On Wed, 2011-02-23 at 07:50 -1000, David Cantrell wrote:
> Followup to previous fixes for /boot on LVM for s390x. When the
> underlying partitions are on a multipath volume, the device found by
> zipl is a dm-X device. The code correctly picks this up so we can pass
> it to chreipl, but we also try to run getDeviceByName() on the dm-X
> name, which is unnecessary here. When getDeviceByName() returns None,
> return a 2-tuple of (dm-X, None) to the caller so we use dm-X as-is.
We don't want to resolve dm-X to a device-mapper name and use that to
look up the device? I guess it might not matter as long as we aren't
writing /dev/dm-X to any files anywhere. If it's just for more-or-less
immediate use this patch looks fine.
Dave
>
> Related: rhbz#670856
> ---
> booty/util.py | 5 ++++-
> 1 files changed, 4 insertions(+), 1 deletions(-)
>
> diff --git a/booty/util.py b/booty/util.py
> index 48c3194..185c2af 100644
> --- a/booty/util.py
> +++ b/booty/util.py
> @@ -2,7 +2,10 @@ import string
> from flags import flags
>
> def getDiskPart(dev, storage):
> - path = storage.devicetree.getDeviceByName(dev).path[5:]
> + device = storage.devicetree.getDeviceByName(dev)
> + if device is None:
> + return (dev, None)
> + path = device.path[5:]
> cut = len(dev)
> if dev[-1] in string.digits:
> if (path.startswith('rd/') or path.startswith('ida/') or
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
