Unassigned Devices - Managing Disk Drives and Remote Shares Outside of The Unraid Array


Recommended Posts

47 minutes ago, johnnie.black said:

I believe pwm is right, you need to specify the partition (p1)

This is what I am trying to figure out.  UD has been doing it this way without specifying the partition for a long time.

Jan  9 11:21:50 BackupServer unassigned.devices: Format disk '/dev/sdg' with 'xfs' filesystem result: meta-data=/dev/sdg1              isize=512    agcount=4, agsize=19535700 blks          =                       sectsz=512   attr=2, projid32bit=1          =                       crc=1        finobt=1, sparse=0, rmapbt=0, reflink=0 data     =                       bsize=4096   blocks=78142798, imaxpct=25          =                       sunit=0      swidth=0 blks naming   =version 2              bsize=4096   ascii-ci=0 ftype=1 log      =internal log           bsize=4096   blocks=38155, version=2          =                       sectsz=512   sunit=0 blks, lazy-count=1 realtime =none                   extsz=4096   blocks=0, rtextents=0
Jan  9 11:21:53 BackupServer unassigned.devices: Reloading disk '/dev/sdg' partition table.
Jan  9 11:21:53 BackupServer kernel: sdg: sdg1
Jan  9 11:21:53 BackupServer unassigned.devices: Reload partition table result: /dev/sdg:  re-reading partition table

Ok.  I stand corrected.  The log message shows the drive and not the partition.  It is actually formatting /dev/sdg1.  This tells me what is wrong with the nvme format command now.  It needs the partition as nvne0n1p1.

 

Fix is now coming.

Edited by dlandon
Link to comment
9 hours ago, deaerator said:

The update didn't work.

 

here is the disk log



Jan 8 22:17:01 Tower kernel: nvme0n1: p1
Jan 8 22:17:32 Tower emhttp: Samsung_SSD_960_EVO_250GB_S3ESNX0JC76289K (nvme0n1) 244198552
Jan 8 22:34:08 Tower kernel: nvme0n1: p1
Jan 8 22:39:47 Tower unassigned.devices: Removing partition '1' from disk '/dev/nvme0n1'.
Jan 8 22:55:21 Tower unassigned.devices: Removing partition '1' from disk '/dev/nvme0n1'.
Jan 9 11:11:36 Tower unassigned.devices: Adding disk '/dev/nvme0n1p1'...
Jan 9 11:11:36 Tower unassigned.devices: Mount drive command: /sbin/mount -t precleared -o auto,async,noatime,nodiratime '/dev/nvme0n1p1' '/mnt/disks/Samsung_SSD_960_EVO_250GB_S3ESNX0JC76289K'
Jan 9 11:11:36 Tower unassigned.devices: Mount of '/dev/nvme0n1p1' failed. Error message: mount: unknown filesystem type 'precleared'
Jan 9 11:13:00 Tower unassigned.devices: Device '/dev/nvme0n1' block size: 488397168
Jan 9 11:13:00 Tower unassigned.devices: Clearing partition table of disk '/dev/nvme0n1'.
Jan 9 11:13:00 Tower unassigned.devices: Reloading disk '/dev/nvme0n1' partition table.
Jan 9 11:13:00 Tower unassigned.devices: Reload partition table result: /dev/nvme0n1: re-reading partition table
Jan 9 11:13:00 Tower unassigned.devices: Creating a primary partition on disk '/dev/nvme0n1'.
Jan 9 11:13:00 Tower unassigned.devices: Creating unRAID compatible mbr on disk '/dev/nvme0n1'.
Jan 9 11:13:01 Tower kernel: nvme0n1: p1
Jan 9 11:13:01 Tower unassigned.devices: Reloading disk '/dev/nvme0n1' partition table.
Jan 9 11:13:01 Tower kernel: nvme0n1: p1
Jan 9 11:13:01 Tower kernel: nvme0n1: p1
Jan 9 11:13:01 Tower unassigned.devices: Reload partition table result: /dev/nvme0n1: re-reading partition table
Jan 9 11:13:01 Tower unassigned.devices: Formatting disk '/dev/nvme0n1' with 'xfs' filesystem.
Jan 9 11:13:01 Tower unassigned.devices: Format disk '/dev/nvme0n1' with 'xfs' filesystem failed!

