dlandon

I guess I misunderstood what you were doing. Your first screen shot is the stock Unassigned Devices display without the UD plugin being installed. Your second screen shot is with the UD plugin installed. The devices are being discovered properly. You have not setup any array devices and of course your array is not started. You are in a weird state and I'm not sure that UD is detecting unassigned devices. Unraid dynamically sets the unassigned devices for UD by assigning the 'devX' designation. I don't believe UD is keeping up with the Unraid changes in this condition because UD thinks some disks are still assigned to the array. This is not a situation I've tested before - i.e. no array disks assigned. Uninstall the UD plugin and set up your array. Once you'e done that. start the array and then install UD to manage the unassigned disks. Post back here if you still see issues after doing that.

Understood, but I need to see what udev is seeing in each case. I cannot reproduce this issue and will need your diagnostis to try to figure it out. UD+ is only some additional packages installed for additional support. I doesn't change anything that I know of that would affect what devices udev sees.

Post diagnostics when all show and when only some show.

Try doing a force update to refresh the docker container.

I was finally able to reprodice this issue, That's the good news; the bad news is there is no fix that can be done to fix this in Unraid. UD uses some built in functions to perform the script file operations. I can find no solution for Unraid to fix this. This only occurs when Private Browsing in Chrome and appears to be a bug in one of Chrome's security features causing POST failures in specific circumstances. There are no other instances of doing things like UD does in Unraid, so this should not show up anywhere else. There are several options to deal with this: Turn off security in Chrome when Private Browsing. Turn off security for Unraid only when Private Browsing in Chrome. Don't use Chrome Private Browsing when using Unraid. This comment was not very helpful at all, and was honestly quite annoying. There was an issue when some characters (e.g. '<', '>', etc) are used in file names that will cause a php warning and fail a file save or edit. I've added some code to deal with this.

Show a screen shot of the default file name in the settings after clocking on the "Default" button. I'd like to see if there are any characters in the default name causing issues with php.

Go to the Tools->PHP Settings and set the error reporting level to all caegories and the n create the issue where yoou can't save a script. The see if any warnings or errors show in the PHP log.

I'm using Brave (also based on Chromium) and don't see the issue in incognito mode. Do a few things and see if there are errors/warnings when this happens: Click on Tools->PHP Settings and see if there are any log entries. While on the UD Device Settings page hold down Ctrl-Shift keys and press 'J'. This will show the Java console. See if there are any errors there.

What browser are you using?

UD creates those mount points: disks - used for mounted UD disk devices. remotes - used for mounted UD remote shares. addons - used to mount user mounted devices so FCP will not flag the mount as eroneous. rootshare - used for rootshare mounts. Why are these there if I don't use them you might ask? Those mountpoints are mounted themselves with a max size limit so misconfigured Docker Containers and VMs don't crash Unraid by writing lots of data to that mount point (tmp file system) instead of the actual device. /mnt/ is in the tmp file system and doesn't have a lot of room for eroneous data.

Yes, when the remote share has issues, I've tried to set up UD to cleanly unmount and remount the remote share so it will not hang. That's generally from a name resolution issue. Is your remote server on a static IP address? You might try using the IP address of the remote server but it needs to be static. I have put a lot of time on sorting out name resolution issues and changing IP addresses when the server is not on a static IP address. It is pretty hard to test and it may be missing a server IP address change.

I'm sure the drives have unique numbers, but the dock is not reporting them that way as udev sees one serial number for three drives. It's not UD that is duplicating the serial numbers. It's udev. There are others here that have more experience than i do.

It appears you are mounting a remote server share manually: Apr 7 13:03:36 apollo sudo: root : TTY=pts/0 ; PWD=/root ; USER=root ; COMMAND=/sbin/mount -t nfs4 -o proto=tcp,port=2049 100.69.1.200:/export/users /mnt Use UD to mount it.

Your USB dock is reporting three of the drives as the same serial number. That's why you see only one Hitachi disk. The other two, I don't understand why they are not showing. Unraid and UD require unique serial numbers to be recognized properly.

Post diagnostics. It'll help me understand your issue better.

Can't tell you how many times I've done things like that. No worries.

No. Each entry in the exports file has a unique fsid. What is the client trying to connect to Unraid?

A couple ideas for you to look at: UD uses this as the rule to mount another NFS share on another Unraid server: "*(rw,sec=sys,insecure,anongid=100,anonuid=99,no_root_squash,lock)". UD uses this command to mount a remote NFS share: "/sbin/mount -t 'nfs' -o rw,soft,relatime,retrans=4,timeo=300 'MEDIASERVER:/mnt/user/Public' '/mnt/remotes/MEDIASERVER_Public'" If you use "-t 'nfs'", nfs will negotiate the version to use, starting with v4.. This is the setup of an Unraid NFS share set to "Public": "/mnt/user/Public" -fsid=106,async,no_subtree_check *(rw,sec=sys,insecure,anongid=100,anonuid=99,all_squash)" These may give you some help in solving your issues.

03.03.2024 is the latest. The 04.01.2024 is UD+. It is working for me. Post your diagnostics so I can take a look.

The fix was already implemented. You just have to have the latest version of UD or Unraid. It only affects saving the scripts after editing them. It doesn't have anything to do with execution of the script when a device is connected. Just turn on the "Automount" switch and the script will run when the devkice is connected. Also, be sure to use the default script (click the "Default" button) and modifty it to your needs.

I'll concentrate on network issues first: You have IPV6 enabled. Do you need it? If not turn it off. I see multiple vlans. Be sure they are configured properly for your situation. Turn off NetBIOS. Use a static IP address on your server. It's set right now for DHCP. Verify network settings on your PCs. Add a second NIC and set it up with a bonding mode to spread the load over both NICs. Other factors: Upgrade to 6.12.10. Remove un-get and NerdTools plugins then reboot. Make sure no packages will install after reboot. Do this before upgrading to 6.12.10. There may be some package interference.

Give me some time to review.

Start by posting your diagnostics.

On the UI, click on the 'Boot' device label and then the "Syslinux Configuration" tab. Romove any instances of this text "root=sda" you see, save it, and then reboot. You should be able to preclear without memory issues.

Start with a fresh diagnostics.