Docker swarm based elastic agents for GoCD

Home Page: https://www.go.cd

License: Apache License 2.0

Java 93.14% HTML 5.58% Shell 0.92% Dockerfile 0.36%

docker-plugin gocd gocd-agent elastic-agents

docker-swarm-elastic-agent-plugin's Introduction

GoCD Elastic agent plugin for Docker Swarm

Installation
Building the code base
Using your own docker image with elastic agents
Troubleshooting
License

Installation

Documentation for installation is available here.

Building the code base

To build the jar, run ./gradlew clean test assemble

Using your own docker image with elastic agents

More information to build custom GoCD agent docker image is available here

Troubleshooting

If you already have it running it on a mac, make sure to restart it (see docker/for-mac#17 (comment)). Time drift is known to cause the plugin to not work, because the timestamps returned by the docker API has drifted from the host.

A good way to know if there's a time drift is to run docker ps —

```
CONTAINER ID        IMAGE               COMMAND             CREATED             STATUS              PORTS               NAMES
e0754c9f4cdb        alpine:latest       "/bin/sh"           32 minutes ago      Up 17 seconds                           test
809f310ba1e4        ubuntu:trusty       "/bin/bash"         33 minutes ago      Up About a minute                       reverent_raman
```

Notice how the CREATED and STATUS are several minutes apart for a recently created container.

Enabling debug level logging

If you are on GoCD version 19.6 and above:

Edit the file wrapper-properties.conf on your GoCD server and add the following options. The location of the wrapper-properties.conf can be found in the installation documentation of the GoCD server.

# We recommend that you begin with the index `100` and increment the index for each system property
wrapper.java.additional.100=-Dplugin.cd.go.contrib.elastic-agent.docker-swarm.log.level=debug

If you're running with GoCD server 19.6 and above on docker using one of the supported GoCD server images, set the environment variable GOCD_SERVER_JVM_OPTIONS:

docker run -e "GOCD_SERVER_JVM_OPTIONS=-Dplugin.cd.go.contrib.elastic-agent.docker-swarm.log.level=debug" ...

If you are on GoCD version 19.5 and lower:

Enabling debug level logging can help you troubleshoot an issue with the elastic agent plugin. To enable debug level logs, edit the /etc/default/go-server (for Linux) to add:

export GO_SERVER_SYSTEM_PROPERTIES="$GO_SERVER_SYSTEM_PROPERTIES -Dplugin.cd.go.contrib.elastic-agent.docker-swarm.log.level=debug"

If you're running the server via ./server.sh script —

$ GO_SERVER_SYSTEM_PROPERTIES="-Dplugin.cd.go.contrib.elastic-agent.docker-swarm.log.level=debug" ./server.sh

License

Copyright 2019, Thoughtworks, Inc.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

   http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

docker-swarm-elastic-agent-plugin's People

Contributors

Stargazers

Watchers

Forkers

ketan arvindsv bdpiprava mattlk13 markoradinovic skizz xxbedy ganeshspatil akshaydewan akshaymayekar varshavaradarajan aiminickwong kadirtaskiran sloukam dhanasp dlethin isabella232

docker-swarm-elastic-agent-plugin's Issues

Upgrade spotify docker client library

Add ability to read configuration from external file

The idea is to have a external file, something like a swarm-elastic-agents.conf that would go along with the GoCD configuration and the plugin could merge the configuration made through the UI and what is in the file.

The idea came from the implementation made by the slack notifier plugin (https://github.com/ashwanthkumar/gocd-slack-build-notifier), not sure if there are other ways of doing that.

Add support for constraints in profile

I'm running a swarm with nodes running on centos 7 and windows 2016 server. When I schedule a service, it usually "bounces" between various hosts because a windows container when scheduled on a linux host fails and gets re-assigned to another host until it is assigned to a windows host. Similar thing happens when a linux container is scheduled on a windows host. Depending on the number of bounces, it can take several seconds for the container to come up.

In such mixed-os swarm clusters, it is recommended that:

each node be "tagged" with a label:
docker node update --label-add "os=[windows|linux]"
when scheduling a particular service, apply a constraint to run on a specific node(s):
docker service create --constraint 'node.labels.os==windows'

Use the server info endpoint to map swarm services to a particular gocd server

Upgrade version of spotify docker client to V 8.3.1

Spotify docker client library is now in version 8.3.1 which supports the usage of secrets in swarm. This is important when you need to pull materials from private repos where you need an ssh private key in elastic agents for example.

