i realized shortly after i posted this to the FC test list that it
really deserved to go here instead. thoughts?
---------- Forwarded message ----------
Date: Wed, 22 Sep 2004 07:34:24 -0400 (EDT)
From: Robert P. J. Day <rpjday@xxxxxxxxxxxxxx>
Reply-To: For testers of Fedora Core development releases
<fedora-test-list@xxxxxxxxxx>
To: Fedora Test List <fedora-test-list@xxxxxxxxxx>
Subject: python exception trying to reuse LVM partitioning
yesterday, i started an FC3t2 install on a little latitude laptop, and
formatted the single drive as a couple physical volumes, and various logical
volumes within. part of it:
LV "system":
/tmp 256M (LV name "tmp")
/var 256M (LV name "var")
and so on. nothing fancy, your basic /, /usr and so on LVs. the installation
was partway through when the NFS server serving the ISO images was shut down.
bummer.
so i came in this morning, started over from scratch (NFS server was running
again), got all the way up to manually partitioning with disk druid and, of
course, the LVM partitioning was still there from the day before.
the LVs were still there, but they had obviously lost the idea of their own
mount point, so i figured, maybe i can just double click on each LV, an edit
window will come up, i can add the mount point back in and the fact that it
should be formatted. nothing to lose, i figured.
so i double clicked on the "tmp" entry in the "Disk Setup" window ... and got
a python exception traceback. just copying from the display manually, the
traceback routine list is:
self.editCb()
self.editLVMVolumeGroup(vgrequest)
...
now, even if i can't reuse the LVM partitioning *that* way (by double
clicking on each LV and restoring the missing info), i'm pretty sure i
shouldn't get a traceback. surely there's a more graceful response to this,
no?
thoughts?
rday
