3 months ago
Hello Railway Support Team,
I am experiencing recurring deployment failures across multiple projects, including repositories that were previously working fine. Every time I attempt to deploy, the deployment status becomes “Crashed”, and the logs show the following error:
AH00534: apache2: Configuration error: No MPM loaded.
This error suggests an Apache configuration issue, but it is happening on all my deployments — even simple PHP APIs and custom Dockerfile setups — and I have confirmed there were no changes made to the repositories that should cause this issue.
Details:
Multiple repositories and services affected
Error occurs during deployment start
All deploys end in Crashed
Using Dockerfiles from GitHub repositories
Could you please investigate if this is related to a platform issue on Railway, or if there has been a recent change in the build/runtime environment affecting Apache? I can share the specific deployment IDs and logs if needed.
Thank you in advance for your assistance.
2 Replies
3 months ago
Multiple applications, including basic phpMyAdmin setups Template
3 months ago
The AH00534 error is coming from Apache inside your Docker image, not from Railway itself. It means Apache starts but no MPM module is loaded, so it exits and the service crashes.
Please check your Dockerfiles and Apache config:
Make sure at least one MPM is enabled, for example by adding something like:
a2enmod mpm_event (or mpm_prefork) in your Dockerfile if you use apache2 on Debian/Ubuntu images, or the matching LoadModule mpm_*_module line in httpd.conf if you use httpd.
Confirm you aren’t overriding the default Apache config with a custom config that removes or comments out all the mpm_* modules.
If the same image used to work and now all projects with that image fail, share a sample Dockerfile and a recent deploy ID so Railway can check if a base image/package update changed Apache defaults, but the direct fix is to explicitly enable one MPM in your image.