Overview
Datadog can collect and display errors, response time, duration, latency, as well as monitor the performance of API traffic in Tyk to discover issues in your APIs or consumers.
Tyk has a built-in Datadog integration that collects metrics from Tyk API gateway.
Tyk API gateway records all the traffic that it’s processing. It sends that information to Datadog and builds dashboards around it.
How it works
Tyk pump writes custom application metrics and sends them into Datadog by sending them to DogStatsD, a metrics aggregation service bundled with the Datadog Agent. DogStatsD implements the StatsD protocol which adds a few Datadog-specific extensions including the Histogram metric type, that is in use by Tyk-gateway
.
Tyk-gateway
uses Tyk-pump
to send the analytics it generated to Datadog.
When running the Datadog Agent, DogstatsD gets the request_time
metric from Tyk-pump
in real time, per request, so you can understand the usage of your APIs and get the flexibility of aggregating by various parameters such as date, version, returned code, method etc.
The custom metric Tyk is using is of type DD_HISTOGRAM_AGGREGATES.
Setup
Tyk’s integration is included in the tyk-pump
package, so you only need to set configuration in the pump.conf
(and there’s no need to install anything on your Tyk platform).
Installation
Installation
For this integration you need to have a running Tyk installation. You can install Tyk self managed or Tyk OSS. Both options include the tyk-pump
.
Install Datadog Agent
Install the Datadog Agent in your environment.
Run the Datadog Agent in your K8s cluster, as a Docker container, on your Mac, or any way as long as Tyk pump
is able to access it.
For containerized environments, see the Autodiscovery Integration Templates for more guidance. To validate that the changes are applied, run the Agent’s status subcommands
Configuration
Tyk-pump
To set a Datadog pump follow the instructions in the DogstatsD section of the pump README.
The following is an example of Datadog pump configuration in pump.conf
:
pump.conf:
...
"dogstatsd": {
"type": "dogstatsd",
"meta": {
"address": "dd-agent:8126",
"namespace": "tyk",
"async_uds": true,
"async_uds_write_timeout_seconds": 2,
"buffered": true,
"buffered_max_messages": 32,
"sample_rate": 0.9999999999,
"tags": [
"method",
"response_code",
"api_version",
"api_name",
"api_id",
"org_id",
"tracked",
"path",
"oauth_id"
]
}
},
This example was taken from Tyk-demo project, an open source project that spins up a full Tyk platform in one command and offers ready-made examples, including the Datadog example. To run this integration, use up.sh analytics-datadog
.
Setup Datadog Agent
Tyk’s integration uses DogstatsD. It is a metrics aggregation service bundled with the Datadog Agent. DogStatsD implements the StatsD
protocol and adds a few Datadog-specific extensions. Tyk is using Histogram metric type
.
Set up the following Datadog and DogStatsD environment variables in your environment:
DD Environment variable | Value | Description |
---|
DD_API_KEY | {your-datadog-api-key} | For the Datadog Agent to connect the DD portal. Your API key can be found in Account Settings. |
DD_ENV | tyk-demo-env | Sets the environment name. |
DD_DOGSTATSD_TAGS | “env:tyk-demo” | Additional tags to append to all metrics, events, and service checks received by this DogStatsD server. |
DD_LOGS_ENABLED | true | Enables log collection for the Datadog Agent. |
DD_LOGS_CONFIG_CONTAINER_COLLECT_ALL | true | Collects logs from containers. |
DD_DOGSTATSD_SOCKET | /var/run/docker.sock | Path to the Unix socket to listen to. Docker compose mounts this path. |
DD_DOGSTATSD_ORIGIN_DETECTION | true | Enables container detection and tagging for Unix socket metrics. |
DD_DOGSTATSD_NON_LOCAL_TRAFFIC | true | Listens for DogStatsD packets from other containers. (Required to send custom metrics). |
DD_AGENT_HOST | dd-agent | Name of the agent host in Docker. |
DD_AC_EXCLUDE | redis | Excludes Datadog redis checks. (Optional) |
DD_CONTAINER_EXCLUDE | true | Excludes docker checks for the Datadog Agent. |
After setting environment variables listed above, set up the agent with DogstatsD.
Restart the Agent after setup.
Validation
Create a dashboard or import the sample and add a widget. In the section Graph your data under the metric option, start typing the namespace you chose for the pump in the config pump.conf
under dogstatsd.namespace
.
In the example above, it’s tyk
. Once you start typing, all the available metrics are displayed.
Data Collected
Metrics
tyk.request_time.95percentile (gauge) | the 95th percentile of request time values in the time interval Shown as millisecond |
tyk.request_time.count (rate) | Represents the number of values submitted during the interval. Represents the number of values submitted during the interval, X. The Agent submits this number as a RATE so it would show in app the value of X/interval Shown as millisecond |
tyk.request_time.avg (gauge) | Represents the average of request time values in the time interval Shown as millisecond |
tyk.request_time.max (gauge) | Represents the maximum value of request time values sent during the time interval Shown as millisecond |
tyk.request_time.median (gauge) | Represents the median of request time values in the time interval Shown as millisecond |
Dashboards
With Datadog, you can create dashboards that display statistics about your API services and their consumption.
Here’s an example for such a dashboard:
Note: You can import the above dashboard and use it as an example or baseline for your own dashboard.
Events
The Tyk integration does not include any events.
Service Checks
The Tyk integration does not include any service checks.
Troubleshooting
Need help? Contact Datadog support.