"No disk space left on server, please contact the server administrator to continue."

Hi everyone,
I’m new to using nextcloud or linux server.
After I upgraded my nextcloud to 18.0.3, installed in Ubuntu 18.04, when I save a file using Collabora Online I having error message “No disk space left on server, please contact the server administrator to continue.”.
What can I check?
I found some system disks are fulls: they are named loop1, loop2 and loop3. I don’t know what they are use for…
could they be the cause?

I have the same problem for months now. I’ll present the symptoms below, and then describe a workaround that I’ve been using.

The Problem:
Whenever I restart my server (a few times a week), I receive two error messages the the first time (and only the first time) I attempt to open a file with Collabora Code in Nextcloud:

No disk space left on server, please contact the server administrator to continue.

l click OK to clear that error message, revealing a second error message under the first error message:

Failed to load document. Please ensure the file type is supported and not corrupted, and try again.

The Workaround:
Click OK on the error messages.

I can then open files in Nextcloud using Collabora Code without issue until the next time I restart the server, at which point I see the same errors again the first time I try to open a file.

For me, this is just an annoyance, but if anyone knows what the cause may be, I’m happy to troubleshoot.


v18 has OnlyOffice built-in (although it’s a separate manual download), won’t you guys want to give it a try (replace LibreOffice)? I did that on mine when going from 17 to 18 and I’ve not seen that message.


By “built-in,” do you mean I only have to download the OnlyOffice app, and OnlyOffice just works in NC18?

Collabora has a NC18 app as well, but there is a bit of setup needed to make it work. You make it sound like since OnlyOffice is “built-in” to NC18, I just need to download the OnlyOffice app and it will just work. Is this what you’re saying?


I’d first remove the reference to LibreOffice URL, then follow this page: https://nextcloud.com/blog/how-to-install-onlyoffice-in-nextcloud-hub-and-new-integration-feature/

@GiacomoG: I updated collabora/code to jovial_nightingale (4.2.2-2), and I no longer have this problem. More importantly, I changed the partition sizes on my server. I believe this was actually the root cause of the error, even though the partition was (only) about 85% full, that may have been too full.

@copenhaus: There are two reasons I’m not switching to OnlyOffice at this time
1. The community Document Server seems to have trouble with data loss, according to user reviews and GitHub. Collabora has a robust versioning scheme that works really well. The Community Document Server seems like a good concept, but not reliable enough to trust my files with yet.
2. I’ve got a solution running now (Collabora) that is more highly rated than OnlyOffice by Nextcloud app reviewers. While I don’t make every decision based on others’ reviews, Collabora is reliable and working well for me, so I’ll need something big to move me to OnlyOffice.

I may be willing to try it in the future, but for now Collabora is good enough.


i don’t find how to update collabora/code to jovial_nightingale (4.2.2-2).
could you explain o link any instructions?!


I run collabora/code on Docker, so update the collabora/code image on Docker, just run this command:

$ docker pull collabora/code

The current version of collabora/code as of this writing is “angry_curran.”


This command is not enough to update colabora.
You need to pull the new image, stop and remove the old and start the new image.

Watch this (at the end) :

Here’s everything you need to do to update the docker image for collabora/code for Nextcloud:

List all docker images; note the container ID of the collabora/code image to update:
docker ps -a

Stop the collabora/code container (use CONTAINER_ID from above step):
docker stop <CONTAINER_ID>

Delete the collabora/code container:
docker rm <CONTAINER_ID>

Confirm that the collabora/code container was removed:
docker ps -a

Get the new collabora/code image into docker:
docker pull collabora/code

Start the new collabora/code image (ENSURE YOU INSERT YOUR DOMAIN AS NOTED BELOW!):
docker run -t -d -p -e 'domain=<your>\\.<domain>\\.com' --restart always --cap-add MKNOD collabora/code

View/verify the collabora/code image is running:
docker ps -a

That’s it. I’ve been using these exact steps for a year or so without issue.

As always, check the documentation to ensure you’re keeping up with changes:

Collabora Setup for nginx:

Nextcloud Collabora Integration Page: