Received another Error 22 on da4

[EvanVanVan]

Last month I received emails stating an unretryable error 22 had occurred on da4. With some helpful (https://forums.freenas.org/index.php?threads/error-22-failing-drive.42292/) advice I made sure everything was OK with the drive and even removed encryption from my pool.

A couple of days ago I received a similar error again (attached at the bottom). It was suggested last time that Error 22 may have to do with my LSI card. I'm wondering if my first step should be swapping out my SAS-Sata breakout cables.

But before I do anything I like to check here and ask several rapid fire questions.

1. Is it relatively safe to change cables?
2. I'm assuming I'll need to verify that the breakout cables are plugged into the drives in the same order. Hopefully they they are labeled from the manufacturer 1 through 4?
3. I currently have 2 breakout cables to handle my either drives, both are from monoprice. Should I replace both of them and/or does it matter if I go with a different brand and just replace 1 of them?

Code:

freenas.local kernel log messages:
> (da4:mps0:0:4:0): WRITE(10). CDB: 2a 00 3f 7b fe 10 00 00 08 00
> (da4:mps0:0:4:0): CAM status: SCSI Status Error
> (da4:mps0:0:4:0): SCSI status: Check Condition
> (da4:mps0:0:4:0): SCSI sense: ILLEGAL REQUEST asc:21,0 (Logical block address out of range)
> (da4:mps0:0:4:0): Info: 0x3f7bfe10
> (da4:mps0:0:4:0): Error 22, Unretryable error

-- End of security output --

Code:

Checking status of zfs pools:
NAME           SIZE  ALLOC   FREE  EXPANDSZ   FRAG    CAP  DEDUP  HEALTH  ALTROOT
freenas-boot  7.19G  5.62G  1.57G         -      -    78%  1.00x  ONLINE  -
volume1       21.8T  11.0T  10.7T         -     7%    50%  1.00x  ONLINE  /mnt

  pool: volume1
state: ONLINE
status: One or more devices has experienced an unrecoverable error.  An
        attempt was made to correct the error.  Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
        using 'zpool clear' or replace the device with 'zpool replace'.
   see: http://illumos.org/msg/ZFS-8000-9P
  scan: resilvered 8K in 0h0m with 0 errors on Wed Apr 27 02:43:55 2016
config:

        NAME                                            STATE     READ WRITE CKSUM
        volume1                                         ONLINE       0     0     0
          raidz2-0                                      ONLINE       0     0     0
            gptid/2a0e67aa-b790-11e3-b9da-002590d65107  ONLINE       0     0     0
            gptid/2a82eeea-b790-11e3-b9da-002590d65107  ONLINE       0     0     0
            gptid/2afb7434-b790-11e3-b9da-002590d65107  ONLINE       0     0     0
            gptid/2b79a9e5-b790-11e3-b9da-002590d65107  ONLINE       0     0     0
            gptid/2bf1472f-b790-11e3-b9da-002590d65107  ONLINE       0     0     0
            gptid/2c67faef-b790-11e3-b9da-002590d65107  ONLINE       0     1     0
            gptid/2cedc0a0-b790-11e3-b9da-002590d65107  ONLINE       0     0     0
            gptid/2d7a9191-b790-11e3-b9da-002590d65107  ONLINE       0     0     0

errors: No known data errors

-- End of daily output --

 

[ethereal]

it doesn't matter in which order the drives are connected - freenas doesn't care

 

[EvanVanVan]

One more question, I only bought one new breakout cable and want to make sure I replace the correct one. I'm wondering if I can determine which hard drive is da4 without having to physically remove the drive and match up the serial numbers. I know my LSI card has two SAS ports, labeled 0 and 1 and the breakout cables are labeled p1-p4.

Can I assume that the drives are connected as follows?

SAS Port 0:

P1 = da0
P2 = da1
P3 = da2
P4 = da3​

SAS Port 1:

P1 = da4
P2 = da5
P3 = da6
p4 = da7​

Or do I have to remove the drive and verify the serial number matches up?

Thanks

​

 

[ethereal]

i don't think that you can guarantee that it work's this way.

if you use the gui you can find the serial number associated with the device number

you can also use -

camcontrol identify da5 |grep ^serial

smartctl -i /dev/da5 | grep ^Serial

change da5 to the drive your interested in

 

[Ericloewe]

I'm wondering if I can determine which hard drive is da4 without having to physically remove the drive and match up the serial numbers.

No.
There is no guarantee they will be consistent across reboots, never mind map nicely to physical ports.

 

[EvanVanVan]

Thank you both for the confirmation.

I guess while I'm at it, it would be helpful if I p-touch some labels on each hard drive to make figuring out the serial numbers easier in the future...

 

[Ericloewe]

Thank you both for the confirmation.

I guess while I'm at it, it would be helpful if I p-touch some labels on each hard drive to make figuring out the serial numbers easier in the future...

That may be unnecessary. WD drives, for instance, have end labels on the side opposite the SATA connectors, with the serial number.

 

[EvanVanVan]

That may be unnecessary. WD drives, for instance, have end labels on the side opposite the SATA connectors, with the serial number.

That was much more convenient although I did have to take my LSI card out and *carefully* remove the back cover of the box (with the motherboard still attached to it) to make just enough room to see the front S/Ns. In the end I p-touched labels on to the back/sata side of the drives anyway to make it even easier next time.

Anyway, like the last time...just restarting FreeNAS cleared the critical alert, so hopefully with the new cable it doesn't come back.

Thanks for the help.