repository on an NT box and then check out to individual Powerbooks running Mac OS X (ie. BSD Unix under the hood) and then commit changes back into the NT-based repository. In order to gain access to the NT files system we are using a product called Sharity that allows us to map NT disk volumes so that the Macs can get to them. The one down side is that Sharity does not support file locking at all (read or write file locks are ignored by it entirely). Now to the question at hand: Does CVS at any time rely on read or write locks on files in the repository to run properly? If it uses some other method (IE. Existent of lock files rather than actual locks on files) then we might be able to use this combination but if not then maybe we'll want to use a Mac OS X server to hold the repository so that file locks work as CVS would need them. Thanks in advance for any info you might have on this! Thanks, Guy Gardner Chief RD Scientist Dynamic Healthcare Technologies 51 Sawyer Rd. Waltham, MA 02453 781.642.6200x3246 http://www.dht.com ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: CVS file locking question
Thanks, Guy -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 21, 2001 2:49 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: CVS file locking question [ On Thursday, June 21, 2001 at 13:58:26 (-0400), Larry Jones wrote: ] Subject: Re: CVS file locking question Using a network filesystem to share a repository is a recipe for disaster; we've had many reports of corrupted repository files due to network file system bugs. I strongly advise you to set up some form of client/server CVS instead. You can find a version of CVS that will run as a server on NT at http://www.cvsnt.com/, but you'd likely have fewer problems running the server on a Unix platform. I often share my working directories over NFS, but even then I never do any CVS operations on the client Sharing of working directories works particularly well if either all build machines are the same architecture, or you've got an architecture- independent build system (eg. the native *BSD build -- I'm building NetBSD/sparc from an NFS-mounted working directory right now, and I built the i386 variant from the same directory at the same time (past tense because my i386 machine is faster than the sparc, particularly since it didn't have to pull the sources over NFS over my 10Mbps ethernet :-)). -- Greg A. Woods +1 416 218-0098 VE3TCP [EMAIL PROTECTED] [EMAIL PROTECTED] Planix, Inc. [EMAIL PROTECTED]; Secrets of the Weird [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Concept question.
I have 4 programmers working on 2 different projects. All of them are using cvs off the server. I set up cvs originally as 2 different projects that each team could checkout. Now they tell me that they have Shared files that each project needs. How would you set it up? I thought I could set it up as one project with three different modules so they could checkout what they needed. Or Setup a third project with the shared files or is there another idea? Del Nash ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
AW: CVS file locking question
Using a network filesystem to share a repository is a recipe for disaster; we've had many reports of corrupted repository files due to network file system bugs. I strongly advise you to set up some form of client/server CVS instead. You can find a version of CVS that will run as a server on NT at http://www.cvsnt.com/, but you'd likely have fewer problems running the server on a Unix platform. Is the same also true for sharing a sandbox? I plan to do the following: I have a notebook and a pc, both running Win 2000. I mainly work on the pc, but when I'm abroad, I want to work on the notebook. So I want to keep the sandbox on the notebook and share it (as a Windows share) to the pc. I will never work on both computers at the same time, so there should not be any issues with concurrent accesses. Or should I rather create two sandboxes and use commit/update to synchronize them? Jens ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS file locking question
On Thu, Jun 21, 2001 at 11:25:19PM +0200, Jens Vonderheide wrote: sandbox on the notebook and share it (as a Windows share) to the pc. I will never work on both computers at the same time, so there should not be any So you say now. But at some point, you will probably want to do that. issues with concurrent accesses. Or should I rather create two sandboxes and use commit/update to synchronize them? commit/update. If you have to go onto a branch to do that, fine. But commit/update always. Commit early, commit often. mrc -- Mike Castle [EMAIL PROTECTED] www.netcom.com/~dalgoda/ We are all of us living in the shadow of Manhattan. -- Watchmen fatal (You are in a maze of twisty compiler features, all different); -- gcc ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS file locking question
On Thu, Jun 21, 2001 at 11:25:19PM +0200, Jens Vonderheide wrote: [...] I want to keep [a] sandbox on the notebook and share it (as a Windows share) to the pc. I will never work on both computers at the same time, so there should not be any issues with concurrent accesses. Or should I rather create two sandboxes and use commit/update to synchronize them? Creating two sandboxes would be much better. I don't think it's nearly as likely that your sandbox will be corrupted due to network bugs as that such a thing would happen to a shared repo ... but it's a lot *more* likely that it'll be corrupted due to pilot error -- copying the out-of-date version over top of the current one instead of vice versa, for example, or forgetting to sync them up and finding out too late that you've made major changes in both. This is one of the problems CVS was invented to solve (though in a slightly different context). Considering how good it is at keeping such things straight, and how poor humans are at it, why *not* let CVS do the bookkeeping for you? (CVS can't prevent the forgetting-to-sync problem, of course, but it sure makes recovering easier -- with CVS, it's just a standard merge). -- | | /\ |-_|/ Eric Siegerman, Toronto, Ont.[EMAIL PROTECTED] | | / With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. - RFC 1925 (quoting an unnamed source) ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: CVS file locking question
Guy Gardner said: We are thinking of having the main repository on an NT box and then check out to individual Powerbooks running Mac OS X (ie. BSD Unix under the hood) and then commit changes back into the NT-based repository. In order to gain access to the NT files system we are using a product called Sharity that allows us to map NT disk volumes so that the Macs can get to them. The one down side is that Sharity does not support file locking at all (read or write file locks are ignored by it entirely). I'd like to second Larry's comment that using CVS over a file share is a bad idea. The CVS client/server model is one of the advantages of CVS and can greatly reduce network traffic. Presumably, you bought those PowerBooks not just 'cause they looked cool, but also 'cause they're portable. You'll notice the difference between client/server and file shares in a hurry over a modem! If you're taking the plunge to Mac OS X anyway, why not use Mac OS X as the CVS server? We're a primarily Windows shop here, but we installed OS X on an old Beige G3 to serve a CVS repository. It does just fine, but if you're worried about performance, make sure it has plenty of RAM and a fast hard-drive; the CPU is not the bottleneck... There is an server for CVS that runs on NT, but I've never tried it. I do notice that WinCVS has a special option to use when accessing it. I'm not sure if this is an optional enhancement, or if it's required, but it does make me worry about compatibility. Cheers, -dete ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Question?
Hi, my name is MaX. I have a question. Is it possible to have two instances of WINCVS that permanently map, on two different repositories? If yes, how can I make it? Besides, into WinCVS what does it serve 'save as' in the menu FILE? Thanks in advance MF Entra in www.omnitel.it . Ti aspetta un mondo di servizi on line ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Taginfo question
Must be a pretty fundamental change to get the result Mark wants. Is there any documentation for this patch? I'd like to apply it but I'd rather appreciate what it does before applying it. Andy -Original Message- From: Derek R. Price [mailto:[EMAIL PROTECTED]] There's still a patch up at http://alumni.engin.umich.edu/~oberon/ccvs.tmpltfilter.final.diff that provides this functionality in addition to attempting to standardize the *info interfaces. It's fairly complete and works well, but I wanted some more opinions before updating the main tree. I have since reached the conclusion that the better direction is the XML pipe interface I brought up recently on bug-cvs but I haven't received any comments. Some of the XML pipe code has been implemented but not nearly enough to make it generally useful yet. Derek NOTICE AND DISCLAIMER: This email (including attachments) is confidential. If you have received this email in error please notify the sender immediately and delete this email from your system without copying or disseminating it or placing any reliance upon its contents. We cannot accept liability for any breaches of confidence arising through use of email. Any opinions expressed in this email (including attachments) are those of the author and do not necessarily reflect our opinions. We will not accept responsibility for any commitments made by our employees outside the scope of our business. We do not warrant the accuracy or completeness of such information. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Taginfo question
Andy Baker wrote: Must be a pretty fundamental change to get the result Mark wants. Is there any documentation for this patch? I'd like to apply it but I'd rather appreciate what it does before applying it. First, I just noticed that I posted the link to the wrong patch. The proper link was: http://alumni.engin.umich.edu/~oberon/ccvs.newfmtstrings.1-11.diff The first link I posted was to another patch that provides functionality on top of this one and doesn't do everything it should, so I'd leave it alone unless you really want the additional functionality as it stands or feel like dooing some hacking. Anyhow, there isn't much external documentation. You can review my previous posts - here's the results of searches for ccvs.newfmtstrings.1-11.diff on the bug-cvs and info-cvs archive on mail-archive.com: http://www.mail-archive.com/bug-cvs@gnu.org/msg00598.html http://www.mail-archive.com/info-cvs@gnu.org/msg07456.html http://www.mail-archive.com/info-cvs@gnu.org/msg06744.html http://www.mail-archive.com/info-cvs@gnu.org/msg05603.html Use those if you want abstract descriptions. Be sure to follow the threads. The best thing to do is probably review the portions of the patch which affect doc/cvs.texinfo and maybe the individual ChangeLogs in the patch file itself if you want itemized functionality descriptions. Hope that's useful. Derek -- Derek Price CVS Solutions Architect ( http://CVSHome.org ) mailto:[EMAIL PROTECTED] CollabNet ( http://collab.net ) -- I have sworn upon the alter of God eternal hostility against every form of tyranny over the mind of man. - Thomas Jefferson ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Taginfo question
Thanks Derek. I'll give that a go Andy -Original Message- From: Derek R. Price [mailto:[EMAIL PROTECTED]] http://alumni.engin.umich.edu/~oberon/ccvs.newfmtstrings.1-11.diff NOTICE AND DISCLAIMER: This email (including attachments) is confidential. If you have received this email in error please notify the sender immediately and delete this email from your system without copying or disseminating it or placing any reliance upon its contents. We cannot accept liability for any breaches of confidence arising through use of email. Any opinions expressed in this email (including attachments) are those of the author and do not necessarily reflect our opinions. We will not accept responsibility for any commitments made by our employees outside the scope of our business. We do not warrant the accuracy or completeness of such information. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Taginfo question
Ah. I see (next time I'll read the posting properly!). Good 'un that. Nothing obvious springs to mind. -Original Message- From: Mark [mailto:[EMAIL PROTECTED]] Since taginfo is in a sense, a pre-op trigger, the tag won't be placed in the RCSfile until the taginfo script successfully exists. But the rlog command will, of course, work in helping to identify existing tags. NOTICE AND DISCLAIMER: This email (including attachments) is confidential. If you have received this email in error please notify the sender immediately and delete this email from your system without copying or disseminating it or placing any reliance upon its contents. We cannot accept liability for any breaches of confidence arising through use of email. Any opinions expressed in this email (including attachments) are those of the author and do not necessarily reflect our opinions. We will not accept responsibility for any commitments made by our employees outside the scope of our business. We do not warrant the accuracy or completeness of such information. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Taginfo question
Is there any way to determine, in a taginfo script, whether or not a tag being created/applied is a branch tag or not? Mark __ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail - only $35 a year! http://personal.mail.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Windows-Linux question.
