Date: Fri, 26 May 2000 20:10:02 -0400 From: Laurence Berland <stuyman@confusion.net> To: John Lawson <mage@channelpoint.com>, questions@freebsd.org Subject: Re: Can't login to the GUI ... login box comes right back Message-ID: <392F1259.30865AA8@confusion.net> References: <03E742431696D311BD1B000629382517D43519@cpex1.channelpoint.com>
next in thread | previous in thread | raw e-mail | index | archive | help
I've always been a fan of tar. cd / tar -clf - * |(cd /mnt/root;tar -xvf -) should do the same thing, but I'm not sure about everything. I don't use cpio much but I'm very confident in the handling of permissions and the like in tar. If you're able to (since the old virtual drive is still there) try this instead, let me know. So you don't have to man page thats... in the root directory. create a tar archive (c) don't cross mountpoints (l) the filename is next (f) - means stdout. Then send it through a pipe which after cding to /mnt/root will then extract a tar archive (x) and be verbose about it (v, don't use on first tar) and the file is next (f) - means stdin. let me know how it goes. Hoping this is helping, I'll send this to the list also and see what others say... Laurence John Lawson wrote: > > Thanks for the followup ... I will check into those files (they're new to > me). *NOTHING* should have changed, however, except, perhaps protections > and/or modification dates, which I thought were supposed to be taken care of > thus: > > $ find -x / -print | cpio -pdamv /mnt/root > > This says (according to MAN), list all files on the "/" file-system, without > dropping into other mount points (-x), and copy those files, with > protections, ownerships, and modification dates intact, creating directories > as needed (oooh, could that be the problem?), and list the names as they're > copied. > > Light bulb alert .... There was an option to CPIO which affects the order in > which directories are traversed ... I wonder if the command I set up caused > the subdirectories to be traversed *before* their parents were *copied*, > causing the -d option to create the directories as needed, which might cause > the directories to be created as if I'd done mkdir, which is different from > *copy*ing the directory. Hmmmmmm! > > I've found numerous incorrect protection masks (most blatantly, for > instance, the new /tmp came out with no g or o access). > > In VMS (where I'm a guru, in Unix, I'm only an egg), you copy a complete > disk, including proper-handling of softlinks, with one simple command > (BACKUP/IMAGE). Is there some equivalent in UNIX, or is what I did the > closest thing? > > John Lawson > > | -----Original Message----- > | From: Laurence Berland [mailto:stuyman@confusion.net] > | Sent: Thursday, May 25, 2000 6:09 PM > | To: John Lawson > | Subject: Re: Can't login to the GUI ... login box comes right back > | > | > | Has something changed in the shared init files for the shell? > | (such as > | /etc/csh.cshrc) That's all I can think of...I checked something and > | found that I don't have such a variable set, and yet kde > | works fine for > | me. What are you setting it to? (/usr/lib or the like?) I'm at a > | loss. Hmmm... > | > | Laurence > | > | John Lawson wrote: > | > > | > Thanks for responding ... > | > > | > Yes, there is a .xsession, and it contains one line: > | > > | > exec startkde > | > > | > I have since found .xsession-errors, and found errors > | indicating that my > | > LD_LIBRARY_PATH environment variable was not set. I looked > | through all the > | > various rc and profile files I could find, and even search > | the entire disk, > | > for anything setting the LD_LIBRARY_PATH variable, and I > | found nothing. > | > > | > When I hard-coded an appropriate setting for > | LD_LIBRARY_PATH into .xsession, > | > it started working ... for the account that I set it in. > | Of course, I'd > | > have to do that to all the .xsession's for all other users > | (including root). > | > > | > So that's clearly not the correct solution. And what > | troubles me is that > | > this all worked *before* I copied my system disk. > | > > | > My conclusion (as naiive as it may be) is that I broke something in > | > duplicating my system disk, which is, somehow, causing some of my > | > environment variables/setup to abort prematurely. > | > > | > It turns out that I did not, in fact, delete my original > | virtual system > | > disk, so I can go back and compare things to see what's > | different. One of > | > the first things that I've found is that the /tmp directory > | on the new > | > system disk had lost its g and o protections ... so, what > | else could I have > | > botched ... I thought that the cpio -pdamv would have > | copied the protections > | > correctly. > | > > | > Ideas? > | > > | > Mage > | > > | > | -----Original Message----- > | > | From: Laurence Berland [mailto:stuyman@confusion.net] > | > | Sent: Wednesday, May 24, 2000 6:12 PM > | > | To: John Lawson > | > | Cc: 'freebsd-questions@freebsd.org' > | > | Subject: Re: Can't login to the GUI ... login box comes right back > | > | > | > | > | > | very cursory guess...does the user logging in (presumably > | you) have a > | > | .xsession file? What's in it? That's the first thing > | that comes to > | > | mind. Also, permissions on the file... > | > | > | > | Laurence > | > | > | > | John Lawson wrote: > | > | > > | > | > I installed FreeBSD (current) on a 1Gb vmware virtual disk, > | > | and realized > | > | > before very long that I'd run out of space very quiickly. > | > | Rather than go to > | > | > two virtual disks, I decided to create a new empty 2Gb > | > | virtual disk, and > | > | > copy the contents of the original disk to the new disk, > | > | eventually deleting > | > | > the old disk. > | > | > > | > | > I did this ... the trickier parts were easy to accomplish using > | > | > /stand/systeminstall (nice utility!). The actual copying > | > | of the files I did > | > | > with the following command (I had only one, the root, > | filesystem): > | > | > > | > | > find -x / -print | cpio -pdamv /mnt/root > | > | > > | > | > I can boot just fine off the new disk, and have since > | > | deleted the old > | > | > virtual disk. The trouble I'm having, however, is with the GUI. > | > | > > | > | > If I comment out the launch of xdm on ttyv8 in /etc/ttys, > | > | and log into the > | > | > console, and then type startx, I get a nice GUI and > | > | everything seems to be > | > | > working fine ... but, I want the GUI to launch at boot > | > | time, of course. > | > | > > | > | > So ... I put the line for ttyv8 and xdm back into /etc/ttys > | > | (as documented > | > | > somewhere on http://www.freebsd.org/, and now I get the > | > | login/password > | > | > window just after booting. > | > | > > | > | > When I log into that window, however, the window disappears > | > | (as you'd > | > | > expect), the system thinks for a little while (20 seconds), > | > | and then the > | > | > login box comes right back. > | > | > > | > | > I can't find a log file anywhere which might be displaying > | > | the error ... > | > | > > | > | > Can anyone give me a pointer to where to look? If this > | > | were VMS, I could > | > | > definitely chase down the various log files to find the > | > | solution, but, alas, > | > | > I am new to this U*ix world, and need some help. > | > | > > | > | > John Lawson > | > | > > | > | > To Unsubscribe: send mail to majordomo@FreeBSD.org > | > | > with "unsubscribe freebsd-questions" in the body of the message > | > | > | > | -- > | > | Laurence Berland, Stuyvesant HS Debate > | > | > | <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> > | > | Windows 98: n. > | > | useless extension to a minor patch release for > | > | 32-bit extensions and a graphical shell for a > | > | 16-bit patch to an 8-bit operating system > | > | originally coded for a 4-bit microprocessor, > | > | written by a 2-bit company that can't stand for > | > | 1 bit of competition. > | > | http://stuy.debate.net > | > | icq #7434346 aol imer E1101 > | > | The above email Copyright (C) 2000 Laurence Berland > | > | All rights reserved > | > | > | > | -- > | Laurence Berland, Stuyvesant HS Debate > | <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> > | Windows 98: n. > | useless extension to a minor patch release for > | 32-bit extensions and a graphical shell for a > | 16-bit patch to an 8-bit operating system > | originally coded for a 4-bit microprocessor, > | written by a 2-bit company that can't stand for > | 1 bit of competition. > | http://stuy.debate.net > | icq #7434346 aol imer E1101 > | The above email Copyright (C) 2000 Laurence Berland > | All rights reserved > | -- Laurence Berland, Stuyvesant HS Debate <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> Windows 98: n. useless extension to a minor patch release for 32-bit extensions and a graphical shell for a 16-bit patch to an 8-bit operating system originally coded for a 4-bit microprocessor, written by a 2-bit company that can't stand for 1 bit of competition. http://stuy.debate.net icq #7434346 aol imer E1101 The above email Copyright (C) 2000 Laurence Berland All rights reserved To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?392F1259.30865AA8>