Computing desk | ||
---|---|---|
< September 22 | << Aug | September | Oct >> | September 24 > |
Welcome to the Wikipedia Computing Reference Desk Archives |
---|
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages. |
Error checking required for my drives, they are way to big (over 1TB each) and will take far too long normally. Partitioning is impossible. What I require is a "Play|Pause" (start from where it was last left) button function, - if you know what I mean - an opensource software that I could use even after I restart my PC the next day or so... Could someone help me please. 103.67.156.38 (talk) 02:42, 23 September 2017 (UTC)
If it's not a laptop or some weird difficult to access OEM computer, it's maybe 5-10 minutes if you know what you're doing. I guess this is a lot of work to you, but to the rest of us, this is less time than I'm sure the OP has spent with their many, many questions asking for random software. As for damage, again if you know what you're doing the risk is minimal.
The biggest risk by far is actually running the check, although that just means the disk should be binned or RMAed anyway. That said, I'm not sure if I agree that a PC is unusable when it's running a comprehensive disk test program. Unless you're doing something really intensive, if you aren't actually trying to use the disk, and if you have a decent multi core processor and RAM, you're not likely to notice nowadays. An exception is if there's really something physically wrong, in that case slow downs can sometimes affect the whole system (although often not). But in that case I don't see a point checking, just get the data off it if you actually need to then either RMA or bin the drive. (Getting the data off may be where having a second computer will be useful.)
I'd note that for a 1TB disk, if it's IO limited it'll take maybe 2-3 hours to read the whole disk. If it's not IO limited, what exactly are you concerned about and testing? If it's a file system issue, that may take a while but it shouldn't have to read the whole disk. If you want to read and write, the whole disk, you can maybe double to triple that. Ideally of course you should just clear the disk in which case the writing and read steps can be separate.
If you're leaving the data there, remember what I said earlier, if there really is a physical problem and you need the data and don't have a backup, doing this could easily mean further damage. You really should be getting the data off ASAP, so concentrate on that first before worrying about checking the disk. And then next time make sure you have backups do don't have to do this again.
Not likely. If you're simply reading the entire hard disk, you will be physically IO limited whether on Windows or *nix. There is no magic here. For a 1 TB hard disk, 90 minutes is simply not realistic. To read 1TB in that time, your HD will need an average transfer speed of ~186MBb/s. This is one of the fastest HDs out there [2]. Yet even it can only achieve an average 179MB/s over the whole disk. Now if short stroke it, you will have a higher average speed and some 1TBs may be short stroked, but they'll often also have lower platter densities. New 1TB hard disks in the 3.5" category are fairly rare anyway. So more likely you'll get maybe 150MB/s or lower.
Incidentally, you recommend manufacturer tools then bring up *nix. There are only two manufacturers of 3.5" hard disks that survive, Seagate and Western Digital. SeaTools for DOS hasn't been updated since 2010. It may still work particularly on a SATA drive or obviously with an old computer like that recommended by Guy Macon above but I wouldn't guarantee it. If you have compatibility problems, SeaTools for Windows is your only option, and no it will not be slower unless there's something majorly messed up with your computer or if you're trying to check an use drive (especially a system drive).
Western Digital is updating their Data Lifeguard Diagnostics for DOS [3] [4] although I'm not sure how well they've kept up with compatibility especially for USB. Again, whether you use the Windows version or DOS version, you should not expect significant speed differences unless you're trying to test an in use drive, or there's something majorly wrong with your computer. (Or I should mention also if it's USB3 on a USB3 controller but limited to high speed probably because of some DOS compatibility issue.)
Mind you nowadays the manufacturer tools don't tend to do that much special except perhaps give you an RMA code. The extended/long test is simply part of S.M.A.R.T.. Likewise, for the SMART values, from my experience it's often actually better to use better SMART utilities which can keep track of changes. The manufacturer tools tend to just give you a pass or fail. So generally, there's nothing wrong with using the manufacturer tools (and actually I normally do that myself for the tests, but not the SMART values), but they often don't actually give you that much more than other tools. So if you did want to use something else on *nix, Windows or whatever, go ahead just don't expect it to be any faster. The main things which are semi unique from the manufacturer tools are stuff like checking for new firmware (which for SeaTools and I think WD requires the Windows version because it simply sends you to the website) or maybe the few feature controls (especially short stroking) unrelated to diagnostics.