Red Hat Bugzilla – Bug 63439
Bad: up2date corrupted rpm database
Last modified: 2008-05-01 11:38:02 EDT
From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0; Q312461) Description of problem: I had about 15 updates in up2date. I updated them. Right after some gnome component was installed the db problems started. Right now I'm at a loss. The rpm dbs seem trashed. Version-Release number of selected component (if applicable): How reproducible: Didn't try Additional info:
Created attachment 53764 [details] zipped tar archive of corrupt /var/lib/rpm directory
Unfortunately, an rpmdb is too large for bugzilla attachments. Describe "trashed" in detail please, so that I can tell you what to do.
Well, the upload worked well. Must've been about 5 megs. Anyway, this is what I get: 1) When I do an rpm -qa: kudzu-0.99.23-1 error: rpmdb: damaged header instance #116 retrieved, skipping. apmd-3.0final-34 lokkit-0.50-6 vixie-cron-3.0.1-63 ghostscript-fonts-5.50-3 mpage-2.5.1-7 perl-HTML-Tagset-3.03-3 perl-Parse-Yapp-1.04-3 perl-XML-Encoding-1.01-2 perl-libxml-enno-1.02-5 error: rpmdb: damaged header instance #403 retrieved, skipping. error: rpmdb: damaged header instance #415 retrieved, skipping. ... ... Finally, it segfaults. 2) rpm --rebuilddb -v gives me this: error: rpmdb: damaged header instance #116 retrieved, skipping. error: rpmdb: damaged header instance #403 retrieved, skipping. error: rpmdb: damaged header instance #415 retrieved, skipping. *** dbiUpdateRecord: uh is NULL Segmentation fault Cool, eh? 3) Trying a db_dump Packages-ORIG | db_load Packages just hangs (at least it was running at 100% for the last 15 hours without a result) Thanks for helping
OK. If you give me a pointer (i.e. URL, attachments won't work) then I'll try to save what I can. Meanwhile, what's different about your machine from other? Is it an SMP machine? What CPU? What kernel? What else was happening while you were running up2date? I'm trying for info that might allow reprodcing the problem.
I don't really know what might've been different, except for that on this system I didn't run it for a while. Right now it's giving me creeps when thinking about running up2date on more important production systems with RH 7.2, since I read about similar problems already elsewhere Actually up2date and everything else ran just fine until the error occured. Here is the file: http://roth.lu/rpm.tar.gz +++ CPUINFO processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 8 model name : Pentium III (Coppermine) stepping : 6 cpu MHz : 804.034 cache size : 256 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx fxsr sse bogomips : 1605.63 +++ KERNEL Linux version 2.4.9-31 (bhcompile@daffy.perf.redhat.com) (gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-98)) #1 Tue Feb 26 07:11:02 EST 2002 You can have a login for that machine if you like. Mail me privately in that case. Thanks
Thanks for the info. Hmmm, db3_dump and db3_load had problems with rpm/Packages. So did my my delete program. db4 worked fine, removing instance #582 (ghostscript?) eliminated the segfault. A rpm --rebuilddb fixed the rest. There's a repaired copy of your rpm/Packages at ftp://people.redhat.com/jbj/rpmdb-63439-FIX.tar.gz Download and install by doing cd /var/lib mv rpm/Packages rpm/Packages-ORIG tar xzvf rpmdb-63439-FIX.tar.gz rpm --rebuilddb -vv Nope, dunno what happened.