September 2026 Domotz Release Notes

3 MIN

Release Notes

HOME PAGE

3 min

This page will be updated every week during this month.

Items marked as Beta are part of an early access program. Contact us to learn more or request access.

Items marked as Limited are available to a subset of customers. Contact support to learn more.

Items marked as Coming Soon are in development and not yet available.

September 2, 2026

New Features & Improvements

MCP

Power on and off — Individual PDU outlets and PoE ports can now be switched off and back on via MCP, not just power-cycled, so a device can be deliberately powered down and brought back up later.
Collector SNMP defaults — A Collector’s default SNMP credentials and polling period can now be read and changed via MCP, so a site can be prepared for SNMP monitoring without opening the app.

Public API

Alert identifier in webhook payloads — Alert webhook payloads now include a unique identifier for the alert, so a receiving system can correlate deliveries with the alert that produced them and recognize repeat notifications.
Collector SNMP defaults via the Public API — The same default SNMP credentials and polling period can now be configured programmatically through the Public API, for provisioning a fleet of Collectors from an external system.

Coming Soon

The following are in development and not yet available.

Collaboration for RBAC — You’ll be able to invite someone from a different Domotz account to collaborate on your own — even if they already have their own Domotz account — and grant them exactly the permissions and Collectors you choose.

Bug Fixes

Alerts

Fixed: An alert rule on a driver or SNMP variable now triggers no matter where it was applied from — previously, rules bound through a device profile or the device Alerts tab weren’t firing, even though the same rule applied directly from the variable table was.
Fixed: Applying an alert rule from a filtered device table now binds it only to the devices the filter actually shows, instead of spilling over onto other devices — and rules that had already been applied no longer disappear from the list intermittently.

RBAC & Organizations

Fixed: An API key created for MCP by a user with limited permissions is now scoped to that user’s own permissions, instead of inheriting the full permissions of the account owner.
Fixed: The built-in Team Member role no longer has access to the audit log by default — only Administrators and the Account Owner get it by default, though the permission can be extended to other roles via RBAC.
Fixed: A Field Operator can no longer open, edit, or apply device profiles, and permission checks on the alerts contained in a profile have been corrected.
Fixed: Loading the monitoring table no longer fails with a permission error for users whose role does grant access to it.
Fixed: Creating a user with the email address of a previously deleted one now works — a stale cached entry was still being served.

Device Management

Fixed: Device profiles that had been created are listed again — they had stopped appearing, and a new profile couldn’t be created reusing one of those names.
Fixed: Restored the tabs that were missing from the Collector Dashboard.
Fixed: The selection checkboxes for bulk device actions on the Inventory dashboard now work correctly.
Fixed: Turning off Advanced TCP Scan on a routed network now actually stops TCP port checks against its devices, instead of continuing to probe ports 80, 443, and 3389.
Fixed: IP cameras from a number of manufacturers are now recognized as ONVIF devices during analysis, so the ONVIF driver is applied and their details and streams are collected.
Fixed: A Windows Collector no longer reports an offline period far longer than the update itself when its scheduled update task restarts the service.
Fixed: A crash of the Windows Collector under heavy external-host and discovery-check load, which previously took the whole Collector down.
Fixed: SNMP templates included in a device profile are now applied when the profile is applied to an External Host device.

Monitoring

Fixed: An External Host monitored by ping only now goes down when it stops responding, instead of staying up indefinitely once its first check succeeded.
Fixed: Custom driver inspections no longer fail in bulk on large Collectors — a watchdog was mistakenly terminating busy (not stuck) driver sandboxes, failing every inspection running in them.