+++ This bug was initially created as a clone of Bug #228221 +++ I did a clone of 228221 because it sounds very similar, I'll try to describe my problem in some detail. I installed FC6 fresh on a Dell Latitude C600 recently. Everything went smoothly. When I rebooted, I got the expected message that I had 106 updates available and asked it to start installing them. The dependencies check runs fine, the packages get downloaded, and the system says everything installed successfully. But when I try to run it again (or the "Install Software" app), I get the "rpmdb open fail" error below: " Component: Software Updater Summary: TB6b5e0e87 config.py:661:_getsysver:TypeError: rpmdb open failed Traceback (most recent call last): File "/usr/sbin/pup", line 659, in ? main() File "/usr/sbin/pup", line 652, in main pup = PackageUpdater() File "/usr/sbin/pup", line 90, in __init__ GraphicalYumBase.__init__(self, False) File "/usr/lib/python2.4/site-packages/pirut/__init__.py", line 124, in __init__ plugin_types=(yum.plugins.TYPE_CORE,)) File "/usr/lib/python2.4/site-packages/yum/__init__.py", line 125, in doConfigSetup self.conf = config.readMainConfig(startupconf) File "/usr/lib/python2.4/site-packages/yum/config.py", line 598, in readMainConfig yumvars['releasever'] = _getsysver(startupconf.installroot, startupconf.distroverpkg) File "/usr/lib/python2.4/site-packages/yum/config.py", line 661, in _getsysver idx = ts.dbMatch('provides', distroverpkg) TypeError: rpmdb open failed Local variables in innermost frame: installroot: / ts: <rpmUtils.transaction.TransactionWrapper instance at 0x9f25a4c> distroverpkg: redhat-release " If I reboot, I get a fresh message saying exactly [N - 1] updates are available (i.e. 105, 104, etc.). I run it and it downloads everything (again) and "successfully" installs, but apparently only one package. I tried selecting only the pirut and yum packages, which seemed to update OK, but I still have the same problem. As described in the comments for the other bug, I tried deleting the db files, which had the same affect as a reboot (allows me to go, but only one package installs, then next time I try to run I crash). Needless to say I don't want to manually do each of 102 (left) packages, especially since it seems to re-download them each time. I've attached my db files in case they're getting corrupted somehow. Help in fixing or at least getting around this bug would be much appreciated. I'm not a savvy Linux guy but I can get by, so if you need more info, let me know how to get at it and I will. *** OLDER BUG (LIKELY RELATED) *** Description of problem: i get this error when i updated awhile back but it went away but still no updates have succeded in installation even though it says it does i also get this error message when i install an rpm Version-Release number of selected component (if applicable): How reproducible: i think it happened when i did a shutdown after a update yum install didnt shutdown Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: -- Additional comment from thatfarmerguy on 2007-02-11 16:32 EST -- Created an attachment (id=147865) bug dump -- Additional comment from n3npq on 2007-02-12 07:46 EST -- Try doing rm -f /var/lib/rpm/__db* -- Additional comment from pnasrat on 2007-03-13 07:07 EST -- Is this still occuring.
Created attachment 150550 [details] Zipped tar of db files that may be causing problems
Running yum with the -t (tolerate errors) option seemed to get through more than one package before throwing an exception. I think it got about 6 (went from 100 updates remaining to 94), then threw the same type of PANIC messages. Also, the exception occurs immediately after installing a package successfully if I run Add/Remove Software rather than the auto-update selection. Removing a package has the same symptoms.
I also ran fsck and badblocks on the disk to make sure I don't have any bad sectors thrown into the mix.
This is fixed with newer kernels
Newer as in within the last week? I ask because I just downloaded and installed FC6 from the ISO CD images on the mirror sites two weeks ago, so I'm surprised I'm working with an out-of-date kernel... I'm not hugely savvy on the Linux kernel topic in general, so is that normal to have an old buggy kernel downloaded from the site? Is there a specific upgrade process I need to follow that I might have missed? Glad to hear it's a known, fixed bug though.
Problem solved! Turns out the problem was that Anaconda installed the i586 kernel the first time around. I tried updating it, but that didn't seem to resolve the issue. So I did a clean re-install and Anaconda selected i686 correctly this time, and everything else seems to have worked out fine. The first thing I did was upgrade the kernel using yum, which had to install multiple packages and didn't have any problem, so I'm pretty sure it was the i586 kernel that caused the grief.