 

Update to 2018.01.09b.  It should format the disk now.

Edited by dlandon
Link to comment
36 minutes ago, jonathanm said:

I don't think that card is currently recommended. I'd do some searching on the forum to see what issues you may encounter with a marvell based HBA like that.

 

Found it...

 

Good catch, thanks.

 

Some users remarked that Marvell may have patched this problem in later versions of its firmware. I'll give it a shot and report back. 

Link to comment

I'm not entirely sure this is the right place to post this but here we go.

 

I'm setting up my first Unraid Server from a machine that used to run Ubuntu server. I'm a bit of a noob and I'm stuck at a critical part of the migration process.

 

The Ubuntu server had around 6 TB on it so I added two 4 TB drives for the transition and copied over up my 6 TB of data on there. I added the two 3 TB drives to the array.

 

I planned to use the Unassigned Devices plugin with Krusader to copy the data from the two 4 TB drive to the array but I can't seem to mount the drives. I need a hand to figure out how to mount the devices or migrate the data another way.

 

Screenshot of the unassigned devices | Logs of the two disks I can't mount

  • The 1 TB disk used to be the ubuntu boot drive, it's mounted properly and I can access the data
  • The two 4 TB disks are the one I'm looking to access, the filesystem isn't displayed on the UI but they're ext4 drives formatted in Ubuntu

Thanks in advance to anyone willing to help :)

Edited by odrel
Link to comment
4 hours ago, odrel said:

I'm not entirely sure this is the right place to post this but here we go.

 

I'm setting up my first Unraid Server from a machine that used to run Ubuntu server. I'm a bit of a noob and I'm stuck at a critical part of the migration process.

 

The Ubuntu server had around 6 TB on it so I added two 4 TB drives for the transition and copied over up my 6 TB of data on there. I added the two 3 TB drives to the array.

 

I planned to use the Unassigned Devices plugin with Krusader to copy the data from the two 4 TB drive to the array but I can't seem to mount the drives. I need a hand to figure out how to mount the devices or migrate the data another way.

 

Screenshot of the unassigned devices | Logs of the two disks I can't mount

  • The 1 TB disk used to be the ubuntu boot drive, it's mounted properly and I can access the data
  • The two 4 TB disks are the one I'm looking to access, the filesystem isn't displayed on the UI but they're ext4 drives formatted in Ubuntu

Thanks in advance to anyone willing to help :)

UD does not recognize the file system on the drives.  Were the drives in a raid array?

Link to comment

So I booted on a live USB of lubuntu and couldn't mount the drive because of a permission issue, put the drives on 777 and was able to mount them and browse their content, it didn't help for Unraid though.

 

I did however notice that for some reason, Ubuntu apparently didn't create a partition table for the disks, here is how GParted displays it. I don't know much but I think this could be the cause of the issue?

 

Can I create a partition table without compromising the data on the disk?

 

EDIT: An alternative solution could be to just copy the content of the 4 TB disks to the 3 TB disks already in the array from lubuntu, is there any risk to that?

Edited by odrel
Link to comment
9 hours ago, odrel said:

Can I create a partition table without compromising the data on the disk?

No.

 

9 hours ago, odrel said:

EDIT: An alternative solution could be to just copy the content of the 4 TB disks to the 3 TB disks already in the array from lubuntu, is there any risk to that?

This is a viable solution.  No problem if you do the copy correctly.

Link to comment

There is a strange problem when I mount a remote samba share.  When I run rsync  ( rsync -avzP --delete /mnt/user/tv/ /mnt/disks/10.0.0.120_500GB-2.5/tv )  it copies every time the same files that already exist on that share. I have different rsync parameters but nothing works. I haven't encountered such a problem running rsync in general other than using unassigned devices plugin with a samba shared folder. I would appreciate if you have any ideas what I'm doing wrong. 

Link to comment
22 minutes ago, etsi said:

There is a strange problem when I mount a remote samba share.  When I run rsync  ( rsync -avzP --delete /mnt/user/tv/ /mnt/disks/10.0.0.120_500GB-2.5/tv )  it copies every time the same files that already exist on that share. I have different rsync parameters but nothing works. I haven't encountered such a problem running rsync in general other than using unassigned devices plugin with a samba shared folder. I would appreciate if you have any ideas what I'm doing wrong. 

