プライベートロケーション 🔗
A private location is a name you create in Splunk Synthetic Monitoring to represent a custom location from which you can run synthetic tests. The name you give to a private location allows you to specify that name in a synthetic test’s Locations field. To run synthetic tests from private locations, you must also set up one or more private runners within the private location to do the actual communication with your test targets and with Splunk Synthetic Monitoring.
プライベートロケーションのユースケース 🔗
Private locations provide a way for you to find, fix, and prevent performance problems in internal applications in any environment, inside or outside of your firewalls. You can use private locations to run tests earlier in your development cycle against internal sites or applications that aren’t available to the public. You can also use private locations to test public endpoints from locations that aren’t included in the list of Splunk Synthetic Monitoring public locations.
To summarize, here are sample use cases for private locations:
一般に公開されていないプライベートアプリケーションをテストします。
公開ステージングサイトを持たない本番前のアプリケーションをテストします。
Gain a higher level of flexibility by giving Splunk Synthetic Monitoring access to applications.
Splunk Synthetic Monitoring のパブリックロケーションで現在サポートされていないロケーションからテストします。
新しいプライベートロケーションをセットアップする 🔗
以下の手順に従って、新しいプライベートロケーションをセットアップしてください：
Splunk Synthetic Monitoring で、設定の歯車アイコン、Private locations の順に選択します。
Select Create private location and add a name.
Follow the steps in the guided setup to set up one or more private runners for that private location.
プライベートロケーションを保存します。
プライベートロケーションIDでできること 🔗
各プライベートロケーションには、対応するプライベートロケーションIDがあります。このIDを使って、以下のことができます：
チャートまたはダッシュボードを作成する
プライベートロケーションごとにメトリクスを検索する
Splunk Synthetics Monitoring API とやり取りする場合は、プライベートロケーション ID を参照してください。
トークンを管理する 🔗
トークンの更新と管理は各自の責任で行ってください。トークンは1年間有効です。セキュリティを高めるために、Dockerでトークン用のシークレット環境変数を作成してください。最初のトークンの有効期限が切れる前に、2つ目のトークンを作成することを検討してください。トークンの有効期限切れは通知されません。
Assess the health of a private location 🔗
プライベートロケーションの健全性は3つの要因に左右されます：
要因 |
説明 |
解決策 |
|---|---|---|
アクティブなランナー |
少なくとも1つのランナーがアクティブにチェックインしています。 |
チェックインするランナーがない場合は、プライベートロケーションに新しいランナーを設定します。 |
テストで使用中 |
プライベートロケーションは現在、1つ以上のテストで使用されています。 |
プライベートロケーションを削除する必要がある場合は、まずすべてのテストから削除する必要があります。 |
キューをクリアする |
指定された場所のキューは定期的にクリアされており、バックアップされていません。 |
キューがバックアップされている場合は、プライベートロケーションに新しいランナーを追加します。 |
キューの長さとレイテンシをトラブルシューティングする 🔗
もしキューのレイテンシと長さの両方が時間の経過とともに増加するようであれば、パフォーマンスを向上させるためにランナーを追加します。
If your queue latency increases but your queue length doesn’t, try these troubleshooting methods:
Check to see if a step is delaying the rest of the test.
あなたのマシンでプライベートロケーションのランナーを実行するのに十分なリソースがあるかどうかを調査してください。
キュー内の最大実行回数は100,000回。
1時間以上前の実行はキューから削除されます。
Private runners 🔗
A private runner queries Splunk Synthetic Monitoring for tests configured to run in its inherent private location, performs the test’s steps on your private target, and reports the results back to Splunk Synthetic Monitoring. Because a private runner must have access to your private target, it is a Docker image which you deploy on your own infrastructure, within your own internal network. See プライベートロケーション.
If you deploy multiple private runners on behalf of a single private location, they can process that location’s test queue faster. Splunk Synthetic Monitoring doesn’t track how many private runners you’ve deployed for a given private location. It’s up to you to manage your own fleet of private runners.
