Created attachment 800158 [details]
.tgz of various data: GPT repair log, .bash_history tail, fstab,
/boot/default/grub, grub.cfg, efibootmgr -v, parted -l, recent journal files,
dmidecode, more

The repair log contained in the attachment contains a visual record of sector 0
prior to successfully regenerating the partition table. Several sectors
following looked similar.

Original summary:
15.1beta zypper dup from 416.2 (20190219) to 426.2 (20190311) trashed GPT

For the record, I'm filing this so that the cause might be determined, and
fixed. I'm guessing intentionally not having the ESP partition mounted plays a
significant part. Intentional recreation I'd be unlikely to attempt even if
possible. I've never before had a partition table trashed by an upgrade. Fixing
was easy only after considerable time was spent determining how to fix.

Short form "creation" procedure and results:
1-zypper dup 15.1beta on sda9 from 416.2 to 426.2
2-reboot
3-select 15.1 stanza from default UEFI TW Grub custom.cfg on sda7

Actual behavior:
1-UEFI detours into setup
2-GPT sectors are mostly filled with something other than boot code
3-no 55 AA at end of sector 0
4-no protective table entry
5-nothing to boot from

https://forums.opensuse.org/showthread.php/535286-15-1-dup-to-426-2-from-416-2-trashed-GPT-table
continues my problem description initialized at
https://forums.opensuse.org/showthread.php/535180-Leap-15-1-Beta-Please-post-any-initial-comments-here?p=2897023#post2897023

Motherboard: https://www.asus.com/us/Motherboards/PRIME-B250M-C-CSM/