Trying to use the virt-clone script results in: # virt-clone -o centos4-i386 -n centos4-i386-clone -f /home/vm/centos4-i386-clone.img ERROR: Disk size must be an int or a float. What would you like to use as the disk (path)? python-virtinst-0.300.2-2.fc8.noarch
Applied a fix for this upstream: http://hg.et.redhat.com/virt/applications/virtinst--devel?cs=611782118e76
Excellent. :) Any idea when a new fedora update will be out? The status of POST here seems odd... isn't that for kernel patches?
Yeah the whole bug state thing confuses me. It was recommended to me to put bugs in POST when they are fixed upstream but not released yet. Seems to me to be a sensible state between ASSIGNED (developer is aware and potentially working on it) and MODIFIED (fix is in cvs). Since we really only track virtinst releases and usually don't apply specific fixes to cvs, MODIFIED doesn't seem appropriate. As for the update, not sure. In the interim, the fix is a one liner, just make the above change to /usr/lib/python2.5/site-packages/virtinst/CloneManager.py
Yeah, no biggie. I would probibly just leave it ASSIGNED in that situation. ok. Easy to do, and glad it's an easy fix. :)
The BZ docs saying POST is only for kernel patches is wrong. Since we have a close relationship between Fedora packages & upstream for all virtualization userspace, we always put tickets into POST when the patch is posted & merged upstream. This lets us clearly distinguish BZ which have patches available now (POST) and those which still require work (ASSIGNED). The normal workflow for all virtualization bugs is NEW -> ASSIGNED (dev working) -> POST (patch ready) -> MODIFIED (patch in CVS & Koji) -> ON_QA (in updates-testing) -> CLOSED (in updates/rawhide)
This was fixed way back in March in 0.300.2-4.fc8, so closing.