Requirements for private runners 🔗
要件 |
説明 |
|---|---|
Docker |
|
許可リスト |
|
オペレーティングシステム |
Linux、Windows、または macOS |
For optimal performance when running browser tests:
Linux
2.3GHzデュアルコアIntel Xeon（または同等の）プロセッサー
8 GB RAM、2コア
Supported platforms 🔗
Docker
Docker Compose
AWS ECS
Docker for Mac or Windows
Helm
Kubernetes
OpenShift
Podman
Podman for MacOS or Windows
ARM64 machines on AWS and GCP
Browser compatibility 🔗
The private runner uses a headless browser to run the browser tests. The private runner Docker image for AMD64 architecture contains the Chrome browser, and the private runner Docker image for ARM64 architecture contains the Chromium browser, because Chrome is unavailable for ARM64. Chrome and Chromium versions might not be the same.
To find the browser type and version, look at the labels browser-type and browser-version in the Docker image. You can find these labels either at http://quay.io/ or in the output of the following commands:
docker pull quay.io/signalfx/splunk-synthetics-runner:latest
docker inspect -f '{{ index .Config.Labels "browser-type" }}' quay.io/signalfx/splunk-synthetics-runner:latest
docker inspect -f '{{ index .Config.Labels "browser-version" }}' quay.io/signalfx/splunk-synthetics-runner:latest
Required container permissions 🔗
This section outlines the minimum requirements for the container on which the private runner Docker image is running.
Minimum container permissions 🔗
The container must have the following permissions at a minimum. The private runner Docker image already has these permissions set up by default, so if you don’t change the container runtime user, you don’t need to take any action:
Read and write access to the application’s home or working directory (usually this is
/home/pptruser/)Read and write access to
/tmp(the system grants this permission to all users by default)
注釈
Don’t set the container’s root filesystem to read-only (the readOnlyRootFilesystem flag), because this prevents the container from starting up.
Optional permissions for custom CA certificates 🔗
If the tests you send to the private runner need to use a custom CA certificate for API and uptime tests, the container must support privilege escalation in an init container, which adds the custom certificate to the runner’s system CA certificates. To allow privilege escalation, set containers.securityContext.allowPrivilegeEscalation to true:
containers:
securityContext:
allowPrivilegeEscalation: true
To verify that the container allows privilege escalation, see if it runs the sudo command.
Optional permissions for network shaping 🔗
If the tests you send to the private runner need to simulate different network throughputs, the Docker container must support privilege escalation along with the NET_ADMIN capability (for network shaping).
containers:
securityContext:
capabilities:
add:
- NET_ADMIN
allowPrivilegeEscalation: true
If you see the following warning message sudo: unable to send audit message: Operation not permitted, also add the AUDIT_WRITE capability:
containers:
securityContext:
capabilities:
add:
- NET_ADMIN
- AUDIT_WRITE
allowPrivilegeEscalation: true
Required container resources 🔗
The Docker container on which you deploy the private runner must have the following resources.
Increase the resources allocated to the private runner’s pod when you see:
Browser crashes or errors
Log messages indicating that there are errors spawning the browser
Minimum container resources 🔗
CPUs: 1
Memory: 2GB
Recommended container resources 🔗
CPUs: 2
Memory: 8GB
Resource intensive tests 🔗
The CPU and memory required for each test are heavily dependent on the test being executed. If the tests you send to the private runner are resource-intensive browser tests such as those listed below, the Docker container must have at least the recommended resources instead of the minimum resources.
Loading complex webpages with high resolution images or complex JavaScript
Media playback such as video streaming
Heavy JavaScript execution such as extensive DOM manipulations or memory hogging
Loading and interacting with large data sets (for example, sorting, filtering, or searching operations)
Uploading or downloading large files
These are only some examples of resource-intensive browser tests; your test cases may vary.
