Red Hat Bugzilla – Bug 12281
(python traceback) filesystems not re-mounted on a second attempt upgrade
Last modified: 2007-04-18 12:27:16 EDT
After one attempt at an upgrade that failed (see bug 12278), I killed off
some stuff in /mnt/sysimage from vc2 (note that in bug 12278 the unmounting
worked for all filesystems except /mnt/sysimage itself) then clicked "Next"
again at the "About to upgrade" screen that the GUI install was in. This
re-ran the "Transferring image to hard drive" but the GUI install "locked"
after that with the buttons (Hide Help, Back, Next) all greyed out and
inactive. The status shows 420 packages to upgrade of 623MB in size.
vc2 df shows that the other filesystems never got re-mounted on the second
pass attempt. vc1 shows a python traceback:
Exception in thread Thread-4:
Traceback (innermost last):
File "/usr/lib/python1.5/threading.py", line 376, in __bootstrap
File "/usr/lib/anaconda/iw/progress_gui.py", line 16, in run
rc = self.todo.doInstall ()
File "/usr/lib/anaconda/todo.py", line 1368, in doInstall
File "/usr/lib/anaconda/fstab.py", line 434, in umountFilesystems
isys.umount(mntPoint, removeDir = 0)
File "/usr/lib/anaconda/isys.py", line 113, in umount
mountCount[what] = mountCount - 1
TypeError: bad operand type(s) for -
verified ... added to the winston scrub list ... thanks for the report!
*** This bug has been marked as a duplicate of 12106 ***