Relay agents
This feature is available with an Enterprise plan
Relay agents allow you to securely connect to data sources inside your own network. For information on downloading, deploying and upgrading a Relay Agent see the following, depending on your platform:
How do Relay Agents work?
An agent is a service that runs on your host inside your own network that has access to the on-prem data source. Ideally, you'd install it on a host that has the fastest connection to the on-prem data source, based on its location.
Agents are organized into agent groups, which allows for benefits such as redundancy through not having a single point of failure. The agent listens queue requests from SquaredUp and picks up tasks, for example connecting to local APIs that SquaredUp has asked for.
What is an agent group?
Agent groups dynamically determine which agents are used to retrieve data. When you're configuring an on-prem data source, you select the agent group the data source will use.
Tips for using agents and agent groups
Agents need to be deployed in the data center that has access to the on-prem data source.
One agent with one group can be used to connect to multiple on-prem data sources. This is the most common setup scenario because it is the most simple to maintain. If you are running multiple agents on the same machine, consider consolidating those agents into one if they can run under the same service authentication account.
Agents and agent groups are versatile in that they can scale to your needs to support high-availability and network segregation requirements. In this case, you may need more than one agent, agent group or a combination. Be sure to understand the implication of this architecture before undertaking its setup.
Check out this SQUPCAST episode for more details on best practices with agents and agent groups.
- Access to data:
Agents need to be deployed in the data center that has access to the on-prem data source will use. If you have multiple groups, remember to set the data source to use the correct group, i.e. the group that contains agents with access to the on-prem data source. Tip: Agents can be part of multiple groups. - Multiple agents
For high-availability and load balancing, you can deploy multiple agents on the same network and put them all in the same agent group. This way, there is always an agent to pick up the task in the queue, and the workload can be divided between all agents.
If you want to use just one agent you can do that, but you still need to put it into a group and set your data source to use that group. - Structuring your agent groups:
You can organize your agents into groups that make sense to you. For example, if you have a worldwide network of data centers, you might create agent groups based on the locations of the machines. You can then pick the group closest to the on-prem service and therefore get the fastest connection.
If you have multiple groups, remember to set the data source to use the correct group, i.e. the group that contains agents with access to the on-prem data source. Tip: Agents can be part of multiple groups.