What is Tyk On-Premises?
Tyk On-Premises is the way to install our entire Tyk Pro solution in your own infrastructure, it enables you to have full control over every element of the Tyk stack as well as no external dependency on our cloud solution or infrastructure.
The full Tyk On-Premises Pro system consists of:
- Tyk API Gateway: The API Gateway that proxies and manages your traffic.
- Tyk Dashboard: The management Dashboard and integration API for managing a cluster of Tyk Gateways, also shows analytics and features the Developer portal.
- Tyk Pump: Handles moving analytics data between your gateways and your Dashboard (amongst other data sinks).
- Tyk Identity Broker (Optional): Handles integrations with third-party IDP’s.
- Tyk Multi-Data-Center Bridge (Optional, Enterprise-only): Allows for the configuration of a Tyk ecosystem that spans many data centers and clouds.
For a production environment, we recommend that the Gateway, Dashboard and Pump are installed on separate machines. If installing multiple Gateways, you should install each on a separate machine. For more information on deploying to a production environment, see Planning for Production.
Installing Tyk On-Premises:
Install Tyk On-Premises with Docker.
Install Tyk On-Premises with Kubernetes.
Install Tyk On-Premises with Ansible.
Install Tyk On-Premises on Red Hat.
Install Tyk On-Premises on Ubuntu.
Install Tyk On-Premises on Amazon AWS.
Install Tyk On-Premises on Heroku.
Install Tyk On-Premises on Microsoft Azure.
We distribute Tyk via Packagecloud.io APT and Yum repositories, as well as via our Github repository for the Tarballs.
For licence queries please contact your account manager. For free trial licences, please visit our get started page on the web site.
By default the Tyk Gateway uses MongoDB. You can also use the following:
The Tyk Dashboard and Portal use Redis.
Supported MongoDB and Redis versions
Tyk has been tested on the following versions:
- MongoDB 3.x and 4.0.x
- Redis 2.8.x to 5.0.x
- CentOS 6, RHEL 6, Amazon Linux ship with Upstart 0.6.x
- Ubuntu 14.04, Debian Jessie with Upstart 1.x
- CentOS 7, RHEL 7, Ubuntu 16.04, Debian Stretch are running with systemd
- Certain older distros may only provide SysVinit but all of them typically provide compatibility with its scripts
Note that any init scripts of your choosing can be used instead of automatically detected ones by copying them from the
install/inits directory inside the package directory.
This init system variance implies there are different ways to manage the services and collect service logs.
For Upstart, service management can be performed through the
initctl or a set of
status commands. Upstart 1.x also works with the
For systemd, either
service commands may be utilised.
service command can usually be used with SysVinit scripts, as well as invoking them directly.
Service logs availability
- Upstart 0.6.x and SysVinit: log files are located in
/var/logsfor every respective service, e.g.
- Upstart 1.x: by default everything is stored in
- systemd utilises its own logging mechanism called journald, which is usable via the
journalctl -u tyk-gateway
Please consult with respective init system documentation for more details on how to use and configure it.