Re: Access to old "historic" CVS releases ?

2004-07-26 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: >> And as far as I understand the CVS archives of CVS itself aren't >> available either. > > They are. You can browse them on-line at cvshome.org by picking the > "Version control" link, or you can register as a user and then set up an > SSH tunnel for dire

sylinks in fs pointing to cvs files [owned by root]

2002-05-30 Thread Harry Putnam
I'm only a light user of cvs, and use it to keep track of my config files on home machines, as well as my home projects. One thing that seems convienient to me is to have basic system config files like /etc/hosts /etc/syslog.conf and quite a few more in cvs and only symlinks in the fs pointing to

automate some of it.. cvs commit

2002-02-05 Thread Harry Putnam
Setup: Single user repository home setup Light script devel and tracking of rc files Probably going at this all wrong. My usage has grown up kind of by accident. Whenever I needed to keep track of something, I'd figure out enough about cvs to get it committed and available. Overtime I

Re: obtain revision information

2002-01-28 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: >> >> How can I get the current revision number of a package (module)? > > CVS is a *file* versioning system, it does not have any concept of a > revision number for any larger package. Most people use t

obtain revision information

2002-01-28 Thread Harry Putnam
How can I get the current revision number of a package (module)? And given that information, check out a version several versions older? Running `cvs status ' in the checked out module gives me piles of data on every single file in the distro. If I run `cvs status gnus' (This is the gnus newsrea

Re: Import output - how to interpret

2001-12-29 Thread Harry Putnam
Duncan Sommerville <[EMAIL PROTECTED]> writes: >> cvs import produces various output line begining with different key >> letters (like N or I). None of the sections in my manual concerning >> `import' discuss this aspect. > > Hi, > >N -> New file (ie successfully imported) >I -> Ignored

Re: Removing directory from repository - where is its record kept?

2001-12-29 Thread Harry Putnam
Michael Sims <[EMAIL PROTECTED]> writes: > At 08:13 PM 12/28/2001 -0800, Harry Putnam wrote: >>Now when I run cvs update I get a message that the deleted directroy >>cannot be found. So thinking to edit the source of that message. > [...] >>Still it comes up. Wh

Import output - how to interpret

2001-12-28 Thread Harry Putnam
cvs import produces various output line begining with different key letters (like N or I). None of the sections in my manual concerning `import' discuss this aspect. I see a handy guide to interpreting `update' output. Is there something similar for `import'. __

Removing directory from repository - where is its record kept?

2001-12-28 Thread Harry Putnam
The cvs manual tells up we can't really remove directories from our repository but then goes on to tell how to do it. First advice is to cvs remove the files then just ignore the directory. Well I took the other approach. The one that isn't advocated... First delete then cvs remove any file in

More on root commits

2001-12-12 Thread Harry Putnam
Running a very recent (3days) ccvs. I want to commit as root. Its the easiest way on my setup. However following Larrys' advice about editing src/options.h is not working for me here. Run ./configure Then edit out the section in src/options.h about BADROOT. Completely remove those three lin

Re: cvs commit as root

2001-12-10 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: [...] > root is just another user as far as file ownership goes. If a file is > owned by joe and readable only by the owner, then no one other than joe > can read it (except for root) just like no one other than root can read > a root-owned file that is

Re: cvs commit as root

2001-12-10 Thread Harry Putnam
"Ralph A. Mack" <[EMAIL PROTECTED]> writes: [...] > Keeping a layer of makefile/script between your source control system > and your deployment areas gives you a greater control of file system > attributes of deployed files. If you are concerned about security you > may want this layer in place,

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: >> >> No, this connection is over ssh using CVS_RSH set to ssh and the ext >> syntax. Like: >> >>cvs -d :ext:HOST:/usr/local/cvsroot >> >> I don't actually use that synta

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
[EMAIL PROTECTED] (Greg A. Woods) writes: > [ On Saturday, December 8, 2001 at 07:19:10 (-0800), Harry Putnam wrote: ] >> Subject: Re: cvs commit as root >> >> Well, its not really true that it `isn't much different'. The copying >> step is really a ma

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: >> >> But cvs commit still results in: >> "/tmp/cvsc_ayxh" 8 lines, 296 characters >> cvs [server aborted]: cannot commit files as 'root' >> cvs commit: saving log message

