Hello,
i have a problem with one zfs pool in this configuration
where the first ssd fails regulary and the pool gets degraded. This are the last few lines from dmesg, but there are much more with this kind of errors
Here is the related part from syslog
I replaced the 860 ssd with a new 870 on march 19 but now the errors are back. So i assume the old 860 ssd is ok and so is the 870 but the controller, cable or the mainboard might be broken.
I am running proxmox on an old Asus Maximus IV GENE-Z mainboard with 32GB RAM and a SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] controller (10Gtek® Internal SAS/SATA RAID Controller PCI Express Host Bus Adapter für LSI 9211-8I, LSI SAS2008 Chip, 8-Port 6Gb/s, High Profile & Low Profile) with another 4x ssd pool in the same configuration on the other port. This pool never showed any errors. This setup is three years old
How can i test the controller and the mainboard while waiting for the new cable to arrive?
i have a problem with one zfs pool in this configuration
Code:
zpool2
mirror-0
ata-Samsung_SSD_870_EVO_1TB
ata-Samsung_SSD_860_EVO_1TB
mirror-1
ata-Samsung_SSD_860_EVO_1TB
ata-Samsung_SSD_860_EVO_1TB
Code:
[1070650.370590] blk_update_request: I/O error, dev sde, sector 902089617 op 0x0:(READ) flags 0x700 phys_seg 6 prio class 0
[1070650.371078] zio pool=zpool2 vdev=/dev/disk/by-id/ata-Samsung_SSD_870_EVO_1TB-part1 error=5 type=1 offset=461868835328 size=82432 flags=180980
[1104078.756568] sd 0:0:4:0: [sde] Unaligned partial completion (resid=1020, sector_sz=512)
[1104078.756585] sd 0:0:4:0: [sde] tag#440 CDB: Read(10) 28 00 3b 02 dc a6 00 00 10 00
[1104078.756591] sd 0:0:4:0: [sde] tag#440 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE cmd_age=0s
[1104078.756594] sd 0:0:4:0: [sde] tag#440 Sense Key : Aborted Command [current]
[1104078.756596] sd 0:0:4:0: [sde] tag#440 Add. Sense: Information unit iuCRC error detected
[1104078.756598] sd 0:0:4:0: [sde] tag#440 CDB: Read(10) 28 00 3b 02 dc a6 00 00 10 00
[1104078.756600] blk_update_request: I/O error, dev sde, sector 990043302 op 0x0:(READ) flags 0x700 phys_seg 1 prio class 0
[1104078.757110] zio pool=zpool2 vdev=/dev/disk/by-id/ata-Samsung_SSD_870_EVO_1TB-part1 error=5 type=1 offset=506901122048 size=8192 flags=180880
[1283475.485182] mpt2sas_cm0: log_info(0x31080000): originator(PL), code(0x08), sub_code(0x0000)
[1283475.485206] sd 0:0:4:0: [sde] tag#2314 FAILED Result: hostbyte=DID_SOFT_ERROR driverbyte=DRIVER_OK cmd_age=0s
[1283475.485209] sd 0:0:4:0: [sde] tag#2314 CDB: Read(10) 28 00 27 c7 c7 6f 00 00 0c 00
[1283475.485211] blk_update_request: I/O error, dev sde, sector 667404143 op 0x0:(READ) flags 0x700 phys_seg 1 prio class 0
[1283475.485717] zio pool=zpool2 vdev=/dev/disk/by-id/ata-Samsung_SSD_870_EVO_1TB-part1 error=5 type=1 offset=341709872640 size=6144 flags=180880
[1283475.485728] sd 0:0:4:0: [sde] tag#2318 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE cmd_age=0s
