Jump to content
  • 0

New drive not balancing


Scuro

Question

Got a new drive and noticed that no matter how many times I tell it to balance, it won't balance fully.

https://imgur.com/a/t4Dgi

Service log shows no errors but it does say "0:10:38.3: Information: 0 : [Rebalance] Cannot balance. Immediate re-balance throttled."

Regardless, if I try a manual rebalance it only runs for a few minutes and then shows the same balance as shown in picture.

 

I am running the latest release client.

Link to comment
Share on other sites

13 answers to this question

Recommended Posts

  • 0

The event you've posted most likely has nothing to do with the issue that you're seeing.  If you have the "not more often than every X hours" option enabled, then you'll see this. 

 

That said, could you enable tracing and try to enable file system logging and initiate rebalancing?
http://wiki.covecube.com/StableBit_DrivePool_2.x_Log_Collection

And after that, run the StableBit Troubleshooter?
http://wiki.covecube.com/StableBit_Troubleshooter

Use "3451" as the Contact ID

Link to comment
Share on other sites

  • 0
8 hours ago, Christopher (Drashna) said:

The event you've posted most likely has nothing to do with the issue that you're seeing.  If you have the "not more often than every X hours" option enabled, then you'll see this. 

 

That said, could you enable tracing and try to enable file system logging and initiate rebalancing?
http://wiki.covecube.com/StableBit_DrivePool_2.x_Log_Collection

And after that, run the StableBit Troubleshooter?
http://wiki.covecube.com/StableBit_Troubleshooter

Use "3451" as the Contact ID

I will run that for you and submit the troubleshooter for you shortly, but I have been having serious issues since upgrading to the RC builds. Any program that puts IO stress on the pool is hard locking windows 10 after throwing disk errors into event history ("The IO operation at logical block address 0x1 for Disk 2 (PDO name: \Device\00000039) was retried."). Disk 2 is drivepool.

I had first thought a drive may be dying but after replacing two drives that sometimes showed errors in event viewer during these lockups but a lockup happened again this morning.

I have (had as I now moved them off drivepool) a hyper-v VM and a Maria database in drivepool. Any extensive IO operations by these and windows would hard lock (no BSOD, computer would be unresponsive to everything). Is there a recommended stable version for me to try downgrading to?

Link to comment
Share on other sites

  • 0
14 hours ago, Scuro said:

I will run that for you and submit the troubleshooter for you shortly, but I have been having serious issues since upgrading to the RC builds. Any program that puts IO stress on the pool is hard locking windows 10 after throwing disk errors into event history ("The IO operation at logical block address 0x1 for Disk 2 (PDO name: \Device\00000039) was retried."). Disk 2 is drivepool.

LBA errors are normal on the Pool drive, and shouldn't be causing any sort of issues.  Period.  Basically, it's requesting data from a specific block on the disk. Which doesn't exist, since it's not a physical disk and has no blocks.  So the pool returns an error.  This is normal and fairly common.

As for the hard locks, I have Win 10 FCU (1709), and the latest RC build on my system. I'm not seeing this behavior, at all. 

If you could, get a memory dump from when it is locking up:
http://wiki.covecube.com/StableBit_DrivePool_System_Freeze

15 hours ago, Scuro said:

Is there a recommended stable version for me to try downgrading to?

Unfortunately, the recommendation would be the RC build.  

Otherwise: 2.1.1.561. 

Link to comment
Share on other sites

  • 0

I appreciate the help. I have sent the troubleshoot report in.

 

Do you have any VMs or SQL databases on drivepool? The hardlocks appeared to happen after any intense IO operations from either or of these services. It may be better for me to remove these services form the pool regardless.

Still having issues with the drives not balancing correctly.

Link to comment
Share on other sites

  • 0

On Windows 10, I do run Outlook from the pool.  It's a couple of large PST files, and are handled like databases.

And in case it would trigger it (not as likely), copied a number of large (multiple GB) files to the pool.

Also, Alex (the Developer) actually runs his VMs off of a pool, as well. 

 

That said, getting a crash dump from when the system locks up would be the most useful, as we could see what was going on at that time, and what was causing the lockup.   It may not be DrivePool related and could be hardware related.  BUt if it is DrivePool related, this should make it very obvious. 

Link to comment
Share on other sites

  • 0

Just as an update, I believe I found the root cause of the hard lockups.

IO errors in event log showed IO hardware failures on almost every drive. I replaced two that had the most errors thinking that they were failing. The IO errors and lockups continued. After doing further searching in event logs, I found that the IO errors all started an hour after I had replaced the PSU on the server. This leads me to what I would now believe a faulty PSU, not drivepool (which I had updated around the same time). I will be returning the PSU and getting a different model to verify.

On another note however, I still can't get drivepool to balance the new drive even though drivepool knows that it is not balanced. Troubleshooting is still ongoing with stablebit.

I've disabled write caching on the drives to avoid corruption of files until crash is fully diagnosed.

Link to comment
Share on other sites

  • 0

Just an update for this thread, I've updated to the latest version (2.2.0.906) and still unable to get balance to finish. Drivepool will recognize it's unbalanced but will randomly stop balancing in the middle of it's balance task.

Link to comment
Share on other sites

  • 0
2 hours ago, Christopher (Drashna) said:

Have you tried setting the balancing ration to 100%

Still not finishing it's balance. I believe I can tell why.

Based on the balance targets, it is trying to balance my SSD that it is not able to add more data because of file placement restrictions. I only allow a few folders on my SSD. The balance plugin seems unaware of this and tries to balance regardless.

Link to comment
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...