disk-burnin-and-testing VS stressdisk

Verify authenticity or integrity? I run this on all of the drives I get in: https://github.com/Spearfoot/disk-burnin-and-testing

You could check them using badblocks or the script available at this github repo https://github.com/Spearfoot/disk-burnin-and-testing. If they fail, send the seller the log file as evidence and getting a refund/replace should be pretty straightforward

I used this script fine on 18TB drives. I don't know why it wouldn't work on 20TB https://github.com/Spearfoot/disk-burnin-and-testing . It uses badblocks, but honestly I don't know much about this stuff. If it passes I consider it good to deploy. It does look like it uses a 8192 block size.

Hello, I'm using Spearfoot's disk-burnin script for smart short, smart long and bad blocks tests. I read here and there that we shouldn't test bad blocks on SSDs and I'm curious why. I'd also like to know how to test SSDs with the Spearfoot script since I already know how to test multiple disks at the same time with screens in Putty. I've read the documentation and scrolled through the script and I can't find a way to deactivate the bad blocks test and only do smart short and smart long.

I'm starting this process tonight and this is the best ressource I've found so far to do disk burnin and testing.

You can do a 'burn-in' test for 48 hours in an attempt to give yourself some confidence that they are okay. In my experience unless they were going into a high-redundancy array the risk associated with keeping them is much higher than the downside of replacing them right away, if you can. Shouldnt be any guilt with returning them, the shop that sold them should be able to make a claim against the shipping company and should sell them as used/returned drives and the person who does buy them understands and accepts that risk.

I use this for new drives https://github.com/Spearfoot/disk-burnin-and-testing

So I'm currently testing a new pair of drives in my TrueNAS Server. I'm running Spearfoot's burnin script and it has finished running badblocks and has launched the final smartctl long-test.

I actually played with that software for a few minutes and tested an NVME SSD and got no issues. Sucks up space very quickly (depending on disk speed), so it does seem to be a very useful tool. For ease of use, the ready to run binary files are here: https://github.com/ncw/stressdisk/releases, and on windows, the CMD prompt you would use is

I never have issues with amazon and counterfeit SDs, then again i also tend to run burn in tests on any SDs i buy before actually putting them into service. Stressdisk is a really good one i like to go with.

Whatever you decide, I'd run stressdisk for about 24h(the default time) before putting it into service, but maybe that's just the paranoia talking. It's pretty good at what it does.

Consider the possibility of the drive being not entirely reliable. Careful use of StressDisk might reveal something.

And as for the stress test: I've had them all in the enclosure. I disabled raid mode, so all 4 drives showed up individually, and then I ran https://github.com/ncw/stressdisk this script twice parallel on each drive. It did not die then :\

I like to use short smart, then stressdisk for 72h, then long smart. Or there's Hard Disk Sentinel that will do all that in one program. Sentinel costs money, but with a google search for a coupon, I think it's like $12. Great investment if you are going to be working with a lot of disks for, I dunno, a NAS. :D

If you're running Linux, I'd install stressdisk and see if any errors occur. Could save a lot of grief later.