Private runners on Docker 🔗
Install a private runner 🔗
Start your container using the Docker run command and the following flags:
docker run --cap-add NET_ADMIN \
-e "RUNNER_TOKEN=YOUR_TOKEN_HERE" \
quay.io/signalfx/splunk-synthetics-runner:latest
Upgrade a private runner 🔗
Manual upgrades 🔗
To upgrade the Docker image manually, follow these steps:
Pull the latest image:
docker pull http://quay.io/signalfx/splunk-synthetics-runner:latest
Stop the running container:
docker stop <container_id_or_name>
Remove the old container:
docker rm <container_id_or_name>
Start a new container with the updated image:
docker run --cap-add NET_ADMIN -e "RUNNER_TOKEN=YOUR_TOKEN_HERE" http://quay.io/signalfx/splunk-synthetics-runner:latest
Automatic upgrades 🔗
You can automate the upgrade of the private location Docker images by using an automated upgrade solution such as Watchtower , a third party open-source Docker container that connects to remote Docker repositories on a schedule and checks for updates. This section explains how to use Watchtower, but if your operations team already has a mechanism established for deploying updates to Docker images you can use your existing mechanism without making any configuration changes to the private runner. The best practice is to run your upgrade automation at least once every 24 hours. Failing to update the private runner to the latest available image may result in inconsistent data and loss of functionality.
When Watchtower finds an updated image, it instructs your Docker host to pull the newest image from the repository, stop the container, and start it again. It also ensures that environment variables, network settings, and links between containers are intact.
On your Docker host, launch the Watchtower container from command line:
docker run -d \
--name watchtower \
-v /var/run/docker.sock:/var/run/docker.sock \
v2tec/watchtower --label-enable --cleanup
Using the label-enable flag ensures that only containers with the correct label, like the Splunk private runner, are auto-updated.
There are additional options available in the Watchtower documentation that you can explore, including auto-cleanup of old images to ensure that your Docker host does not hold onto outdated images.
注釈
For Watchtower to issue commands to the Docker host, it requires the docker.sock volume or TCP connection, which provides Watchtower with full administrative access to your Docker host. If you are unable to provide Watchtower with this level of access, you can explore other options for automating updates.
Uninstall a private runner 🔗
List all containers:
docker ps -a
Remove a stopped container by ID or name:
docker rm <container_id_or_name>
Force-remove a running container:
docker rm -f my_running_container
Private runners on Docker Compose 🔗
The private runner should work on all the latest versions of Docker Compose.
Install a private runner 🔗
Create a
docker-compose.ymlfile with the following definition:version: '3' services: runner: image: quay.io/signalfx/splunk-synthetics-runner:latest environment: RUNNER_TOKEN: YOUR_TOKEN_HERE cap_add: - NET_ADMIN restart: always
Run the following command to start the container:
docker-compose up
Upgrade a private runner 🔗
Manual upgrades 🔗
To upgrade the Docker image manually, follow these steps:
Navigate to the directory containing your
docker-compose.ymlcd /path/to/your/docker-compose-file
Pull the latest images:
docker-compose pullRecreate containers with the updated images:
docker-compose up -d
Automatic upgrades 🔗
You can automate the upgrade process by using your CI/CD pipelines or by using Watchtower.
Uninstall a private runner 🔗
Navigate to your project directory:
cd /path/to/your/docker-compose-directory
Run the docker-compose down command:
docker-compose down
Private runners on Docker for Mac or Windows 🔗
Install a private runner 🔗
Upgrade a private runner 🔗
Manual upgrades 🔗
To upgrade the Docker image manually, follow these steps:
Pull the latest image:
docker pull http://quay.io/signalfx/splunk-synthetics-runner:latest
Stop the running container:
docker stop <container_id_or_name>
Remove the old container:
docker rm <container_id_or_name>
Start a new container with the updated image:
docker run --cap-add NET_ADMIN -e "RUNNER_TOKEN=YOUR_TOKEN_HERE" \ http://quay.io/signalfx/splunk-synthetics-runner:latest
Automatic upgrades 🔗
You can automate the upgrade of the private location Docker images by using an automated upgrade solution such as Watchtower <https://github.com/v2tec/watchtower>, a third party open source Docker container that connects to remote Docker repositories on a schedule and checks for updates. This section explains how to use Watchtower, but if your operations team already has a mechanism established for deploying updates to Docker images you can use your existing mechanism without making any configuration changes to the private runner. The best practice is to run your upgrade automation at least once every 24 hours. Failing to update the private runner to the latest available image may result in inconsistent data and loss of functionality.
