(SOLVED) print_req_error: critical target error


Leondre

Recommended Posts

  • 1 year later...

Got the same problem.

And I get an email with the error "fstrim: /mnt/cache: FITRIM ioctl failed: Remote I/O error" from the server everytime the trim is scheduled to run.

 

But the thing is that I have 4 SSDs on the same HBA, and 3 of them runs trim without a hatch. The faulty one is an INTEL SSDSC2BW120H6 120G drive.

 

One note is that I did upgrade the firmware of my Logic SAS2308 PCI-Express Fusion-MPT SAS-2, and it was after that that the problems started occurring.

 

 

I have no clue what to do about this, not to keen on flashing back to old firmware either >.>

unraid-diagnostics-20190225-0007.zip

Link to comment
7 hours ago, Nischi said:

But the thing is that I have 4 SSDs on the same HBA, and 3 of them runs trim without a hatch. The faulty one is an INTEL SSDSC2BW120H6 120G drive.

That's not correct, trim isn't working in any of the them, the only one that would be trimmed by the LSI HBA is the Intel since it's only one that supports read zeros after trim, hence the different error, but it also doesn't work with latest Linux drivers, see here for more info:

https://forums.unraid.net/topic/74493-samsung-ssd-hba-trim/?do=findComment&comment=720627

https://forums.unraid.net/topic/74493-samsung-ssd-hba-trim/?do=findComment&comment=720637

 

Link to comment

Thanks for the links to that thread, really insightful.

My version info:

modinfo mpt3sas
filename:       /lib/modules/4.19.24-Unraid/kernel/drivers/scsi/mpt3sas/mpt3sas.ko.xz
alias:          mpt2sas
version:        26.100.00.00

Also tried some commands from the mentioned thread.

root@unraid:~# hdparm -I /dev/sdm | grep TRIM
       *    Data Set Management TRIM supported (limit 8 blocks)
       *    Deterministic read ZEROs after TRIM
root@unraid:~# hdparm -I /dev/sdj | grep TRIM
       *    Data Set Management TRIM supported (limit 1 block)
       *    Deterministic read data after TRIM
root@unraid:~# hdparm -I /dev/sdl | grep TRIM
       *    Data Set Management TRIM supported (limit 8 blocks)
root@unraid:~# hdparm -I /dev/sdk | grep TRIM
       *    Data Set Management TRIM supported (limit 1 block)
       *    Deterministic read data after TRIM
root@unraid:~# fstrim -av
/etc/libvirt: 926.5 MiB (971513856 bytes) trimmed on /dev/loop3
/var/lib/docker: 13.7 GiB (14724616192 bytes) trimmed on /dev/loop2

fstrim: /mnt/cache: FITRIM ioctl failed: Remote I/O error

The reason I say only the intel doesn't work is because the unraid logs only mentions the intel one(sdj) as having problems. 

Feb 25 16:50:01 unraid kernel: print_req_error: critical target error, dev sdj, sector 232785982
Feb 25 16:50:01 unraid kernel: BTRFS warning (device sdj1): failed to trim 1 device(s), last error -121

 

I see that the /mnt/cache/ is mounted from /dev/sdj1. could that be why it complains about sdj above? Thought it would have said sdj1  on both errors.

 

Seems like it's time to look for one of those 9300-8i soon. Thanks for your help.

Edited by Nischi
Link to comment
  • 1 year later...

Good Morning,

 

im getting these errors

Quote

Feb 28 07:20:00 T30 kernel: print_req_error: critical target error, dev sde, sector 1971719480

on my normal WD RED (WDC_WD40EFRX-68N32N0_WD-WCC7K7JKSKT0 - 4 TB (sde))

 

This drive is connected to my LSI2008 I guess (havent looked into the server yet), disabled atm.

Im having 4 errors at the moment and im helpless what to do now..

Link to comment
24 minutes ago, johnnie.black said:

Disk looks fine, might be a good idea to replace/swap cables just to rule them out if it happens again to the same disk.

Thx for helping.

As the disk is installed to the LSI Card, its connected with this cable set:

https://www.amazon.de/gp/product/B00S7KU0VO/ref=ppx_yo_dt_b_asin_title_o09_s00?ie=UTF8&psc=1

 

bought brand new - so there is no way to change just one cable :(

Link to comment
  • 9 months later...
On 2/4/2018 at 1:40 AM, JorgeB said:

LSI2008 doesn't support trim on most SSD, try connecting it to the onboard SATA ports.

I'm actually running into this same exact problem...if connecting to SATA ports isn't an option do I need to switch back to spinning disks for my cache instead of SSD's? Or can the SSD's run fine w/o trim?

Link to comment
10 minutes ago, JorgeB said:

Still better to have SSDs even without trim.

Ok good to know.  Are these trim errors going to blow anything up in the future? or it's just saying it's unable to trim and going from there?  

 

Also, someone in another thread recommended installing one of these: Amazon.com: I/O CREST 2 Port SATA III PCI-e 2.0 x1 Controller Card Asmedia ASM1061 Non-Raid with Low Profile Bracket SY-PEX40039: Electronics and plugging the drives into that.  If I shut the server down completely and install this, then move the drives to this...will that destroy my array because I moved the cache drives? Or will unraid detect all that on it's own?

 

Or should I go through the process of completely moving cache to array, removing cache drives, shut down, install card, plug drives into the card, boot back up and reconfigure cache?

 

Thanks for the quick response!

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.