My experience is that cvs convert file to the client OS format. I work on a project on sourceforge and when I checkout files on windows, they end with CR+LF. When other members of the team check them out on Linux, the files end with LF only. Though, I don't know if that behavior apply when the server is running on Windows. -Message d'origine- De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]De la part de Mikael Aronsson Envoyé : mercredi 13 juin 2001 13:03 À : [EMAIL PROTECTED] Objet : Windows-Linux question. Hi ! I have cvs running on a Windows computer, but I need to compile the files on a Linux box, when I checkout the files on Windows all lines end with CR+LF (normal MS format), is there a way to tell cvs to checkout the files as unix format (LF only) ? Mikael ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Windows-Linux question.
Hi ! I have cvs running on a Windows computer, but I need to compile the files on a Linux box, when I checkout the files on Windows all lines end with CR+LF (normal MS format), is there a way to tell cvs to checkout the files as unix format (LF only) ? Mikael ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Windows-Linux question.
If you're compiling the files on a Linux box, why aren't you doing the checkout on the Linux box? -Original Message- From: Mikael Aronsson [mailto:[EMAIL PROTECTED]] Sent: Wednesday, June 13, 2001 4:04 AM To: [EMAIL PROTECTED] Subject: Windows-Linux question. Hi ! I have cvs running on a Windows computer, but I need to compile the files on a Linux box, when I checkout the files on Windows all lines end with CR+LF (normal MS format), is there a way to tell cvs to checkout the files as unix format (LF only) ? Mikael ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Windows-Linux question.
You could try installing cygwin on your Windows box, choosing the Unix end of line option when you install and then checking out from a bash shell. I get Unix-style end of lines on my PC box when I do that. Matt -Original Message- From: Kostur, Andre [mailto:[EMAIL PROTECTED]] Sent: 13 June 2001 15:50 To: 'Mikael Aronsson'; [EMAIL PROTECTED] Subject: RE: Windows-Linux question. If you're compiling the files on a Linux box, why aren't you doing the checkout on the Linux box? -Original Message- From: Mikael Aronsson [mailto:[EMAIL PROTECTED]] Sent: Wednesday, June 13, 2001 4:04 AM To: [EMAIL PROTECTED] Subject: Windows-Linux question. Hi ! I have cvs running on a Windows computer, but I need to compile the files on a Linux box, when I checkout the files on Windows all lines end with CR+LF (normal MS format), is there a way to tell cvs to checkout the files as unix format (LF only) ? Mikael ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs -- If you have received this e-mail in error or wish to read our e-mail disclaimer statement and monitoring policy, please refer to http://www.drkw.com/disc/email/ or contact the sender. -- ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Windows-Linux question.
My ADSL internet connection does not work with Linux (yet) Mikael - Original Message - From: Kostur, Andre [EMAIL PROTECTED] To: 'Mikael Aronsson' [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Wednesday, June 13, 2001 3:49 PM Subject: RE: Windows-Linux question. If you're compiling the files on a Linux box, why aren't you doing the checkout on the Linux box? -Original Message- From: Mikael Aronsson [mailto:[EMAIL PROTECTED]] Sent: Wednesday, June 13, 2001 4:04 AM To: [EMAIL PROTECTED] Subject: Windows-Linux question. Hi ! I have cvs running on a Windows computer, but I need to compile the files on a Linux box, when I checkout the files on Windows all lines end with CR+LF (normal MS format), is there a way to tell cvs to checkout the files as unix format (LF only) ? Mikael ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Windows-Linux question.
Mikael Aronsson wrote: My ADSL internet connection does not work with Linux (yet) If possible, you want to get yourself one of these: http://www.ars-technica.com/reviews/01q2/netgear/rt314-1.html ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Stupid newbie question
Matt Keyes wrote: Hey all... If I'm using Debian Linux, should I download the .gz file or the .rpm?I thought Debian didn't use rpm, rather apt, but I'm not sure... You'll want the .gz, though if you want to figure out how to build a debian package and submit the conf file, we'd welcome the patch. Derek -- Derek Price CVS Solutions Architect ( http://CVSHome.org ) mailto:[EMAIL PROTECTED] CollabNet ( http://collab.net ) -- He who laughs last thinks slowest. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: can't commit! permissions question
Schwenk, Jeanie wrote:
I've included two examples below complete with error messages. The first is
me logged into cvs as user cvsadmin in the cvsgrp group. There are two
modules: rat and DEV. The second example is me logged in as user jschwenk
in the users group. Unfortunately, I did not set up the initial repository
nor did I install cvs. Are the files supposed to be read-only?
Yes. You need write permissions in the directory or in the LockDir (read about
the 'config' file in the CVS manual) even to check out files.
It's possible that the user you have CVS running as isn't a member of the
'user' group it appears you $CVSROOT is owned by. Either change the
owner/group of the files and directories under $CVSROOT or change the
user/group that the cvs server runs as. As Larry mentioned, sometimes setting
the setgid bit ('man chmod') helps too.
Derek
--
Derek Price CVS Solutions Architect ( http://CVSHome.org )
mailto:[EMAIL PROTECTED] CollabNet ( http://collab.net )
--
If nature has made any one thing less susceptible than all others of exclusive
property, it is the action of the thinking power called an idea, which an
individual may exclusively possess as long as he keeps it to himself; but the
moment it is divulged, it forces itself into the possession of everyone, and
the receiver cannot dispossess himself of it. Its peculiar character, too, is
that no one possesses the less, because every other possesses the whole of it.
He who receives an idea from me, receives instruction himself without lessening
mine; as he who lights his taper at mine, receives light without darkening me.
That ideas should freely spread from one to another over the globe, for the
moral and mutual instruction of man, and improvement of his condition, seems to
have been peculiarly and benevolently designed by nature, when she made them,
like fire, expansible over all space, without lessening their density at any
point, and like the air in which we breathe, move, and have our physical being,
incapable of confinement or exclusive appropriation. Inventions then cannot, in
nature, be a subject of property.
