Continuous deployment gives eu.gcr.io not found

[Jeroenen]

Hey Guys,

So I've built my dockerfile which makes a simple website.
The build succeeds in Docker local and in Cloud Build. But when It goes to step 2: push it fails.

The logs:

Already have image (with digest): gcr.io/cloud-builders/docker

The push refers to repository [eu.gcr.io/<private>]

c3bbf6265c89: Preparing

d934f04ddb49: Preparing

0362de5e0486: Preparing

594b6d3b100a: Preparing

243c9bb49410: Preparing

bc53c21c3fdd: Preparing

d926290317ff: Preparing

408cde48ea57: Preparing

a790f937a6ae: Preparing

bc53c21c3fdd: Waiting

d926290317ff: Waiting

408cde48ea57: Waiting

a790f937a6ae: Waiting

name unknown: Repository "eu.gcr.io" not found

I can't find anything on the internet about it.

Any suggestions would help, thank you!

[ErnestoC]

Can you explain a bit more about how you received this error? Which was the command you ran to push the Docker image to GCP? Are you using a Cloud Build configuration file (cloudbuild.yaml) that you can share to replicate the issue?

For your reference, the documentation provides starter guides to push images to both Artifact Registry and Container Registry.

[Jeroenen]

Hi @ErnestoC ,

My apologies for the late reply,

I've built a docker image so that I can host my website on Cloud Run.  Hereby my steps te replicate the problem.

- Create an Cloud Run instance
- Give it a name
- Select Europe-west4 region
- Select Continuous deployment from a source repository
- Setup with Cloud Build
- Repository builder : Github
- Select private repository(already authenticated)
- Select Main branch
- Select Dockerfile
- Fill in /Dockerfile
- Save
- Minimal of 1 instance
- Allow unauthenticated invocations
- Listen to port 80
- Create

When I do these steps the instance gets created and after 30 seconds I get the error from Cloud build. I've tried it again and again, but the same result over multiple days.

Dockerfile:

FROM ubuntu/apache2

COPY script /var
RUN chmod +x /var/script
RUN /var/script

COPY /src/html /var/www/html
COPY /src/certificates /var/certificates

RUN service apache2 restart

My script:

#!bin/bash
apt update
apt install software-properties-common -y
add-apt-repository ppa:ondrej/php
apt update

apt install php8.0 -y
apt install php8.0-mysqli -y

rm -rf /var/www/html/*
rm /etc/apache2/sites-available/000-default.conf
rm /etc/apache2/sites-available/default-ssl.conf
a2dissite 000-default.conf

cat > /etc/apache2/sites-available/<private>.nl.conf << EOF
<VirtualHost *:80>
ServerAdmin<private>
Servername <private>
ServerAlias <private
DocumentRoot /var/www/html/
DirectoryIndex index.php
ErrorLog ${APACHE_LOG_DIR}/error.log
CustomLog ${APACHE_LOG_DIR}/access.log combined
</Virtualhost>
EOF

a2ensite site.conf

a2enmod rewrite

Thank you in advance for taking a look at my question.

[ErnestoC]

Since there are multiple internal details about your pipeline that would need to be taken into account, the best way to proceed would be to create a support case directly with GCP. This so your project can be internally reviewed and you can get an accurate analysis.

[ernst]

We have the exact same issue. We use cloudbuild to push images to gcr.io.

We recently changed from gcr.io to explixitly use eu.gcr.io, and ever since, some of our cloudbuild jobs fail. The URL change is the only one we made in our cloudbuild jobs.