Hi,
One more thing.
On 01/21/2010 01:03 AM, Peter Jones wrote:
<snip>
@@ -3026,21 +3024,71 @@ class MultipathDevice(DMDevice):
else:
self.parents.append(parent)
- def setup(self, intf=None):
- if self.status:
- self.teardown()
- self._isUp = True
- parents = []
- for p in self.parents:
- parents.append(p.path)
- self._pyBlockMultiPath = block.device.MultiPath(*parents)
+ def setupPartitions(self):
+ log_method_call(self, name=self.name, kids=self.kids)
+ rc = iutil.execWithRedirect("kpartx",
+ ["-a", "-p", "p", "/dev/mapper/%s" % self.name],
+ stdout = "/dev/tty5",
+ stderr = "/dev/tty5")
+ if rc:
+ raise MPathError("multipath partition activation failed for '%s'" %
+ self.name)
+ udev_settle()
+
In case you didn't see it in BZ already kpartx currently has a bug where it
sets up logical partition in such a way that parted cannot (successfully) modify
the mappings when making changes to the table. See:
https://bugzilla.redhat.com/show_bug.cgi?id=557065
So either you need to use pyparted + pyblock here to DIY (it can be advocated
this is best anyways as it also means that the same partition table parsing code
gets used as for other disk types, where as kpartx has its own
partition table parsing code), or clone this bug for RHEL-6 and make it block
the Beta I guess.
Regards,
Hans
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