When Watchtower finds an updated image, it instructs your Docker host to pull the newest image from the repository, stop the container, and start it again. It also ensures that environment variables, network settings, and links between containers are intact.
On your Docker host, launch the Watchtower container from the command line:
docker run -d \
--name watchtower \
-v /var/run/docker.sock:/var/run/docker.sock \
v2tec/watchtower --label-enable --cleanup
Using the label-enable flag ensures that only containers with the correct label, like the Splunk private runner, are be auto-updated.
There are additional options available in the Watchtower documentation <https://github.com/v2tec/watchtower#options> that you can explore, including auto-cleanup of old images to ensure that your Docker host does not hold onto outdated images.
注釈
In order for Watchtower to issue commands to the Docker host, it requires the docker.sock volume or TCP connection, which provides Watchtower with full administrative access to your Docker host. If you are unable to provide Watchtower with this level of access you can explore other options for automating updates.
Uninstall a private runner 🔗
To upgrade the Docker image manually, follow these steps:
List all containers:
docker ps -a
Remove a stopped container by ID or name:
docker rm <container_id_or_name>
Force-remove a running container:
docker rm -f my_running_container
Private runners on AWS ECS 🔗
Install a private runner 🔗
In your AWS ECS console, go to Task definitions.
Select Create new task definition with JSON from the yellow dropdown menu.
Copy the following JSON and paste it into the console:
{ "requiresCompatibilities": [ "EC2" ], "containerDefinitions": [ { "name": "splunk-synthetics-runner", "image": "quay.io/signalfx/splunk-synthetics-runner:latest", "memory": 7680, "cpu": 2048, "essential": true, "environment": [ { "name": "RUNNER_TOKEN", "value": "YOUR_TOKEN_HERE" } ], "linuxParameters": { "capabilities": { "add": ["NET_ADMIN"] } } } ], "volumes": [], "networkMode": "none", "memory": "7680", "cpu": "2048", "placementConstraints": [], "family": "splunk-synthetics" }
Select Save to close the JSON input panel.
Select Create to create the task.
Create a service in your cluster using the splunk-synthetics-runner. For steps, see the AWS documentation.
Upgrade a private runner 🔗
Manual upgrades 🔗
You can upgrade the runner by using the forceNewDeployment option. This shuts down the existing container and brings up a new container by pulling the latest image from the repository.
Automatic upgrades 🔗
You can automate the upgrade of the private location Docker images by using an automated upgrade solution such as Watchtower , a third party open source Docker container that connects to remote Docker repositories on a schedule and checks for updates. This section explains how to use Watchtower, but if your operations team already has a mechanism established for deploying updates to Docker images you can use your existing mechanism without making any configuration changes to the private runner. The best practice is to run your upgrade automation at least once every 24 hours. Failing to update the private runner to the latest available image may result in inconsistent data and loss of functionality.
When Watchtower finds an updated image, it instructs your Docker host to pull the newest image from the repository, stop the container, and start it again. It also ensures that environment variables, network settings, and links between containers are intact.
To use Watchtower with Amazon’s Elastic Container Service, you need to create a task definition for it. For example, here is a sample task definition that you can run as a DAEMON within your cluster.
{
"requiresCompatibilities": [
"EC2"
],
"containerDefinitions": [
{
"command": [
"--label-enable",
"--cleanup"
],
"name": "watchtower",
"image": "v2tec/watchtower:latest",
"memory": "512",
"essential": true,
"environment": [],
"linuxParameters": null,
"cpu": "256",
"mountPoints": [
{
"readOnly": null,
"containerPath": "/var/run/docker.sock",
"sourceVolume": "dockerHost"
}
]
}
],
"volumes": [
{
"name": "dockerHost",
"host": {
"sourcePath": "/var/run/docker.sock"
},
"dockerVolumeConfiguration": null
}
],
"networkMode": null,
"memory": "512",
"cpu": "256",
"placementConstraints": [],
"family": "watchtower"
}
Uninstall a private runner 🔗
Open the console at https://console.aws.amazon.com/ecs/v2.
