Seeing this very issue on a pair of 1 TB 990 Pro with Heatsink (4B2QJXD7) since they were new. Running Mint 22, current kernel:
6.8.0-57-generic #59-Ubuntu SMP PREEMPT_DYNAMIC Sat Mar 15 17:40:59 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux 2 identical servers each using 2 of these (all 4 same FW, bought together) in RAID-1 config. One machine always had this issue with both drives, one would fail after a couple of weeks (but the system ploughed on with the 2nd disk) then after maybe a month the 2nd would disappear. The other system had no issues with either disk after being up for a few months. Moved one disk from the affected server to the non-affected server and the problem moved with the disk so now the same disk is also failing in the 2nd server, whilst the disk moved from the good server to the one that originally had the issue seems OK so far (whilst the disk that stayed in the 'bad' server went offline last night after 3 weeks). The systems are Supermicro AS-2025HS-TNR with dual Epyc 9474F. Maybe not relevant but the serial numbers of the good and bad disks aren't in any order, i.e. one bad disk has the highest serial number, but the other bad disk has the lowest, and the 2 good ones lie between these. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2097618 Title: NVME Unable to change power state from D3cold to D0, device inaccessible To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2097618/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
