2 hours ago, itimpi said:

If this is working you can use the ‘diagnostics’ command to get a zip file generated in the ‘logs’ folder on the flash drive which you can then post to help us see what is going on.

 

Hi itimpi,

diagnostics in attach.

 

I get this two error on boot:

error: failed to connect to the hypervisor

error: Operation not supported: Cannot use direct socket mode if no URI is set

 

webgui still not working

unraid-diagnostics-20220518-1439.zip

Two dockers no longer work, Whoogle and PhotoStructure. Both seem to have a permissions error now. Log from PhotoStructure:

 

Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.
    at L.assertReadWriteExecutable (/ps/app/bin/main.js:9:191324)
    at async Object.t.setupLibraryDataDir_ (/ps/app/bin/main.js:9:387513)
    at async /ps/app/bin/main.js:9:626510
{"fatal":true,"exit":true,"status":14,"pid":25,"ppid":18,"error":"main setup failed: Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.: Library setup failed: ¹"}
This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). The promise rejected with the reason:
Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.
    at L.assertReadWriteExecutable (/ps/app/bin/main.js:9:191324)
    at async Object.t.setupLibraryDataDir_ (/ps/app/bin/main.js:9:387513)
    at async /ps/app/bin/main.js:9:626510
usermod: invalid user ID 'XXXXXX'
{"fatal":true,"exit":true,"status":12,"pid":25,"ppid":18,"error":"Library setup failed: ¹: Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000."}
Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.: Library setup failed: ¹
Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.
    at L.assertReadWriteExecutable (/ps/app/bin/main.js:9:191324)
    at async Object.t.setupLibraryDataDir_ (/ps/app/bin/main.js:9:387513)
    at async /ps/app/bin/main.js:9:626510
{"fatal":true,"exit":true,"status":14,"pid":25,"ppid":18,"error":"main setup failed: Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.: Library setup failed: ¹"}
This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). The promise rejected with the reason:
Error: Please check directory permissions for /ps/library/.photostructure: it must be read/write/execute by user 1000/group 1000.
    at L.assertReadWriteExecutable (/ps/app/bin/main.js:9:191324)
    at async Object.t.setupLibraryDataDir_ (/ps/app/bin/main.js:9:387513)
    at async /ps/app/bin/main.js:9:626510

** Press ANY KEY to close this window **

 

And Whoogle's error:

 

May 18 17:33:42.000 [notice] New control connection opened from 127.0.0.1.
May 18 17:33:42.000 [notice] Heartbeat: Tor's uptime is 7:45 hours, with 0 circuits open. I've sent 129 kB and received 319 kB. I've received 1013 connections on IPv4 and 0 on IPv6. I've made 5 connections with IPv4 and 0 with IPv6.
May 18 17:33:42.000 [notice] While not bootstrapping, fetched this many bytes: 167859 (consensus network-status fetch)
May 18 17:33:42.000 [notice] Average packaged cell fullness: 71.436%. TLS write overhead: 6%
WARNING:root:Exception raised while handling cache file '/config/session/e2f2cc11b5ff5c8870688683df98d3b6'
Traceback (most recent call last):
  File "/usr/local/lib/python3.11/site-packages/cachelib/file.py", line 222, in set
    fd, tmp = tempfile.mkstemp(
              ^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/tempfile.py", line 341, in mkstemp
    return _mkstemp_inner(dir, prefix, suffix, flags, output_type)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/tempfile.py", line 256, in _mkstemp_inner
    fd = _os.open(file, flags, 0o600)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
PermissionError: [Errno 13] Permission denied: '/config/session/tmpxv6wtr0r.__wz_cache'

 

Any ideas how to correct?

9 hours ago, ezhik said:

Upgraded two systems, no issues. To those who are running nvidia cards and the nvidia plugin, it was a smooth sail to upgrade. MUCH better than earlier versions. 

 

Cheers.

Yea, I was running the plugin but it wasn't smooth sailing for me. Needed to remove it, re-install, and then select the version that would be applicable to my card as the latest one doesn't support it. (I will be upgrading the graphics card soon). But once that was sorted everything is going super well for me for me so far. Happy to see I can use WSD again without getting cores stuck to 100%. Now all I'm missing from Unraid is ZFS, VM snapshots/backups, and ARC and/or L2ARC #myUnraidWishlist.

Edited May 18, 20224 yr by JustOverride
Typo.

6 hours ago, Frank1940 said:

Quick Google found this:

    https://discussions.apple.com/thread/2241666

 

A lot of files are replaced during a version upgrade so it not truly a coincidence.  All the activity can cause a flaky flash drive to crash.  Or it could be something going on.  IF worse comes to worse, it is a simple procedure to update the Unraid license to a new flash drive.

