I was setting up a Debian server with all SSDs and I kept getting a strange error that caused mdadm to not be able to flush data which made the entire system lock up.
I decided to check the SMART log and I saw a strange error. All of the SMART attributes are good with no reallocated sectors. I’m using all Intel SSDs and I’ve never had a problem with them until this one.
The error does not appear to be the drive itself as I tested another Intel 730 SSD and even moved a known good drive to the port and it still ocurred. It’s actually a problem with the cable. Most likely it was just loose and not making perfect contact with the SATA port on the motherboard. I moved the cables around and I have yet to see it happen again.
One strange thing is the other 730 drive I tested already had a bunch of these errors. I do not recall having issues with it so maybe it was from unplugging it while it was on?
As you can see I also ran an extended SMART test and it passed just fine.
TL;DR Check the SATA cable and reseat it.