From the navigation bar, select the region that contains your task definition.
In the navigation pane, select Task definitions.
On the Task definitions page, select the task definition family that contains one or more revisions that you want to deregister.
On the Task definition name page, select the revisions to delete, and then select Actions and Deregister.
Verify the information in the Deregister window, and then select Deregister to finish.
Private runners deployed with Helm 🔗
Install a private runner 🔗
For Helm deployments, you can either use the latest image and pullPolicy or a tagged image.
To install the chart with the release name my-splunk-synthetics-runner, run the following commands. For more information, see https://github.com/splunk/synthetics-helm-charts/tree/main/charts/splunk-synthetics-runner#installing-the-chart:
helm repo add synthetics-helm-charts https://splunk.github.io/synthetics-helm-charts/
helm repo update
helm install my-splunk-synthetics-runner synthetics-helm-charts/splunk-synthetics-runner \
--set=synthetics.secret.runnerToken=YOUR_TOKEN_HERE \
--set synthetics.secret.create=true
Upgrade a private runner 🔗
Run the helm upgrade command:
helm upgrade my-splunk-synthetics-runner synthetics-helm-charts/splunk-synthetics-runner --reuse-values
If you’re upgrading to an image that has major version change, you must also upgrade your Helm chart.
Uninstall a private runner 🔗
Run the helm uninstall command:
helm uninstall my-splunk-synthetics-runner
Private runners on Kubernetes 🔗
Install a private runner 🔗
Create a Kubernetes Secret with the runner token:
kubectl create secret generic runner-token-secret \ --from-literal=RUNNER_TOKEN=YOUR_TOKEN_HERE
Create the deployment YAML:
apiVersion: apps/v1 kind: Deployment metadata: name: splunk-o11y-synthetics-runner spec: replicas: 3 selector: matchLabels: app: splunk-o11y-synthetics-runner template: metadata: labels: app: splunk-o11y-synthetics-runner spec: containers: - name: splunk-o11y-synthetics-runner image: quay.io/signalfx/splunk-synthetics-runner:latest imagePullPolicy: Always env: - name: RUNNER_TOKEN valueFrom: secretKeyRef: name: runner-token-secret key: RUNNER_TOKEN securityContext: capabilities: add: - NET_ADMIN allowPrivilegeEscalation: true resources: limits: cpu: "2" memory: 8Gi requests: cpu: "2" memory: 8Gi
Apply the Deployment YAML:
kubectl apply -f deployment.yaml
Upgrade a private runner 🔗
Run the kubectl apply command:
kubectl apply -f deployment.yaml
Since you’re using the latest tag with imagePullPolicy: Always, you don’t need to make changes to the deployment.yaml file. Reapplying the deployment pulls the latest image.
Uninstall a private runner 🔗
Run the kubectl delete command:
kubectl delete -f deployment.yaml
Private runners on OpenShift 🔗
Install a private runner 🔗
Create a OpenShift Secret with the runner token:
oc create secret generic runner-token-secret --from-literal=RUNNER_TOKEN=YOUR_TOKEN_HERE
Create the deployment YAML:
apiVersion: apps/v1 kind: Deployment metadata: name: splunk-o11y-synthetics-runner spec: replicas: 3 selector: matchLabels: app: splunk-o11y-synthetics-runner template: metadata: labels: app: splunk-o11y-synthetics-runner spec: containers: - name: splunk-o11y-synthetics-runner image: quay.io/signalfx/splunk-synthetics-runner:latest imagePullPolicy: Always env: - name: RUNNER_TOKEN valueFrom: secretKeyRef: name: runner-token-secret key: RUNNER_TOKEN securityContext: capabilities: add: - NET_ADMIN allowPrivilegeEscalation: true resources: limits: cpu: "2" memory: 8Gi requests: cpu: "2" memory: 8Gi
Apply the deployment YAML:
oc apply -f deployment.yaml
Upgrade a private runner 🔗
Run the oc apply command:
oc apply -f deployment.yaml
Since you’re using the latest tag with imagePullPolicy: Always, you don’t need to make changes to the deployment.yaml file. Reapplying the deployment pulls the latest image.
Uninstall a private runner 🔗
Run the oc delete command:
oc delete -f deployment.yaml
Private runners on Podman 🔗
Install a private runner 🔗
Start your container using the Podman run command and the following flags.
podman run --cap-add NET_ADMIN -e "RUNNER_TOKEN=YOUR_TOKEN_HERE" \
quay.io/signalfx/splunk-synthetics-runner:latest
Upgrade a private runner 🔗
Pull the latest image:
podman pull http://quay.io/signalfx/splunk-synthetics-runner:latest
Stop the running container:
podman stop <container_id_or_name>
Remove the old container:
podman rm <container_id_or_name>
Start a new container with the updated image:
podman run --cap-add NET_ADMIN -e "RUNNER_TOKEN=YOUR_TOKEN_HERE" \ http://quay.io/signalfx/splunk-synthetics-runner:latest
Uninstall a private runner 🔗
List all containers:
podman ps -a
Remove a stopped container by ID or name:
podman rm <container_id_or_name>
Force remove a running container:
podman rm -f my_running_container
Private runners on Podman for MacOS or Windows 🔗
Install a private runner 🔗
podman run -e "DISABLE_NETWORK_SHAPING=true" -e "RUNNER_TOKEN=YOUR_TOKEN_HERE" \
quay.io/signalfx/splunk-synthetics-runner:latest
Upgrade a private runner 🔗
Pull the latest image:
podman pull http://quay.io/signalfx/splunk-synthetics-runner:latest
Stop the running container:
podman stop <container_id_or_name>
Remove the old container:
podman rm <container_id_or_name>
Start a new container with the updated image:
podman run --cap-add NET_ADMIN -e "RUNNER_TOKEN=YOUR_TOKEN_HERE" http://quay.io/signalfx/splunk-synthetics-runner:latest
Uninstall a private runner 🔗
List all containers:
podman ps -a
Remove a stopped container by ID or name:
podman rm <container_id_or_name>
Force remove a running container:
podman rm -f my_running_container
Private runners on ARM64 machines on AWS and GCP 🔗
There are no special instructions to install or upgrade a Docker image running on an ARM64-based machine. You can deploy this image manually with Docker or Docker Compose, deploy it to Kubernetes hosted on AWS EKS, GCP GKE, self-hosted, or deploy it on AWS ECS.
The Docker manifest contains information about available platforms and links to the correct images. So, for example, when you run the command docker run http://quay.io/signalfx/splunk-synthetics-runner:latest Docker downloads the correct image based on the architecture of your machine.
Troubleshoot a private runner 🔗
Docker health check 🔗
The private location Docker image utilizes the Docker health check to communicate when its container has entered an unhealthy state. The container state is healthy if the private runner is able to authenticate with the API and has successfully fetched a synthetics test in the last 30 minutes. If the container state is unhealthy, try the following troubleshooting tips in this order:
Check the logs of the container to see if there is an authentication error. If there is, confirm that you specified the correct value for the
RUNNER_TOKENenvironment variable at pod startup.Restart the container.
Automatically restart unhealthy Docker containers 🔗
If you plan on running a private location for an extended period of time, it can be helpful to allow the container to automatically restart in the event that it becomes unhealthy.
