ZFS Pool I/O Read Errors

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

optimans

Member
Feb 20, 2015
61
77
18
Hi, hoping someone can shed some light on this for me.

I seem to be getting pool read errors on a couple of pools. Not on the devices themselves (top-level). Running a scrub gives no errors and no repairs. however the read i/o count keeps increasing.

Smart tests show ok.
Looking at pool info in napp-it it show error for the ssd's
Looking at data on drive, nothing appears to be corrupted.

Drives failing? HBA failing?

Does anyone else have the same issue or a fix?

zpool status.png zpool status 2.png

Solaris 11.2
ZPool V6 35

8GB RAM for VM
Xeon E3-1265L-V3
VMware ESXi 5.5 Update 2
LSI Logic 9211-8i HBA
 

optimans

Member
Feb 20, 2015
61
77
18
Bit more info:

replaced the 9211-8i with 9207-8i
replaced sff-8087 to sata cables with different brand
ran memtestx86 with no errors

Found out that P20 firmare was causing lots of ereport.io.scsi.cmd.disk.dev.rqs.derr and transport errors. Downgrading to P19 has stopped this. (!SOLVED! - Lots of trouble with LSI 9207-8e HBA... | Nexenta Community

after downgrading to P19, /var/adm/messages doesn't show any command errors. fmdump -eV no longer shows errors

However still getting the read io count on pools as putty output above

only things i can think of maybe psu (RM550w) or hdd's are starting to fail???
 
  • Like
Reactions: OBasel

OBasel

Active Member
Dec 28, 2010
494
62
28
I had issues with a bad PSU but that caused drives to fall out of the array.