Mounting a samba share shouldn't make any difference to the rsync command from local to local share mount.  There is probably a permissions issue on the remote samba share that confuses rsync with regards to dates.

 

Where is the remote samba share?  What type of server?  Could you rsync directly to that share?

Link to comment
34 minutes ago, dlandon said:

Mounting a samba share shouldn't make any difference to the rsync command from local to local share mount.  There is probably a permissions issue on the remote samba share that confuses rsync with regards to dates.

 

Where is the remote samba share?  What type of server?  Could you rsync directly to that share?

 

Using --modify-window=5 fixed my problem. There is always  1 second  difference timestamp between source and target files.  The remote samba share is a wdtv media player but also I have a problem with QNAP samba share.

Edited by etsi
Link to comment
21 minutes ago, etsi said:

 

Using --modify-window=5 fixed my problem. There is always  1 second  difference timestamp between source and target files.  The remote samba share is a wdtv media player but also I have a problem with QNAP samba share.

What is your problem with the QNAP share?

Link to comment

I'm using the plugin now to mount a NFS share hosted on another linux box in my house.  Just wondering if there is any way to change the local path that share gets mounted at to something other than /mnt/disks/<IP of server>_<name of share>. It would be nicer if I could just have it mounted at /mnt/disks/plexdl.  This feature exists for mounted local drives.

 

TIA for any help, and I apologize if I missed something really obvious here

Link to comment
2 hours ago, devros said:

I'm using the plugin now to mount a NFS share hosted on another linux box in my house.  Just wondering if there is any way to change the local path that share gets mounted at to something other than /mnt/disks/<IP of server>_<name of share>. It would be nicer if I could just have it mounted at /mnt/disks/plexdl.  This feature exists for mounted local drives.

 

TIA for any help, and I apologize if I missed something really obvious here

Click on the mount point name when the share is not mounted and change the name.  Press enter to accept.

Link to comment
1 minute ago, deaerator said:

I'm starting to see these errors in my syslog

Unsure how to fix this.

 

I can't collect any syslog or diagnostics file.  It's seem the system is hanging or stuck.

 

 

Did you try searching the forum? wrong csrf_token has been answered many times.

Link to comment

Hi all, first time posting here. I've been using unRAID for ~1 year, and have been fortunate enough to have your descriptive posts for troubleshooting and answering my many questions. A question that I can't locate the answer to brings me to my first post:

 

I have a single drive mounted by unassigned devices, it's shared as an SMB. The UD share is titled 'Downloads'. I am running unRAID 3.5. 

 

Problem: unRAID persistently adds the UD share 'Downloads' as a user share. Meaning I have two shares with the same name (Downloads), one a user share, the other a UD share. This causes issues with SMB, when I navigate on a Windows machine through windows explorer to my UD Downloads share I get an empty folder, I believe because SMB maps \\(IP address of my unRAID box)\Downloads to the user share, which I've never populated and is therefore empty, not to the UD share. This only started happening after upgrading to 3.5. 

 

If I manually delete the user share (been doing this via the GUI) then after some time (1-10mins typical) the SMB share remaps, and I can access the UD share using the same path, \\(IP address of my unRAID box)\Downloads, but eventually unRAID recreates the user Downloads share. This re-creation of the user share happens without rebooting, meaning I delete the user share, don't restart the system nor array, and within 24 hours unRAID re-creates the user share during the same boot. 

 

Apologies if this issue has already been identified and worked through, just wasn't able to locate the solution in my search. Please let me know if I can provide any further screens/details.

Link to comment
1 minute ago, itimpi said:

It sounds as if you have a top level folder called Downloads on either the cache drive or one of the array drives.     A User Share is automatically created for all top level folders on the cache drives or array drives.

Yes, this is almost certainly the reason for the User Share. Maybe you have a docker misconfigured and it is creating the folder.

Link to comment
2 hours ago, trurl said:

Yes, this is almost certainly the reason for the User Share. Maybe you have a docker misconfigured and it is creating the folder.

 

Thanks itimpi and trurl, spot on, misconfigured docker. One of my containers was set to mnt/user/Downloads. Also checked the disks, they were clear, probably from the last time I deleted the user share. Thanks for the quick fix!

Link to comment
  • trurl pinned this topic

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  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.