Colin Watson wrote: > You explain to the common man not to use unstable. :) > > It *certainly* shouldn't have broken in the first place, but accidents > happen. If one doesn't have enough system administration experience to > cope with this kind of thing (after all, it was "just" everything > written in C++ that broke, not, say, the dynamic linker as I'm told used > to happen, or PAM preventing all logins, or ...), then unstable is > really not the distribution one should be running.
Yes, I remember the PAM incident back in early 2001. That was much nastier than this C++ problem. One had to switch to single user mode, then download and install a corrected package from incoming. > This may sound callous, but those "some people" - or at least those > people who *can* fix it, perhaps not trivially easily - are the only > people who should be using unstable. I have to agree. I hit the C++ problem yesterday afternoon when I did my daily Sid update. It was annoying, but not that hard to work around. I considered symlinking it to the nearest-match version of the same library, but decided it was probably a more certain fix to just downgrade to the previous libstdc++2.10-glibc2.2 package. Anyone who couldn't figure out how to do that without help probably should stick to testing rather than unstable. Otherwise, the next time something nasty happens in Sid, they may find that they don't know what to do, and all the usual resources (Debian lists and web pages) are unreachable because of the breakage. What then? Craig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
