Red Hat Bugzilla – Bug 468479
Extra output seems to confuse resizing in anaconda
Last modified: 2013-01-09 22:24:17 EST
When attempting to resize an LVM LV (shrink it down) I got this error screen. Unfortunately it's a live install and I don't think it's logged anywhere.
Running... ['resize2fs', '-p', '/dev/VolGroup01/LogVol00', '27344M']
Resizing the filesystem on /dev/VolGroup01/LogVol00 to 7000064 (4k) blocks.
Begin pass 3 (max = 292)
Scanning inode table ----------------------------------------XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
The filesystem on /dev/VolGroup01/LogVol00 is now 7000064 blocks long.
Anaconda thinks the resize failed.
*** Bug 468478 has been marked as a duplicate of this bug. ***
The output isn't the problem -- we say it failed based on the exit status being non-zero. I'm not seeing any cases of a successful resize returning zero from a quick look through the resize2fs code. Eric -- any thoughts?
(Also, adjusted the error so that it will at least tell us what was returned)
After the message above, in resize's main():
printf(_("The filesystem on %s is now %u blocks long.\n\n"),
the only possible return is (0), and I don't see any called functions between the two points which could lead to an exit().
So I don't get why it thinks it failed...? I guess it'll be interesting to see the error code ...
I can't duplicate this anymore :/