7.3.1 upgrade from 7.3 destroyed all my Dockers.... All of them!

The docker section is empty, and I mean completely empty, simply no dockers...

All previous upgrades installed flawlessly. Not this time.

Now my server is for all intent and purposes dead for 3 days. now.

• All data seems there. Restoring backup = nothing, no dockers.

  I'm a relatively new user of Unraid. A paid licensed user.

17 minutes ago, Master_Yoda said:

7.3.1 upgrade from 7.3 destroyed all my Dockers.... All of them

Start a new thread in General Support with your Diagnostics.

1 hour ago, remblack said:

I upgraded from 7.3.0 to 7.3.1 and now the disk on the main array don't seems to be spinning down. Anyone seeing the same?

NVM ... it seems like they spun down now ... I will keep on eye, the server was unused over night and they should have been down.

What HBA card are you using? I have not had this issue at all. Also just double check you settings are correct:

26 minutes ago, Master_Yoda said:

7.3.1 upgrade from 7.3 destroyed all my Dockers.... All of them!

The docker section is empty, and I mean completely empty, simply no dockers...

All previous upgrades installed flawlessly. Not this time.

Now my server is for all intent and purposes dead for 3 days. now.

• All data seems there. Restoring backup = nothing, no dockers.

  I'm a relatively new user of Unraid. A paid licensed user.

But also look back as this has happened to quite a few people and it's been solved...

Here you are:

and before that if you need more info:

Edited June 2Jun 2 by craigr

Successfully upgraded my Ugreen DXP4800 Pro from 7.2.7 to 7.3.1.

There was one difference from previous updates: This time, there was no notification indicating that the system was now safe to restart. After waiting for quite a while, I went ahead and restarted it anyway.

Update from 7.2.3 to 7.3.1 without any issues. Thanks.

Hi @craigr ,

I have that same config for a long time.
I though it was working after becasue i noticed they did stop ... but right after started again

I created this topic under support and I added logs

https://forums.unraid.net/topic/199233-disk-not-spinning-down-after-upgrade-730-to-731/

20 hours ago, craigr said:

What HBA card are you using? I have not had this issue at all. Also just double check you settings are correct:

On 6/1/2026 at 2:10 PM, Frank1940 said:

Us folks who are fortunate to have a good quality flash drive from the 2010 era or to have a USB SD memory card adapter are in a much better position than those using 2026 era flash drives!!!

Yeah, I’m still firmly in that old-school club too.

Want to stay on USB and avoid all the modern headaches of internal boot (QLC, fTPM, license slots, M.2 requirements)?

The Innodisk 3ME is your best bet.

Industrial MLC at $4 each -- you can just grab one and get back into the club.

It’s a proper high-quality stick.

P/N: DEUA1-64GI61BW1SC

eBay item number: 326046070546

Edited June 6Jun 6 by Lolight

intel optanes 16GB ssd is where i go for a nvme interal boot now as ists a quick low strage long lasting althernative for usb access.

bought a 10 pack a while back when usinga asrock b250 miners for hive and other. converting them into fedor bazite installs...

there quite a few low end ssd that are great for the internal boot option.

Was able to upgarde 1 system with out issues.

suprsied to see kerneal panics after a update hard shutdown the machine and turned on then no issues... so something to be weary of in some old hardware on 5th gen intel baords potenal kerneal rip
Could npt find a define cause at boot after upgrade.

ram was good. usb boot... againa hard shutdown and power up no issues... but defint somthing to look for during the upgrade path...

I upgraded 2 of my 3 servers to 7.3.1 and noticed one issue that can hopefully be easily corrected. When running my ubuntu VM, I keep getting kernel panic causing complete stop of my remote desktop session. I checked online and it said to use the latest Machine Q35-10.2 but between 10-15 minutes after running the vm, it stops. Any hints with this? Before 7.3.1 I had 7.2.3 and no issue. Any ideas? If you need my diag, let me know but this one sounds like an easy fix?

5 hours ago, GeorgeJetson20 said:

I keep getting kernel panic causing complete stop of my remote desktop session.

Please create a new thread in the KVM section and add the diagnostics

On 6/6/2026 at 12:12 PM, Lolight said:

Yeah, I’m still firmly in that old-school club too.

Want to stay on USB and avoid all the modern headaches of internal boot (QLC, fTPM, license slots, M.2 requirements)?

The Innodisk 3ME is your best bet.

Industrial MLC at $4 each -- you can just grab one and get back into the club.

It’s a proper high-quality stick.

P/N: DEUA1-64GI61BW1SC

eBay item number: 326046070546

Wow, doing a quick Google search and that model ranges from £175 to like £1000+ over here in the UK...

2 hours ago, jademonkee said:

Wow, doing a quick Google search and that model ranges from £175 to like £1000+ over here in the UK...

Yeah, you aren't wrong about that price point -- the wow factor on that listing is absolutely real. Finding a deal like that right now feels wild, especially considering how badly the market has been hit by MLC industrial NAND shortages lately.

The markups on industrial USB flash drives have been brutal over the last year, so seeing one listed at $3.99 is definitely a steal if you can actually grab it.

Edited June 9Jun 9 by Lolight

On 6/2/2026 at 6:36 PM, remblack said:

I upgraded from 7.3.0 to 7.3.1 and now the disk on the main array don't seems to be spinning down. Anyone seeing the same?

NVM ... it seems like they spun down now ... I will keep on eye, the server was unused over night and they should have been down.

