Hermetry 2.0 is out today. The release notes cover devices, telemetry, dashboards, SCADA screens, alarms and flows. I work on security and on where your data lives, so this post is about those two lines of the 2.0.0 changelog.
Organisations are kept apart in the database itself. One organisation never sees another’s devices, readings or settings. Members, roles and groups are per organisation as well. The same person can be an Owner in one organisation and a User in another.
Hermetry runs as separate deployments in the EU, the US and Türkiye. You pick the home region when the organisation is created, and its devices, telemetry, dashboards, flows, members’ work and audit log are stored there. The global directory that routes requests knows which region and node an organisation lives on, and holds no name, address, owner or setting of it.
Choose the region with some care. Moving an organisation to another region later is not something you can do yourself; it goes through support.
Also in 2.0 on the access side: roles and attribute-based access policies, an audit log, two-factor sign-in, passkeys and single sign-on.
Docs: Regions and data residency, How Hermetry protects your data, Create your account and choose a region.
That is the answer I wanted for our vendor questionnaire. Two follow-ups from the same form: where do backups go, and can your own operators get into a node?
Both are on the Residency page of the organisation settings (/admin/settings/residency, owners and admins).
Backups: it shows the class of place where backups are kept. Until that is reported for your node it says not attested, rather than guessing.
Operator access: there is an emergency (“break-glass”) procedure. A grant is time-limited and needs two signatures, every action taken under it is marked in the audit log of the organisation it touches, and the grants of the last 365 days are listed on that same page. The page also lists the region’s sub-processors and its region facts (residency class, legal area, jurisdiction, legal frameworks). The region facts are signed by our control plane, so the page can’t show facts that weren’t issued for your node.
We have a plant in Europe and a warehouse in the US. One organisation or two? Ideally I’d like one login for my team.
An organisation has one home region, so all of its data sits in one place. If the warehouse data has to stay in the US and the plant data in the EU, that’s two organisations. Each organisation has its own members and roles, which is often what you want anyway. Each region is also a separate deployment, so ask support how sign-in works across the two before you plan around one login.
Do the devices also connect per region, or is there one global broker hostname?
Per region. Devices connect to the device host of the organisation’s region. Use the host on the device’s Connection details card rather than typing one from memory.
- You must be logged in to reply to this topic.
Join the conversation
Sign in to ask and answer, like helpful posts and mark the reply that solved your problem. Every contribution earns points and levels.