On Tue, 14 Oct 1997, Dale Scheetz wrote: > On Tue, 14 Oct 1997, Scott Ellis wrote: > > > Someone messed up with the libc5 in the latest stable. It conflicts with > > libc6 for no reason, if you install libc6 with --force-conflicts it should > > work fine. > This isn't strictly speaking true. Libc6 only conflicts with the libc5 in > bo. This is because there is a difference in the format of utmp/wtmp lock > files between the old libc5 and the new libc6. The version of libc5 in > hamm is patched to do the lock file format the same way that libc6 does > them. This seems more than trivial to me since forcing libc6 install will > corrupt utmp/wtmp lock files for one or the other process. Seems to me > this would be good to point out in the HOWTO. There really is a good > reason to upgrade to libc5 in unstable.
OTOH, utmp/wtmp corruption doesn't matter to some people, there are only a few programs that care. Yes the libc5 in stable has a different utmp/wtmp format, but other than that it works for people without them being forced into the altdev environment which can be confusing. > > You could also install the latest libc5 from unstable instead > > (better alternative if you intend to upgrade completely to unstable), but > > that is likely to force you down the altgcc/libc5-altdev path if you want > > to be able to compile libc5 stuff. > This shouldn't make any difference. Installing libc6-dev will force you > down that path though. The problem is that the libc5 from unstable conflicts with libc5-dev (needing libc5-altdev instead). That's as good a way as any to force you into altgcc/altdev usage. -- Scott K. Ellis <[EMAIL PROTECTED]> http://www.gate.net/~storm/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .