-
Notifications
You must be signed in to change notification settings - Fork 26
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Container instances never end up running #18
Comments
you run |
I'm tryiing to use my docker image as a JNLP agent (windows). Same problem. In Linux it is working. |
Seems it's a connection issue, but I can't tell why connection timed out. |
I've tried this with an SSH connection. I see similar behaviour. The container is started and appears to run, but exits before doing anything useful.
I also see the ALPN callback dropped message in the logs
|
@jamesabbottsmith Can you provide more logs about SSHLauncher in Jenkins logs. Such as:
|
I've noticed a lot of issues with ACI even in just manually creating containers. It seems like the service has a lot of performance and consistency issues. Definitely to the extent that a plugin like this wouldn't make sense until the start time of any Windows or Linux container can be brought to under ~15 seconds. |
I'm trying to setup an Jenkins deployment using the Solution Template together with an ACI based build environment. I'm running into very similar problems with the this setup and it is impossible to debug. jenkins.log shows many entries like these:
|
So I'm finding my way around the Azure Portal and I finally found some relevant error feedback in the Deployments view of my ResourceGroup. Turned out I needed to issue this command and then it started working!! |
@atrauzzi We recommend to use SSH to launch container as it's much more stable than JNLP. Please have a try and let us know if there is any issue. |
I don't have SSH access to launch nodes on certain services. So this is still an issue. Why did you close this issue before I could reply? |
@atrauzzi Please provide more logs and configurations about using SSH to launch ACI as you haven't succeeded. From our test, SSH is more stable than JNLP. Please keep in contact or we don't know what going on now. |
I am facing similar issue with jenkins/ssh-slave setup with this plugin. I see the agent gets deployed in ACI but gets deleted and another instance keeps coming up. 2020-03-29T17:18:15.716749905Z: [ERROR] 2020-03-29 17:18:15.716+0000 [id=112036] INFO c.m.j.c.aci.AciService#createDeployment: Deployment ztest-r3tkt376 succeed 2020-03-29T17:18:16.422975930Z: [ERROR] 2020-03-29 17:18:16.422+0000 [id=112209] INFO c.m.j.c.remote.SSHLauncher#launch: SSHLauncher: Copy slave.jar to remote host successfully 2020-03-29T17:18:17.241691129Z: [ERROR] 2020-03-29 17:18:17.241+0000 [id=112215] INFO o.internal.platform.Platform#log: ALPN callback dropped: HTTP/2 is disabled. Is alpn-boot on the boot class path? 2020-03-29T17:18:24.527343702Z: [ERROR] 2020-03-29 17:18:24.526+0000 [id=88] WARNING hudson.slaves.NodeProvisioner#lambda$update$6: Unexpected exception encountered while provisioning agent ztest 2020-03-29T17:18:28.893307318Z: [ERROR] 2020-03-29 17:18:28.893+0000 [id=112245] INFO hudson.model.AsyncPeriodicWork#lambda$doRun$0: Started DockerContainerWatchdog Asynchronous Periodic Work 2020-03-29T17:18:34.526268101Z: [ERROR] 2020-03-29 17:18:34.525+0000 [id=96] INFO c.m.j.c.aci.AciCloud#provision: Start ACI container for label ztest workLoad 1 2020-03-29T17:18:35.578729453Z: [ERROR] 2020-03-29 17:18:35.578+0000 [id=112036] INFO c.m.j.c.aci.AciService#createDeployment: Waiting for deployment ztest-6rmxcrfl |
Logs from the container before it dies out TX1-XX:aci$ az container logs --name ztest-23n25 --resource-group XX-01 --subscription XX
|
I'm trying out the ACI based cloud but am encountering an issue with both Windows and Linux based containers. While I do end up seeing that the plugin attempts to create instances, my tasks never seem to run. Instead, the instances linger for a while and then disappear, having done nothing.
Feedback on the containers both in the portal and via
az
seems sparse:Looking at the Jenkins logs, I do see one exception per type if I leave things to run long enough:
Nothing fancy about my templates, just trying the defaults for now to make sure everything works:
The text was updated successfully, but these errors were encountered: