Unassigned Devices Preclear - a utility to preclear disks before adding them to the array

[jimeez]

1 hour ago, dlandon said:

Start by going to the cache_dirs settings and only inculding shares that need to be cached.  Normally any share on an SSD pool device doesn't need to be cached by cache_dirs.  Shares like 'isos' don't need to be cached.  Cache_dirs by default will cache everything it finds.  It can be a real memory hog.

Thank you for the replay.

 

I don't have any shares on the SSD Pool.  At least not intentionally.  And I wasn't quite sure how to do this, but I went into the settings and excluded every folder that IS NOT a normal media folder.  Hopefully this is what you were suggesting?

 

If I did do it properly, it did not fix it.  Same error on preclear attempt.

Edited October 27, 2023 by jimeez

[dlandon]

17 minutes ago, jimeez said:

Thank you for the replay.

 

I don't have any shares on the SSD Pool.  At least not intentionally.  And I wasn't quite sure how to do this, but I went into the settings and excluded every folder that IS NOT a normal media folder.  Hopefully this is what you were suggesting?

I actually prefer the opposite.  Only include the shares you want cached.  If I remember correctly, cache_dirs will cache remote shares and that is not necessry and is wasteful of memory.

[jimeez]

13 hours ago, dlandon said:

I actually prefer the opposite.  Only include the shares you want cached.  If I remember correctly, cache_dirs will cache remote shares and that is not necessry and is wasteful of memory.

Made the changes.  Still same error.  Is it possible it's a hardware issue?  Not the drive, but the mobo or sata ports?

[dlandon]

1 hour ago, jimeez said:

Made the changes.  Still same error.  Is it possible it's a hardware issue?  Not the drive, but the mobo or sata ports?

Maybe, but the error was a memory issue.  That's why I had you check the cache_dirs setup.  At this point I would run a memory test.

[jimeez]

2 hours ago, dlandon said:

Maybe, but the error was a memory issue.  That's why I had you check the cache_dirs setup.  At this point I would run a memory test.

Thank you for the responses and attempts to help.  I'm switching my setup back over to an Intel board in a few days.  Seeing some other wonkiness since switching to AMD.   Will wait until I make that change and run a preclear again; see what happens.  Thanks again.

[InReasonNotFish]

When I run the plugin from the Tools screen I get a message that says "There are no disks that can be precleared.".  But, I have a mounted disc in the unassigned Devices screen.  

When I look at the unassigned devices screen I don't see an icon for the preclear.

I've tried the plugin with the drive mounted & unmounted.  Not sure what I'm missing.

***

Update:  And I found my own answer.  I had to delete the partition first.

Edited October 29, 2023 by InReasonNotFish
more info

[jimeez]

On 10/27/2023 at 11:16 AM, dlandon said:

Maybe, but the error was a memory issue.  That's why I had you check the cache_dirs setup.  At this point I would run a memory test.

So I ran a memory test on both the old memory I was using and the new memory I just installed during the switchover (back) to an Intel setup.  All momory checked out fine; no errors.

 

But on the "new" setup, once I got everything up and running I attempted to preclear some discs.  Four discs to be exact.  They all failed and produced the same low memory error.  So I'm now using a new mobo, new processor, and new memory and am still getting the same error.  Also, during these new attempts to preclear, I stopped all dockers and VMs and shut down both services.  As you can see in the stats screenshot the system has plenty of available memory.

 

I'm at a loss.

Edited November 1, 2023 by jimeez

[dlandon]

13 minutes ago, jimeez said:

As you can see in the stats screenshot the system has plenty of available memory.

Go to a command line and show the output of this command:

df /

 

[jimeez]

7 minutes ago, dlandon said:

Go to a command line and show the output of this command:

df /

 

df / output:

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 Did you mean just:

df

 

 

[dlandon]

25 minutes ago, jimeez said:

df / output:

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 Did you mean just:

df

 

 

You got it right.  So that is the problem.  It shows no space on rootfs.  I'll have to do some research and see if I can figure out why.

[dlandon]

33 minutes ago, jimeez said:

df / output:

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 Did you mean just:

df

 

 

Give me the output of this command:

mount | grep rootfs

 

[jimeez]

2 minutes ago, dlandon said:

Give me the output of this command:

mount | grep rootfs

 

 

rootfs on / type rootfs (rw)
rootfs on /mnt type rootfs (rw)

 

[jimeez]

2 hours ago, dlandon said:

Go to a command line and show the output of this command:

df /

 

By the way, when I boot in safe mode I get this:

 

Filesystem     1K-blocks    Used Available Use% Mounted on
rootfs          16404780 1090788  15313992   7% /

 

And this:

rootfs on / type rootfs (rw,size=16404780k,nr_inodes=4101195,inode64)
rootfs on /mnt type rootfs (rw,size=16404780k,nr_inodes=4101195,inode64)

 

Edited November 1, 2023 by jimeez

[dlandon]

I was going to ask you to do that next.  That's what it should look like.  A plugin or docker is causing an issue.  Add them back one at a time until you see the issue.  Let me know what you find.

[jimeez]

1 minute ago, dlandon said:

I was going to ask you to do that next.  That's what it should look like.  A plugin or docker is causing an issue.  Add them back one at a time until you see the issue.  Let me know what you find.

I verified it was not a docker or VM.  Same results with all dockers and VM running.  No issues. 

 

How does one go about adding plugins one at a time? 

[dlandon]

Just go to the apps and add them back.

[jimeez]

3 minutes ago, dlandon said:

Just go to the apps and add them back.

So this is interesting.  I just booted back up under normal mode, but prior to doing so I removed .plg files from the \flash\config\plugins folder.  So no plugins are currently installed/running.  Docker is shut down and VM is shut down.  I am back to:

 

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 

[dlandon]

43 minutes ago, jimeez said:

So this is interesting.  I just booted back up under normal mode, but prior to doing so I removed .plg files from the \flash\config\plugins folder.  So no plugins are currently installed/running.  Docker is shut down and VM is shut down.  I am back to:

 

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 

Reboot with docker set to not start.  I suspect you have a docker container mapping issue.

[jimeez]

23 minutes ago, dlandon said:

Reboot with docker set to not start.  I suspect you have a docker container mapping issue.

Yep.  That's the condition I am currently sitting in.  Rebooted to a normal boot, but both Docker and VM services are turned off.  Plugins are all loaded and active.  Same rootfs condition.

 

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 

[dlandon]

Just now, jimeez said:

Yep.  That's the condition I am currently sitting in.  Rebooted to a normal boot, but both Docker and VM services are turned off.  Plugins are all loaded and active.  Same rootfs condition.

 

Filesystem     1K-blocks  Used Available Use% Mounted on
rootfs                 0     0         0    - /

 

Give me a new diagnostics in that condition.

[jimeez]

2 minutes ago, dlandon said:

Give me a new diagnostics in that condition.

Many thanks for the continued help.  Sent you a DM.

server-nas-diagnostics-20231101-1220.zip

[dlandon]

13 minutes ago, jimeez said:

Many thanks for the continued help.  Sent you a DM.

server-nas-diagnostics-20231101-1220.zip 140.76 kB · 1 download

Looking at the newest diagnistics, I have several suggestions for plugins to remove as a test:

• Compose manager.
• NerdTools - you may have a package being installed causing an issue.
• Gpu stat.
• Nvidia-driver.
• disk location.

 

The reboot.  I don't know if any of these are causing an issue, but let's eliminate the possibility.

[dlandon]

4 minutes ago, jimeez said:

Dan - after rebooting the server (same old flash drive, but on a USB 2.0 port) I am not able to mount the SMB shares now.  Did you change something in the SMB setting that would prevent these from starting?

 

 

UD thinks 'SERVER-NEW' is off-line.  Turn 'NETBIOS' back on and see if UD sees the server is online.  It may be needed to recognize that server.

[dlandon]

Just now, jimeez said:

Yep, that fixed it.

That older server needs it enabled so UD can resolve the IP address.

 

Did the USB2.0 port change let the 'df /' work?

[jimeez]

1 minute ago, dlandon said:

That older server needs it enabled so UD can resolve the IP address.

 

Did the USB2.0 port change let the 'df /' work?

Replying in/switching to DM.  Apologies.