[1283475.485730] sd 0:0:4:0: [sde] tag#2318 Sense Key : Aborted Command [current]
[1283475.485732] sd 0:0:4:0: [sde] tag#2318 Add. Sense: Information unit iuCRC error detected
[1283475.485733] sd 0:0:4:0: [sde] tag#2318 CDB: Read(10) 28 00 27 c7 c7 99 00 00 0d 00
[1283475.485734] blk_update_request: I/O error, dev sde, sector 667404185 op 0x0:(READ) flags 0x700 phys_seg 1 prio class 0
[1283475.486221] zio pool=zpool2 vdev=/dev/disk/by-id/ata-Samsung_SSD_870_EVO_1TB-part1 error=5 type=1 offset=341709894144 size=6656 flags=180880
Code:
Apr 3 11:40:04 proxmox kernel: [1283475.485182] mpt2sas_cm0: log_info(0x31080000): originator(PL), code(0x08), sub_code(0x0000)
Apr 3 11:40:04 proxmox kernel: [1283475.485206] sd 0:0:4:0: [sde] tag#2314 FAILED Result: hostbyte=DID_SOFT_ERROR driverbyte=DRIVER_OK cmd_age=0s
Apr 3 11:40:04 proxmox kernel: [1283475.485209] sd 0:0:4:0: [sde] tag#2314 CDB: Read(10) 28 00 27 c7 c7 6f 00 00 0c 00
Apr 3 11:40:04 proxmox kernel: [1283475.485211] blk_update_request: I/O error, dev sde, sector 667404143 op 0x0:(READ) flags 0x700 phys_seg 1 prio class 0
Apr 3 11:40:04 proxmox kernel: [1283475.485717] zio pool=zpool2 vdev=/dev/disk/by-id/ata-Samsung_SSD_870_EVO_1TB-part1 error=5 type=1 offset=341709872640 size=6144 flags=180880
Apr 3 11:40:04 proxmox kernel: [1283475.485728] sd 0:0:4:0: [sde] tag#2318 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE cmd_age=0s
Apr 3 11:40:04 proxmox kernel: [1283475.485730] sd 0:0:4:0: [sde] tag#2318 Sense Key : Aborted Command [current]
Apr 3 11:40:04 proxmox kernel: [1283475.485732] sd 0:0:4:0: [sde] tag#2318 Add. Sense: Information unit iuCRC error detected
Apr 3 11:40:04 proxmox kernel: [1283475.485733] sd 0:0:4:0: [sde] tag#2318 CDB: Read(10) 28 00 27 c7 c7 99 00 00 0d 00
Apr 3 11:40:04 proxmox kernel: [1283475.485734] blk_update_request: I/O error, dev sde, sector 667404185 op 0x0:(READ) flags 0x700 phys_seg 1 prio class 0
Apr 3 11:40:04 proxmox kernel: [1283475.486221] zio pool=zpool2 vdev=/dev/disk/by-id/ata-Samsung_SSD_870_EVO_1TB-part1 error=5 type=1 offset=341709894144 size=6656 flags=180880
Apr 3 11:40:04 proxmox zed: eid=1240 class=io pool='zpool2' vdev=ata-Samsung_SSD_870_EVO_1TB-part1 size=6656 offset=341709894144 priority=0 err=5 flags=0x180880 delay=60ms bookmark=18245:1:0:1691100
Apr 3 11:40:04 proxmox zed: eid=1241 class=io pool='zpool2' vdev=ata-Samsung_SSD_870_EVO_1TB-part1 size=6144 offset=341709872640 priority=0 err=5 flags=0x180880 delay=60ms bookmark=18245:1:0:1691096
I am running proxmox on an old Asus Maximus IV GENE-Z mainboard with 32GB RAM and a SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] controller (10Gtek® Internal SAS/SATA RAID Controller PCI Express Host Bus Adapter für LSI 9211-8I, LSI SAS2008 Chip, 8-Port 6Gb/s, High Profile & Low Profile) with another 4x ssd pool in the same configuration on the other port. This pool never showed any errors. This setup is three years old
How can i test the controller and the mainboard while waiting for the new cable to arrive?