- Thomas Jefferson
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
Re: can't commit! permissions question
Schwenk, Jeanie writes: Unfortunately, I did not set up the initial repository nor did I install cvs. Are the files supposed to be read-only? Yes. cvs -z9 checkout -P rat (in directory C:\) cvs checkout: in directory rat: cvs checkout: cannot open CVS/Entries for reading: No such file or directory That looks like the directory you're trying to checkout into (rat) already exists -- CVS really wants to create it. cvs server: failed to create lock directory for `/export/cvsroot/rat/scripts' (/export/cvsroot/rat/scripts/#cvs.lock): Permission denied That means that you don't have write permission in the /export/cvsroot/rat/scripts directory. cvs -z9 checkout -P DEV (in directory C:\) cvs checkout: in directory dev: cvs checkout: cannot open CVS/Entries for reading: No such file or directory Same as above. cvs server: failed to create lock directory for `/export/cvsroot/systema/environments/dev' (/export/cvsroot/systema/environments/dev/#cvs.lock): Permission denied Ditto. cvs server: failed to create lock directory for `/export/cvsroot/systema/environments/dev' (/export/cvsroot/systema/environments/dev/#cvs.lock): Permission denied Ditto. -Larry Jones I'm getting disillusioned with these New Years. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
cvs update question
I searched the CVS book I have, and have not been able to find what the P means beside a file after I do a cvs update. Anyone know? Thanks, Vicki Nelson Sr. Software Engineer Alldata, LLC [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: cvs update question
Hi Vicki, This is explained in page 110 of the Cederquist manual which essentially says that, P file : Like 'U' but the CVS server sends a patch instead of an entire file. These two things (U and P) accomplish the same thing. Rgs Sriram --- Nelson, Vicki [EMAIL PROTECTED] wrote: I searched the CVS book I have, and have not been able to find what the P means beside a file after I do a cvs update. Anyone know? Thanks, Vicki Nelson Sr. Software Engineer Alldata, LLC [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs __ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail - only $35 a year! http://personal.mail.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: cvs update question
That file has been updated by Patch (and not a full file-copy) See Appendix A.16.2 in the Cederqvist manual. -Original Message- From: Nelson, Vicki [mailto:[EMAIL PROTECTED]] Sent: Friday, June 8, 2001 4:51 PM To: '[EMAIL PROTECTED]' Subject: cvs update question I searched the CVS book I have, and have not been able to find what the P means beside a file after I do a cvs update. Anyone know? Thanks, Vicki Nelson Sr. Software Engineer Alldata, LLC [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
can't commit! permissions question
Who is supposed to be the owner of the files that are in the respository? I can update and checkout but not commit. Obviously the permissions are wrong somewhere. I've been looking in the OpenSourceDevelopment with CVS by Fogel and in the Cederqvist but I must conclude I've just not run across the needed info or it's not there. The accounts on the unix box are cvsadmin,sc, and jschwenk. The files in the repository are owned by sc but I'm logged in as cvsadmin. I imported a new directory of files as user jschwenk. I just checked in the repository and those files are owned by jschwenk. Does that mean that no one but jschwenk can commit these files? How do other individuals ever checkout/commit then? How can permissions be controlled? Does one need to give files a different user before checking in? Here's the passwd file. anonymous:!: cvsadmin:6Ji: dkennedy:Jul:sc pkuah:Jul:sc jschwenk:Jul:cvsadmin As cvsadmin, I can't commit the changes that were sent to me from the contractors because I don't have proper permissions. If any of this is unclear, please email me directly cause I need help! Jeanie ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: can't commit! permissions question
Schwenk, Jeanie writes: Who is supposed to be the owner of the files that are in the respository? I can update and checkout but not commit. Exactly what error messageare you getting? In general, you need read access to the files and write access to the directory. Files are owned by the last user to update them, directories are owned by the user who added them. If you're on a SysV-like system (most notably Linux), you may find it useful to set the SGID bit on all the repository directories (chmod g+s) -- that will cause newly-created directories to inherit their groupid from the parent directory instead of using the groupid of the creator. -Larry Jones Hmm... That might not be politic. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: can't commit! permissions question
I've included two examples below complete with error messages. The first is me logged into cvs as user cvsadmin in the cvsgrp group. There are two modules: rat and DEV. The second example is me logged in as user jschwenk in the users group. Unfortunately, I did not set up the initial repository nor did I install cvs. Are the files supposed to be read-only? Example 1: Here's the top level directory of the DEV module drwxrwxr-x 6 sc users 1024 May 25 13:53 environments drwxrwxr-x 11 sc users 1024 May 25 13:57 source all the files are read-only (just listed a few) -r--r--r-- 1 sc users19489 May 25 12:53 IDTGUI_Bundle_de_DE.properties,v -r--r--r-- 1 sc users18044 May 25 12:53 IDTGUI_Bundle_en_US.properties,v -r--r--r-- 1 sc users 1201 May 25 12:53 IDTGUI_CustomisedBundle_de_DE.properties,v Error while accessing C:\Macros (error 2) CVSROOT: :pserver:cvsadmin@pilot:/export/cvsroot (password authentication) TCL is available, shell is enabled : help (select and press enter) cvs -z9 login (Logging in to cvsadmin@pilot) *CVS exited normally with code 0* cvs -z9 checkout -P rat (in directory C:\) cvs checkout: in directory rat: cvs checkout: cannot open CVS/Entries for reading: No such file or directory cvs server: Updating rat U rat/RatComm.java cvs checkout: cannot open CVS/Entries.Log: No such file or directory U rat/RatGuiPilot.java cvs checkout: cannot open CVS/Entries.Log: No such file or directory cvs server: Updating rat/scripts cvs server: failed to create lock directory for `/export/cvsroot/rat/scripts' (/export/cvsroot/rat/scripts/#cvs.lock): Permission denied cvs server: failed to obtain dir lock in repository `/export/cvsroot/rat/scripts' cvs [server aborted]: read lock failed - giving up *CVS exited normally with code 1* cvs -z9 checkout -P DEV (in directory C:\) cvs checkout: in directory dev: cvs checkout: cannot open CVS/Entries for reading: No such file or directory cvs server: Updating dev cvs server: failed to create lock directory for `/export/cvsroot/systema/environments/dev' (/export/cvsroot/systema/environments/dev/#cvs.lock): Permission denied cvs server: failed to obtain dir lock in repository `/export/cvsroot/systema/environments/dev' cvs [server aborted]: read lock failed - giving up *CVS exited normally with code 1* Example 2: cvs -z9 checkout -P DEV (in directory C:\) cvs server: Updating dev cvs server: failed to create lock directory for `/export/cvsroot/systema/environments/dev' (/export/cvsroot/systema/environments/dev/#cvs.lock): Permission denied cvs server: failed to obtain dir lock in repository `/export/cvsroot/systema/environments/dev' cvs [server aborted]: read lock failed - giving up *CVS exited normally with code 1* -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Thursday, June 07, 2001 11:56 AM To: Schwenk, Jeanie Cc: [EMAIL PROTECTED] Subject: Re: can't commit! permissions question Schwenk, Jeanie writes: Who is supposed to be the owner of the files that are in the respository? I can update and checkout but not commit. Exactly what error messageare you getting? In general, you need read access to the files and write access to the directory. Files are owned by the last user to update them, directories are owned by the user who added them. If you're on a SysV-like system (most notably Linux), you may find it useful to set the SGID bit on all the repository directories (chmod g+s) -- that will cause newly-created directories to inherit their groupid from the parent directory instead of using the groupid of the creator. -Larry Jones Hmm... That might not be politic. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Question regarding CVS
Hello everybody, My name is Tim Wensink and I'm a Uniface developer from the Netherlands. We're currently doing a project in which we want to use CVS for version control, but I'm having a small problem. If I use Wincvs everything is OK, I can checkout/in etc. If I try to use CVS from the dos-prompt, i get a *panic* administration files missing error. Last week, CVS worked perfectly from the prompt, so I don't think it's a bug in the software. Does anyone have an idea what's wrong? Thanks in advance, Tim WensinkGet Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Question about CVS
Cheng Ie-Sun [EMAIL PROTECTED] writes: Got 2 questions regarding CVS 1. Is it possible to limit the number of revisions that CVS creates? ie. 5 revisions, 1.1x, 1.2x, 1.3x, 1.4x and 1.5x 2. Is it possible to remove the old revisions ie. remove revisions 1 tru 49 and just leave revision 50 and 51 from a 51 revision file Appreciate your help. Thanks 1. Not unless you stop checking in changes. As noted just about any time anyone asks anything about revisions on this list: Revision numbers are the magic smoke that make CVS work, mess with them and CVS will no longer do its work reliably. Do not confuse revision numbers and releases, more details on this issue at URL:http://cvshome.org/docs/manual/cvs_4.html#SEC44. 2. This question is mostly just a matter of saving disk space and if you really are that bad off on disk space, go ahead using 'cvs admin -o' as described in URL:http://cvshome.org/docs/manual/cvs_16.html#SEC119. Note that this is a really dangerous operation and really only worth doing if you are handling large binary files (these tend to be real space hogs). Good luck /Torbjörn ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Question regarding CVS
Hi, If I use Wincvs everything is OK, I can checkout/in etc. If I try to use CVS from the dos-prompt, i get a *panic* administration files missing error. Last week, CVS worked perfectly from the prompt, so I don't think it's a bug in the software. Does anyone have an idea what's wrong? Can you show the complete error message here... It seemed to me someone has deleted some of the administration files, what about your backup... have checked it which files are backup'ed from CVSROOT directory last week and which this week? Kind regards. -- Dipl.-Ing. Karl Heinz Marbaise | Phone: +49 (241) 4 13 26 - 48 QIS Systemhaus GmbH Aachen | Fax : +49 (241) 4 13 26 - 40 Juelicher Strasse 338 | Internet: http://www.qis-systemhaus.de 52070 Aachen | e-mail : [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Question regarding CVS
HI there again, My name is Tim Wensink and I'm a Uniface developer from the Netherlands. We're currently doing a project in which we want to use CVS for version control, but I'm having a small problem. If I use Wincvs everything is OK, I can checkout/in etc. If I try to use CVS from the dos-prompt, i get a *panic* administration files missing error. Last week, CVS worked perfectly from the prompt, so I don't think it's a bug in the software. Does anyone have an idea what's wrong? It could also be possible that you have a directory which is called CVS you should check your working directory on that or you have deleted files in CVS directory (work space). Kind Regards. -- Dipl.-Ing. Karl Heinz Marbaise | Phone: +49 (241) 4 13 26 - 48 QIS Systemhaus GmbH Aachen | Fax : +49 (241) 4 13 26 - 40 Juelicher Strasse 338 | Internet: http://www.qis-systemhaus.de 52070 Aachen | e-mail : [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Question about CVS
