Red Hat Bugzilla – Bug 89519
rpm hangs followed by rpm segfault on -qa, --rebuildb,
Last modified: 2007-04-18 12:53:16 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20021003 Description of problem: This is a RH8.0 systems with a "complete" installation. I was experiencing similar symptoms to previous bugs in which rpm would hang on -Uvh or Fvh. I killed the rpm process each time and only later learned that I also needed to ensure the db lock files were removed for proper operation. However, I have since done that... [root@kokua i386]# rm /var/lib/rpm/__db* rm: remove regular file `/var/lib/rpm/__db.001'? y rm: remove regular file `/var/lib/rpm/__db.002'? y rm: remove regular file `/var/lib/rpm/__db.003'? y ...and now any operation with rpm results in a segmentation fault. For example: rpm -vv --rebuilddb Segmentation fault rpm -vv -qa Segmentation fault An strace can be seen at http://www.ldeo.columbia.edu/~vschmidt/rpmqatrace if that's any help. Version-Release number of selected component (if applicable): rpm-4.1-1.06 How reproducible: Always Steps to Reproduce: 1.rpm -qa is all that's required 2. 3. Additional info: Although I see the problem every time, I'm not sure what steps were required to get me here - so in that sense I haven't been able to reproduce it. I'm grasping at straws to understand what causes these kinds of errors.
Can you give me a pointer (i.e. URL, attachments won't work) to a tarball of your database cd /var/lib tar xzvf /tmp/rpmdb-89519.tar.gz rpm and I'll take a look? Thanks.
Yes. /var/lib/rpm tar'd and zipped and can be found at www.ldeo.columbia.edu/~vschmidt/rpmdb.tar.gz
OK, your database looks fine, all header signatures verify, so something else is going on. The strace indicates segfault before reaching main in rpm, which seems to point at something other than rpm as the cause. Do any rpm commads work? Try rpm -V glibc kernel possibly other commands as well, and tell me what works or not, please.
It turns out the cause of this didn't have anything to do with rpm directly. Early on, rpm (4.1.1) had hung and had to be killed. It appears that some, but not all, of the glib* rpms were updated during that process. This seems to have caused the ensuing segmentation faults with rpm and (other processes on the system it would turn out). I force updated the glib* rpms from the RH8.0 basic distro with rpm -Uvh --force glib* The segmentation faults are gone. OF NOTE: There were two errors when executing the above (in a trigger script and a post install script). Although this may bite me in the bottom, at first look, they appear to be inconsequential. To be sure, I've executed an rpm - qa --info > rpmlonglist, and when I have time I'll try to figure out if my forced install has broken any dependancies. Thanks for everyone's help. -Val