 

I hope you followed the advice given to make a complete backup of the flash drive before you started.  IF you did not, try to make one before you start the repair.  In particular, the   /config     directory is very important. 

 

Thanks. Unfortunately, still not working. I ran disk utility, which immediately finished without errors. But issue persits.

8 hours ago, itimpi said:

An upgrade is one time quite a bit of data is written to the flash drive so I guess it is the most likely time to trigger a flash drive failure.

 

I have also noticed that after an upgrade there seems to be a higher than normal instance of getting read errors from the flash drive.  Often simply rewriting the flash drive on a PC/Mac can fix such an issue if the flash drive has not really failed.

 

I THINK a Mac has an option to check the flash - but quite where it is in Finder I am not sure.

Unfortunately, disk utility (Mac) didn't show any errors. What do you mean by "simply rewriting the flash drive"? You suggesting I delete the files and then copy them back on? Or even format the USB?

Edited May 19, 20224 yr by steve1977

Upgrade did not go smooth for me. Thankfully I backed up my flash before upgrading. I upgraded through the webgui from the previous stable version. I received an error about waiting 30 seconds for device labeled UNRAID to load and after 30 seconds just rebooted. I ended up having to reflash the usb stick using the official unRaid usb creator and cooy back over my config folder to the flash. I then had to turn off SSL through SSH to be able to get back to the webgui.  For some reason PCIe ACS Overide setting was changed to disabled and I had to change that back to downstream to get my Sata HBA and GPU working again. I am on 6.10 now though.

Upgrade went smooth for me and things are going well, but I did have a parity check triggered post-reboot/update.  The server was safely rebooted after applying the update, so not sure what that's about.  No harm, no foul though.  It's about halfway done and should finish the rest tonight.

-Upgraded ✅

-Cert upgraded ✅

 

make sure you add to pfsense (Services > DNS Resolver > General Settings > Custom Options)

private-domain: "myunraid.net"

 

-if you can't access the GUI after cert upgrade.

1 ssh root@SERVERIP
2 use_ssl no

then you can log in using the normal ip way.

 

-Root Shares ✅

-File Manager ✅

 

Thank you for all the hard work, everything is working GUCCI. Disks are finally sleeping!😴😇🤫

7 hours ago, Rejserr said:

Upgrade from 6.9.2 to 6.10., webgui is not work , can't connect.

any solution ?

 

ssh is working, docker and VM is working.

On network I can see IP.

 

spec of server:

CPU i7 12700k

MBO Z690

 

 

 

Edited May 19, 20224 yr by craigr

5 hours ago, SimonF said:

Have a look at this thread 

 

 

Thanks thats fixed my issue. 

10 hours ago, gundamguy said:

JorgeB got it... but if I never set a password before the upgrade how would I go about doing that now?

 

Take a closer look at the screen, are you sure it isn't asking you to set a root password?

6 hours ago, thunderclap said:

Two dockers no longer work, Whoogle and PhotoStructure. Both seem to have a permissions error now. 

Any ideas how to correct?

 

I also had a couple of broken containers. I would use WinSCP to go into the folders for those containers in appdata, and see if there are any files that have stricter permissions than other. For example some which are owned by "root" while the others is owned by other users. I don't know why upgrading would break this though.

On 5/17/2022 at 11:23 PM, exico said:

Unfortunately I can't seem to update from rc8 to stable.

 

When i try from Tools -> update OS it doesnt find any new release in Next or Stable.

 

If i try to install it manually like told in the first post in Plugins -> Install Plugin i get this:

 

plugin: installing: https://unraid-dl.sfo2.cdn.digitaloceanspaces.com/stable/unRAIDServer.plg
plugin: downloading https://unraid-dl.sfo2.cdn.digitaloceanspaces.com/stable/unRAIDServer.plg
plugin: downloading: https://unraid-dl.sfo2.cdn.digitaloceanspaces.com/stable/unRAIDServer.plg ... done
plugin: not installing older version

 

Same here

39 minutes ago, jluerken said:

Same here

 

You can always use the procedure described here to handle the upgrade if you are not able to do it via the GUI.

13 hours ago, gundamguy said:

but if I never set a password before the upgrade how would I go about doing that now?

If you never set a password it should be asking to set one up:

 

 

If you don't see this do the recovery password procedure and you should see that at next boot.

8 hours ago, sadkisson said:

Upgrade did not go smooth for me. Thankfully I backed up my flash before upgrading. I upgraded through the webgui from the previous stable version. I received an error about waiting 30 seconds for device labeled UNRAID to load and after 30 seconds just rebooted. I ended up having to reflash the usb stick using the official unRaid usb creator and cooy back over my config folder to the flash. I then had to turn off SSL through SSH to be able to get back to the webgui.  For some reason PCIe ACS Overide setting was changed to disabled and I had to change that back to downstream to get my Sata HBA and GPU working again. I am on 6.10 now though.

