›

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google

Sign In Register
Our Sites

covecube.com 
community.covecube.com 
blog.covecube.com 
wiki.covecube.com 
bitflock.com 
stablebit.com 

Categories

Poll

No poll attached to this discussion.

In order to better support our growing community we've set up a new more powerful forum.


The new forum is at: http://community.covecube.com


The new forum is running IP.Board and will be our primary forum from now on.


This forum is being retired, but will remain online indefinitely in order to preserve its contents. This forum is now read only.


Thank you,


M4: Heavy parallel activity takes drives offline?

evanr76evanr76
edited February 2012 in DrivePool
Build 5677

I am using a multi-core batch music converter (dbPowerAmp) to convert lots of files from one lossless format to another (average file size ~30MB).  My set-up has four cores.

I am converting files on one pool folder and storing them on another (different) pool folder.

I have a 6-drive pool connected via port multiplier (Rosewill enclosure) to an ASMedia 106x eSATA add-in card.

So, it's four cores doing sustained I/O across two pool folders on a 6-drive pool connected via port multiplier.

The conversion process starts smoothly enough, but after about 8-12 files, half the drive pool literally goes offline!  HD activity LEDs go black and the processing stalls.  Eventually (after a minute or so), the drives come back and the conversion resumes, only to fail again a few files later.

Share on Twitter

Comments

  • ShaneShane
    Resident Guru
    Hmm. In disk management, when this happens, are the physical drives still present and accessible? Any SMART errors? High drive temperatures?
    Share on Twitter
  • PaalandPaaland
    Member
    I experience something similar. Both when ripping CD's via iTunes and importing pictures via Google Picasa the process freeze for 5-10 seconds every now and again. Sometimes the application just hangs unresponsive, other times it gives a "write error" or "access denied" error. Just waiting a bit and retrying fixes the issue. Seems DrivePool is unable to handle the load and starts locking up. I've got 6 SATA drives in the pool as well on a quad core intel processor with 8 GB ram system. So the load should not be any problem at all.
    Share on Twitter
  • evanr76evanr76
    Member
    So I can make this happen with two different eSATA add-in cards.  The symptoms are the same - some kind of failure is taking an entire eSATA bus offline, e.g. this message in the system event log:

    "The driver for device \Device\Scsi\SI31241 detected a port timeout due to prolonged inactivity. All associated busses were reset in an effort to clear the condition."

    From what I understand, DrivePool operates on top of higher-level layers like NTFS, so in theory it shouldn't be able to crash the controllers - on the other hand, I'm not sure how an entire bus could get taken out.

    I'm going to submit some logs next time this happens (which should be in a few minutes since I'm running the batch music converter now).

    Share on Twitter
Sign In or Register to comment.