Jump to content
Covecube Inc.
  • 0

DrivePool and Drive power-down's


bblue

Question

I've read that DrivePool should correctly handle bringing up drives in the pool as needed, but if a full directory is done, all drives may have to be powered up and this can take some five seconds per drive.

 

Right now I'm on my second server build with W8.1 Pro as the host. Right now this server has 18 drives, ten of which are in one drive pool, containing audio and video media, to the tune of about 34T, mirrored.

 

Most of the time there is no issue, but occasionally after several hours away from the computer I'll still have a window open to one of the media directories, and then proceed to drag-'n-drop copy to that folder. If the copy contains a bunch of smaller files like an album, it will copy one or two, and then stop. I've waited up to 15 minutes for it to do something, but nothing happens and I cannot close the window of the open folder, though I can kill the copy process.

 

I see something similar occasionally when after a long time (sufficient for all-drive power down) I open a folder to the top level of the pool. I can see its directory (probably cached) but trying to access a sub directory, I get the same behavior.

 

This suggests one or more drives not powering up appropriately, or DP somehow not aware that it/they did.

 

Is any of this familiar? I'm not running the scanner, and the service for it is currently disabled (that'll be another issue I'll post about in a few days). Both of these are the latest V2 BETA's (on the web site).

 

The drives are all Seagate 3T or 4T and one WD 3T. Mostly 4T's are used.

 

Any comments or advice?

 

--Bill

Link to post
Share on other sites

6 answers to this question

Recommended Posts

  • 0

The Ordered File Placement balancer may help with this issue. It fills one disk at a time, and tries to keep the contents of a folder together on one disk. This would help minimize the "wake drive" issue.

 
Additionally, since you're using the latest beta (2.1.0.46x?), could you click on the gear (top right corner), select "troubleshooting" and then "enable file system logging". Then try to reproduce this issue, and note the approximate time? And then upload the logs to us?

 

Additionally, if you have plenty of RAM, there is a "hack" that tells Windows to use more memory for file system cache, which means quicker access for repeated commends (such as listing drive contents).

http://technet.microsoft.com/en-us/library/cc785435.aspx

"fsutil behavior set memoryusage 2". 

 

And as I said, this is a "hack".

Link to post
Share on other sites
  • 0

The Ordered File Placement balancer may help with this issue. It fills one disk at a time, and tries to keep the contents of a folder together on one disk. This would help minimize the "wake drive" issue.

My Placement balancer shows all drives present, and both duplicated and non-duplicated checkboxes checked for each. That was the default, is it not correct?

 

Additionally, since you're using the latest beta (2.1.0.46x?), could you click on the gear (top right corner), select "troubleshooting" and then "enable file system logging". Then try to reproduce this issue, and note the approximate time? And then upload the logs to us?

I'm using 2.1.0.432 which is the latest Beta I can find anywhere. Where do I download 46x? There is no troubleshooting command that I can see in 432.

 

Additionally, if you have plenty of RAM, there is a "hack" that tells Windows to use more memory for file system cache, which means quicker access for repeated commends (such as listing drive contents).

http://technet.microsoft.com/en-us/library/cc785435.aspx

"fsutil behavior set memoryusage 2". 

 

And as I said, this is a "hack".

Why is it a hack? It seems like a good adjustment to make for a system that has numerous drives with high directory and file counts? I would run out of directory caching on certain drives frequently. While it updates and shuffles things around to even complete the directory listing on the requested drive, it can take a minute or two, and that was before I added DrivePool for testing with ten more drives.

 

Seems quite logical to me. Any further details on the 'hackness' of it?

 

Thanks.

 

--Bill

Link to post
Share on other sites
  • 0

Yes, that should be the correct setup.

 

 

Yeah, the "troubleshooting" option isn't on build 432.

As for any newer builds, all the relevant links can be found here:

http://wiki.covecube.com/Downloads

(keep in mind, these are internal beta's and may not be as stable. Also, newest at the bottom)

 

 

As for why the FSUTIL command is a hack? Because it's "not standard", and falls under the category of "you shouldn't do this if you don't understand it's implications". That said... I've done this on most of my systems. My server has a LOT of spare memory at any given time, and most of the other systems have 8GBs of RAM and maybe use half of it.

But again, any usage of "FSUTIL" is basically a "hack".  But it definitely sounds like it may be something worth doing on your system. Just make sure you have plenty of memory available for this.

Link to post
Share on other sites
  • 0

...

Yeah, the "troubleshooting" option isn't on build 432.

As for any newer builds, all the relevant links can be found here:

http://wiki.covecube.com/Downloads

(keep in mind, these are internal beta's and may not be as stable. Also, newest at the bottom)

These almost look like auto-nightly builds. Is 472 known to be in reasonable condition?

 

As for why the FSUTIL command is a hack? Because it's "not standard", and falls under the category of "you shouldn't do this if you don't understand it's implications". That said... I've done this on most of my systems. My server has a LOT of spare memory at any given time, and most of the other systems have 8GBs of RAM and maybe use half of it.

But again, any usage of "FSUTIL" is basically a "hack".  But it definitely sounds like it may be something worth doing on your system. Just make sure you have plenty of memory available for this.

It seems to be working fine, and watching the memory, there hasn't been any significant increase that I notice. So far no directory stalls, either. So that's a good thing.

 

Do you ever find yourself needing to modify mftzone in fsutil? It seems not to apply unless you have gazillions of very small files. Not usually something you would see in a media oriented server.

 

--Bill

Link to post
Share on other sites
  • 0

These almost look like auto-nightly builds. Is 472 known to be in reasonable condition?

 

 

These are not auto nightly builds but are deliberately built after a Code Change Request (CCR) is completed (see: http://wiki.covecube.com/Development_Status#Development_Workflow). The build process is automatic and the publish is automatic.

 

The difference between these internal BETAs and the BETAs that are published on StableBit.com is that the internal BETAs don't undergo any kind of functional testing before a build is published. The BETAs published on StableBit.com undergo basic functional testing and the installer's upgrade procedure is tested on every supported OS. This takes a bit of time to accomplish so it's only done for public BETAs.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...