JorgeB

Also note that that one is PCIe x1, they also exist with PCIe x2, x1 can be a bottleneck with 6 disks.

Should be. No. It will support any current size available now and in the near feature.

Not really, just wouldn't try anything else before using a different PSU or at least replace/check all the cables/connections, after that check the syslog for errors like these, there's still a problem if they continue to appear. Aug 21 12:03:58 Tower kernel: ata3.00: status: { DRDY } Aug 21 12:03:58 Tower kernel: ata3: hard resetting link Aug 21 12:04:04 Tower kernel: ata3: link is slow to respond, please be patient (ready=0) Aug 21 12:04:08 Tower kernel: ata3: COMRESET failed (errno=-16) Aug 21 12:04:08 Tower kernel: ata3: hard resetting link Aug 21 12:04:14 Tower kernel: ata3: link is slow to respond, please be patient (ready=0) Aug 21 12:04:18 Tower kernel: ata3: COMRESET failed (errno=-16) Aug 21 12:04:18 Tower kernel: ata3: hard resetting link Aug 21 12:04:23 Tower kernel: ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300) Aug 21 12:04:23 Tower kernel: ata3.00: configured for UDMA/133 Aug 21 12:04:23 Tower kernel: ata3: EH complete Aug 21 12:04:46 Tower kernel: ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen Aug 21 12:04:46 Tower kernel: ata4.00: failed command: FLUSH CACHE EXT Aug 21 12:04:46 Tower kernel: ata4.00: cmd ea/00:00:00:00:00/00:00:00:00:00/a0 tag 2 Aug 21 12:04:46 Tower kernel: res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) Aug 21 12:04:46 Tower kernel: ata4.00: status: { DRDY } Aug 21 12:04:46 Tower kernel: ata4: hard resetting link Aug 21 12:04:47 Tower kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 SControl 300) Aug 21 12:04:47 Tower kernel: ata4.00: configured for UDMA/133 Aug 21 12:04:47 Tower kernel: ata4.00: retrying FLUSH 0xea Emask 0x4 Aug 21 12:04:47 Tower kernel: ata4: EH complete Aug 21 12:05:54 Tower kernel: ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen Aug 21 12:05:54 Tower kernel: ata4.00: failed command: READ DMA EXT Aug 21 12:05:54 Tower kernel: ata4.00: cmd 25/00:40:30:86:79/00:05:1e:00:00/e0 tag 29 dma 688128 in Aug 21 12:05:54 Tower kernel: res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)

Aug 21 22:46:50 Perplex kernel: mpt2sas_cm0: failure at drivers/scsi/mpt3sas/mpt3sas_scsih.c:11045/_scsih_probe()! LSI driver is crashing, 9300-8i is using a very old firmware, it should be updated, note that it will require doing a new config after since the disks will be detected with different names, one of the problems of that firmware, then test again, or test with just the 9201-16e to see if it works by itself.

Depending on the issue there might not even be a notification, I suggest using a script to monitor any pool.

Please post the diagnostics: Tools -> Diagnostics

There are ATA errors in multiple disks, these are usually a power/connection problem.

Run it again the this option.

Yes, that might fix it, and if it doesn't there's the rebuild option.

Pool looks OK now, just need to wait for the balance to finish.

Balance *should* resume from where is was after a restart.

You need to format the array device first, next to array start/stop buttons.

Possibly, try with an ad-don NIC if available.

Sector size is the same, just the alignment is different, could be filesystem related, i.e., xfs might be more efficient with very small files, but don't remember this coming up before, and certainly not by this much: See if moving the data back to cache takes up all the space again.

If it didn't find errors in the beginning likely won't find any.

First the ASM1062 is a two port controller, so if there are more than two ports it has to have multiple chips or port multipliers, second the JMB575 is a port multiplier.

Diags are after rebooting so not much to see, if it happens again see if you get get the diags on the console, if got get at least the syslog.

Yes, missed that, but that's not because of the pool: Aug 18 23:08:20 Tower shfs: shfs: ../lib/fuse.c:1451: unlink_node: Assertion `node->nlookup > 1' failed. It's this issue: A reboot will fix it.

This just means it supports port multipliers, they can be external. This one includes a port multiplier and should not be used.

If it only boots in safe mode it suggests a plugin problem, remove all plugins.

Where are you seeing that? ASM1166 supports 6 SATA ports, no reason to have an additional port multiplier unless the controller has more than 6 ports.

Default mode is raid1, you can convert to others.

👍

Do you have another NIC you could try with, just to confirm if it's a NIC driver issue or not?

This is the current usage: Used: 859.18GiB Free (estimated): 71.58GiB (min: 71.58GiB) Something is using that space, balance won't change anything, if you don't know what it is start moving data to the array until you find out.