Well, not necessarily. The swap issue is the same as the rootfs one-- at some point multipath times out, and other init jobs would also timeout trying to bring up the swap (which can't be identified from the UUID) whereas the rootfs got brought up on a single path using the UUID.
The debdiff looks quite good; I'll add in the removal of scsi_wait_device which is just useless (that module no longer exists).
Well, not necessarily. The swap issue is the same as the rootfs one-- at some point multipath times out, and other init jobs would also timeout trying to bring up the swap (which can't be identified from the UUID) whereas the rootfs got brought up on a single path using the UUID.
The debdiff looks quite good; I'll add in the removal of scsi_wait_device which is just useless (that module no longer exists).