Hacker News Reader

Comment by geocrasher | Hacker News Reader

Funny thing. This article prompted me to check the health of my two Samsung SSD's (a 250GB 850 EVO SATA III, and a 970 EVO Plus 1TB NVMe), which were fine.

But Samsung's Magician also listed my Seagate ST2000DM008-2FR102 2TB spinny disk. It found a SMART error. I ran a performance test and looked at SMART again, and the "Hardware ECC Recovered" value went from 80 to 81, with a threshold of 64. My other software labels this as "good". Nevertheless, this drive is now being replaced by a 4TB WD Blue. Thanks, article. Saved me some future troubles!

View on HN

The value rising from 80 to 81 is an improvement. The calculated value decreases when the raw value of "Hardware ECC Recovered" worsens.

Is this the case for all SMART values? Higher=better?

Yes, the non-raw values are always reported with higher meaning better. The raw values are the raw measurements/counts and they each mean different things.

Is the number the percentage of a max 100%?

You did the right thing by replacing the Seagate ST2000DM008. I do data recovery professionally and that's not one of my favorite drives! Lots of issues IMHO.

No, sometimes it's in the 0-200 range. I believe the device always reports the maximum and minimum possible values.

Oh, now I feel like a total idiot. Thank you for clarifying that. Good thing is, I still needed a bigger drive so now that's on the way :)

Hardware ECC Recovered represents the amount of time between error correction events, so a higher number is better.

You might want to double-check that SMART status in CrystalDiskInfo

I don't trust Magician to report correctly on other vendor's storage

Yep, I was wrong on this. I should have used my google-fu instead of jumping to a conclusion. I guess I just really wanted an excuse to upgrade to a 4TB :)

How do I check health of a Samsung SSD on Linux?

LMGTFY sudo smartctl -t long -a /dev/sdX

That's not the way for a modern SSD. Try `sudo nvme smart-log /dev/nvmeN`

Seems to give the same output as the second section "smartctl --all" gives... so, less information.

Aside, any idea why it thinks my drive is 208% used?

No idea on that one. Mine are all three indicating 0%, but I've seen wacky stuff from SMART indicators over the years.

Completely normal

Reply on news.ycombinator.com