To automatically restart the container you must add --restart unless-stopped and -e ALWAYS_HEALTHY=true to the pod startup command (docker run or podman run and so on). The ALWAYS_HEALTHY=true environment variable terminates the Docker container as soon as it fails a health check. This option works on any Docker restart policy.
docker run --restart unless-stopped -e ALWAYS_HEALTHY=true --cap-add NET_ADMIN \
-e "RUNNER_TOKEN=YOUR_TOKEN_HERE" \
quay.io/signalfx/splunk-synthetics-runner:latest
Dockerでの作業 🔗
Follow these steps to limit logging in Docker:
以下のようなディレクトリにファイルを作成します：
/etc/docker/daemon.json.ファイルに追加します：
{ "log-driver": "local", "log-opts": { "max-size": "10m", "max-file": "3" } }
docker サービスを再起動します：
sudo systemctl docker.service restart.
Add certificates 🔗
Splunk Synthetic Monitoring supports injecting custom root CA certificates for Uptime tests running from your private locations. Client keys and certificates aren’t supported at this time.
ホストマシン上に
certsというフォルダを作成し、その中に CA 証明書（CRT 形式）を配置します。コンテナ
(-v ./certs:/usr/local/share/ca-certificates/my_certs/)にボリュームとしてcertsフォルダを追加します。コンテナ起動時に使用するコマンドを変更し、エージェントバイナリ
(bash -c "sudo update-ca-certificates && bundle exec bin/start_runner)を起動する前に CA 証明書キャッシュを更新します。
例えば、あるコマンドを自分の環境に合わせて修正した後のイメージは以下のとおりです：
docker run -e "RUNNER_TOKEN=<insert-token>" --volume=`pwd`/certs:/usr/local/share/ca-certificates/my_certs/ quay.io/signalfx/splunk-synthetics-runner:latest bash -c "sudo update-ca-certificates && bundle exec bin/start_runner"
注釈
カスタムルート CA 証明書は、ブラウザテストではサポートされていません。ブラウザテストでは、正確なテストのために SSL/TLS 検証が必要です。必要に応じて、ブラウザテストのSSL/TLS検証を無効にできます。
Configure proxy settings for a private runner 🔗
インターネットへの直接アクセスが制限されている環境では、以下の環境変数を設定することで、合成テストのトラフィックをプロキシサーバー経由でルーティングすることができます：
http_proxy: Specifies the proxy server for HTTP traffic.Example:
export http_proxy="http://proxy.example.com:8443"
https_proxy: Specifies the proxy server for HTTPS traffic.Example:
export https_proxy="http://proxy.example.com:8443"
no_proxy: Specifies a comma-separated list of domains or IP addresses that should bypass the proxy.Example:
export no_proxy="localhost,127.0.0.1,.internal-domain.com"
例えば、あるコマンドを自分の環境に合わせて修正した後のイメージは以下のとおりです：
docker run --cap-add NET_ADMIN -e "RUNNER_TOKEN=*****" -e "no_proxy=.signalfx.com,.amazonaws.com,127.0.0.1,localhost" -e "https_proxy=http://172.17.0.1:1234" -e "http_proxy=http://172.17.0.1:1234" quay.io/signalfx/splunk-synthetics-runner:latest
In this example:
http_proxy and https_proxy are set to route traffic through a proxy at http://172.17.0.1:1234.
no_proxy is configured to bypass the proxy for local addresses and specific domains like .signalfx.com and .amazonaws.com.
これらの変数がネットワークポリシーに従って正しく設定されていることを確認してください。この設定により、制御されたネットワーク環境で、合成テストが安全かつ効率的に通信できるようになります。
When using a private runner, it’s important to correctly configure the proxy settings to avoid issues with browser-based tests. Follow these steps when setting up the environment of the private runners:
Ensure proper no_proxy setup:
When configuring
no_proxyalways include the following addresses:127.0.0.1(localhost通信用)localhost(ローカルテストの解決用)
これらのアドレスは、プロキシを経由せずに内部サービスやテストが正しく実行されることを保証し、潜在的な障害を防ぎます。
Understand Dockerfile defaults:
By default, the private runner sets the
no_proxyvariable in the Dockerfile to include127.0.0.1. If you overrideno_proxy, you must ensure that127.0.0.1andlocalhostare still present, or browser tests may fail.
注釈
Lower case variable names take precedence and are best practice.