Something broken in get.docker.com script

so i was nicely itterating on one of our machines,
and the latest rebuild suddenly didn’t install docker anymore

when running the script from https://get.docker.com during the automated install i get:

E: Failed to fetch https://download.docker.com/linux/debian/dists/trixie/pool/stable/amd64/containerd.io_2.2.1-1~debian.13~trixie_amd64.deb 404 Not Found [IP: 13.227.219.37 443]

so far i can’t seem to find any file called that,

i see:
containerd.io_2.1.5-1~debian.13~trixie_amd64.deb
and
containerd.io_2.2.0-2~debian.13~trixie_amd64.deb

did someone change the file sooner than the new update is synced? or i there something else going on?

1 Like

i have the same issuse on a new install linux server ubuntu i am useing apt install

1 Like

then luckily i’m not the only one :slight_smile:

about an hour ago everything worked fine …

anyway,
steps to reproduce:

root@osbourne:/tmp# wget -O /tmp/get-docker.sh https://get.docker.com

Saving to: '/tmp/get-docker.sh'

/tmp/get-docker.sh                           100%[============================================================================================>]  21.87K  --.-KB/s    in 0s      

2026-01-02 12:17:43 (133 MB/s) - '/tmp/get-docker.sh' saved [22396/22396]

root@osbourne:/tmp# sh get-docker.sh 
# Executing docker install script, commit: 8b33a64d28ec86a1121623f1d349801b48f2837b
+ sh -c apt-get -qq update >/dev/null
+ sh -c DEBIAN_FRONTEND=noninteractive apt-get -y -qq install ca-certificates curl >/dev/null
+ sh -c install -m 0755 -d /etc/apt/keyrings
+ sh -c curl -fsSL "url_removed_becauseimtoyoung/linux/debian/gpg" -o /etc/apt/keyrings/docker.asc
+ sh -c chmod a+r /etc/apt/keyrings/docker.asc
+ sh -c echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/docker.asc] url_removed_becauseimtoyoung/linux/debian trixie stable" > /etc/apt/sources.list.d/docker.list
+ sh -c apt-get -qq update >/dev/null
+ sh -c DEBIAN_FRONTEND=noninteractive apt-get -y -qq install docker-ce docker-ce-cli containerd docker-compose-plugin docker-ce-rootless-extras docker-buildx-plugin docker-model-plugin >/dev/null
E: Failed to fetch url_removed_becauseimtoyoung/linux/debian/dists/trixie/pool/stable/amd64/containerd.io_2.2.1-1%7edebian.13%7etrixie_amd64.deb  404  Not Found [IP: 13.227.219.58 443]
E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?
root@osbourne:/tmp#

It ws reported on GitHub as well:

Workaround for time being: Installation issues on ubuntu - #7 by meyay

The package did exist in the repos for a period of time. I have this package version installed on one of my docker hosts, but it doesn’t exist anymore. Seems like it was deleted, but the repo index was not updated to reflect it.

well, as I said, I was iterating and reinstalling machines the whole time and i think about 15 minutes before my post I suddenly noticed it couldn’t pull the image.
initially i thought i messed up, but reverted to a previous build where it still happened … so that was the main reason for creating an account here and posting about it.
now seeing al the talks about it, I guess this will be fixed soon :slight_smile:

A lot of user with this issue see: containerd.io 2.2.1-1 missing from download.docker.com · Issue #12741 · containerd/containerd · GitHub So yes, hopefully a quick fix :slight_smile:

Thank you for sharing the github issue, but I am surprised the forum didn’t warn you that it was already shared

https://forums.docker.com/t/something-broken-in-get-docker-com-script/150737/5?u=rimelek

Right before @meyay recommended his workaround

This is not a containerd issue. It was a packaging issue in the docker repos, which is tracked here:

The fix is already roled out. I tested it successfully on Debian 12 and Ubuntu 24.04.

i am quite confidently that i noticed the problem before the github issue was created, i think it was about 20 minutes after this initial post that the GitHub issue was created , and i think i noticed this within 10 minutes of happening, it only took 10 minutes extra to create an account here and figure out it was the script ,
( but now typing the above i suddenly start doubting myself..)

anyway, the issue has been resolved i believe, so on Monday 'll continue the work i was doing and i’d like to thank everyone replying here