JorgeB

One thing to do is to make sure RAM isn't overclocked, see here for more info.

9400-8i is IT mode only.

4 DIMMs @ 3200 is overclock, see here for more info: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/?tab=comments#comment-543490

Try this.

There are write errors on both disks: Feb 19 16:51:38 Tower kernel: sd 5:0:1:0: [sdc] tag#7520 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08 Feb 19 16:51:38 Tower kernel: sd 5:0:1:0: [sdc] tag#7520 Sense Key : 0x4 [current] [descriptor] Feb 19 16:51:38 Tower kernel: sd 5:0:1:0: [sdc] tag#7520 <<vendor>>ASC=0x81 ASCQ=0x0 Feb 19 16:51:38 Tower kernel: sd 5:0:1:0: [sdc] tag#7520 CDB: opcode=0x8a 8a 00 00 00 00 01 5d 50 a2 88 00 00 01 00 00 00 Feb 19 16:51:38 Tower kernel: print_req_error: critical target error, dev sdc, sector 5860532872 Feb 19 16:51:38 Tower kernel: md: disk2 write error, sector=5860532808 Feb 19 16:51:38 Tower kernel: md: disk2 write error, sector=5860532816 Feb 19 16:51:38 Tower kernel: md: disk2 write error, sector=5860532824 Feb 19 16:51:38 Tower kernel: sd 5:0:2:0: [sdd] tag#7531 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08 Feb 19 16:51:38 Tower kernel: sd 5:0:2:0: [sdd] tag#7531 Sense Key : 0x4 [current] [descriptor] Feb 19 16:51:38 Tower kernel: sd 5:0:2:0: [sdd] tag#7531 <<vendor>>ASC=0x81 ASCQ=0x0 Feb 19 16:51:38 Tower kernel: sd 5:0:2:0: [sdd] tag#7531 CDB: opcode=0x8a 8a 00 00 00 00 01 5d 50 a2 88 00 00 01 00 00 00 Feb 19 16:51:38 Tower kernel: print_req_error: critical target error, dev sdd, sector 5860532872 Feb 19 16:51:38 Tower kernel: md: disk3 write error, sector=5860532808 Don't know what's causing that, but it's not a software problem, try updating LSI firmware, though doubt it's that.

Try a manual test: smartctl -t long /dev/sdX And post the output if you get an error.

HBA problem: Feb 19 21:30:04 Tower kernel: mpt2sas_cm0: SAS host is non-operational !!!! Make sure it's well seated and sufficiently cooled, you can also try a different PCIe slot if available.

IIRC there are known performance issues with that 3ware controller.

Check filesystem on disk1: https://wiki.unraid.net/Check_Disk_Filesystems#Checking_and_fixing_drives_in_the_webGui

Can you post a link to the model you're using? I knew there are 6 and 10 ports models, but they all look the same on the syslog, they appear as 12 ports with 2 dummy ports. Feb 19 19:08:51 Server kernel: ahci 0000:03:00.0: AHCI 0001.0301 32 slots 12 ports 6 Gbps 0xff3 impl SATA mode ... Feb 19 19:08:51 Server kernel: ata9: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580100 irq 49 Feb 19 19:08:51 Server kernel: ata10: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580180 irq 49 Feb 19 19:08:51 Server kernel: ata11: DUMMY Feb 19 19:08:51 Server kernel: ata12: DUMMY Feb 19 19:08:51 Server kernel: ata13: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580300 irq 49 Feb 19 19:08:51 Server kernel: ata14: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580380 irq 49 Feb 19 19:08:51 Server kernel: ata15: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580400 irq 49 Feb 19 19:08:51 Server kernel: ata16: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580480 irq 49 Feb 19 19:08:51 Server kernel: ata17: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580500 irq 49 Feb 19 19:08:51 Server kernel: ata18: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580580 irq 49 Feb 19 19:08:51 Server kernel: ata19: SATA max UDMA/133 abar m8192@0xf6580000 port 0xf6580600 irq 49

Select all, then re-assign the missing ones, double check all assignments, then check "parity is already valid" before starting the array.

Yep, for example disk1 was ST12000NM0007-2A1101_ZJV1HE5B and now is ST12000NM0007-2A_ZJV1HE5B_35000c500b1c28799 You could do a new config, but flashing the LSI to current firmware, it's using a very old one, should fix it.

While still true just noticed it uses the 3408 chip, so it might be flashable to IT mode, similar to the 9240-8i which is also a MegaRAID by default but can be flashed to 9211 IT.

At least that problem yes, note that the LSI won't trim the SSDs, so you should connect them onboard, use the LSI for HDDs.

That's a MegaRAID controller, unlikely to be flashable to IT mode, you might still be able to use it by creating a raid0 for each disk, but it's not recommended, for example if you change controller in the future Unraid won't accept the disks.

That won't matter, likely the disk ID string changed slightly, you should post the diags.

You're using a 6 or 10 port Asmedia controller with SATA port multipliers, they are a known problem.

--repair can only fix a small number of issues, and sometimes it makes things even worse, you should backup any data on cache and re-format the pool, there are some recovery options in the FAQ if needed.

That's a server board, so unlikely to support overclock, but it should tell what frequency RAM is running. Yep, that was my bad, it was the CPU clock i saw, not the RAM, so I immediately deleted the reply.

Cache="prefer" moves from array to cache, and your cache is now full, you want cache="yes" for that share, see the GUI help.

You can stop it after a single error, it confirms RAM is the problem.

That looks like an Nginx problem, and outside my knowledge, maybe someone else can help, rebooting should fix it, but parity check will restart.

You can try and get extra SATA cables if it's a modular PSU, or just get a new one with enough cables, a new chassis is also an option, budget allowing.

If the dropped disks are OK, and you can run a long SMART test to be sure, you can do a new config with those disks also assigned, to any slot, then just re-sync parity.

SATA splitter should be used for max 2 drives, more than that can easily go over the max 12v rating for a SATA plug, 4.5 Amps.