Got 2 questions regarding CVS 1. Is it possible to limit the number of revisions that CVS creates? ie. 5 revisions, 1.1x, 1.2x, 1.3x, 1.4x and 1.5x 2. Is it possible to remove the old revisions ie. remove revisions 1 tru 49 and just leave revision 50 and 51 from a 51 revision file Appreciate your help. Thanks
CVS - SSH question
Greetings, My question is how to get WinCVS to work with SSH without having to enter the password in the SSH window with every command that is executed. Thanks, Bob ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS - SSH question
On Fri, Jun 01, 2001 at 10:17:21AM -0400, Robert Briggs wrote: Greetings, My question is how to get WinCVS to work with SSH without having to enter the password in the SSH window with every command that is executed. This is more of an ssh question really, and might be better answered in an ssh specific forum (say, the newsgroup comp.security.ssh). But, one way is to use ssh-keygen to create a public/private key pair, and put the public key on the remote side (don't use a passphrase, of course). But this doesn't realy offer a lot of security. Anyone with physical access can then take your private key and use it wherever they like. I'm not sure how useful something like ssh-agent is in this case, I have no experience with it. But maybe that's what you need. mrc -- Mike Castle [EMAIL PROTECTED] www.netcom.com/~dalgoda/ We are all of us living in the shadow of Manhattan. -- Watchmen fatal (You are in a maze of twisty compiler features, all different); -- gcc ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS - SSH question
My question is how to get WinCVS to work with SSH without having to enter the password in the SSH window with every command that is executed. Following are instructions I wrote some time ago to get WinCVS and passwordless ssh working. You can find ssh-1.2.14-win32bin.zip at: http://opensores.thebunker.net/pub/mirrors/ssh/contrib/ssh-1.2.14-win32bin.zip gwen Requirements: \\hal\software\sshcliwin32\ssh-1.2.14-win32bin.zip \\hal\software\wincvs\wincvs.zip account on cvs server with proper permissions Steps: 1. Install windows command line ssh client a. unzip ssh-1.2.14-win32bin.zip anywhere (e.g. C:\ssh); make sure it is in its permanent location, as there is no setup.exe to run afterwards. b. mkdir \etc located within directory used in last step (e.g. C:\ssh\etc) c. add directory in Step 1a to %PATH% environmental variable --Note: At this point, you should be able to access the ssh.exe and scp.exe from any location using the command line. d. set %HOME% environmental variable to any location; this location should be someplace specific to the individual user (e.g. C:\home\username) e. mkdir %HOME%\.ssh (e.g. C:\home\username\.ssh) f. ssh -l username cvs_host (the next few commands will be on the unix remote host) g. ssh-keygen1 (accept default locations and when prompted for a password, leave blank) h. cat $HOME/.ssh/identity.pub $HOME/.ssh/authorized_keys i. chmod 644 .ssh/authorized_keys --Note: You will need to ensure that there are no group or other write permissions through $HOME/.ssh. If there are, passwordless ssh will not work properly. j. logout (return back to local Windows command prompt) k. cd %HOME%\.ssh l. scp username@cvs_host:.ssh/* . --Note: At this point, you should be able to log in to cvs_host without using a password. This is critical for WinCVS to operate properly. If this doesn't work, go back and make sure you did all the steps properly. 2. Install wincvs a. unzip wincvs.zip b. run setup.exe; all the default settings are fine (alternatively, you may choose to run installer from hal) c. launch wincvs.exe (it should be located under Start/Programs) d. Enter :ext:username@cvs_host:/usr/local/cvsroot in text box e. Select SSH server in Authentication drop box f. Press OK. g. Select Checkout Module under CVS Admin menu h. Choose an appropriate location to keep your cvs working copy (e.g. %HOME%) and press OK i. Enter the module name in text box j. Press OK k. Select %HOME% as your home directory l. Press OK --Note: At this point, you should see standard cvs checkout output printing in the white area the the bottom right. If nothing happens after a minute, you should first check if cvs_host is having problems. If the server seems okay, make sure passwordless ssh is working. If you didn't set up passwordless ssh correctly, WinCVS will not checkout files b/c it is waiting for the user to enter a password (WinCVS doesn't allow the user to interactively enter one). If those things are fine, make sure you followed the steps properly. --Note: These next steps are optional. m. Select Change browser location under View menu. n. Select working copy location. o. Press OK. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS - SSH question
[ On Friday, June 1, 2001 at 10:17:21 (-0400), Robert Briggs wrote: ] Subject: CVS - SSH question My question is how to get WinCVS to work with SSH without having to enter the password in the SSH window with every command that is executed. That's an SSH question, the answer to which depends on which SSH client and server implementations you are using and what security policies are in effect on the server. I'd susggest asking on the [EMAIL PROTECTED] mailing list. You might also try reading the instructions written to help users of sourceforge (under their Site Docs project): http://sourceforge.net/docman/display_doc.php?docid=761group_id=1 There's a bit more basic stuff here: http://sourceforge.net/docman/display_doc.php?docid=3602group_id=1 While you're there you might want to read about SSH security issues too: http://sourceforge.net/forum/forum.php?forum_id=89285 :-) -- Greg A. Woods +1 416 218-0098 VE3TCP [EMAIL PROTECTED] [EMAIL PROTECTED] Planix, Inc. [EMAIL PROTECTED]; Secrets of the Weird [EMAIL PROTECTED] ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
One Question
I'm thinking in use cvs but I have the folowing doubt: If I run the CVS server in aUNIX machine while win32 clients are used. I want that every win32 client had his private project (files), so Can I use passwords for security issues without create a UNIX user for every win32 client? Thanx in advance, Pedro ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: One Question
On Thu, May 31, 2001 at 03:25:40PM +0200, Pedro wrote: I'm thinking in use cvs but I have the folowing doubt: If I run the CVS server in aUNIX machine while win32 clients are used. I want that every win32 client had his private project (files), so Can I use passwords for security issues without create a UNIX user for every win32 client? Thanx in advance, Pedro Yes, you can. Look at the manual for details, you can write an specific password file for CVS under the CVSROOT directory (which is not the same as the root of the cvs repository but is located under it). ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: [Newbie question] When LOGIN fails
David A. Cobb wrote:
Today I was working with a repository at the other end of an SSH tunnel.
My CVS LOGIN was rejected: server denied access to ${CVSROOT}
However, while experimenting with Xemacs/PCL-CVS I requested to STATUS,
and then to UPDATE my directories. Both functions apparently worked.
Does this indicate that the login only applies to a comit?
Is this by design, or is it a big security hole?
No. You were most likely using different logins/passwords/roots somehow for the
different cases you describe above.
Derek
--
Derek Price CVS Solutions Architect ( http://CVSHome.org )
mailto:[EMAIL PROTECTED] CollabNet ( http://collab.net )
--
151. H lp! S m b dy st l ll th v w ls fr m my k yb rd!
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
Re: question about cvs!
I have downloaded a version of cvs for solaris 2.5.1 and now it works
Thanks,
wangc
- Original Message -
From: Donald Sharp [EMAIL PROTECTED]
To: wangc [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Sent: Tuesday, May 29, 2001 9:40 PM
Subject: Re: question about cvs!
Your using a version of cvs that is compiled for a later version( 2.6,2.8? )
of solaris than what you are using.
Get the source and compile.
donald
On Tue, May 29, 2001 at 10:51:44AM +0800, wangc wrote:
hi,
I just want to use cvs for a project's version control.And now
I have installed the cvs1.11p1 on a sun workstation(Solaris 2.5.1) as
cvs server and Wincvs on my PC(win98) as a client.But there are some
errs.The Wincvs show this hint :
ld.so.1: cvs: fatal: relocation error: symbol not found: setsockopt: referenced in
cvs
I dont know the reason!
Thanks for your help!
ú ¾ÉX§X¬´ß¡Ëì{¨®m¶ÿ¨¥{¨®æj)fjåËbú?wèrû
Re: question about cvs!
Your using a version of cvs that is compiled for a later version( 2.6,2.8? ) of solaris than what you are using. Get the source and compile. donald On Tue, May 29, 2001 at 10:51:44AM +0800, wangc wrote: hi, I just want to use cvs for a project's version control.And now I have installed the cvs1.11p1 on a sun workstation(Solaris 2.5.1) as cvs server and Wincvs on my PC(win98) as a client.But there are some errs.The Wincvs show this hint : ld.so.1: cvs: fatal: relocation error: symbol not found: setsockopt: referenced in cvs I dont know the reason! Thanks for your help! ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
[Newbie Question] Flag records
I understand from reading here that CVS includes all of RCS, in one way or
another.
Are lines such as $ID$ something and $DATE$ something features of RCS?
And how do they get there in the first place?
David A. Cobb, Software Engineer, Public Access Advocate, All around nice guy.
Get my PGP key at
:http://pgpkeys.mit.edu:11371/pks/lookup?op=getsearch=superbiskit
Fingerprint=0x{6E3E_DB8C_2E8C_4248_62B2_FE29_08EE_CF0A_3629_E954}
By God's Grace I am a Christian man, by my actions a great sinner.
--The Way of a Pilgrim, R. M. French [tr.]
---.!.!.!.!.!.!.
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
[Newbie question] When LOGIN fails
Today I was working with a repository at the other end of an SSH tunnel.
My CVS LOGIN was rejected: server denied access to ${CVSROOT}
However, while experimenting with Xemacs/PCL-CVS I requested to STATUS,
and then to UPDATE my directories. Both functions apparently worked.
Does this indicate that the login only applies to a comit?
Is this by design, or is it a big security hole?
David A. Cobb, Software Engineer, Public Access Advocate, All around nice guy.
Get my PGP key at
:http://pgpkeys.mit.edu:11371/pks/lookup?op=getsearch=superbiskit
Fingerprint=0x{6E3E_DB8C_2E8C_4248_62B2_FE29_08EE_CF0A_3629_E954}
By God's Grace I am a Christian man, by my actions a great sinner.
--The Way of a Pilgrim, R. M. French [tr.]
---.!.!.!.!.!.!.
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
question about cvs!
hi, I just want to use cvs for a project's version control.And now I have installed the cvs1.11p1 on a sun workstation(Solaris 2.5.1) as cvs serverand Wincvs on my PC(win98) as a client.But there are some errs.The Wincvs show this hint: ld.so.1: cvs: fatal: relocation error: symbol not found: setsockopt: referenced in cvs I dont know the reason! Thanks for your help!
Gotta cvs admin files (CVSROOT) question...
