Jump to content
Covecube Inc.
FjYskcwknon

RocketRAID 3540 - no SMART despite UnsafeDirectIo, DirectIoTest crashes

Recommended Posts

Hello,

 

I recently migrated from hardware RAID arrays on the Highpoint RocketRAID 3540 controller to DrivePool, using the legacy drive passthrough of said controller as well as additional disks directly connected to the motherboard.

 

I've set the UnsafeDirectIo option to True in the (properly renamed) config file, but SMART data cannot be read by Scanner. The DirectIoTest tool crashes when I select "Unsafe Direct I/O" and on all other methods, it detects the disk as being an SSD (which it isn't).

  Version: 
40
  Model:   HPTDISK 0-6
  Serial number: 

  Bus type:   RAID
  Command queuing:   True
  Device type:   0x00
  Raw device properties:   0
  Removable media: 
False
  Vendor ID:   HPT
  Product ID:   DISK 0-6
  Product revision:   4.00

I'd really like to get this to work and I'm available for any testing necessary. The email notification feature of the controller does not support SSL and I'd also have to parse SMART data from the Highpoint web interface to get at least some idea of the health of the drives otherwise.

Share this post


Link to post
Share on other sites

If any drives are getting SMART data, could you click on it, select "SMART details", scroll to the bottom, and check the "Submit to Bitflock" box? This may take a while, but it should create an ID. Once it does, post that here.

Share this post


Link to post
Share on other sites

Okay, I've flagged it for Alex, and he'll look at the backend.

https://stablebit.com/Admin/IssueAnalysis/6279

 

In the meanwhile, could you use this file:

https://www.dropbox.com/s/0krodbweqz1rrc4/Scanner.Service.exe.config

This is taken from my personal system, when I was using a HighPoint RocketRAID 2720SGL card. So it should work.

If it causes any issues, then delete the file.

 

ALso, with the directIoTest tool, could you check the "specific method" option and see if you can find a method that gets the "SMART Attributes" setting to "light up" (well show a green check/tick next to it, instead of the red "x"), and report which method does, if any?

Share this post


Link to post
Share on other sites

I forgot to mention I already tried your config (found in another topic) and the only difference to my config is that I have Smart_DoNotQueryRemovableMedia and Smart_DoNotQueryUnknownMedia

 set to True. With both configs, I get very frequent crashes of the Scanner service. I can provide event logs or error reports of the crashes later.

 

I think these crashes, as well as the crash of the test tool that I mentioned, might be from the same bug in the underlying library.

 

None of the methods in the test tool is successful in reading SMART attributes.

Share this post


Link to post
Share on other sites

The config file that I've linked is definitely an older one one.

 

If you're getting frequent crashs with the config file, then it's definitely having an issue with the UnsafeDirectIo setting.

So please remove the config file, so that it doesn't crash the service.

 

And none of the methods work at all? That's very unusual.

 

As for the card, are the disks set to "Legacy" mode (I believe that's the term that HighPoint uses)? Or are they part of an array or "jbod"?

Share this post


Link to post
Share on other sites

The crashes are definitely from the UnsafeDirectIo setting, yes. I have found error report files in C:\ProgramData for both Scanner and DrivePool and I went ahead and sent them via the contact form, referencing this topic. Maybe they help.

 

All disks are in legacy mode.

Share this post


Link to post
Share on other sites

Okay, remove the advanced config file for now (if you haven't already). Stability is more important.

 

Could you try one method at a time by checking "Specific method" in Direct I/O Test and selecting methods in this order:
  • SmartIoctl
  • AtaPassThrough
  • ScsiPassthrough
  • ScsiPassthrough48
  • ScsiMiniportClassDriver
  • ScsiMiniportPortDriver
Starting at the top with the "safest" method and going down to the most "unsafest" one.
 
After you select a method wait for it to query the selected drive and report if that method works. If you find one that works then let me know and I'll see if we can add something to the code to force that method.

Share this post


Link to post
Share on other sites

The first one there, "Windows Server 2012, 2008 / Windows 8, 7, Vista (32/64-bit)", v1.2.25.9 . I could try the Vista or WHQL one, but it's the one the OS preferred.

 

None of the methods do any more than this:

removed

 

If I switch between the methods randomly, I can sometimes provoke a crash. it says "unknown software error" and causes a .NET runtime event log entry like this: removed

Share this post


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
Reply to this topic...

×   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...