Skip to content

Latest commit

 

History

History
96 lines (70 loc) · 6.24 KB

load-balancing-failover-and-health-checks.md

File metadata and controls

96 lines (70 loc) · 6.24 KB
Raw
description
This section focuses on configuring load-balancing, failover, and health-checks as Gravitee backend services

Load-balancing, Failover, and Health-checks

Overview

APIM offers three main backend services for managing your APIs that are built into the Gravitee platform:

  • Load-balancing: A technique that distributes incoming traffic across multiple backend servers to optimize resource utilization, maximize throughput, minimize response time, and avoid overloading any single server.
  • Failover: Ensures high availability and reliability by redirecting incoming traffic to a secondary server or backup system in the event of a primary server failure.
  • Health-checks: A health check is a mechanism used to monitor the availability and health of your endpoints and/or API Gateways.

Load-balancing

Gravitee load-balancing relies on:

{% tabs %} {% tab title="Round robin" %} Maintains a list of backend servers and assigns each incoming request to the next server on the list. Once the last server has been reached, the algorithm starts again from the beginning of the list, cycling through the servers in a circular manner. {% endtab %}

{% tab title="Random" %} Selects a backend server at random for each incoming request. Each server has an equal chance of being selected, regardless of its current load or processing capacity. {% endtab %}

{% tab title="Weighted round robin" %} Works similarly to round robin, but instead of assigning incoming requests in a circular manner, requests are assigned based on the specified weight given to each backend server.

Example: If endpoint 1 has a weight of 9 and endpoint 2 has a weight of 1, endpoint 1 is selected 9 times out of 10, whereas endpoint 2 is selected only 1 time out of 10. {% endtab %}

{% tab title="Weighted random" %} Distributes incoming traffic across multiple backend servers based on the predefined weight assigned to each server. The weight represents relative capacity or processing power, where higher weights indicate greater ability to handle incoming requests. The algorithm generates a random number within a defined range based on the total sum of all server weights. This number is used to select one of the backend servers for processing the request.

Example: If three backend servers, A, B, and C, have weights of 1, 2, and 3, respectively, the total weight of all servers is 6. When a request arrives, the load-balancer generates a random number between 1 and 6. If the number is between 1 and 1 (inclusive), server A is selected. If the number is between 2 and 3, server B is selected. If the number is between 4 and 6, server C is selected. {% endtab %} {% endtabs %}

To configure load-balancing:

  1. Log in to your APIM Console

  2. Select APIs from the left nav

  3. Select your API

  4. From the inner left nav, select Endpoints under Backend services

    Endpoint configuration

  5. To confirm the load-balancing algorithm (chosen when your endpoint's group was created), click Edit group and select the General tab. Click the arrow to Go back to the endpoint configuration

    Edit endpoint group

  6. Click the pencil icon for your endpoint and select the General tab to edit the load-balancing weight

    Configure load-balancing weight

  7. Click Save

Failover

To configure failover:

  1. Log in to your APIM Console

  2. Select APIs from the left nav

  3. Select your API

  4. From the inner left nav, select Failover under Backend services

    Configure failover

  5. Configure the following:

    • Toggle Enable Failover ON
    • Max Attempts: Define the upper limit for the number of possible Gravitee API Gateway attempts to find a suitable endpoint, according to the load-balancing algorithm, before returning an error
    • Timeout: Defines the upper limit for time (in ms) between successive attempts before timing out

  6. Click Save

Health-checks

To configure health-checks:

  1. Log in to your APIM Console

  2. Select APIs from the left nav

  3. Select your API

  4. From the inner left nav, select Health-check under Backend services

    Configure health-checks

  5. Configure the following:

    • Toggle Enable health-check ON
    • Define the Trigger Schedule to establish the time interval between successive health-checks
    • Select the HTTP Method that will trigger the health-check
    • Define the Path that will trigger the health check
    • Toggle From root path ('/') ON to apply the path specified at the root URL level, e.g., for the endpoint URL www.test.com/api, this option removes /api before appending the path
    • Specify the HTTP Headers that will trigger a health check (supports Gravitee Expression Language)
    • Use Gravitee Expression Language to define an Assertion that specifies conditions to test for in the API response that will trigger a health-check, then click + Add assertion
    • Click Save, which also generates a visual summary of the health-check configuration