Create Agent Fails when optional `Memory soft limit` or `Maximum hard limit` is not specified on EA profile

Create Agent request fails with following exception when MaxMemory or ReservedMemory is not specified:

2018-01-16 12:13:26,191 ERROR [1144@MessageListener for CreateAgentListener] DockerPlugin:82 - Failed to handle request go.cd.elastic-agent.create-agent due to:
java.lang.IllegalArgumentException: Invalid size:
	at com.google.common.base.Preconditions.checkArgument(Preconditions.java:122)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.utils.Size.parse(Size.java:83)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerService.requirements(DockerService.java:138)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerService.create(DockerService.java:111)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerServices.create(DockerServices.java:49)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerServices.create(DockerServices.java:34)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.executors.CreateAgentRequestExecutor.execute(CreateAgentRequestExecutor.java:44)
	at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.handle(DockerPlugin.java:55)
	at com.thoughtworks.go.plugin.infra.DefaultPluginManager$2.execute(DefaultPluginManager.java:167)
	at com.thoughtworks.go.plugin.infra.DefaultPluginManager$2.execute(DefaultPluginManager.java:162)
	at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.executeActionOnTheService(FelixGoPluginOSGiFramework.java:316)
	at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.doOn(FelixGoPluginOSGiFramework.java:246)
	at com.thoughtworks.go.plugin.infra.DefaultPluginManager.submitTo(DefaultPluginManager.java:162)
	at com.thoughtworks.go.plugin.access.PluginRequestHelper.submitRequest(PluginRequestHelper.java:48)
	at com.thoughtworks.go.plugin.access.elastic.ElasticAgentExtension.createAgent(ElasticAgentExtension.java:71)
	at com.thoughtworks.go.plugin.access.elastic.ElasticAgentPluginRegistry.createAgent(ElasticAgentPluginRegistry.java:41)
	at com.thoughtworks.go.server.messaging.elasticagents.CreateAgentListener.onMessage(CreateAgentListener.java:32)
	at com.thoughtworks.go.server.messaging.elasticagents.CreateAgentListener.onMessage(CreateAgentListener.java:22)
	at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.runImpl(JMSMessageListenerAdapter.java:73)
	at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.run(JMSMessageListenerAdapter.java:53)
	at java.lang.Thread.run(Thread.java:745)

Possible Fix

Do not parse resource limits and resource reservations when not specified in the EA profile.

Option to map volumes

I need to map /var/run/docker.sock into agent container. Your agents probably do not execute docker commands?

Bind mounts only ignored without existing volume

https://github.com/gocd-contrib/docker-swarm-elastic-agents/blob/7325415399455e4590119aa6b23ae05eeb1dd430/src/main/java/cd/go/contrib/elasticagents/dockerswarm/elasticagent/DockerMounts.java#L93

If there is no existing volume, then nothing gets mounted. This is a problem in the case when profile contains only "bind" type mounts

Do not throw NPE when plugin is not configured

The ping message will inevitably cause the plugin to connect to a swarm instance and cause a NPE.

Ideally if the plugin is not configured, all messages should be ignored and logged at "WARN" level (without a stacktrace)

Show the max container limit exceeded error on server health messages

Right now, there is no way of knowing why an elastic agent isn't assigned to the job. We have to look at the plugin logs to figure out that the max container limit has been exceeded. I think that showing this message with the plugin id will help a bit when users don't understand why the agent hasn't been assigned.

Docker host hostname

I don't know if this is an issue or intended behaviour.

I'd like an agent container to know the hostname of the docker host it's running on.

Then agents docker-entrypoint.sh as specified in Dockerfile.erb should be able to have a CMD line passed to it. I thought that I could pass the same idea, like $HOSTNAME or something of the sort via https://hostname/go/admin/elastic_profiles under Docker command but it doesn't pass arg to ENTRYPOINT infact it actually overwrites it and tried to start the container with $HOSTNAME as command.

Should mention here I've built my own image from a fork of your repo and I'm exporting ${SWARM_NODE_HOSTNAME} from $1 when it's passed in

The way the agents are brough up I can't seem to specify additional service or run options like --args or env. I can't even use -v to mount a file from the host.

