Error: Failed to start containers

after that command, did you update again umbrelOS to latest version?
Try reflashing the mSD card now after you did that.
Those startup strings doesn’t sound good.
variable is not set. Defaulting to a blank string.

Did the OS flash first. then the commands. Will do a reflash. Keep you posted. Thanks for the fast reply

Dear DarthCoin

Bought a new SD card because I was having some trouble reflashing the OS on the SD card.
Flashed the OS on the new SD card, no issues there with the new card.
But I am still getting the same errors when trying to start umbrel
Any further suggestions or does this not look good…


I find myself in the same situation. I`ve rebooted the system, have already also tried " sudo systemctl stop umbrel-startup.service && docker system prune --force --all && sudo systemctl start umbrel-startup.service" and reflashed the SD card.

Still stuck in the /start page, loading…

If you already reset those services and reflashed the mSD, sincerely, I don’t have other idea what could be. I’ve looked into the logs and seems that is the same error with not starting the docker containers.
Maybe is something corrupted on your SSD.
If you do not have funds in that node wallet (onchain and/or LN) I sugest to start over from scratch with erasing the SSD disk also, maybe the drive have some bad sectors.

@louneskmt do you have any other idea?

I managed to restart my node completely following these steps:

You have error that a specific Umbrel App do not start or it keeps staying in “installing”, maybe the docker container for that app it wasn’t installed correctly or it hangs.

Step 1 - Enter using SSH into your Umbrel node:

Step 2 - stop Umbrel node with:

cd umbrel

sudo ./scripts/stop

Step 3 - clean / reconfigure the broken container:

sudo rm -f .env && sudo ./scripts/configure

Step 4 - Restart your node:

sudo ./scripts/start


Gad you could sort it out by yourself. And you @ocap?

Also, on UmbrelOS, you should use sudo systemctl <start | stop> umbrel-startup instead of those scripts.


Hi @louneskmt

No I haven’t had any luck.
I run into error while stopping service --> Couldn’t find env file: /home/umbrel/umbrel/.env
as well as sudo rm .env --> cannot remove ‘.env’: No such file or directory

pastebin with sudo ./scripts/stop

pastebin with logs is in the above messages

any help is greatly appreciated.


Please try again, I’ve updated @21isenough message to fix this.

Hello @louneskmt

These modified pieces of code work like magic :wink:
Node is back up and running. Bitcoin core is already 100% synced. BTC wallet and lightning wallet are still not showing the transactions and balance. But i guess this can take a while?

Many thanks! Is there a place where I can send the umbrel team some sats? (once my lightning wallet comes back online)


1 Like

If nothing appears within the next 24 hours, please send me a DM.

No worries! We do not accept donations, but we really appreciate your continuous support :grin:


Hi @louneskmt, I am having the same issue after a short power out. My node is currently running the configure script and is stuck at ‘Generating Tor Password’. How long does the reconfiguration usually take?

Thanks for your help!

I continued playing around… it worked now. Thanks for this threat and the help!

I had an uncontrolled power down last night and had the broken container screen after powering Raspberry Pi Umbrel back on.

In case it helps anyone I only needed to do:

cd umbrel
sudo ./scripts/stop

and after that the front end came back to life and everything was there. Posting in case that helps someone else.

Hey! Please do not use this command on Umbrel OS. You should use instead:

sudo systemctl restart umbrel-startup
1 Like

Hi, I have seen your post here.
I am still getting the error: Failed to start container…
I have tried:
1.-sudo systemctl stop umbrel-startup
2.-sudo rm -f .env && sudo ./scripts/configure. here I am getting the below:

=========== RECONFIGURING ============
========= UMBREL (mainnet) ===========

Generating auth credentials

Generating Tor password

docker: Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?.
See 'docker run --help'.

I have tried to re-flash the sd-card, but same issue. I think I have an issue with the SSD disk. Any recommendation for unmounting and fixing it?

I am trying to unmount it using the command:

sudo umount /dev/sda1
umount: /swap: umount failed: Operation not permitted.

sudo umount -f /dev/sda1
umount: /swap: target is busy.

Thanks in advance


I’m also getting the “Failed to start containers” error after several months of trouble-free operation. I’ve tried the steps above and end up with the following:

Job for umbrel-startup.service failed because the control process exited with error code.
See “systemctl status umbrel-startup.service” and “journalctl -xe” for details.

I suspect the SD card has got corrupted somehow. Is there something else worth trying before I flash a new card and start from scratch with a backup?

1 Like

Turns out that a new SD card was all that was required. I flashed the new card with the latest version of the OS and turned it on again. All is back to normal, just waiting for blocks to resync. All funds and channels still there. Phew.


I think I have the same problem. Did you just purchase a new SD card and reflash it? Mine was working fine for months too, but recently I had to reflash several times, looks like the SD card got corrupt. Thanks for posting

1 Like

Yes, just bought a new one (about £7) and started it up with that after flashing the latest version of the OS on to it. I didn’t realise that only the OS is on the card, the important stuff is on the drive, so flashing a new card gets all back to normal provided the hard drive is ok, which mine is. Only downside for me is that my lightning node was down for 3 days and I got one of my channels force closed on me.