dlandon

It looks like the disk is formatted with a file system without a partition. UD will only mount a disk that has a file system on a partition. Can you move the drive to a system that can mount it? If you can, you can then mount it as a remote mount in UD and deal with it that way.

Try this: Click on 'Default' on the device settings page. Replace the default script section when the disk mounts with the desired script, or reference the desired script in the section that executes when the device is mounted. Or, place your desired script at '/boot/config/plugins/unassigned.devices/' and choose it from the dropdown. '/boot/' is the Linux reference, '/flash/' is the SMB reference. Click 'Save'. The only thing you can change in the 'Device Script:' field is the file name. You can't change the path. I would just use the default file name. You can run a script at another location by referencing it in the UD script, but UD wants all device scripts at '/boot/config/plugins/unassigned.devices/'. Any other path will be rejected.

Unraid will attempt to spin down a disk if it has a partition. If there is no partition, it will not attempt to spin down the disk. Once a disk is precelared, the preclear signature and a blank partition are created on the disk. Unraid will attempt to spin down the disk during post read. As you noted, the disk will spin back up. The reason this happens is Unraid is detecting file reads and writes to determine if a disk should be spun down and because there is no file system, there will not be any file reads and writes. The spin downs can be ignored. I did release a new version of UD Preclear today to address some nginx issues I saw in your log.

I'm really confused here. You said: and the same day you said: How can that be? And stop using the script. Make this change as suggested: We need to have answers that we can implement in Unraid. The script is not the answer to address this issue.

Post Unraid diagnostics and UD disgnostics,. Get UD diagnostics by entering 'ud_diagnostics" on a command line, then post the /flash/logs/ud_diagnostics.zip file here. Also, go to the UD Settings and click on the "Remote Servers" tab and enter the server name or IP address and then the 'Test' button and post the output here.

This will be added to the next release of the AD Extras plugin. Please follow this post for updates:

I am working on an upgrade to this plugin: Additional settings for Samba related to AD permissions. Backup and restoration of the database so permissions will remain persistent on a reboot. Set initial share and disk permissons specifically for AD. Only 'tdb' will be available as a backend database. The idea here is to get AD working properly and then incorporate these changes into Unraid. Keep in mind this is beta/test software so be prepared to deal with issues and provide feedback so we can get any kinks worked out. The one thing that can't be done right now is keeping the vfs object acl_xattr on shares when a share change is made. You can fix that by clicking on the 'Clear Cache' or 'Default' buttons or make a change on the page and click 'Apply' on the AD Extras page. That will re-apply the 'vfs object = acl_xattr' back to all the shares. When the changes are made in Unraid, this will no longer be an issue. I do not have a Windows Server and can only do limited testing. I rely on AI to give me direction. I will need feedback from the community to verify the changes.

The UI is at IP:9000. 9000 is a common docker container port so check for conflicts.

Post diagnostics.

What browser do you use?

I see an issue and will update UD Preclear after my testing. For now it is fairly harmless.

What version of Unraid are you using? This has been fixed since Unraid 6.12.5. If you are using a later version of Unraid, please post the full diagnostics.

Where are you trying to use special characters?

Does the log show the: [Dec19 21:13] receive_cb_reply: Got unrecognized reply: calldir 0x1 xpt_bc_xprt 00000000c25205b7 xid 783fc85a messages. This is on 7.0.0 rc2?

If you click on the "Clear Cache" it will clear the cache and restart Samba for you. Let me see what I can do.

I thought you were running 7.0.0-rc2? Please run rc2 with the addition to the share.cfg file. You can edit this file, update to rc2, then reboot and it will take effect.

Post diagnostics when the array won't stop.

You need to post on the LMS Forum as this is a coere LMS issue.

Some CSS changes in UD may have caused this. I would suggest: Try another browser. Clear your browser cache. Upgrade to 6.12. And see if it clears up.

We have identified and corrected the issue with NFS Daemon not stopping properly. This fix will be in the 7.0 rc2 release and will be backported to 6.12. Please retest with this release and see if the issue is corrected. This is interesting. I will look into this further.

Follow this procedure: Update UD and UD Preclear. Reboot. This insures there are no mounts that show in 'df'. Remove partitions from the disks to be precleared. Preclear the disks.

I see an issue with a disk that has a partition but no file system. I'll work on sorting that out.

This is extremely confusing for the following reasons: Preclear should never allow preclearing a disk with a partition. In your case, none of the partitions have a file system. This may be a bug in preclear. I will investigate. Preclear should never try to preclear a disk that is mounted. How did these disks get mounted? Dev 2 is showing disks /dev/sdk twice with one preclearing and the other with the 'Mount' button grayed out. I don't know how this could happen.

It looks like those disks are mounted with UD. That's why you can't preclear them. Please show a screen shot of the UD page showing those two disks. If you use the preclear page to start a preclear, show a screen shot of that page. You shouldn't be able to start a preclear on mounted disks.

If that is disk /dev/sdj, it appears to have cleared.