Hi
I recently bought a Penta SATA Hat from Radxa to host a RAID 5 array on a Pi5. This appears to work well apart from a couple of strange happenings.
The raid set was originally created on an x86_64 desktop running Fedora 30 something a couple of years ago. It consists of 4 x WD Red 6TB 3.5" drives.
I copied the contents of /etc/mdadm/mdadm.conf over to the Pi5 before connecting the Penta SATA hat and drives so RaspiOS knew about the array from the start. Plus I added 'dtparam=pciex1' to /boot/firmware/config.txt
1st strange occurrence: when the OS examines the array on boot it does some enquiries and the the 2nd disk in the array looks like it does a lengthy resync. However neither 'cat /proc/mdstat' nor 'mdadm --detail /dev/md0' show any sign of a resync.
These screenshots were taken while the apparent resync was happening:
The 2nd oddity is a difference in the reported Used Dev Size : 5860390400 (5.46 TiB 6.00 TB) and the used size approx 15TiB as shown by 'du -sh /mnt/raid' The size shown by du -sh is correct as far as I can determine.
Anyone have any idea why this is occurring?
I recently bought a Penta SATA Hat from Radxa to host a RAID 5 array on a Pi5. This appears to work well apart from a couple of strange happenings.
The raid set was originally created on an x86_64 desktop running Fedora 30 something a couple of years ago. It consists of 4 x WD Red 6TB 3.5" drives.
I copied the contents of /etc/mdadm/mdadm.conf over to the Pi5 before connecting the Penta SATA hat and drives so RaspiOS knew about the array from the start. Plus I added 'dtparam=pciex1' to /boot/firmware/config.txt
1st strange occurrence: when the OS examines the array on boot it does some enquiries and the the 2nd disk in the array looks like it does a lengthy resync. However neither 'cat /proc/mdstat' nor 'mdadm --detail /dev/md0' show any sign of a resync.
These screenshots were taken while the apparent resync was happening:
The 2nd oddity is a difference in the reported Used Dev Size : 5860390400 (5.46 TiB 6.00 TB) and the used size approx 15TiB as shown by 'du -sh /mnt/raid' The size shown by du -sh is correct as far as I can determine.
Anyone have any idea why this is occurring?
Statistics: Posted by Smudged — Tue Aug 13, 2024 10:25 am