Can Scanner hold drives hostage?

[fleggett1]

This question might already be answered, but I'm on the verge of a nervous breakdown with this system and could use some help along with a hefty bit of hand-holding.

Had to restart the system today because Windows claimed there was a filesystem error that needed fixing.  I have NO idea what drive it was referring to and, to make matters even more mysterious, no check happened upon reboot.  However, when the system did reboot, Drivepool said that a drive was missing and no longer part of my pool.

A big drive.  An important drive.  A drive storing terabytes of data.

Upon looking at Drivepool, I identified the drive that had dropped out and then looked at what Scanner was doing.  It was scanning the disk and still is given its size.  The thing is, Scanner isn't reporting anything wrong with it.  It can read the drive info, is reporting the sector map, there are no SMART warnings, and everything looks hunky-dory.

So, has Scanner put the drive into some sort of "It's mine and no one else can touch it!" mode?  Windows' own disk management isn't even reading the drive correctly, which has me REALLY concerned.

The scan is probably going to take all night to complete and I don't want to abort it in case it really does need a complete read.

So...help?

[fleggett1]

Well, some more info.  It's now at the brink of dawn here and I can't sleep, so I'm back at the computer.  Scanner completed the scan on that dropped-out drive.  Thinking I might be able to salvage the drive via a power cycle, I rebooted the system a second time and, this time, Windows found and fixed whatever filesystem problem it had encountered earlier.  Why it didn't do that during the first reboot is anyone's guess.  I have no idea if any actual files were affected nor, for that matter, which drive was repaired since Windows is stubborn about releasing such obviously critical information to the user.  I know in the old manual chkdsk days that recovered files would be saved in a FOUND.000 directory, but I don't see anything like that.  Maybe that's a good thing, but all this weirdness with intermittent SMART errors and out-of-the-blue filesystem problems is making me go more than a little batty.

In any event, after the second reboot, the drive reappeared in the pool, thank the Gods.  I don't know why the drive dropped in the first place and I suspect I'll never know, but I figured a follow-up was in order.

[Christopher (Drashna)]

Hold hostage, no. 

But if there was a file system error, the repair process may take the volume offline to fix it.  This is normal, and part of how CHKDSK (and related API) works.  The disk scan from windows, CHKDSK and the file system check in StableBit Scanner all use the same API/method, and can do this. 

 

[fleggett1]

Urgghh.  I need Sherlock Holmes to figure out what the hell is going on with my system.  Right now, DP is going through an absolutely agonizing rebalancing process that, by my estimation, will take at least 24 more hours to complete.  I don't know what spurred this, as it looks like it's evacuating an entire drive for no discernible reason.

It's a good thing I don't drink.

Anyway, thanks for the quick reply.  I'm ASSuming that, when the filesystem error occurred, that's when Windows took the drive offline, which caused it to drop out of the pool.  I'd still like to know what, exactly, happened to cause the error in the first place, but that's probably buried in some arcane unreadable logfile in a directory that's twenty levels deep.

I'm 99.9999% convinced I just need to shuck the entire system and start from scratch.  Save what drives look to be salvageable and throw some serious money at a new platform.  My system is so ancient that I can't even upgrade to Windows 11 (well, at least not without some Rufus-style hacks).

[Christopher (Drashna)]

Well, file system errors can be caused by a whole bunch of reasons.  But if DrivePool is also evacuating a drive, then it's likely because Scanner also marked that drive as bad.  So, if this is all the same drive, it could be that the drive in question is actively trying to fail.  

And sometimes, new hardware isn't a bad idea.  If you do, I recommend something that supports ECC memory.