FWIW, I have noticed that on very rare occasions rpm has gone into "hang" mode (usually when installing a new glibc). The only way to clear it was to do a "killall -9 rpm" from another screen. This has only happened on the smp boxen running RH 8.0 or Rawhide, however.
After some playing --er-- experimentation, I found the only cure was a reboot. Rebuilding the rpm data base didn't help. If your symptom is the same as mine, "rpm -q" will not return any results unless you exit root and perform the query from an unprivileged user account. After the reboot, rpm returns to normal under the root account. The last time it happened I got interested enough to try to trace it. However, as with all rare intermittents, I couldn't get the bug to reproduce, so no joy at trying to fingerprint it. > On Tue, 2003-02-04 at 15:50, John Salamone wrote: > > > On Tue, 2003-02-04 at 15:29, John Salamone wrote: > > > > The remove command didn't work. It came back and said rm: > > > > invalid > > option -- > > > > / > > > > > > > > rm -f /var/lib/rpm/__db* > > > > > > Try again and make sure you enter the command exactly as it is > > > above. Cut and past it. > > > > same result as before > > Well, then, I'm not sure what to tell you. If I cut and paste that > command, I don't get errors. There is only one option to rm, and that > is "-f". > > It may be that there's some problem cut-n-pasting between an old X > application and a new terminal (with the UTF-8 changes, I've seen it > happen), but somehow or other you're going to need to type that > command and remove those files. -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
