Configure probes to check containers - Elastic Container Instance

Liveness probes and readiness probes are mechanisms that are used to check the status of containers. Liveness probes are used to check whether containers run as expected. Readiness probes are used to check whether containers are ready to serve requests. This topic describes how to configure liveness probes and readiness probes to perform health checks on containers. This way, Elastic Container Instance can better monitor and manage the running status of containers and ensure the high availability and stability of services.

Feature description

Elastic Container Instance allows you to use liveness probes and readiness probes to check the condition and running status of containers.

Probe

Description

Scenario

Liveness probe

Liveness probes are used to check whether a container is working as expected.

If the check is successful, the container is working as expected.
If the check fails, the system determines whether to restart the container based on the configured container restart policy.
By default, if a liveness probe is not configured, the container is considered to be working as expected at all times.

When an application is running but no further operations can be performed on the application, liveness probes can capture the deadlock. Then, the system restarts the container to make the application run as expected regardless of whether bugs exist.
After applications run for an extended period of time, the applications may eventually transition to the broken state. If you want to restore the applications, you must restart the applications. You can use liveness probes to detect and remedy such situations.

Readiness probe

Readiness probes are used to check whether a container is ready to serve requests.

If the check is successful, the container is ready to receive business requests.
If the check fails, the container is not ready and the system does not send requests to the container until the recheck is successful.
By default, if a readiness probe is not configured, the container is considered to be ready.

An application may be temporarily unable to serve external requests when the application loads a large amount of data or configuration files during startup. In this case, if you do not want to terminate the application or send requests to the application, you can use readiness probes to detect and take action to mitigate such situations.

Configuration description (API mode)

When you create an elastic container instance by calling the CreateContainerGroup operation, you can use LivenessProbe-related and ReadinessProbe-related parameters to configure liveness probes and readiness probes accordingly. The following table describes the parameters. For more information, see CreateContainerGroup.

LivenessProbe-related parameters

Important

When you configure LivenessProbe-related parameters, you can select only one of the HttpGet, Exec, and TcpSocket check methods.

Parameter	Type	Sample value	Description
Container.N.LivenessProbe.HttpGet.Path	string	/healthyz	The path to which HTTP GET requests are sent when you use HTTP requests to perform health checks.
Container.N.LivenessProbe.HttpGet.Port	integer	8888	The port to which HTTP GET requests are sent when you use HTTP requests to perform health checks.
Container.N.LivenessProbe.HttpGet.Scheme	string	HTTP	The protocol type of HTTP GET requests when you use HTTP requests to perform health checks. Valid values: HTTP HTTPS
Container.N.LivenessProbe.Exec.Command.N	array	cat /tmp/healthy	Command N to be run in the container when you use the CLI to perform health checks.
Container.N.LivenessProbe.TcpSocket.Port	integer	8000	The port to which Transmission Control Protocol (TCP) socket requests are sent when you use TCP sockets to perform health checks.
Container.N.LivenessProbe.InitialDelaySeconds	integer	5	The number of seconds between the time when the container starts and the time when the check starts.
Container.N.LivenessProbe.PeriodSeconds	integer	1	The interval at which the container is checked. Default value: 10. Minimum value: 1. Unit: seconds.
Container.N.LivenessProbe.SuccessThreshold	integer	1	The minimum number of consecutive successes that must occur before the check is considered as successful. Default value: 1. Set the value to 1.
Container.N.LivenessProbe.FailureThreshold	integer	3	The minimum number of consecutive failures that must occur before the check is considered as failed. Default value: 3.
Container.N.LivenessProbe.TimeoutSeconds	integer	1	The timeout period of the health check. Default value: 1. Minimum value: 1. Unit: seconds.

ReadinessProbe-related parameters

Important

When you configure ReadinessProbe-related parameters, you can select only one of the HttpGet, Exec, and TcpSocket check methods.

Parameter	Type	Sample value	Description
Container.N.ReadinessProbe.HttpGet.Path	string	/healthyz	The path to which HTTP GET requests are sent when you use HTTP requests to perform health checks.
Container.N.ReadinessProbe.HttpGet.Port	integer	8888	The port to which HTTP GET requests are sent when you use HTTP requests to perform health checks.
Container.N.ReadinessProbe.HttpGet.Scheme	string	HTTP	The protocol type of HTTP GET requests when you use HTTP requests to perform health checks. Valid values: HTTP HTTPS
Container.N.ReadinessProbe.Exec.Command.N	array	cat /tmp/healthy	Command N to be run in the container when you use the CLI to perform health checks.
Container.N.ReadinessProbe.TcpSocket.Port	integer	8000	The port to which TCP socket requests are sent when you use TCP sockets to perform health checks.
Container.N.ReadinessProbe.InitialDelaySeconds	integer	5	The number of seconds between the time when the container starts and the time when the check starts.
Container.N.ReadinessProbe.PeriodSeconds	integer	1	The interval at which the container is checked. Default value: 10. Minimum value: 1. Unit: seconds.
Container.N.ReadinessProbe.SuccessThreshold	integer	1	The minimum number of consecutive successes that must occur before the check is considered as successful. Default value: 1. Set the value to 1.
Container.N.ReadinessProbe.FailureThreshold	integer	3	The minimum number of consecutive failures that must occur before the check is considered as failed. Default value: 3.
Container.N.ReadinessProbe.TimeoutSeconds	integer	1	The timeout period of the health check. Default value: 1. Minimum value: 1. Unit: seconds.