 

Seems I was facing same situation as you do. This is really a pity as Unraid had been working well for many years prior to the upgrade.

 

I tried to run "disk utility", but this did not show any error and didn't change anything.

 

I then followed your instruction above. Created a new USB device with the usb creator and copied the config folder. It boots and shows more lines. It stops at "vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem

 

Any thoughts?

Hi !

 

The upgrade went well (i think) but the root password did not work, so i reseted it, no worries. but when i close the browser and

try to login the new password in not working?

 

I have not tested it via ssh, but web gui not working, smb pwd works fine.

Now I got to the GUI, but seeing the "GUID error". I guess I can fix this with what is described above.

 

I am also seeing a screen saying "purchase key", so I guess I need to somehow fix this as well? Any thoughts above?

I just replaced the key and now things seem to be working, Thanks for the help in this forum!

Unfortunately, the upgrade from 6.9.2 to 6.10.0 failed for me and I suspect it was due to the way I had DNS configured within UnRAID.  I had the DNS Server settings within UnRAID configured to point to two Windows Server Virtual Machines (Windows Domain Controllers/DNS Servers) running within UnRAID - which are not available if the Array is down.  I manually edited the NETWORK.CFG file on the USB stick to point to a public DNS Server (e.g. 8.8.8.8) and then restarted my UnRaid "Server" - bringing it back online.

 

I also had a minor issue with one Docker Application which required deleting the "perms.txt" file within the AppData folder for the Application to reset the file system permissions.  All other Docker Applications worked/started without issue.

 

I also attempted to log on at the Console and remotely but both appeared to timeout after having entering the "root" username - the prompt for a password did not appear.  I did leave it for a while but my main focus was restoring access to UnRAID.

 

Diagnostics file attached.

bigbox-diagnostics-20220518-0926.zip

Edited May 19, 20224 yr by PPH
Added additional information

9 hours ago, Ademar said:

 

I also had a couple of broken containers. I would use WinSCP to go into the folders for those containers in appdata, and see if there are any files that have stricter permissions than other. For example some which are owned by "root" while the others is owned by other users. I don't know why upgrading would break this though.

Thanks for the suggestions. Whoogle I had to set to 0777 while PhotoStructure I had to re-set ownership to 1000/1000. Not sure if this changed or if it wasn't why it didn't cause a problem before. Maybe Unraid tightened up a security a bit? Anyway, appreciate the help.

Good job team and thanks for all the hard work! The only issue I'm seeing so far is that Plex seems to not be able to access any media now and is unable to play anything.

updated 2 machines. first one no problems. Second one had all disks in the 2 pools "missing" as the device names had changed (those using an H240). I noted the disk positions/assignments, used new config and preserved the array disks, went back to the main tab and re-assigned the pool disks, marked parity as correct, and started the array. Normal operation as expected with no loss of data (also expected).

15 hours ago, craigr said:

 

 

Already setup to USE_SSL="no"

When write /etc/rc.d/rc.nginx reload , get message Nginx is not running.

 

Still cannot connect to webgui.

Any another solution ?

12 minutes ago, 1812 said:

updated 2 machines. first one no problems. Second one had all disks in the 2 pools "missing" as the device names had changed (those using an H240). I noted the disk positions/assignments, used new config and preserved the array disks, went back to the main tab and re-assigned the pool disks, marked parity as correct, and started the array. Normal operation as expected with no loss of data (also expected).

 

soke too soon, having a problem with my gpu, which went from functional to not. Log shows:

 

May 19 13:46:23 Tower kernel: nvidia-nvlink: Unregistered the Nvlink Core, major device number 245
May 19 13:46:23 Tower kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 245
May 19 13:46:23 Tower kernel: NVRM: The NVIDIA GeForce GT 730 GPU installed in this system is
May 19 13:46:23 Tower kernel: NVRM:  supported through the NVIDIA 470.xx Legacy drivers. Please
May 19 13:46:23 Tower kernel: NVRM:  visit http://www.nvidia.com/object/unix.html for more
May 19 13:46:23 Tower kernel: NVRM:  information.  The 510.73.05 NVIDIA driver will ignore
May 19 13:46:23 Tower kernel: NVRM:  this GPU.  Continuing probe...
May 19 13:46:23 Tower kernel: NVRM: No NVIDIA GPU found.

 

It appears that my GPU driver was automatically updated and now I have to roll back to the 470.xx option in the Nvidia plugin. Ok I guess... a bit of an annoyance to have to fix something like this that was working just fine.