which drive?

The drive that dropped offline, aren't you posting about that? Cache2

Yes just wasnt sure which of the two cache drives was having the issue.

I moved my unraid to a new case, new PSU, new cables. same drives. its still happening.

plexserver-diagnostics-20230312-1644.zip

cache2 has disconnected

If it keeps happening to the same device with new cables replace the device.

So I removed that drive, but now im getting this.  Cache pool BTRFS too many profiles

plexserver-diagnostics-20230313-0938.zip

I guess there was a balance running.

That's normal during a balance, it even says so in the notification.

So I replaced the the cache drive that was failing, changed the ram and it still happens. Even reformatted the both cache drives.

plexserver-diagnostics-20230328-0726 (1).zip

Device is still dropping offline, this is usually a power/connection problem, what was replaced so far?

all cables, new PSU, new motherboard, ram,

Is it connected directly to the PSU and board directly or is there some enclosure/cable extension/splitter?

its connected directly, I had splitter before and I've had it on a sas card but now its directly. Had same issue on all 3

If it's always the same device swap cables (both power and SATA) with the other one, and see if the problem follows the device or not.

Ok, I'll try that. What do you search for on the log to find which drive is failing?

See here to setup the scrip, and you can then see which drive has the issue.

thanks, Il ready have that running, and it does inform me when the issue happens but Id like to learn how to search my own diag file to identify which cache drive is the one failing. Should I consider changing to XFS for the two cache drives?

10 hours ago, fespinoza831 said:

but Id like to learn how to search my own diag file to identify which cache drive is the one failing

Look in the syslog for these:

 

Mar 27 20:46:42 PlexServer kernel: ata5: reset failed, giving up
Mar 27 20:46:42 PlexServer kernel: ata5.00: disable device
Mar 27 20:46:42 PlexServer kernel: ata5: EH complete
Mar 27 20:46:42 PlexServer kernel: sd 6:0:0:0: [sdf] tag#29 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=DRIVER_OK cmd_age=90s
Mar 27 20:46:42 PlexServer kernel: sd 6:0:0:0: [sdf] tag#29 CDB: opcode=0x93 93 08 00 00 00 00 04 9a 40 70 00 00 01 c0 00 00

 

 

10 hours ago, fespinoza831 said:

Should I consider changing to XFS for the two cache drives?

xfs won't help with dropping devices.

 

 

Was good for a few day and it happened again yesterday. so I replace the power cables again last night. ran the scrub and woke up to docker being crashed and cache drives disabled again. is there anything else that could be causing this. Any of my hardware not recommended. Not sure what else to do here.

plexserver-diagnostics-20230407-0706.zip

Is this still my cache drive?

 

Apr  7 05:47:52 PlexServer kernel: ata3: hard resetting link
Apr  7 05:47:56 PlexServer kernel: ata5: link is slow to respond, please be patient (ready=0)
Apr  7 05:47:58 PlexServer kernel: ata3: link is slow to respond, please be patient (ready=0)
Apr  7 05:48:01 PlexServer kernel: ata5: COMRESET failed (errno=-16)
Apr  7 05:48:01 PlexServer kernel: ata5: hard resetting link
Apr  7 05:48:02 PlexServer kernel: ata3: COMRESET failed (errno=-16)
Apr  7 05:48:02 PlexServer kernel: ata3: hard resetting link
Apr  7 05:48:06 PlexServer kernel: ata5: link is slow to respond, please be patient (ready=0)
Apr  7 05:48:08 PlexServer kernel: ata3: link is slow to respond, please be patient (ready=0)
Apr  7 05:48:11 PlexServer kernel: ata5: COMRESET failed (errno=-16)
Apr  7 05:48:11 PlexServer kernel: ata5: hard resetting link
Apr  7 05:48:12 PlexServer kernel: ata3: COMRESET failed (errno=-16)
Apr  7 05:48:12 PlexServer kernel: ata3: hard resetting link
Apr  7 05:48:16 PlexServer kernel: ata5: link is slow to respond, please be patient (ready=0)
Apr  7 05:48:18 PlexServer kernel: ata3: link is slow to respond, please be patient (ready=0)
Apr  7 05:48:46 PlexServer kernel: ata5: COMRESET failed (errno=-16)
Apr  7 05:48:46 PlexServer kernel: ata5: limiting SATA link speed to 3.0 Gbps
Apr  7 05:48:46 PlexServer kernel: ata5: hard resetting link
Apr  7 05:48:47 PlexServer kernel: ata3: COMRESET failed (errno=-16)
Apr  7 05:48:47 PlexServer kernel: ata3: limiting SATA link speed to 3.0 Gbps
Apr  7 05:48:47 PlexServer kernel: ata3: hard resetting link
Apr  7 05:48:51 PlexServer kernel: ata5: COMRESET failed (errno=-16)
Apr  7 05:48:51 PlexServer kernel: ata5: reset failed, giving up
Apr  7 05:48:51 PlexServer kernel: ata5.00: disable device
Apr  7 05:48:51 PlexServer kernel: ata5: EH complete
Apr  7 05:48:51 PlexServer kernel: sd 6:0:0:0: [sdf] tag#4 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=DRIVER_OK cmd_age=89s
Apr  7 05:48:51 PlexServer kernel: sd 6:0:0:0: [sdf] tag#4 CDB: opcode=0x2a 2a 00 1c 97 0a a8 00 00 70 00
Apr  7 05:48:51 PlexServer kernel: I/O error, dev sdf, sector 479660712 op 0x1:(WRITE) flags 0x8800 phys_seg 14 prio class 0

Also noticed that when this happens I cannot restart the server. I stop dock and VMs. check for open file then reboot but it will hang at a black screen. I have to hard reset to actually bring it back up from that. but if I do a shutdown and then start it it works fine.

brought it back up and running the scrub but it doesn't show any errors 

 

Still running Parity Check...

 

14 minutes ago, fespinoza831 said:

Is this still my cache drive?

ata3 and ata5 are both cache devices, maybe you board doesn't like the MX500, connect both devices to the HBA, swap if other disks if necessary, trim won't work but at least see if they stop dropping.

OK will do this, but should mention that I had same issue on my previous board which was a MSI MPGZ490 now using  a ASUS Prime Z590-A. Are the MX500 known to work?

Edited April 7, 20233 yr by fespinoza831