My question centers around the val-tags and history files in the CVSROOT directory Leading up to the question. We have code in two cvs repositories on two separate unix machines: alpha:/data/cvs beta:/data/cvs The alpha:/data/cvs repository has a root - bubba - we want to move to the beta machine. The beta:/data/cvs repository does not have a bubba root. Within bubba there are several directories, some of which will remain behind on the alpha machine. For example, alpha:/data/cvs/bubba/dir_a alpha:/data/cvs/bubba/dir_b alpha:/data/cvs/bubba/dir_X (will remain with alpha) alpha:/data/cvs/bubba/dir_Y (will remain with alpha) What we did was that we tar'd a copy of alpha's entire repository to a holding area on the beta machine. We then copied the bubba root (minus the dir_X and dir_Y directories) from the holding area to the beta:/data/cvs and renamed it dubya. So now we have, alpha:/data/cvs/bubba/dir_X alpha:/data/cvs/bubba/dir_Y beta:/data/cvs/dubya/dir_a beta:/data/cvs/dubya/dir_b Of course both repositories have a CVSROOT directory and each has a history and a val-tags file. On the beta machine (which is what we care aboutthe alpha machine holds code belonging a vendorCOTS to us), is it a simple matter to merge the contents of these two files from the two repositories? The merged files will be housed on the beta machine. The vendor will most likely remove our code from his repository. We want to retain history for our code that was under the bubba regime. We do use tags. CVS 1.9 is what we use (due to bureaucracy we haven't gotten around to upgrading yet). We've done some simple testing (cvs log filename cvs history filename) but have yet to perform more extensive testing (spanning back several tagged versions). I hoping it's a simple manner of just merging these two admin files. Thanks Glen Luna ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Gotta cvs admin files (CVSROOT) question...
Luna, Glen writes [about merging repositories]: Of course both repositories have a CVSROOT directory and each has a history and a val-tags file. On the beta machine (which is what we care aboutthe alpha machine holds code belonging a vendorCOTS to us), is it a simple matter to merge the contents of these two files from the two repositories? The merged files will be housed on the beta machine. You don't need to merge the val-tags files -- the missing tags will be discovered and added whenever they're used for the first time. On the other hand, they're easy to merge: sort -u should do nicely. The history file is used only by the history subcommand -- if you don't care about that, there's no need to merge the files. If you do, then it should suffice to simply concatenate them together (of course, you could remove any entries for foreign stuff first if you want). -Larry Jones I've got to start listening to those quiet, nagging doubts. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
cvs question
I have a question regarding how CVS can be configured to work with multiple development sites. Let's say there are three development sites in the US, Russia and India. US site has the source code repository and developers here have access to all of the source code(C/C++/java). Russians can only access C++ source code and cannot access C or java source code. Whereas Indians can only access java source code(checkin checkout etc) and cannot access C++ or C source code. Is it possible to implement a model like this using CVS? Your reply is appreciated. Thanks, Hem ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: cvs question
Yes, the simplest way is to create seperate modules with seperate group permissions. I know the names below are dumb, they are just for the sake of example :) groups : c cpp java The modules are named c, cpp and java and are read/write/setgid to their individual group. US users are members of all three groups, Indian users are a member of the Java group and Russians are members of the cpp group. HTH, Rob Helmer On Wed, May 23, 2001 at 05:11:10PM -0400, Hem Bapat wrote: I have a question regarding how CVS can be configured to work with multiple development sites. Let's say there are three development sites in the US, Russia and India. US site has the source code repository and developers here have access to all of the source code(C/C++/java). Russians can only access C++ source code and cannot access C or java source code. Whereas Indians can only access java source code(checkin checkout etc) and cannot access C++ or C source code. Is it possible to implement a model like this using CVS? Your reply is appreciated. Thanks, Hem ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
commitinfo script question
Hi, Is it possible to run a series of tests on files with a commitinfo script, and then, when they have finished successfully, run another test on all the files, rather than on a directory by directory basis? What I'm trying to do is check each of my source files for proper formatting, titles, $Id:$ tags, etc, and then after that, make sure the package still compiles before the log message window appears. I've only been able to have it compile each time, which amounts to a whole re-compilation of the project for each directory that has modified files in it. Is there a way around this? This is with cvs 1.11 on a redhat 7.1 machine, if it makes a difference. Thanks for any advice! -matt ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: commitinfo script question
Matthew R. MacIntyre writes: Is it possible to run a series of tests on files with a commitinfo script, and then, when they have finished successfully, run another test on all the files, rather than on a directory by directory basis? Not without redesigning CVS. As it is, it does everything a directory at a time. -Larry Jones It must be sad being a species with so little imagination. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Beginner Question: I wanna see my changes!
Hi, I just started working with CVS and WINCVS on my Windows2000-Workstation (my Computer) and on a Windows-NT 4.0 Server (Development). Everything runs fine so far and I just commited a sample file with some changes on it. In my editor and in WINCVS I can see the changes I've done but I can't see the changes on the server. The file is still the same and doesn't gets updated after my commit! Has somebody any ideas? Thank you very much in advance for your help. Philipp NORDENIA INTERNATIONAL SERVICES GmbH Philipp Krapp Steinfeldstraße 5 * 39179 Barleben Phone:+49 (0) 39 203 / 837 - 71 Fax: ++49 (0) 39 203 / 837 - 89 [EMAIL PROTECTED] www.nordenia-services.de ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Beginner Question: I wanna see my changes!
Krapp, Philipp wrote: but I can't see the changes on the server. The file is still the same and doesn't gets updated after my commit! 'cvs update'? Derek -- Derek Price CVS Solutions Architect ( http://CVSHome.org ) mailto:[EMAIL PROTECTED] CollabNet ( http://collab.net ) -- If at first you don't succeed, skydiving is not for you. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
WinCvs question
Hi, . I am given the charge of version controlling the software products, here in my new small company of9 developers. It is decided that we go with WinCvs for our Windows platform and I did some hands-on-practice with WinCvs with the help of http://www.computas.com/pub/wincvs-howto/#branch, WinCvs - Daily Use Guide. Now, though, I am confident of doing it, I have some starting trouble. 1. In my company, currently, 2nd release of the product is going on. To start with how do I go doing it orwhat files do I have to put in the repository? The 8-9 developers are in different locations. But we are not going in for clint/server mode. My boss likes to be under control in one local machine, so that modified files can be got over the net and updated/committed in the repository by a single person. 2. I understand each time we modify and save a file, version(revision) is automatically incremented for each individual file. But how do we indentify Release 1 from Release 2. Is it the module name or tag name which makes the difference in Release name?. Please update me or any other site/address where I can ask questions related to winCvs. Thank you.MohamedGet Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Question about cvsignore file
George Mathew writes: I am trying not to ignore the core directory during cvs import. I added a cvsignore file in $CVSROOT/$CVSROOT directory of server. [...] This did not work. But if I add this to my $HOME/.cvsignore file it works. The documentation says $CVSROOT/$CVSROOT is the first place it looks. Is it a bug? or the documentation is not correct? It's a long-standing bug: the ignore logic in client/server mode does not work correctly. The client and server each apply their own ignore lists separately, which means that a server ignore specification cannot override the client's ignore specification like it should. Unfortunately, it requires significant redesign effort to fix this, which is why it hasn't been fixed yet. Someone was working on it, but I haven't heard anything about it in quite a while, so I presume the effort is dead. -Larry Jones My life needs a rewind/erase button. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Question about cvsignore file
I am trying not to ignore the core directory during cvs import. I added a cvsignore file in $CVSROOT/$CVSROOT directory of server. This cvsignore file contains: ! RCS SCCSCVS CVS.adm RCSLOG cvslog.* tagsTAGS .make.state .nse_depinfo *~ #* .#* ,* _$* *$ *.old *.bak *.BAK *.orig *.rej .del-* *.a *.olb *.o *.obj *.so*.exe *.Z *.elc *.ln .*.o.flags This did not work. But if I add this to my $HOME/.cvsignore file it works. The documentation says $CVSROOT/$CVSROOT is the first place it looks. Is it a bug? or the documentation is not correct? Thanks ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Configuration Question
Eric Siegerman wrote: Also, where can I find files to create a pserver? Either RPM or source will do. There are no extra files; the main CVS binary will suffice. Though someone's mentioned a program for maintainging pserver password files; search the mailing-list archives. CVSPwd2 http://www.pajamian.dhs.org Regards, Peter ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: rsh access method question
Hi Mark, the rsh for NT supplied by Cygwin (www.cygwin.com) works pretty well. Chris Mark wrote: Hi, I am reading up on the rsh access method. We have meet strong resistance from the UNIX folks with regards to setting up a inetd service as root. This is expected as the production network and development network are one and the same. We are looking to have the repositories on a unix solaris machine. The clients will be local, others on solaris, mainly windows NT and possibly some win2k. In the Cederqvist book is says the Windows NT rsh is no good. Is there another rsh for Windows that can be used? Should we use :server: or :ext:? We will start with WINCvs but may migrate to tkcvs or a java cvs client in the distant future. Any ideas, input, and/or advice is appricated. Thanks, = \===\ \==\ Mark O'Brien \==\ CM Consultant \==\ [EMAIL PROTECTED] \===\ __ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: rsh access method question