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
[EMAIL PROTECTED] (Greg A. Woods) writes: > In the simplest case you could just blindly copy all the files from your > working directory (or exported release) and be done with it, which > really isn't much different than what you've been trying to do, except > that now you have to edit them in a

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
Harry Putnam <[EMAIL PROTECTED]> writes: > [EMAIL PROTECTED] (Larry Jones) writes: [...] >> Yes, that's a configuration problem. I recommend you fix that rather >> than changing options.h to allow anonymous root commits. > > That turned out to be not such a

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
Harry Putnam <[EMAIL PROTECTED]> writes: > [EMAIL PROTECTED] (Larry Jones) writes: > >> Harry Putnam writes: >> > >>> server.c:31:21: krb5.h: No such file or directory >>> make[3]: *** [server.o] Error 1 >> >> That's a

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: > > >> server.c:31:21: krb5.h: No such file or directory >> make[3]: *** [server.o] Error 1 > > That's a bug in the configure script that has been fixed in the current > development version.

Re: cvs commit as root

2001-12-08 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: > > >> Yes, it works like you describe on my linux boxes. But on a solaris 8 >> (intel) I have problems with it. Probably more due to local config >> than anything else. Plain su root on the solaris bo

Re: cvs commit as root

2001-12-07 Thread Harry Putnam
[EMAIL PROTECTED] (Greg A. Woods) writes: > [ On Thursday, December 6, 2001 at 22:17:21 (-0800), Harry Putnam wrote: ] > > Subject: cvs commit as root >> >> I'd like to override the taboo on committing as root. I imagine it is >> possible but not clear how to do

Re: cvs commit as root

2001-12-07 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: >> >> I'd like to override the taboo on committing as root. I imagine it is >> possible but not clear how to do it. > > You're only not allowed to commit as root if CVS can't figure out who

Re: cvs commit as root

2001-12-07 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > Harry Putnam writes: > > >> I'd like to override the taboo on committing as root. I imagine it is >> possible but not clear how to do it. > > You're only not allowed to commit as root if CVS can't figure out w

cvs commit as root

2001-12-06 Thread Harry Putnam
Setup: OS = Redhat Linux 7.1 CVS = Concurrent Versions System (CVS) 1.11.1p1 (client/server) Standalone home user situation I'd like to override the taboo on committing as root. I imagine it is possible but not clear how to do it. On my setup, I use cvs as much for having versi

Usage pattern - $HOME as cvs co - a broad outline please

2001-11-10 Thread Harry Putnam
Setup: OS = Redhat Linux cvs = 1.11.1p1 (client/server) Only an amatuer user of cvs, mainly keeping such things as machine config files, home made scripts, *rc files and the like in cvs repository. So far, I use this repository across three home machines. Over time, more and more things

Re: remote cvs connection analasys (debug)

2001-11-01 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: > That implies that the server has some kind of access control (perhaps > tcpd?) that is denying you access. You can find out how people are > connecting by asking them to tell you what's in CVS/Root in their > working directories. Yeah, that already came

Re: remote cvs connection analasys (debug)

2001-11-01 Thread Harry Putnam
[EMAIL PROTECTED] (Larry Jones) writes: [...] > Let me spell it out more completely: It's a server problem. Either the And did you ever... A very nice help, thank you. [...] > telnet cvs.myhost.com 2401 > > If the connection is immediately closed, then the server is > misconfigured (

Re: remote cvs connection analasys (debug)

2001-11-01 Thread Harry Putnam
"Gianni Mariani" <[EMAIL PROTECTED]> writes: > maybe you can try a tcp forwarding thingy. > >http://freshmeat.net/projects/tcp_forward/ > > and tweak it to dump the transaction. That looks promising .. thanks. If I knew more about tcpdump I guess I could find out something that way too.

Re: remote cvs connection analasys (debug)

2001-10-31 Thread Harry Putnam
Harry Putnam <[EMAIL PROTECTED]> writes: [...] > I wondered if there is a way to force verbosity, where the negotiation > get printed to tty or something similar so that I can see what is the > problem. The normal flags for that kind of stuff -[vV] are taken for > other t

remote cvs connection analasys (debug)

2001-10-30 Thread Harry Putnam
Setup: Redhat Linux 7.1 Cvs : 1.11.1p1 (client/server) Having trouble with a remote connection that has worked in the past. The server moved and ever since I cannot connect to it. I'm using the right address far as I know and other people are successfully connecting remotely to it. I am able t