Again it'd be nice for the container to know which node it's on. The reason being that the container agent is building containers and has the docker client on it. It's connnecting through to the hosts docker daemon via tcp://. It's fine if you have one worker and know the hostname. But if you have a few workers this has to be dynamic.

I've also tried querying the remote docker api but came up short.

Any ideas?

btw I should mention that this works manually with

docker run -it --rm agent_image:ver $HOSTNAME

as this is passed as arg

It's also not enough to add the whole start up line

/docker-entrypoint.sh $HOSTNAME

to the agent_profile as it looks for /docker-entrypoint.sh locally lol and not as pid1 for the container.

Can't even find where it's been placed on the gocd /go/api/agents/ side.

Layout breaks on smaller screens

On my 15 inch mac. The layout seems to break. If I squeeze the width further, it wraps completely.

Add support for configs

Just like we have a need to provide agents with secrets, we need to have an option to provide configs.

Will look into it, when I have more time. Should be easy to implement.

bad `placeholder` for host file field

Allow for maximum time to keep a container alive

Currently the plugin will aggressively delete the container as soon as it sees that the agent is idle. Would like to see the ability to terminate the container only if it remains unused for X amount of period, or just put a soft and hard upper bound on how long a container should remain alive.

Soft upper bound being a point where the agent will be disabled. Hard upper bound being the point where the container will be terminated (even if it is running a build). This would be useful in cases where the agent or build that is running on the container hangs up.

Show Elastic Agent Starting state on agent status page

Duplicate of gocd/kubernetes-elastic-agents#19.

Permission denied

Getting the following error on http://gocd-ip:8153/go/admin/status_reports/cd.go.contrib.elastic-agent.docker-swarm

The plugin sent a response that could not be understood by Go. Plugin returned with code '500' and the following response: 'Failed to handle request go.cd.elastic-agent.status-report due to:java.util.concurrent.ExecutionException: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied'

In go-server.log
2017-12-11 07:28:40,282 ERROR [142@MessageListener for ServerPingListener] JMSMessageListenerAdapter:77 - Exception thrown in message handling by listener com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener@426e2cf2
java.lang.RuntimeException: Interaction with plugin with id 'cd.go.contrib.elastic-agent.docker-swarm' implementing 'elastic-agent' extension failed while requesting for 'go.cd.elastic-agent.server-ping'. Reason: [The plugin sent a response that could not be understood by Go. Plugin returned with code '500' and the following response: 'Failed to handle request go.cd.elastic-agent.server-ping due to:com.spotify.docker.client.exceptions.DockerException: java.util.concurrent.ExecutionException: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied']
at com.thoughtworks.go.plugin.access.PluginRequestHelper.submitRequest(PluginRequestHelper.java:42)
at com.thoughtworks.go.plugin.access.elastic.ElasticAgentExtension.serverPing(ElasticAgentExtension.java:68)
at com.thoughtworks.go.plugin.access.elastic.ElasticAgentPluginRegistry.serverPing(ElasticAgentPluginRegistry.java:49)
at com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener.onMessage(ServerPingListener.java:32)
at com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener.onMessage(ServerPingListener.java:22)
at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.runImpl(JMSMessageListenerAdapter.java:73)
at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.run(JMSMessageListenerAdapter.java:53)
at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.RuntimeException: The plugin sent a response that could not be understood by Go. Plugin returned with code '500' and the following response: 'Failed to handle request go.cd.elastic-agent.server-ping due to:com.spotify.docker.client.exceptions.DockerException: java.util.concurrent.ExecutionException: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied'
at com.thoughtworks.go.plugin.access.PluginRequestHelper.submitRequest(PluginRequestHelper.java:40)
... 7 common frames omitted

