Skip to main content
logoTetrate Service BridgeVersion: 1.9.x

Organization Setting

Organization Setting allows configuring global settings for the organization. Settings such as network reachability or regional failover that apply globally to the organization are configured in the Organizations Setting object.

This is a global object that uniquely configures the organization, and there can be only one Organization Setting object defined for each organization.

The following example shows how these settings can be used to describe the organization's network reachability settings and some regional failover configurations.

kind: OrganizationSetting
name: org-settings
organization: myorg
vpc01: vpc02,vpc03
- from: us-east1
to: us-central1


Settings that apply globally to the entire organization.

FieldDescriptionValidation Rule


Reachability between clusters on various networks.


List of tetrateio.api.tsb.types.v2.RegionalFailover
Default locality routing settings for all gateways. Please use FailoverSettings instead. If FailoverSettings is set, it takes precedence over this field.

Explicitly specify the region traffic will land on when endpoints in local region becomes unhealthy. Should be used together with OutlierDetection to detect unhealthy endpoints. Note: if no OutlierDetection specified, this will not take effect.

Security settings for all proxy workloads in this organization. This can be overridden at TenantSettings, WorkspaceSettings, or security group's SecuritySetting for specific cases. The override strategy used will be driven by the SecuritySetting propagation strategy. The default propagation strategy is REPLACE, in which a lower level SecuritySetting in the configuration hierarchy replaces a higher level SecuritySetting defined in the configuration hierarchy. For instance, a WorkspaceSettings defined SecuritySetting will replace any tenant or organization defined SecuritySetting. Proxy workloads without a specific security group will inherit these settings. If omitted, the following semantics apply:

  1. Sidecars will accept connections from clients using Istio Mutual TLS as well as legacy clients using plaintext (i.e. any traffic not using Istio Mutual TLS authentication), i.e. authentication mode defaults to OPTIONAL.

  2. No authorization will be performed, i.e., authorization mode defaults to DISABLED.


Traffic settings for all proxy workloads in this organization. This can be overridden at TenantSettings or WorkspaceSettings for specific cases. Proxy workloads without a specific traffic group will inherit these settings. If omitted, the following semantics apply:

  1. Sidecars will be able to reach any service in the cluster, i.e. reachability mode defaults to CLUSTER.

  2. Traffic to unknown destinations will be directly routed from the sidecar to the destination.


Failover settings for all proxies connecting to a host exposed in this organization. Note that this is a server side setting.


Network related settings for clusters.

FieldDescriptionValidation Rule


map<string, string>
Reachability between clusters on various networks. Each cluster has a "network" field representing a network boundary like a VPC on AWS/GCP/Azure. All clusters within the same network are assumed to be reachable to each other for multi-cluster routing. In addition, you can specify additional connectivity between various networks in the mesh here. For example on AWS, each VPC can be treated as a distinct network. VPCs that are reachable to one another (through peering or transit gateways) can be listed as reachable networks. The key is the network name and the value is a comma separated list of networks whose clusters are reachable from this network. For instance, vpc01: vpc02,vpc03 means that the clusters in the network can reach those in vpc02 and vpc03.

Note that reachability is not bidirectional. That is, if vpc01: vpc02 is specified, then vpc01 can reach vpc02, but not the other way around. Hence, the workloads in clusters in vpc01 can access the services through the exposed gateway hostnames in clusters in vpc02 . However, the workloads in clusters in vpc02 cannot access the services exposed through the gateway hostnames in vpc01.