Hello,
My Forge container unexpectedly stopped deploying for my forge app (App ID: f8448394-3126-4070-8dbb-8f4a8ad94bed). Unfortunately, there are no additional logs available.
I’m running:
forge deploy -e development --verbose
and getting the error message:
[eu-central-1] Rollout of container service “java-service” is in status “Failed”. Replicas: 1 desired. Rollout aborted update to revision 24
[eu-central-1] Failed to deploy container service “java-service” to Kubernetes
Failed to deploy all container services. Caused by 1 error(s): (1) java-service in eu-central-1: Rollout of container service failed {serviceKey=“java-service”, cause=“Rollout aborted update to revision 24”}
Error: Deployment failed (requestId: 3b7c793d6f044a229195e501c5a1d8d7)
If there are any additional logs or diagnostic information that would help identify the root cause, please let me know and I’ll provide them.
Thank you in advance for your help.
Looks like the rollout is failing at the Kubernetes/container level rather than the Forge CLI itself. I’d check the container image, startup command, health website checks, and resource limits first. Since the logs only show “Rollout aborted,” Atlassian support may need to provide the underlying pod events or deployment logs for that revision.
Hi @AlexandraIvanova,
Thank you for reaching out and for using Forge Container Services.
Looking at our logs for your attempted service deployments, I can see that this deployment is failing on your java-service container not being ready, and responding to your configured health endpoint, within the time threshold at deployment per the Deploy a service DAC documentation, causing the overall deployment to fail.
This specifics of the failed deployment don’t seem to be surfaced in your forge deploy logs, which is another issue which I will continue to investigate and get back to you once resolved.
I hope the above helps unblock you though.
Thanks, Matthew.
Another cause could be that your image is not runnable in the Forge container service runtime environment.
Possible causes could be that the image does not run under UID 1000/GID 1000, or has a hard dependency on some environment variable and crashes.
We are trying to improve the observability of failed deployments to make troubleshooting easier. In the meantime, if you are really stuck, you can raise a support ticket, and we will try to help you out further.
Hey @AlexandraIvanova,
I have reproduced a simple app the never responds to it’s health check endpoint, and have confirmed that these details are in the error logs (similar to what you have shared above).
[us-east-1] Rollout of container service "java-service" is in status "Failed". Replicas: 2 desired, 1 available, 1 ready, 1 updated, 1 unavailable. Rollout aborted update to revision 2: Caused by (1) instance(s) with errors: (1) ContainersNotReady: java-service container is not ready (reason: waiting for health check, restarted 1 times)
[us-east-1] Failed to deploy container service "java-service" to Kubernetes
Failed to deploy all container services. Caused by 1 error(s): (1) java-service in us-east-1: Rollout of container service failed {serviceKey="java-service", cause="Rollout aborted update to revision 2: Caused by (1) instance(s) with errors: (1) ContainersNotReady: java-service container is not ready (reason: waiting for health check, restarted 1 times)"}
I assume your cause is different given that your logs seem to omit any specific failure reasons. If you are able to resolve your issue, let us know what the original issue was and we will add support for better handling of these error cases. In the mean time, I will continue to trial and error possible scenario’s to reproduce your issue, like those @BoZhang suggested above.
Thanks, Matthew.
Hi again @AlexandraIvanova,
I’ve checked our internal logs and the last attempt that I can see for a deployment was on the 8th of July. We have since moved Forge Container services from EAP to Preview with several platform improvements/changes since.
With this in mind, I’d ask you to please re-enroll your app in the Forge Container services Preview and reach out again with any issues you face in the future.
Thanks, Matthew.
Hi Matthew,
thank you for the quick response.
I confirmed my image now runs under UID 1000 / GID 1000 as required per the image security reference, and I fixed the healthcheck problem. Here’s the relevant part of my Dockerfile:
RUN yum install -y shadow-utils
&& groupadd -g 1000 appuser
&& useradd -u 1000 -g 1000 -r -s /sbin/nologin appuser
&& yum clean all
USER 1000:1000
I also updated the Forge CLI to 13.3.0 and ran the deployment again. The error has now changed to the following:
{
"__typename": "AppDeploymentLogEvent",
"stepName": "Release services",
"createdAt": "2026-08-12T12:16:49.940Z",
"message": "Failed to deploy all container services. Caused by 1 error(s): (1) oboard-backend-svc in ap-southeast-1: Rollout of container service failed {serviceKey="oboard-backend-svc", cause="ReplicaSet \"eco-deployment-85679ffbcc\" has timed out progressing.: Caused by (1) instance(s) with errors: (1) ContainersNotReady: java-service container is not ready (reason: CreateContainerError)"}",
"level": "WARNING"
}
Error: Deployment failed (requestId: 4a3857c6a9794e11a212695fd4df0b2d)
Could you help me understand what’s causing the CreateContainerError here?
Thanks,
Alex
Hi @AlexandraIvanova,
No problem!
The logs that I have available are slightly nondescript, however they indicate a mounting issue for your container I believe, not an actual runtime issue.
Could you please share with me your Dockerfile (or a stripped down / redacted version if there are sensitive details in there). For reference as well, here is our reference containers app.
I suspect that your Dockerfile is missing an ENTRYPOINT or CMD operation, however providing your file will be most valuable.
Also, instead of the run on && chain, splitting each operation into an atomic one is the idiomatic and recommended approach for these operations (for this reason).
I look forward to hearing back from you.
Thanks, Matthew.
Hi Matthew,
please find my Dockerfile here:
FROM public.ecr.aws/amazoncorretto/amazoncorretto:17.0.18-al2-native-jdk
LABEL maintainer="test.dev@gmail.com"
ARG JAR_FILE=target/*-exec.jar
ADD ${JAR_FILE} /app.jar
RUN yum install -y shadow-utils
RUN groupadd -g 1000 appuser
RUN useradd -u 1000 -g 1000 -r -s /sbin/nologin appuser
RUN yum clean all
USER 1000:1000
EXPOSE 8080
ENTRYPOINT ["sh", "-c", "exec java $JAVA_OPTS -Djava.io.tmpdir=/tmp -Dspring.profiles.active=dev -jar /app.jar"]
Latest failed deployment :
Error: Deployment failed (requestId: a8e71b89bdd748ccbfe5b06a398e07f3)