plugin-cd.go.contrib.elastic-agent.docker-swarm.log
2017-12-11 07:38:40,216 INFO [142@MessageListener for ServerPingListener] DockerPlugin:57 - Using private docker registry server my-private-registry.
2017-12-11 07:38:40,250 ERROR [142@MessageListener for ServerPingListener] DockerPlugin:82 - Failed to handle request go.cd.elastic-agent.server-ping due to:
java.lang.RuntimeException: com.spotify.docker.client.exceptions.DockerException: java.util.concurrent.ExecutionException: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.refreshInstances(DockerPlugin.java:93)
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.handle(DockerPlugin.java:57)
at com.thoughtworks.go.plugin.infra.DefaultPluginManager$2.execute(DefaultPluginManager.java:167)
at com.thoughtworks.go.plugin.infra.DefaultPluginManager$2.execute(DefaultPluginManager.java:162)
at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.executeActionOnTheService(FelixGoPluginOSGiFramework.java:316)
at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.doOn(FelixGoPluginOSGiFramework.java:246)
at com.thoughtworks.go.plugin.infra.DefaultPluginManager.submitTo(DefaultPluginManager.java:162)
at com.thoughtworks.go.plugin.access.PluginRequestHelper.submitRequest(PluginRequestHelper.java:33)
at com.thoughtworks.go.plugin.access.elastic.ElasticAgentExtension.serverPing(ElasticAgentExtension.java:68)
at com.thoughtworks.go.plugin.access.elastic.ElasticAgentPluginRegistry.serverPing(ElasticAgentPluginRegistry.java:49)
at com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener.onMessage(ServerPingListener.java:32)
at com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener.onMessage(ServerPingListener.java:22)
at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.runImpl(JMSMessageListenerAdapter.java:73)
at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.run(JMSMessageListenerAdapter.java:53)
at java.lang.Thread.run(Thread.java:748)
Caused by: com.spotify.docker.client.exceptions.DockerException: java.util.concurrent.ExecutionException: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied
at com.spotify.docker.client.DefaultDockerClient.propagate(DefaultDockerClient.java:2657)
at com.spotify.docker.client.DefaultDockerClient.request(DefaultDockerClient.java:2588)
at com.spotify.docker.client.DefaultDockerClient.ping(DefaultDockerClient.java:501)
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerClientFactory.createClient(DockerClientFactory.java:72)
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerClientFactory.docker(DockerClientFactory.java:49)
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerServices.docker(DockerServices.java:135)
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerServices.refreshAll(DockerServices.java:118)
at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.refreshInstances(DockerPlugin.java:91)
... 14 common frames omitted
Caused by: java.util.concurrent.ExecutionException: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied
at jersey.repackaged.com.google.common.util.concurrent.AbstractFuture$Sync.getValue(AbstractFuture.java:299)
at jersey.repackaged.com.google.common.util.concurrent.AbstractFuture$Sync.get(AbstractFuture.java:286)
at jersey.repackaged.com.google.common.util.concurrent.AbstractFuture.get(AbstractFuture.java:116)
at com.spotify.docker.client.DefaultDockerClient.request(DefaultDockerClient.java:2586)
... 20 common frames omitted
Caused by: javax.ws.rs.ProcessingException: java.io.IOException: Permission denied
at org.glassfish.jersey.apache.connector.ApacheConnector.apply(ApacheConnector.java:481)
at org.glassfish.jersey.apache.connector.ApacheConnector$1.run(ApacheConnector.java:491)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at jersey.repackaged.com.google.common.util.concurrent.MoreExecutors$DirectExecutorService.execute(MoreExecutors.java:299)
at java.util.concurrent.AbstractExecutorService.submit(AbstractExecutorService.java:112)
at jersey.repackaged.com.google.common.util.concurrent.AbstractListeningExecutorService.submit(AbstractListeningExecutorService.java:50)
at jersey.repackaged.com.google.common.util.concurrent.AbstractListeningExecutorService.submit(AbstractListeningExecutorService.java:37)
at org.glassfish.jersey.apache.connector.ApacheConnector.apply(ApacheConnector.java:487)
at org.glassfish.jersey.client.ClientRuntime$2.run(ClientRuntime.java:178)
at org.glassfish.jersey.internal.Errors$1.call(Errors.java:271)
at org.glassfish.jersey.internal.Errors$1.call(Errors.java:267)
at org.glassfish.jersey.internal.Errors.process(Errors.java:315)
at org.glassfish.jersey.internal.Errors.process(Errors.java:297)
at org.glassfish.jersey.internal.Errors.process(Errors.java:267)
at org.glassfish.jersey.process.internal.RequestScope.runInScope(RequestScope.java:340)
at org.glassfish.jersey.client.ClientRuntime$3.run(ClientRuntime.java:210)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
... 1 common frames omitted
Caused by: java.io.IOException: Permission denied
at jnr.unixsocket.UnixSocketChannel.doConnect(UnixSocketChannel.java:94)
at jnr.unixsocket.UnixSocketChannel.connect(UnixSocketChannel.java:102)
at com.spotify.docker.client.ApacheUnixSocket.connect(ApacheUnixSocket.java:76)
at com.spotify.docker.client.UnixConnectionSocketFactory.connectSocket(UnixConnectionSocketFactory.java:78)
at org.apache.http.impl.conn.DefaultHttpClientConnectionOperator.connect(DefaultHttpClientConnectionOperator.java:134)
at org.apache.http.impl.conn.PoolingHttpClientConnectionManager.connect(PoolingHttpClientConnectionManager.java:353)
at org.apache.http.impl.execchain.MainClientExec.establishRoute(MainClientExec.java:380)
at org.apache.http.impl.execchain.MainClientExec.execute(MainClientExec.java:236)
at org.apache.http.impl.execchain.ProtocolExec.execute(ProtocolExec.java:184)
at org.apache.http.impl.execchain.RetryExec.execute(RetryExec.java:88)
at org.apache.http.impl.execchain.RedirectExec.execute(RedirectExec.java:110)
at org.apache.http.impl.client.InternalHttpClient.doExecute(InternalHttpClient.java:184)
at org.apache.http.impl.client.CloseableHttpClient.execute(CloseableHttpClient.java:71)
at org.glassfish.jersey.apache.connector.ApacheConnector.apply(ApacheConnector.java:435)
... 21 common frames omitted
2017-12-11 07:38:55,271 INFO [qtp81009902-31] DockerPlugin:57 - [status-report] Generating status report