Configuration description (console mode)

When you create an elastic container instance on the buy page in the Elastic Container Instance console, you can select a container in the Container Configurations section, expand the Advanced Settings of the container, and then turn on Health Check. The following figure shows the configuration details.

Note

When you configure health checks in the Elastic Container Instance console, only the CLI and HTTP request methods are supported. The TCP socket method is not supported.

健康检查1

Parameter	Description
Timeout Period	Waiting Period: The number of seconds between the time when the container starts and the time when the health check starts. Timeout Period: The timeout period of the health check. If a timeout error occurs, the health check failed.
Method	Select a check method based on your business requirements. Valid values: Script: The probe runs a command in the container and checks the exit code of the command. If the exit code is 0, the check is successful. HTTP Request: The probe sends an HTTP request to the container. If the returned status code is greater than or equal to 200 and less than 400, the check is successful.
Script	If you set Method to Script, you must configure the command line script to be run in the container.
HTTP Request	When you set Method to HTTP Request, you must configure the path, port, and protocol for HTTP GET requests.

Configuration examples

In this example, an elastic container instance for which a liveness probe and a readiness probe are configured is created. Then, service exceptions are simulated to check whether the configured probes take effect.

Create an elastic container instance by calling the CreateContainerGroup operation.

The following sample code provides an example on how to use TCP sockets to perform health checks:

ContainerGroupName=test-probe
# Use a NGINX image. Create an elastic IP address (EIP) and associate the EIP with the elastic container instance.
Container.1.Name=container-1
Container.1.Image=nginx
AutoCreateEip=true
# Configure a liveness probe. After the container runs for 5 seconds, the kubelet runs the liveness probe on port 80 every 3 seconds. The timeout period of each check is set to 10 seconds. The minimum number of consecutive successes that must occur before the check is considered as successful is set to 1, and the minimum number of consecutive failures that must occur before the check is considered as failed is set to 3.
Container.1.LivenessProbe.TcpSocket.Port=80
Container.1.ReadinessProbe.InitialDelaySeconds=5
Container.1.LivenessProbe.PeriodSeconds=3
Container.1.LivenessProbe.SuccessThreshold=1
Container.1.LivenessProbe.FailureThreshold=3
Container.N.LivenessProbe.TimeoutSeconds=10
# Configure a readiness probe. After the container runs for 5 seconds, the kubelet runs the readiness probe on port 80 every 3 seconds. The timeout period of each check is set to 10 seconds. The minimum number of consecutive successes that must occur before the check is considered as successful is set to 1, and the minimum number of consecutive failures that must occur before the check is considered as failed is set to 3.
Container.1.ReadinessProbe.TcpSocket.Port=80
Container.1.ReadinessProbe.InitialDelaySeconds=5
Container.1.ReadinessProbe.PeriodSeconds=3
Container.1.ReadinessProbe.SuccessThreshold=1
Container.1.ReadinessProbe.FailureThreshold=3
Container.1.ReadinessProbe.TimeoutSeconds=10

View the related events in the Elastic Container Instance console after an elastic container instance is created.
After an elastic container instance is created, click the ID of the elastic container instance in the Container Group ID/Name column of the instance list page to go to the Instance Details page. Click the Events tab to view the events of the elastic container instance. You can see that the containers on the instance start as expected.
Connect to a container. For more information, see Connect to an elastic container instance.
Change the NGINX listening port in the configuration file to simulate a service exception.
1. Change the NGINX listening port.
  Note
  In this example, a NGINX image on which Vim is not installed is used. Before you run the following vi command to change the NGINX listening port, run the apt-get update and apt-get install vim commands to install Vim.
```
vi /etc/nginx/conf.d/default.conf
```
  For example, change the NGINX listening port from port 80 to port 8080, as shown in the following figure:
2. Restart NGINX.
```
nginx -s reload
```
View the effective status of the probes.
The container automatically restarts a few seconds after you restart NGINX. View the events of the elastic container instance. You can see that the container was restarted after the liveness probe and the readiness probe each have check failures.