Thanks, However, I do not know what to install (minimally) to use rsh from cygwin for CVS. I am on the http://mirrors.rcn.net/pub/sourceware/cygwin/latest/ web page. What package is rsh in? What else besides the rsh package do I need? I have the setup.exe, but our firewall will not allow setup.exe access to the interent, so I have to download what I need from the site, then run it. Thanks again for you help. Mark --- chaessig [EMAIL PROTECTED] wrote: Hi Mark, the rsh for NT supplied by Cygwin (www.cygwin.com) works pretty well. Chris Mark wrote: Hi, I am reading up on the rsh access method. We have meet strong resistance from the UNIX folks with regards to setting up a inetd service as root. This is expected as the production network and development network are one and the same. We are looking to have the repositories on a unix solaris machine. The clients will be local, others on solaris, mainly windows NT and possibly some win2k. In the Cederqvist book is says the Windows NT rsh is no good. Is there another rsh for Windows that can be used? Should we use :server: or :ext:? We will start with WINCvs but may migrate to tkcvs or a java cvs client in the distant future. Any ideas, input, and/or advice is appricated. Thanks, = \===\ \==\ Mark O'Brien \==\ CM Consultant \==\ [EMAIL PROTECTED] \===\ __ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: rsh access method question
You just have to get the rsh.exe and cygwin1.dll. They weight 650 KB, but I can send them to you if you want (and if your provider is fast enough). Chris Mark wrote: Thanks, However, I do not know what to install (minimally) to use rsh from cygwin for CVS. I am on the http://mirrors.rcn.net/pub/sourceware/cygwin/latest/ web page. What package is rsh in? What else besides the rsh package do I need? I have the setup.exe, but our firewall will not allow setup.exe access to the interent, so I have to download what I need from the site, then run it. Thanks again for you help. Mark --- chaessig [EMAIL PROTECTED] wrote: Hi Mark, the rsh for NT supplied by Cygwin (www.cygwin.com) works pretty well. Chris Mark wrote: Hi, I am reading up on the rsh access method. We have meet strong resistance from the UNIX folks with regards to setting up a inetd service as root. This is expected as the production network and development network are one and the same. We are looking to have the repositories on a unix solaris machine. The clients will be local, others on solaris, mainly windows NT and possibly some win2k. In the Cederqvist book is says the Windows NT rsh is no good. Is there another rsh for Windows that can be used? Should we use :server: or :ext:? We will start with WINCvs but may migrate to tkcvs or a java cvs client in the distant future. Any ideas, input, and/or advice is appricated. Thanks, = \===\ \==\ Mark O'Brien \==\ CM Consultant \==\ [EMAIL PROTECTED] \===\ __ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
followup to tag question posted 5/11/01
I saw the email below in the CVS archives and wanted to ask a follow up question. I'm new to CVS and would appreciate any help. Would it be possible to get the information regarding the person who created a tag through the taginfo file by passing the information to a script for reporting? For example I noticed in the loginfo file there is a DEFAULT action which starts with (echo ...). Could you enter something like (echo $USER...) in the taginfo file to capture the user who created the tag and send it to a script along with the rest of the taginfo information? I've tried several iterations of this but failed in all attempts. The failures occur before executing the script so I am not sure if it is possible to use the syntax above. I am hoping to get a tip as in try this... or give it up it's hopeless. We're running CVS 1.10.7 under Solaris 2.7 11/99 on a Sparc platform. Thanx in advance. Zanabria, Moises writes: Is possible to know who person created a tag?? CVS does not record that information. -Larry Jones ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: followup to tag question posted 5/11/01
Lance Murayama writes: Would it be possible to get the information regarding the person who created a tag through the taginfo file by passing the information to a script for reporting? Yes, that's certainly possible. Note, however, that rtag has a -n option that suppresses the taginfo script. Whether that's a problem or not is up to you. tried several iterations of this but failed in all attempts. The failures occur before executing the script so I am not sure if it is possible to use the syntax above. I am hoping to get a tip as in try this... or give it up it's hopeless. To supplement the example in the manual, something like: ALL $CVSROOT/CVSROOT/loggit $USER should work (where loggit is the script shown in the manual). http://cvshome.org/docs/manual/cvs_8.html#SEC78 -Larry Jones Even if lives DID hang in the balance, it would depend on whose they were. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
rsh access method question
Hi, I am reading up on the rsh access method. We have meet strong resistance from the UNIX folks with regards to setting up a inetd service as root. This is expected as the production network and development network are one and the same. We are looking to have the repositories on a unix solaris machine. The clients will be local, others on solaris, mainly windows NT and possibly some win2k. In the Cederqvist book is says the Windows NT rsh is no good. Is there another rsh for Windows that can be used? Should we use :server: or :ext:? We will start with WINCvs but may migrate to tkcvs or a java cvs client in the distant future. Any ideas, input, and/or advice is appricated. Thanks, = \===\ \==\ Mark O'Brien \==\ CM Consultant \==\ [EMAIL PROTECTED] \===\ __ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: rsh access method question
On Tue, May 15, 2001 at 12:45:32PM -0700, Mark wrote: there another rsh for Windows that can be used? Should we use :server: or :ext:? I'd suggest :ext: with ssh. While you may not need to security associated with ssh, it does work well. Of course, if they don't want you installing sshd on the servers mrc -- Mike Castle Life is like a clock: You can work constantly [EMAIL PROTECTED] and be right all the time, or not work at all www.netcom.com/~dalgoda/ and be right at least twice a day. -- mrc We are all of us living in the shadow of Manhattan. -- Watchmen ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
I have a question
Hello, I am baesy . I work in Korea. I have a question . I want to check out previous version , not current . How can I see version list ? How can I check out what I want ?Please, help me, I'll be waiting for your answer. best regards baesy
Re: I have a question
On Mon, May 14, 2001 at 04:58:08PM +0900, ¹è»ó¿ë wrote: Hello , I am baesy . I work in Korea. I have a question . I want to check out previous version , not current . How can I see version list ? Try the log command: cvs log ... (help: cvs log --help, or the info or man pages) How can I check out what I want ? using the -r option with cvs checkout, for instance: cvs checkout -r1.3 module-foo Please, help me, I'll be waiting for your answer. best regards baesy Best regards from Spain. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS Question..
Here is my situation... Using CVS under Unix... I have created a scratch folder. I need to commit in CVS. This scratch folder is a sub-folder of a important folder (A). Many people are using A.. but I dont want to give my private folder scratch which is under (A) to everybody who does cvs update -d -P under A folder. Only when I do a cvs update -d -P under A folder , I should be getting scratch folder.. Is there any way to do that.. Ted __ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS Question..
Pavan Seth writes: Using CVS under Unix... I have created a scratch folder. I need to commit in CVS. This scratch folder is a sub-folder of a important folder (A). Many people are using A.. but I dont want to give my private folder scratch which is under (A) to everybody who does cvs update -d -P under A folder. Only when I do a cvs update -d -P under A folder , I should be getting scratch folder.. Put your scratch directory somewhere in the repository other than under A -- your working directory hierarchy doesn't have to match the repository directory hierarchy. -Larry Jones I've never seen a sled catch fire before. -- Hobbes ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
I have a question
Hello, I am baesy . I work in Korea. I want to control binary file . My OS is Linux . First I make tar-file. I cvs import tar-file . But original file and imported file differ . Size of orginal file is 16691200. Size of imported file is 16691124. And I can't executetar xvf imported file . Why does it do ? How can I cvs import binary file (*.tar file ) . please help me . I will be waiting for youranswer. Best regards baesy
Re: I have a question
On Thu, May 10, 2001 at 10:09:01PM +0900, ¹è»ó¿ë wrote: Hello, How can I cvs import binary file (*.tar file ) . The key is -kb option. Look for it in the cvs documentation, but simply using -kb should work. Hope it helps. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: I have a question
Hello , I am baesy . I work in Korea. I want to control binary file . My OS is Linux . First I make tar-file. I cvs import tar-file . But original file and imported file differ . Size of orginal file is 16691200. Size of imported file is 16691124. And I can't execute tar xvf imported file . Why does it do ? How can I cvs import binary file (*.tar file ) . please help me . I will be waiting for your answer. Best regards baesy Importing binary files: http://www.cvshome.org/docs/manual/cvs_9.html#SEC80 Why are you trying to maintain revisions on a tar file? -Matt ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: I have a question
for adding binary files use -kb option. - Manik ¹è»ó¿ë wrote: Hello , I am baesy .I work in Korea.I want to control binary file .My OS is Linux . First I make tar-file.I cvs import tar-file .But original file and imported file differ .Size of orginal file is 16691200.Size of imported file is 16691124.And I can't execute tar xvf imported file .Why does it do ?How can I cvs import binary file (*.tar file ) .please help me .I will be waiting for your answer. Best regardsbaesy ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: I have a question
-Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]Sent: Thursday, May 10, 2001 8:09 AMTo: [EMAIL PROTECTED]Subject: I have a question Hello, I am baesy . I work in Korea. I want to control binary file . My OS is Linux . First I make tar-file. I cvs import tar-file . But original file and imported file differ . Size of orginal file is 16691200. Size of imported file is 16691124. And I can't executetar xvf imported file . Why does it do ? How can I cvs import binary file (*.tar file ) . please help me . I will be waiting for youranswer. [Thornley, David] What do you mean "imported file"? Do you mean the file in the repository, orone checked out later? The file in the repository is not supposed to be used directly. And, for that matter, why are you controlling a tarball? Why not skip the whole tar process and just import the individual files? Importing a tarball makes no sense whatsoever. Importing is a way toallowyour own development on somebody else's code while keeping track of their changes and merging them in. If they're in a tarball, they're not going to merge, so you'regetting nothing out of it. If you must keep a tarball around, use "cvs add -kb" or something like that.