Replace Spotify docker-client with an alternative

The client currently used is effectively EOL: https://github.com/spotify/docker-client

It also has a lot of dependencies on other libraries which cannot be upgraded (e.g Jersey) who themselves have reported vulnerabilities (which may or may not be relevant to this plugin)

Probably should move to https://github.com/docker-java/docker-java ?

See gocd/docker-registry-artifact-plugin#874 and gocd-contrib/docker-elastic-agents-plugin#258

Validation of HTTPs url

I think it's time to disable URL validation on the plugin and allow passing in http:// URLs instead of https:// URLs for the server url field.

Can I execute the agents in another server (not in go-server)?

Hello!

I would like to know if it is possible go-server call another server where will be create the containers to execute the pipelines.

go-server -> server with elastic agents

If possible, how do I configure?

Thanks.
Regards.

Logs not shown

Elastic agents status report page: property content - text alignment

Issue Type

Bug Report

Summary

The properties in status report page is center aligned currently. The text would read better if it is left-aligned.
Existing:

Proposed:

Note: Color of the table header has been changed to be make it consistent with the rest of the application.

Create installation guide

Getting started guide for docker elastic agent.

Do not rely on Agent autoregister timeout value for terminating agents after job completion

Refer to gocd/kubernetes-elastic-agents#54

Support starting agent containers in privileged mode

This will help in using a docker in docker image to run docker related tasks using elastic agents.

Support Agent status report Request

More information: gocd/gocd#4164

extension failed while requesting for 'go.cd.elastic-agent.server-ping'

I cannot get the plugin to work with GoCD server 17.5.0 and plugin version 1.1.4.

I'm getting
2017-06-27 11:51:10,991 DEBUG [132@MessageListener for ServerPingListener] Printer:113 - com.thoughtworks.go.domain.Plugin{configuration={"environment_variables":"","docker_ca_cert":"","docker_uri":"http://192.168.235.30:2375","go_server_url":"https://192.168.235.30:8154/go","docker_client_key":"","max_docker_containers":"10","docker_client_cert":"","auto_register_timeout":"2"}, pluginId=cd.go.contrib.elastic-agent.docker-swarm, id=1} 2017-06-27 11:51:11,001 DEBUG [cancelHungJobThread] GO_MODE_AWARE_TIMER:41 - Server is in active state, Running: class com.thoughtworks.go.server.service.ScheduleService#cancelHungJobs 2017-06-27 11:51:11,012 ERROR [132@MessageListener for ServerPingListener] JMSMessageListenerAdapter:77 - Exception thrown in message handling by listener com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener@333c4971 java.lang.RuntimeException: Interaction with plugin with id 'cd.go.contrib.elastic-agent.docker-swarm' implementing 'elastic-agent' extension failed while requesting for 'go.cd.elastic-agent.server-ping'. Reason: [java.lang.RuntimeException: java.lang.NullPointerException] at com.thoughtworks.go.plugin.access.PluginRequestHelper.submitRequest(PluginRequestHelper.java:41) at com.thoughtworks.go.plugin.access.elastic.ElasticAgentExtension.serverPing(ElasticAgentExtension.java:60) at com.thoughtworks.go.plugin.access.elastic.ElasticAgentPluginRegistry.serverPing(ElasticAgentPluginRegistry.java:49) at com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener.onMessage(ServerPingListener.java:32) at com.thoughtworks.go.server.messaging.elasticagents.ServerPingListener.onMessage(ServerPingListener.java:22) at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.runImpl(JMSMessageListenerAdapter.java:73) at com.thoughtworks.go.server.messaging.activemq.JMSMessageListenerAdapter.run(JMSMessageListenerAdapter.java:53) at java.lang.Thread.run(Thread.java:748) Caused by: java.lang.RuntimeException: java.lang.RuntimeException: java.lang.NullPointerException at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.executeActionOnTheService(FelixGoPluginOSGiFramework.java:317) at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.doOn(FelixGoPluginOSGiFramework.java:245) at com.thoughtworks.go.plugin.infra.DefaultPluginManager.submitTo(DefaultPluginManager.java:173) at com.thoughtworks.go.plugin.access.PluginRequestHelper.submitRequest(PluginRequestHelper.java:32) ... 7 more Caused by: java.lang.RuntimeException: java.lang.RuntimeException: java.lang.NullPointerException at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.handle(DockerPlugin.java:78) at com.thoughtworks.go.plugin.infra.DefaultPluginManager$2.execute(DefaultPluginManager.java:178) at com.thoughtworks.go.plugin.infra.DefaultPluginManager$2.execute(DefaultPluginManager.java:173) at com.thoughtworks.go.plugin.infra.FelixGoPluginOSGiFramework.executeActionOnTheService(FelixGoPluginOSGiFramework.java:315) ... 10 more Caused by: java.lang.RuntimeException: java.lang.NullPointerException at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.refreshInstances(DockerPlugin.java:86) at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.handle(DockerPlugin.java:58) ... 13 more Caused by: java.lang.NullPointerException at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerServices.refreshAll(DockerServices.java:120) at cd.go.contrib.elasticagents.dockerswarm.elasticagent.DockerPlugin.refreshInstances(DockerPlugin.java:84) ... 14 more 2017-06-27 11:51:11,195 DEBUG [consumeBuildCausesThread] GO_MODE_AWARE_TIMER:41 - Server is in active state, Running: class com.thoughtworks.go.server.service.ScheduleService#autoSchedulePipelinesFromRequestBuffer 2017-06-27 11:51:12,195 DEBUG [consumeBuildCausesThread] GO_MODE_AWARE_TIMER:41 - Server is in active state, Running: class com.thoughtworks.go.server.service.ScheduleService#autoSchedulePipelinesFromRequestBuffer

Add ability to add custom host mapping

Docker provides for --add-host host:ip.add.re.ss to allow adding entries to the hosts file in the container

Usage of `unix:///var/run/docker.sock` possible

I want to beef-up my GoCD stack and run the GoCD server in a SWARM cluster on my Macbook. Thus, I use unix:///var/run/docker.sock.

Mounting the socket into the server and using unix:///var/run/docker.sock as DOCKER_HOST in the plugin configuration leads to:
"""
We're sorry, but something went wrong.
If you are the application owner check the logs for more information.
"""
Is it possible to use a unix-socket in general?

Need an option to add network to agent container

When you enter your swarm service name into "Go Server URL (this is passed to the agents, so don't use localhost):*" and try to save it, it will give an error "We're sorry, but something went wrong."

Most likely the validation is too strict - should give better error though. Issue is most likely that swarm service names have _ in domain part, for example https://gocd_server:8154/go

At first I tried just https://server:8154/go, but that returned java.net.UnknownHostException: server: Try again in agents logs.

Argh, I think its an issue that gocd_server is in its own network and I have no option to provide the network name to agents

Plugin settings: Agent auto-register Timeout (in minutes) should be (in seconds)

In the plugin setting view there is a field to configure the agent auto-register Timeout.
It's label says the number in the field should be written in minutes but the plugin is actually using this number as unit of seconds.

gocd-contrib / docker-swarm-elastic-agent-plugin Goto Github PK