Free space is low on hard drives

[shawn149]

Hello,

 

I have a drive pool set up with 4 2TB drivers for a total of 7451.4GB.  

 

I copied about 200GB last night to the pool.  This morning there was an error in my alert viewer:

 

Free space is low on hard drives

The following hard drives have less than 10% of free space: Z:\

 

When I check the pool, it shows the capacity of 7451.4GB, the Used space as 3610.3GB and Free space as 3841.1GB.  Unless my math is really screwy, I have way more than 10% free.  Looking at the Stablebit DrivePool dashboard, I'm just shy of 50%.  I rebooted several times and I rescanned my storage.  Has any one else encountered this before alert before with their drive pool?

 

I'm running Windows Server Essentials 2012, with StableBit DrivePool version 2.0.0.280.

 

Thanks

[shawn149]

Well, maybe I didn't wait long enough...  I just checked and the alert appears to have cleared itself. 

 

Maybe I put too much on the drive too quickly and I needed to wait 24 hours for the system to catch up?

 

In any case, the alert is gone now, but if anyone has any insight, I would be interested to hear!

[Christopher (Drashna)]

If you had the drive 90% filled at any time (like, lets say, you'd just added the one disk, and it filled most of that disk), then it would have triggered this warning. And .... the notification system sometimes doesn't clear the alert until you manually clear them. Or till the next time it does a "server assessment".

 

Otherwise, the assessment may have taken place while the system was still booting, and wasn't able to properly process the storage. Not entirely sure. (not without logs at least)

[shawn149]

So, after I posted on June 28th that the message went away, I left the server alone.  I.e. no new data. 

 

Today, when I signed on the "Free space is low on hard drives" message is now back.  I don't exactly know when the message appeared (since I haven't used the server in a couple of days), but I'll keep an eye on it.  I did click on the refresh button in the alert viewer to see if the message would disappear (since I verified that there is indeed plenty of space and the screen shots I provided earlier are essentially the same) and the message did not disappear.

 

That said, does anyone know if there are any logs I can search for to see when this event was triggered?  How about any logs that detail what Covecube was doing?

 

The message doesn't appear to be impacting anything, but it would be nice if it didn't throw a false positive (which is what it appears to be at this point with the information I have at the moment).

[Christopher (Drashna)]

If the error is happening during boot, then it may be that not all the drives are loaded into the pool....

 

 

But as for the Essentials stuff, all that stuff is logged in "C:\ProgramData\Microsoft\WindowsServer\Logs" (or something very close to that).  I can take a look at that if you want (best would be to upload it to dropbox or another cloud sync site, as the logs can be large.  Otherwise, I'm not sure why it would be triggering that "randomly".

[shawn149]

Okay, in the Dashboard.log file, I found an entry that tells when the event was fired:

 

[7188] 130703.200145.3240: AlertsViewer: UpdateItem COSMOS 7/1/2013 7:29 PM,Free space is low on hard drives

 

Now for the trick question, why?  I'm not seeing anything in the logs (and there are a log of them) around that time that look related...

 

@drashna, I would be glad to upload to Dropbox.  In the meantime, I'm going to email Covecube support.

[shawn149]

And... never mind. 

 

I think I figured out the problem and it was an "interpretation" problem.  The Console Alert wasn't referring to the DrivePool, but to a hard drive on one of the machines that was being backed up.  Both the server and the machine had a "Z" drive.  I wasn't aware of the Z drive on the other computer and I assumed (albeit falsely) that the only Z drive on my network was the one on the server which was assigned to the drive pool.

 

Sorry for the confusion...

[Christopher (Drashna)]

Well, glad you got it sorted out then. 

 

Also, if I may make a suggestion here... don't use Z:\ on the clients. If you need to use the Client Computer restore feature, it will throw a fit, because it expects to be able to use the Z: drive (or at least, WHSv1, and WHS2011 have this issue, and I suspect it wasn't changed in 2012E).