Skip to main content

Tenant Settings

Admin → Tenant Settings

Tenant-wide configuration: who may join, how NudgeBee reads your telemetry labels, and which features are switched on. Everything here applies to every account in the tenant.

Editing requires tenants:Write. Without it the page is readable and the controls are disabled, with a banner naming the missing permission.

General​

SettingWhat it controls
Tenant NameThe display name for the tenant.
Allow self-onboarding via domain loginWhether someone signing in with a matching email domain gets an account automatically, instead of needing an invite.
Allowed DomainsThe email domains self-onboarding accepts. Only meaningful when self-onboarding is on.
Default Auth RoleThe role a self-onboarded user receives.
Self-onboarding is a membership decision

Turning it on means anyone who can obtain an email address at an allowed domain can enter the tenant at the default role. Set the default role to a read-only one unless you have a reason not to.

Label Mapping​

NudgeBee has to know which of your labels mean what. Three sub-tabs, one per signal type:

Logs​

Map your logs backend's label keys onto product concepts, plus a Cluster Label telling NudgeBee which key carries the cluster name.

Traces​

The same mapping for your tracing backend.

Webhook alerts​

For alerts arriving by webhook, name the labels that carry:

  • Subject Name Labels — what the alert is about (the workload, host or resource).
  • Namespace Labels — where it lives.
  • Severity Labels — how bad it is.

Each accepts several keys, because different alert sources name the same thing differently. NudgeBee takes the first one present.

Getting these wrong is the usual reason a webhook alert arrives with no subject, lands in the wrong account, or shows an unexpected severity. If alerts are arriving but look empty, this is the first place to check.

Features​

Per-tenant feature flags, grouped by category with a count per group and an All view. Toggling one here switches the feature on or off for the whole tenant.

Some product surfaces are gated by a flag here rather than by a permission — the LLM Analyser tab is one — so a feature that seems missing for everyone is worth checking on this tab before anything else.

You can reveal the underlying flag ids, which is what you want when comparing a tenant's configuration against a support request or another environment.