Hi Hans, the system just rebooted by itself, no crash dump has been created but now I think I'm in this situation > This looks vaguely similar to what we've seen on DELL H755 controllers, > where the controller just drops dead after a while, needing a full power > cycle to come back to life. (See https://www.illumos.org/issues/15935) because of this: --------------- ------------------------------------  -------------- --------- TIME            EVENT-ID                              MSG-ID         SEVERITY --------------- ------------------------------------  -------------- --------- Feb 16 13:38:31 ba948ca4-681c-4d79-981c-6da3ca6e6b05  PCIEX-8000-DJ  Major Host        : pg-1 Platform    : ProLiant-ML30-Gen10-Plus  Chassis_id  : CZJ2360PLT Product_sn  : Fault class : fault.io.pciex.device-noresp 40%               fault.io.pciex.device-interr 40%               fault.io.pciex.bus-noresp 20% Affects     : dev:////pci@0,0/pci8086,43b8@1c/pci1590,32b@0                   faulted and taken out of service FRU         : "PCI-E Slot 4" (hc://:product-id=ProLiant-ML30-Gen10-Plus:server-id=pg-1:chassis-id=CZJ2360PLT/motherboard=0/hostbridge=2/pciexrc=2/pciexbus=3/pciexdev=0)                   faulty Description : A problem has been detected on one of the specified devices or on               one of the specified connecting buses.               Refer to http://illumos.org/msg/PCIEX-8000-DJ for more               information. Response    : One or more device instances may be disabled Impact      : Loss of services provided by the device instances associated with               this fault Action      : If a plug-in card is involved check for badly-seated cards or               bent pins. Otherwise schedule a repair procedure to replace the               affected device(s).  Use fmadm faulty to identify the devices or               contact your illumos distribution team for support. PCI-E Slot 4 is where the controller is located. zpool status shows the pool as online, it is not zpool status   pool: dati state: ONLINE status: One or more devices are faulted in response to IO failures. action: Make sure the affected devices are connected, then run 'zpool clear'.    see: http://illumos.org/msg/ZFS-8000-HC   scan: scrub repaired 0 in 0 days 01:07:00 with 0 errors on Fri Feb 16 12:30:01 2024 config:         NAME                       STATE     READ WRITE CKSUM         dati                       ONLINE       0    39     0           mirror-0                 ONLINE       0    45     0             c4t5000CCA85EE5ECB0d0  ONLINE       0    49     0             c4t50014EE6B2513C38d0  ONLINE       0    49     0           mirror-2                 ONLINE       0    73     0             c4t5000C500AAF9B0C3d0  ONLINE       0    83     0             c4t5000C500AAF9BF2Fd0  ONLINE       0    83     0         logs           c3t001B448B4A7140BFd0s0  ONLINE       0     0     0         cache           c3t001B448B4A7140BFd0s1  ONLINE       0     0     0 errors: 20 data errors, use '-v' for a list zpool list NAME    SIZE  ALLOC   FREE  CKPOINT  EXPANDSZ   FRAG    CAP  DEDUP    HEALTH  ALTROOT dati    928G  80.1G   848G        -         -    12%     8%  1.00x    ONLINE  - rpool   222G   138G  83.5G        -         -     0%    62%  1.00x    ONLINE  - with format I don't see the vdevs upon which dati is built. format Searching for disks...done AVAILABLE DISK SELECTIONS:        0. c2t3d0           /pci@0,0/pci1590,28d@17/disk@3,0        1. c2t4d0           /pci@0,0/pci1590,28d@17/disk@4,0        2. c3t001B448B4A7140BFd0           /pci@0,0/pci8086,43c4@1b,4/pci15b7,5017@0/blkdev@w001B448B4A7140BF,0        3. c5tACE42E0005CFF480d0           /pci@0,0/pci8086,43b0@1d/pci1c5c,2f3@0/blkdev@wACE42E0005CFF480,0 Specify disk (enter its number): ^C zpool status -v dati gives this error, probably because it can't read from the pool errors: List of errors unavailable (insufficient privileges) The last line of dmesg reads Feb 16 13:38:43 pg-1 genunix: [ID 390243 kern.info] Creating /etc/devices/retire_store which contains the ID for the controller. I'll power it off to see if upon restart it goes back to a working state. Regards. Maurilio