Hi, i, too, wondered where there should be a duplicate serial number. But indeed:
David Wright wrote: > > /dev/sdi 53 /dev/disk/by-id/ata-Gigastone_SSD_GST02TBG221146 > > /dev/sdj1 54 /dev/disk/by-id/ata-Gigastone_SSD_GST02TBG221146-part1 > ↑↑↑↑↑↑↑↑↑ that is /really/ bad! Does the number of 4 device files /dev/sd[h-k] match the number of installed ata-Gigastone_SSD devices ? Gene talked of "5, ordered in 2 separate orders". (Looking at https://lists.debian.org/debian-user/2024/01/msg00667.html) Now we see 3 to 4, depending on what one wants to believe. Wild ideas: One possible reason could be that a device is mapped to both, /dev/sdi and /dev/sdj. udev would then suffer a race condition when creating the /dev/disk/by-id. Another could be that udev's assessment of the drives derails and that serial number information spilled from the assessment of /dev/sdi to the assessment of /dev/sdj*. It would be interesting to see the output of ls -l /dev/sd[ij]* in order to learn about the existence of /dev/sdj and the the device numbers of sdi* and sdj*. Further one should inquire the serial numbers by lsblk -d -o NAME,MAJ:MIN,MODEL,SERIAL,WWN /dev/sd[hijk] Have a nice day :) Thomas