CVS watches - question
Hi, please help me - I run into problem with read-only files and watches. If I add file after watch was turned on and then another user updates his working copy, the user receives this new added file without read-only flag - is it a bug or I miss something? I'm running CVS server version 1.11.1p1 on Red Hat 7.0 and command line Win2K client version 1.11 Here is what I did: 1.I put watch on my whole module: cvs watch on MP 2. User does checkout cvs checkout MP, receives all files in subdirectories as read-only - that's exactly what he needs. 3. I add a new file in /MP/test/newfile cvs add newfile and the commit cvs commit MP 4. User does update cvs update MP and receives a newfile without read-only flag. WHY?? How can I solve this problem? Regards, Diana ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Configuration Question
I have a few questions about CVS on a RH7 box. On my old box (not originally set-up by me --CVS 1.10.7), CVS has directories within the archived directories, for example: /etc/httpd/CVS But on my new box (1.11.1p1), I can't seem to get the directories created in the same way. If I follow the instructions here: http://cvshome.org/docs/manual/cvs_3.html#SEC38, it seems to work, but I still don't get the CVS directory within /etc/httpd. My module is called httpd_conf. What I get instead is this: /etc/httpd/httpd_conf/CVS. What I want it to do is create /etc/httpd/CVS, and store all the files in /usr/cvs/httpd_conf. Any ideas of how I could do this? Also, where can I find files to create a pserver? Either RPM or source will do. Thanks for any help whatsoever. -J.weber ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Revision Numbers Question
cvs rdiff -s -r tagname modulename The output may need to parsed a bit. Jerry From: Anthony E. Glover [mailto:[EMAIL PROTECTED]] Sent: Tuesday, May 08, 2001 8:47 AM To: Info-CVS Subject: Revision Numbers Question I know this must be a very simple thing to do, but I can't seem to find a way of doing it. How do I dump a complete listing of the files in a repository (not a working directory) with the revision numbers shown. Better yet, the list of revision numbers associated with a tagged set of files. Thanks, Tony ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS /etc/xinetd.conf question
Dear Sir:
I am using RedHat LINUX 7.00 version.When I try to login to CVS server
by using cvspserver
login, I got the 2401 failed messages. Please see the following trace
messages:
===
[jhsieh@sitara]% cvs -d
:pserver:[EMAIL PROTECTED]:/home/jhsieh/CVS login
Logging in to :pserver:[EMAIL PROTECTED]:2401/home/jhsieh/CVS
CVS password:
cvs [login aborted]: connect to sitara.metroncomm.com(63.110.69.160):2401
failed: Connection refused
[jhsieh@sitara]%
The /etc/services are defined correctly as follows:
cvspserver 2401/tcp# CVS client/server
operations
cvspserver 2401/udp# CVS client/server
operations
The /etc/xinetd.d/cvspserver deines cvspserver as follows:
service cvspserver
{
socket_type = stream
protocol= tcp
wait= no
user= root
id = cvspserver
passenv = /home/jhsieh/CVS
log_type= FILE /var/log/xinetdlog
port= 2401
log_on_failure += USERID
server = /usr/bin/cvs
server_args = -f --allow-root=/home/jhsieh/CVS pserver
log_on_success += USERID DURATION
log_on_failure += HOST USERID
disable = no
}
When I verify the log message from /var/log/messages, I found the following
messages:
===
May 7 14:38:40 sitara xinetd[1841]: Service cvspserver^M missing attribute
socket_type
May 7 14:38:40 sitara xinetd[1841]: Service cvspserver^M missing attribute
user
May 7 14:38:40 sitara xinetd[1841]: Service cvspserver^M missing attribute
server
===
The log messages explain when xinetd.d start cvspserver, xinetd cannot read
socket_type,
user and server from /etc/xinetd.d/cvspserver.
Does anyone know why cvspserver missing attribute on socket_type,user and
server ?
Please advise, thanks
John Hsieh
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
Question about auto login
Dear Sir: I use CVS for several month, It is good for me. But I have one question about auto-login. Beacuse Ineed to buildprogram everyday in different machines, I should login CVS server, checkout code, build code and logout. I want these process automatically, so I write a shell command file to do this job. But I need login CVS first. How do I write auto-login shell command ? Sincerely Liwie Chour in OFFICE
Re: Question about auto login
Liwei Chour [EMAIL PROTECTED] writes: I use CVS for several month, It is good for me. But I have one question about auto-login. Beacuse I need to build program everyday in different machines, I should login CVS server, checkout code, build code and logout. I want these process automatically, so I write a shell command file to do this job. But I need login CVS first. How do I write auto-login shell command ? You don't need to. When you first login (successfully) to any cvs server, the password information is stored in an encrypted form in the .cvspass file in your home directory. After this initial login, you never need to use the cvs login command for that same server ever again. (Yes, .cvspass saves passwords from multiple servers). (Note: if you want to forget the password for a particular server, you can run the cvs logout command, which will erase that password from .cvspass. You will then need to run cvs login again for that server before you can use it again.) ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
Instead of: env = HOME=/home/cvs it is much better is to specify: passenv = PATH (i.e., don't pass $HOME to the server at all). The daemon is running as root when it starts. Then it drops priviledge to the user's level. Unfortunately, it still trys to look in root's home directory for .cvsignore, but as it is now running as a normal user, it cannot. The PATH has nothing at all to do with this problem. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
David D. Hagood writes [quoting me]: Instead of: env = HOME=/home/cvs it is much better is to specify: passenv = PATH (i.e., don't pass $HOME to the server at all). The daemon is running as root when it starts. Then it drops priviledge to the user's level. Unfortunately, it still trys to look in root's home directory for .cvsignore, but as it is now running as a normal user, it cannot. The PATH has nothing at all to do with this problem. You misunderstand -- when you don't specify passenv, xinetd passes the entire environment to the child process. When you *do* specify passenv, only the environment variables listed there are passed. So the critical part is that HOME is *not* listed, not that PATH is. This results in $HOME not being set at all in the server's environment (which will cause CVS to compute it correctly) rather than being set to a bogus value. -Larry Jones That's the problem with nature. Something's always stinging you or oozing mucus on you. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
[EMAIL PROTECTED] (Larry Jones) wrote: This results in $HOME not being set at all in the server's environment (which will cause CVS to compute it correctly) rather than being set to a bogus value. So, when the CVS server drops priv, it will correctly set the HOME variable to the user's home dir, or will it leave it blank?And if it is left blank, will the server correctly not look for the file, or will is look in the current working directory. Actually, the point is largely moot, since the latest versions of CVS don't check for those files if running in server mode... ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
David D. Hagood writes: So, when the CVS server drops priv, it will correctly set the HOME variable to the user's home dir, or will it leave it blank?And if it is left blank, will the server correctly not look for the file, or will is look in the current working directory. It leaves $HOME unset (which is different than being set to blank). If it wants to know the user's home directory, it will look it up in /etc/passwd (since $HOME isn't set) and get the right answer. -Larry Jones OK, there IS a middle ground, but it's for sissy weasels. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
David D. Hagood writes:
service cvspserver
{
disable = no
id = cvspserver
env = HOME=/home/cvs
socket_type = stream
protocol= tcp
port= 2401
wait= no
user= root
log_on_failure += USERID
server = /usr/bin/cvs
server_args = -f --allow-root=your root here pserver
}
The /home/cvs dirctory created to work around CVS bug #1, the cannot
access /root/.cvsignore error: create this directory, make it owned by
root, readable by all, and not writable by anybody, and CVS will be happy.
Instead of:
env = HOME=/home/cvs
it is much better is to specify:
passenv = PATH
(i.e., don't pass $HOME to the server at all).
-Larry Jones
Please tell me I'm adopted. -- Calvin
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
RE: newbie pserver question
From: Toby Tremayne [SMTP:[EMAIL PROTECTED]] Sent: Wednesday, May 02, 2001 1:09 PM To: info cvs Subject: newbie pserver question I know this is going to seem really naive, but I'm really stuck with this. I've installed CVS pserver on a linux box and it seems to work quite happily. What I'm trying to do now is test a bunch of mac and windows clients. I've started with WinCVS. My situation is this - my predecessor installed CVS on a redhat linux box. The repository is in /cvs which is a symbolic link to a samba mounted drive on a faraway windows server. So, obviously, the repository itself is actually on the windows box. Many people have written to info-cvs saying it's a bad idea getting to repository files via network file systems. Danger of repository corruption. Much better to have repository on local filesystem of machine where the cvs server runs. Now, some of our developer machines have been using wincvs to connect to that repository. What I'm failing to get into my thick skull is exactly how the client know where the server is - the CVSROOT setting is :pserver:steve@devServer2:f:/code (for example). The f: in that name makes me think your predecessor may have used the nt pserver port http://www.cvsnt.org/ to run the pserver on a windows nt machine. So does the client then find that server/folder and use the /CVSROOT files to determine how to talk to the cvs server? Because I'm kind of stumped to be perfectly honest. I'm pretty new to unix I'm afraid, so I'm obviously missing something major here. http://cvshome.org/docs/manual/cvs_2.html#SEC9 To make sure I understand it and can maintain it, I've been setting up a new server, on a different box. The difference is that the repository is local to that cvs server. So can anyone tell me what the CVSROOT variable should be on the client machine, if the cvspserver is on a host called GOA and the repository is in /cvs ??? Assuming your cvs username is tobytremayne, CVSROOT=:pserver:tobytremayne@GOA:/cvs http://cvshome.org/docs/manual/cvs_2.html#SEC26 http://cvshome.org/docs/manual/cvs_2.html#SEC31 I've been through all the documentation I can find so I must just be reading something wrong, I'd really appreciate any pointers I can get. cheers, Toby Toby Tremayne Code Poet and Zen Master of the Heavy Sleep Show Ads Interactive 359 Plummer St Port Melbourne VIC 3207 P +61 3 9245 1247 F +61 3 9646 9814 ICQ UIN 13107913 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS /etc/inetd.conf question
In order to configure the CVS server, modify the /etc/inetd.conf is necessary. But in the RedHat 7.0 version software, there is no /etc/inetd.conf Where is the file of /etc/inetd.conf or similar configuration file in RedHat 7.0? If there is no /etc/inetd.conf, how do I can configure CVS pserver by using Redhat 7.0 ? Thanks John Hsieh ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
John Hsieh writes: But in the RedHat 7.0 version software, there is no /etc/inetd.conf man xinetd. And if you need more help, ask Linux people who will know about xinetd, not CVS people who generally will not. -Larry Jones I don't need to improve! Everyone ELSE does! -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS /etc/inetd.conf question
John Hsieh wrote:
In order to configure the CVS server, modify the /etc/inetd.conf
is necessary.
