Bug 214129
Summary: | yum wedges the rpm database quite often in FC6 | ||
---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Filip Miletic <filmil> |
Component: | yum | Assignee: | Jeremy Katz <katzj> |
Status: | CLOSED DUPLICATE | QA Contact: | |
Severity: | high | Docs Contact: | |
Priority: | medium | ||
Version: | 6 | CC: | james, orion, thorjansen |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | i686 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2007-01-03 18:05:21 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Filip Miletic
2006-11-06 01:57:08 UTC
Does running rpm --rebuilddb help? (Save a copy of /var/lib/rpm first, so that if it does we can take a look at what you previously had and see if we can figure out what happened) Unfortunately, I already ran 'rpm --rebuilddb' without making a backup so I am not able to help further. The --rebuilddb is of course only possible once one removes the __db.* files that left over in the wedged rpm database (/var/lib/rpm/__db.*). I made 4 or 5 test installs after rebuilding the database, which gave no problem. The only thing I can do now is report back should the problem occur again. I can confirm that this happens regularly to me too. FWIW, I also update from livna. (In reply to comment #3) > I can confirm that this happens regularly to me too. FWIW, I also update from livna. Also, this happens whether using pup, yumex or command line yum, especially if I try to update or install too many files at once. Sometimes two files at a time works, sometimes not, and sometimes it bombs on just one file. It's almost as if yum is racing and overruns __db* when it writes to the database. It always shows up after a package that trips the error is installed, but before cleanup. Anyway, here is some output from yum after the database gets corrupted: rpmdb: PANIC: fatal region error detected; run recovery error: db4 error(-30977) from db->close: DB_RUNRECOVERY: Fatal error, run database recovery Error line is repeated several times. OK, after experimenting, I believe this is a rpm/db4 problem, not yum. I tried all the yum wrappers (yumex, pup, etc.) and command-line yum, and get the same problem. Then I tried to install vmware workstation from an rpm off my hard drive, and it bombed pretty much the same way. Since vmware needs post processing to complete installation, I can't get this package installed because that postproc never runs. OK, just one more comment, then I'm done here. This looks like a dupe of bug 212504 because the folks over there are having similar, if not identical, problems with rpm and db4 (and yum). Just in case it helps: 1.- I was suffering this problem, exactly as is described above 2.- No "ulimit -c unlimited" or so has made a core dump that helps to debug 3.- Command "rpm --rebuilddb" has run slowly but has ended properly. It doesn't solve the problem, by the way 4.- Finally, I had use those "rm -f __db*" at /var/lib/rpm and after it, a "yum update" has ended properly. Normal rpm's commands take a lot of time... and don't finish ok.I use defaults repos in Fedora Core 6 plus Freshrpms. My workkstation has been upgraded since Fedora Core I 'till VI (formerly it was an RH9). Possible duplicate: bug #213986 *** This bug has been marked as a duplicate of 212504 *** I haven't seen this recently with updated systems. Can anyone else comment? |