Working RAID now offline and listed N/A possible HBA?

[chspyder]

I have TrueNAS-12.0-U8 running on a Cisco UCS M220 M3, Which has a RAID 1 with 2 1tb SATA (SAS) Hdds. This had a power failure and suddenly the RAID is down, It has a VM that has some important info on it, I have no drive failure notices, just offline under pools and under Disks it shows N/A.
I have not had an issue with this on another exact same setup, but this one is down, I have read about HBAs and I know this has one, but I am afraid I don't have a clue as why this worked and now doesnt.

Can anyone point me in the right direction?

Thanks
CHSpyder

 

[Arwen]

It's probable that the default card & firmware for Cisco LSI controllers will likely be either MegaRAID or IR, (Integrated RAID). Those are hardware RAID, which does not work well, (and at times completely not at all), with TrueNAS.

You mention RAID 1 and "RAID is down", those indicate that you are using hardware RAID. So you may have an un-supportable environment. TrueNAS and any ZFS pool on such a hardware RAID 1, can't fix or potentially detect problems.

Perhaps someone can walk you through the LSI RAID commands, (I have not used those LSI commands in years).

You can sometimes also access the LSI RAID controller through a boot time menu. I don't remember the keystroke, but it would likely show up on the screen during boot. You can then see what is up with your RAID 1.


Please supply the following, in code tags):
zpool list
zpool status
zpool import

 

[chspyder]

I appreciate the help, I can boot into the LSI Raid controller and I have the drives set a 1 drive per raid in the LSI config. It says all is good.
Here is the info:
zpool list

NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
boot-pool 448G 6.43G 442G - - 0% 1% 1.00x ONLINE -


zpool status
pool: boot-pool
state: ONLINE
scan: scrub repaired 0B in 00:00:14 with 0 errors on Tue Jul 12 03:45:14 2022
config:

NAME STATE READ WRITE CKSUM
boot-pool ONLINE 0 0 0
mfid0p2 ONLINE 0 0 0

errors: No known data errors

zpool insert - Command Fails

 

[elvisimprsntr]

This had a power failure and suddenly the RAID is down, It has a VM that has some important info on it

Make sure you add a UPS and add a 3-2-1 backup strategy.

 

[chspyder]

Dual ups, have backups till may, then someone didn't hook it back up and failed to follow procedures. :(

 

[HoneyBadger]

Please double-check - the final command requested is zpool import and should give you output even if the pool fails to import.

If possible please use the "Code" formatting in the forums (More Options - Insert Code, or BBCode tag [-Code])

 

[chspyder]

Thanks for the info about code tags, couldn't find it. Here is the requested again

zpool list

Code:

zpool list

NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
boot-pool 448G 6.43G 442G - - 0% 1% 1.00x ONLINE -

NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
boot-pool 448G 6.43G 442G - - 0% 1% 1.00x ONLINE -

Code:

zpool status
pool: boot-pool
state: ONLINE
scan: scrub repaired 0B in 00:00:14 with 0 errors on Tue Jul 12 03:45:14 2022
config:

NAME STATE READ WRITE CKSUM
boot-pool ONLINE 0 0 0
mfid0p2 ONLINE 0 0 0

errors: No known data errors

Code:

zpool import
   pool: MEGA
     id: 3999383066549738373
  state: FAULTED
status: The pool metadata is corrupted.
 action: The pool cannot be imported due to damaged devices or data.
        The pool may be active on another system, but can be imported using
        the '-f' flag.
   see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-72
 config:

        MEGA                                          FAULTED  corrupted data
          gptid/63f2c47f-731d-11ec-b4ed-a44c112a16de  ONLINE

And you may of saved my bacon I read that there may be some data loss with the

Code:

zpool -f

command, but I cant seem to figure how to utalize this.


Pre Thanks!

 

[Arwen]

Using the information from zpool import, it it shows your ZFS pool MEGA is a single LUN, (per ZFS side), pool. But, you claim it is a RAID-1 of 2 x 1TB disks, (which it probably is, we just can't see it from the OS side).

ZFS is not designed for data recovery of hardware RAIDed LUNs. It can detect problems using it's checksums and any read errors supplied by the driver software. But since the redundancy is at the hardware level RAID controller, that's where any redundancy lives.

Here is a command that may allow you to import your pool. WARNING - Use at your own risk.

zpool import -f -o readonly=on -R /mnt MEGA

Last note, this is why TrueNAS & ZFS don't recommend any hardware RAID for back end storage. ZFS absolutely works fine with EMC, IBM and Hitachi Enterprise disk arrays for my Solaris servers at work. But, few, (or none), home or small office users have that level of reliability in their hardware RAIDs.

 

[chspyder]

I am about to try your zpool import as it seems to be the best (and only) solution.
Just fyi we did find that one drive has failed and didn't show at all in the truenas software, not sure when it failed, but it would not power up at all, I will discontinue the HW RAID SAS card as soon as I get my db off the disk I hope will restore.
Thank you all so much, if there is any extra ideas I will wait till afternoon Sunday before I try your import and until then I have gathered my backups from May and said my prayers.

Thanks again to all!

 

[Arwen]

TrueNAS is not designed to show hardware RAID failures. Using hardware RAID is out side the scope of TrueNAS. Sorry.

 

[chspyder]

Thank you so much, I was able to restore the DB and I am in progress of setting up new RAID on the TrueNAS controller!

 

[Arwen]

Glad you were able to get the data off those disks.

I hope your "new RAID" controller is really a HBA, (Host Bus Adapter). The difference is that the hardware RAID controllers have a lot more smarts, and tend to hide things from the OS. While a HBA just adds disk ports, (SATA or SAS), that TrueNAS & ZFS see as individual disks.