Hi,
On 01/12/2010 09:06 PM, David Lehman wrote:
This becomes important when removing a disklabel from a device
which has a disklabel and partitions but which is not something
we support partitioning (eg: lvm lv). In normal cases the
partitions will already have been removed, but in the case of
an unsupported partitioned device we ignore the partitions when
scanning and so have to remove them this way.
---
storage/formats/disklabel.py | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/storage/formats/disklabel.py b/storage/formats/disklabel.py
index e5fab94..567d26b 100644
--- a/storage/formats/disklabel.py
+++ b/storage/formats/disklabel.py
@@ -229,6 +229,7 @@ class DiskLabel(DeviceFormat):
if not os.access(self.device, os.W_OK):
raise DeviceFormatError("device path does not exist")
+ self.partedDisk.deleteAllPartitions()
self.partedDevice.clobber()
self.commit()
self.exists = False
I would rather not have that deleteAllPartitions() as currently
implemented in pyparted gets used. I've been advocating its removal
for quite some time now.
The problem is that deleteAllPartitions() calls a function in
libparted which not only removes the partitions from the disk, but
also frees the libparted (c-code) partition objects, so if we still
have a reference to any python partedPartition object from this disk
around, that python partedPartition object now has a c-level pointer
to "random" memory.
I realize that in this case we probably don't have any such references
around, but still I'm strongly against using this function (as said it
should be removed).
Regards,
Hans
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