But in the RedHat 7.0 version software, there is no /etc/inetd.conf
Where is the file of /etc/inetd.conf or similar configuration file in RedHat
7.0?
RH7.0 uses xinetd, which stores its configuration in
/etc/xinetd.d/servicename
You add files to that directory to add services: create a file
cvspserver there and fill this in:
# default: on
#
# service cvspserver
#
service cvspserver
{
disable = no
id = cvspserver
env = HOME=/home/cvs
socket_type = stream
protocol= tcp
port= 2401
wait= no
user= root
log_on_failure += USERID
server = /usr/bin/cvs
server_args = -f --allow-root=your root here pserver
}
The /home/cvs dirctory created to work around CVS bug #1, the cannot
access /root/.cvsignore error: create this directory, make it owned by
root, readable by all, and not writable by anybody, and CVS will be happy.
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
A question about checkout
I have a question of CVS. When I checkout a moudle from server using pserver, there is always a message of cvs server: cannot open /root/.cvsignore: Permission denied cvs [server aborted]: can't chdir(/root): Permission denied How does this happen,and how to resolve it? Thanks ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: A question about checkout
[EMAIL PROTECTED] writes: When I checkout a moudle from server using pserver, there is always a message of cvs server: cannot open /root/.cvsignore: Permission denied cvs [server aborted]: can't chdir(/root): Permission denied http://cvshome.org/docs/manual/cvs_21.html#SEC182 If you need more detailed help, see the archives of this list. This is *the* most frequently asked question. -Larry Jones Somebody's always running my life. I never get to do what I want to do. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
NEWBIE question: checkout, sub-modules
Hello, I have a question that I'm sure others have asked but I can't seem to find an answer for. From an existing module, I want to checkout a subset of submodules under a different module name. For simplicity's sake I'll call the existing modulevegetables (not true names). The vegetables module had some fruits that we want to now collect and promote under a new module named fruits. Now, in no way are we desiring to modify the repository itself. To us, the fruits module will be used only as a library. In reality, the actual module and repository belongs to a vendor. For a long time now we have had developed and maintained our code in the vendor's repository. Recently, we yanked our code out and placed it in our own repository with the same module name. Our intent is to be able to checkout our vendor's code (that has the same module name) under a different name. I used the following version of the checkout command cvs -d :ext:[EMAIL PROTECTED]:/home/cvs -d fruits vegetables/grapes vegetables/oranges I had hoped that this would result in fruits/grapes fruits/oranges Instead the result is fruits/vegetables/grapes fruits/vegetables/oranges Problem: vegetables already exists in my working directory and must remain so. There's got to be a way to have checkout place a copy of grapes and oranges under fruits without the pesky vegetables getting in the way. Any help/advice would be greatly appreciated. Glen Luna Dyncorp IS LLC ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: NEWBIE question: checkout, sub-modules
Luna, Glen writes: I used the following version of the checkout command cvs -d :ext:[EMAIL PROTECTED]:/home/cvs -d fruits vegetables/grapes vegetables/oranges I had hoped that this would result in fruits/grapes fruits/oranges Instead the result is fruits/vegetables/grapes fruits/vegetables/oranges There's got to be a way to have checkout place a copy of grapes and oranges under fruits without the pesky vegetables getting in the way. I think it will work the way you want if you only checkout one thing at a time. CVS isn't smart enough to figure out how much it can shorten names without getting into trouble when you checkout more than one thing at a time, so it just doesn't shorten them at all. -Larry Jones Ha! Wild zontars couldn't drag that information out of me! Do your worst! -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: NEWBIE question: checkout, sub-modules
Thanks Larry but it looks like what I want can't be done regardless of the number of submodules I attempt to checkout. I also forgot to mention that grapes and oranges are subdirectories (rather than simple objects) of vegetables and would be subdirs under fruits. We use a simple implementation of cvs (1.9 and 1.10) and have not made full use of the admin files found in CVSROOT. I was hoping for an answer that would lie within the use of those files. I'm bummed but thanks anyway. Glen -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 25, 2001 4:01 PM To: Luna, Glen Cc: [EMAIL PROTECTED] Subject: Re: NEWBIE question: checkout, sub-modules Luna, Glen writes: I used the following version of the checkout command cvs -d :ext:[EMAIL PROTECTED]:/home/cvs -d fruits vegetables/grapes vegetables/oranges I had hoped that this would result in fruits/grapes fruits/oranges Instead the result is fruits/vegetables/grapes fruits/vegetables/oranges There's got to be a way to have checkout place a copy of grapes and oranges under fruits without the pesky vegetables getting in the way. I think it will work the way you want if you only checkout one thing at a time. CVS isn't smart enough to figure out how much it can shorten names without getting into trouble when you checkout more than one thing at a time, so it just doesn't shorten them at all. -Larry Jones Ha! Wild zontars couldn't drag that information out of me! Do your worst! -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: NEWBIE question: checkout, sub-modules
Luna, Glen writes: Thanks Larry but it looks like what I want can't be done regardless of the number of submodules I attempt to checkout. I also forgot to mention that grapes and oranges are subdirectories (rather than simple objects) of vegetables and would be subdirs under fruits. We use a simple implementation of cvs (1.9 and 1.10) and have not made full use of the admin files found in CVSROOT. I was hoping for an answer that would lie within the use of those files. Well, you *can* do it in the modules file. If you add an entry like: fruits vegetables then you can do: cvs co fruits/grapes fruits/oranges and get what you want. If you always want the same subdirectories, you can include them in the module definition as well: fruits vegetables grapes oranges then you can just do: cvs co fruits and you'll get fruits/grapes and fruits/oranges. -Larry Jones I always send Grandma a thank-you note right away. ...Ever since she sent me that empty box with the sarcastic note saying she was just checking to see if the Postal Service was still working. -- Calvin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: Branch question
Commit your changes on the main trunk, then move the branch points to the newly commited release number. This usually only works for a single file. There once was a tool on Molli's CVS page which helped with this kind of applying the same modification to several branches at the same time. I think it's name was cvslines or something similar. And, BTW, moving branch tags is not obvious, but it is described somewhere in the doc. Rolf [EMAIL PROTECTED] -Original Message- From: Prakash Ranade [SMTP:[EMAIL PROTECTED]] Sent: Tuesday, April 24, 2001 2:53 AM To: [EMAIL PROTECTED] Subject: Branch question Hello CVS gurus, I have a question with branches. foo.c /*- Branch 2.5 / / ---*-- Main \ \ \*-- Branch 2.2 I am currently working on Branch 2.2 's foo.c. The file's contents are same as on Main and Branch 2.5, in other words foo.c is never got changed since we branch out. Now, My question is the changes which I made on foo.c needs to be reflected on Main, and Branch 2.5's foo.c, without checkout - update - commit cycle. Thanks in advance. Prakash ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Branch question
Across the ether fly the words of Prakash Ranade: Now, My question is the changes which I made on foo.c needs to be reflected on Main, and Branch 2.5's foo.c, without checkout - update - commit cycle. Another idea is to create a diff of the changes to foo.c, using cvs diff, then patch the checkouts of foo.c from the trunk and branch 2.5 using patch. I think this is essentially what a cvs merge (cvs update -j) would do, except you would not need to commit the foo.c changes first. gary ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: rename question -- urgent!!
It seems like there is only one thing I can do, that is: rename the file on branch A (old.txt - new.txt) before merging A into trunk. Does it make any differences if the content of old.txt is NOT modified on Branch A? Thanks a lot -Susie -Original Message- From: David L. Martin [mailto:[EMAIL PROTECTED]] Sent: Thursday, April 12, 2001 7:38 PM To: Chen, Susie; [EMAIL PROTECTED] Subject: Re: rename question -- urgent!! - Original Message - From: Chen, Susie [EMAIL PROTECTED] I have two branches (A and B) off the HEAD. Both branches have a file old.txt. What I have done is: 1) On branch B, rename the file from old.txt to new.txt by removing,adding and then committing 2) merge branch B into HEAD At this point, the trunk has old.txt removed and new.txt added. Next I am planning to merge branch A into HEAD. But the filename in A is still old.txt , and the file content is modified. I hope to merge the change into HEAD and keep the new.txt filename on the HEAD as well. The branch A change to old.txt will be disregarded in the merge to the trunk since old.txt no longer exists there. [It won't be completely disreguarded: you'll get an error message telling you that old.txt does not exist but is present on branch A. -Larry Jones] Is CVS smart enough to deal with it? No, CVS will not help you here. The smart must come from the developer in this case. CVS makes no association between old.txt and new.txt; as far as CVS is concerned, these are two unrelated files. David Martin ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Branch question
Hello CVS gurus, I have a question with branches. foo.c /*- Branch 2.5 / / ---*-- Main \ \ \*-- Branch 2.2 I am currently working on Branch 2.2 's foo.c. The file's contents are same as on Main and Branch 2.5, in other words foo.c is never got changed since we branch out. Now, My question is the changes which I made on foo.c needs to be reflected on Main, and Branch 2.5's foo.c, without checkout - update - commit cycle. Thanks in advance. Prakash ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