I had the same problem with a Toshiba N300 - the Seagate Exos are not affected, just the Toshiba. Couldn't find the culprit, so I took out the Toshiba of the Array (and the whole system)

Update so smooth as silk. Thank you!

Upgraded from 7.2.7 to 3.1 and got a... on boarding prompt on a decade+ old installation?!


I get that Unraid is a brand-new product and some things fall between the cracks, but come on.

Yep, to set up the new features from 7.3 if you wish.

26 minutes ago, tuxbass said:

I get that Unraid is a brand-new product and some things fall between the cracks, but come on.

Just hit the Skip option if you are not interested and you do not get prompted again.

16 hours ago, tuxbass said:

Upgraded from 7.2.7 to 3.1 and got a... on boarding prompt on a decade+ old installation?!


I get that Unraid is a brand-new product and some things fall between the cracks, but come on.

...and now discovered time config was borked:

I'm EU-based.

# ls -l /etc/localtime

lrwxrwxrwx 1 root root 33 Jun 10 15:18 /etc/localtime -> /usr/share/zoneinfo/Europe/Madrid

# ls -l /usr/share/zoneinfo/Europe/Madrid

ls: cannot access '/usr/share/zoneinfo/Europe/Madrid': No such file or directory


IIRC localtime used to be pointed at Paris node, no idea where Madrid now came from. Did the fancy new onboarding logic maybe deduct I'm somewhere in Spain and cleverly set it to Madrid without being asked to?

Edited June 11Jun 11 by tuxbass

On 6/1/2026 at 4:12 AM, JorgeB said:

To see if it's a config problem, try booting with a flash drive using a stock 7.3.1 install. If that works or you then need help sorting the config, please create a new thread in the general support forum.

Finally had some time to look into it. It seems the culprit was a misconfigured entry in the go file. Spent hours troubleshooting with assistance from Gemini and arrived at conclusion

Gemini indicated:

The Real Culprit: A Broken Initialization Order

Look at the process IDs (PIDs) from your ps printout. Your system booted up, initialized emhttpd (PID 15938), and then successfully spawned all of your background plugins: autofan, cache_dirs, unbalanced, and the internal PHP pollers.

However, the local GUI startup process is completely missing from the execution tree. When Unraid boots into GUI mode, emhttpd is supposed to invoke a script (/etc/rc.d/rc.4 or the internal Unraid GUI initialization loop) that spins up Xorg and launches the local browser. On your system, emhttpd launched, but it never successfully triggered the GUI loop.

This happens when a third-party plugin or script stalls or misbehaves early in the startup sequence, blocking emhttpd from completing its final runlevels.

Notice how many instances of autofan are running right at the same second (PIDs 8509, 8538, 8582, 8611)? You have a massive wall of fan controllers trying to latch onto the motherboard's nct6775 chip immediately at boot. If one of those hardware-probing scripts (like autofan, intel-gpu-top, or even a stuck cache_dirs loop) blocks the main execution thread during initialization, the system stalls before it ever reaches the command to execute Xorg.

Based on that and it suggestion to try editing the go file, determined I had a script for pulse agent that was missing & at the end causing issues with boot up sequence. Added the & at the end and rebooted and issue resolved. Lots of other posts about getting stuck at bootup on 7.3.1 at exact same stage as me so hopefully helps someone narrow down their issue.

#!/bin/bash
# Start the Management Utility
/usr/local/sbin/emhttp

# Pulse Agent (Pushed to background with &)
bash /boot/config/plugins/pulse-agent/start-pulse-agent.sh &

I updated from 7.2.4 to 7.3.1 today. The update process seemed to go fine, but upon restart it reported errors on my second cache pool drive (which is made of 2 NVMe drives on a Supermicro riser card installed on a Supermicro X12SPi-TF motherboard) that had not been reported previously on 7.2.4.

None of the Docker modules in the Arr stack (Prowlarr, Sonarr, Radarr) would start and gave an Execution Error - Bad Parameter. They were also listed as having invalid IP addresses.

Subsequently I reverted to 7.2.4 and the reported errors disappeared and the Arr stack modules started normally.

Just thought I'd report the experience.

Kevin

7 hours ago, kolepard said:

The update process seemed to go fine, but upon restart it reported errors on my second cache pool drive (which is made of 2 NVMe drives on a Supermicro riser card installed on a Supermicro X12SPi-TF motherboard) that had not been reported previously on 7.2.4.

If you mean GUI errors, pools are not monitored before 7.3.x. If you want to troubleshoot this more, please start a new thread in the general support forum and post the diagnostics.

In case it is not clear sounds like there are potentially 2 issues:

• issues with the pool. Errors on it were not reported prior to 7.3.1. It is possible (even likely) that they have been there prior to the upgrades but you were not being told about them.

• Docker containers not starting

It is possible they are related but not necessarily so. It might be worth upgrading to 7.3.1 again and then if things are not working as expected getting the diagnostics for your system to see if we can determine why.

Thank you @JorgeB and @itimpi . I have updated to 7.3.1 again and posted diagnostics in a new thread, as you have recommended.

https://forums.unraid.net/topic/199433-dockers-fail-to-start-with-731/#comment-1627674

Reverting to 7.2.4 again resolved the issues with the Dockers not starting.

Thank you.

Kevin

On 6/13/2026 at 12:54 PM, kolepard said:

Thank you @JorgeB and @itimpi . I have updated to 7.3.1 again and posted diagnostics in a new thread, as you have recommended.

https://forums.unraid.net/topic/199433-dockers-fail-to-start-with-731/#comment-1627674

Just to close this loop, my issues were resolved in the referenced thread.

Kevin