Jump to content

JorgeB

Moderators
  • Posts

    67,514
  • Joined

  • Last visited

  • Days Won

    707

Everything posted by JorgeB

  1. btrfs check --repair should be used with care, it might make things worse, see here for some recovery options.
  2. 860 EVO one of the best SATA SSDs you can get, jut make sure to use the latest beta since Samsung SSDs appear to be sensitive to partition alignment.
  3. No. Yes, as long as it's detected, but disable array auto start to make sure all is well before starting the array. It should.
  4. Some BIOS have issues when there are many other possible boot devices, you don't need the BIOS on the LSI, disable OPT ROM BIOS on the board BIOS, if that doesn't work to suppress the HBA BIOS you can delete them from the HBAs since they are not needed for Unraid, and it will make booting faster anyway.
  5. As long as both are true HBAs there won't be a problem, there could be if any of then is a RAID controller.
  6. Copy form array to array will never be fast, due to parity, and you now can't even use turbo write to help with that.
  7. See if you can wipe it clean with: blkdiscard /dev/nvme0n1 If command is successful try to format after that.
  8. System load average is very high, but nothing on top using CPU.
  9. Run make_bootable again, if that doesn't work see if it's booting on another PC, if not re-create it.
  10. Does the flash drive appear on the BIOS as a boot device?
  11. Yes, with an expander backplane you just need one HBA to access all drives.
  12. In that case I would go with 7 data + parity, ideally use the latest beta since it has a new alignment for SSDs that is better for performance and endurance, you should get at least 300MB/s reads for large files, possibly more.
  13. What model SSDs and what performance are you expecting? Note that array devices can't be trimmed and parity should be a faster device than the others, since it will be overworked, for example I use an NVMe device for parity.
  14. I don't see any format attempt on the diags posted.
  15. If it happens again post new diags, that fix should be unrelated to array stopping issues.
  16. Macvlan call traces are usually cause by dockers with a custom IP address, more info here, also issues with the NICs, but might be related since they start after the macvlan errors, also some out of memory errors.
  17. It can rebuild two drives if you have dual parity.
  18. I'm used to seeing these in several diags, so probably normal.
  19. Syslog is spammed with ACPI errors, see here for the fix, then reboot and post new diags.
  20. Please post the complete diags instead.
×
×
  • Create New...