Hi, I couldn't find any obviously wrong with this, other then that as you said it will need a lot of testing. What I did notice is that in one place you have a udev_settle after a device.format.commit() call, but not in others. This seems like a bad idea to me, the commit() will always cause the partition table to be rescanned and thus all /dev/sdx# device nodes to disappear and re-appear. I think we should try to get on the safe side of things here and move the udev_settle to the DiskLabelFormat commit() method itself. This makes me wonder do we handle things the same way currently (so no udev_settle in commit()) ? That could explain a few bugs caused by missing device nodes. And I agree with Jesse this should not go in until the alpha is out. Regards, Hans On 08/10/2009 11:23 PM, David Lehman wrote:
This patch set moves the partition table code into DiskLabel, a subclass of DeviceFormat. This removes the implicit requirement that all disks (or any other device type) contain a disklabel. For consistency in the short term, we enforce this requirement explicitly from the device tree's device discovery methods. These patches lay the groundwork for partitioned md devices as well as whole-disk formats (eg: a filesystem, pv, or md component that occupies an entire disk without a partition table). I have done some testing, but much more will be needed. _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
_______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
