I have a 10TB drive that's currently failing from my TrueNAS. I'm able to replace the drive with a new one but don't want to pitch it if there's a chance of saving it. TrueNAS has been flagging sectors as "Uncorrectable" and "Unreadable" can Spinrite help with these kinds of errors?
Not sure if it matters but TrueNAS is on ZFS file system so I don't know if I could even run Spinrite on it.
Heard about this through the podcast and haven't used it yet and have been waiting for 6.1 to purchase but if this is something spinrite can help iron out I think I'll push that up a bit.
Thanks.
You should
definitely run SpinRite on that drive... and I'm not saying that only because I have a vested interest in SpinRite (though I certainly do.
I recently had a similar-sounding situation, where a 2TB drive was rejected and removed from service by one of GRC's primary NAS RAID arrays. Now that I had a version of SpinRite (which 660 individuals are currently experimenting with) which could examine such a drive with sufficient detail and in a reasonable time frame, I was extremely curious to know what exactly was going on with it. So I ran the current Alpha release of SpinRite 6.1 on that drive.
SpinRite found and fixed one short contiguous run of sectors on that drive, all which were uncorrectable, but there was nothing actually wrong with any of them. Something had apparently caused them to be miswritten — perhaps a power glitch occurred while they were being written. And their intransigence had spooked the RAID controller into writing off the drive. SpinRite re-wrote those sectors and found nothing whatsoever to be wrong with them or anything else on the drive. All SMART params looked great and there was no longer any reason to mistrust that drive. There was nothing systemic in evidence.
So, if nothing else you'll learn something that you don't already know.
Several additional points from what you later wrote: I would NOT run SpinRite through a VM unless you had no choice. SpinRite works directly with the system's hardware and anything that interposes should be avoided. It might be that a VM could give SpinRite true access to the hardware, but I would never "seek" that mode of operation over running on native hardware. Best in this instance to be sure about what's going on.
Also, until we get to SpinRite v7, which will natively support USB attached devices, USB *must* be avoided at all costs. We recently discovered serious and potentially undetectable bugs in several USB BIOSes so SpinRite v6.1 will now only operate upon the first 137GB of any drive attached through USB. And even without that deliberately imposed safety limit, the BIOS, and thus USB through the BIOS, doesn't provide anywhere near the same level of control and sensitivity as directly connected SATA and PATA drives attached to AHCI and ATA/IDE controllers, which v6.1 now supports.