Red Hat Bugzilla – Bug 601158
PackageKit - TypeError: Can not create packages table: database is locked
Last modified: 2010-07-16 14:48:16 EDT
Description of problem:
on rhel6 machine i noticed traceback (see bellow) regarding locked database
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Traceback (most recent call last):
File "/usr/share/PackageKit/helpers/yum/yumBackend.py", line 2437, in get_updates
ygl = self.yumbase.doPackageLists(pkgnarrow='updates')
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 1914, in doPackageLists
for (n,a,e,v,r) in self.up.getUpdatesList():
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 793, in <lambda>
up = property(fget=lambda self: self._getUpdates(),
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 621, in _getUpdates
self._up = rpmUtils.updates.Updates(self.rpmdb.simplePkgList(), self.pkgSack.simplePkgList())
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 774, in <lambda>
pkgSack = property(fget=lambda self: self._getSacks(),
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 564, in _getSacks
File "/usr/lib/python2.6/site-packages/yum/repos.py", line 279, in populateSack
sack.populate(repo, mdtype, callback, cacheonly)
File "/usr/lib/python2.6/site-packages/yum/yumRepo.py", line 187, in populate
dobj = repo_cache_function(xml, csum)
File "/usr/lib64/python2.6/site-packages/sqlitecachec.py", line 46, in getPrimary
TypeError: Can not create packages table: database is locked
I'm pretty sure this is a bug caused by the huge number of not widely used yum plugins installed. I can't reproduce on the vnc instance you've got open, but if you can work out what plugin is causing this failure we can blacklist it in PackageKit. I'm pretty sure it's a problem in yum-metadata-parser, so I'll reassign to yum for feedback and any ideas.
The only other time I've seen something like this was when /var/cache was on NFS, and rpc.lockd wasn't working ... thus. sqlite couldn't lock the .sqlite file (which it does on creation).
The vnc session seems dead (I can connect, but not type anything) so I can't check if that matches what the machine looks like.
Given it can't be reproduced, and AFAIK it's unlikely to be a yum/y-m-p bug anyway ... going to close this out.