just wondering if anyone has had positive results with any particular hdd testing application over any other. smartctl and smartmontools (combined with gsmartcontrol) showed up quite a bit in my first search and eventually the long version of the self-test was able to confirm that my old drive was having issues, but wanted to see if anyone else had other options that might be helpful in the future.
i ask especially since my new drive (which is working just fine so far) scored in the old age or pre-failure across 18 different attributes when i first put it through both the short and long test at 6 whole lifetime hours. i did notice when i first got the drive that its manufacture date was a couple years ago. i figured that was part of the reason i got a decent price on it, but also wondered if that was the reason for the old age designation in 15 of those 18 attributes.
I think Killdisk still do this, along side actual disk cleaning - I last used it on a windows machine and it did it then - I seem to remember that there is a Linux version or you can make a bootable disk to do. When I last did it, it confirmed my thoughts that the drive was failing. It might be worth checking out.
i appreciate the suggestion. trying to figure out if i am missing something. i see killdisk downloads for macOS and windows, but the description says it comes with a stand-alone iso. i even tried downloading the win version which is just the exe. maybe running it on a win machine extracts the iso? i’ll keep poking around to see if i can figure it out
thank you for the link. @sylvain’s article was helpful in understanding some of the wealth of data that smartctl reports about a drive while keeping in mind that offline backups and/or raid implementations are still very much a good idea
I join the thread lately, but I hope, @01101111, the smartctl article I wrote was helpful. Let me know if there was something unclear.
i ask especially since my new drive (which is working just fine so far) scored in the old age or pre-failure across 18 different attributes […] i did notice when i first got the drive that its manufacture date was a couple years ago. […] also wondered if that was the reason for the old age designation in 15 of those 18 attributes.
AFAIK “old age” attributes are not concerned with the “clock age” of the disk. Actually, I don’t think (?) the HD or SDD electronics embed any kind of real time clock. On a shelf, SSD shouldn’t degrade over time. We may imagine the mechanical parts of a HD to suffer from aging. But after a quick check, I couldn’t find any study about that.
The best suggestion I can give you if you suspect issues with your disk is to monitor the different SMART attributes to see how they evolve.
keeping in mind that offline backups and/or raid implementations are still very much a good idea
Indeed. In my experience, hard drives tend to fail progressively. On the other hand, catastrophic failures are more frequent with SSD drives. But in both case they will fail. Raid (beyond RAID0 of course) will prevent interruption of service in case of one drive failure. And offline backup will secure your data against outage. I suggest you use both of them.
thank you for your reply, @sylvain. i appreciate @abhishek linking to your article so i could bookmark the page for more helpful linux info. overall, you helped confirm the idea that i was coming to regarding smart info and drive testing in general which is that there is no crystal ball like you said. it was interesting to run tests on some of the disks i use for backup purposes to see how they compared to my new one
i have always thought it would be interesting and informative to run some kind of raid setup, but in all honesty my personal computing needs are very modest. for the time being i am more than happy with aiming clonezilla at my wd external on a mostly regular basis for backup purposes.
thanks