Module 6 — Dynatrace management zones, tags and access in Managed: automatic tagging and permissions

Dynatrace management zones, tags and access in Managed: set a tagging convention, build zones and limit each team’s permissions to its own scope.
Module 6 — Tags, zones, accès en Managed : tagging automatique, management zones et permissions

Module 6 of the Dynatrace Managed course covers Dynatrace management zones, tags and access, so that each team in a shared environment works within its own scope.

Course: Dynatrace Managed · Module 6/8 · Duration: ~60 min · Level: Intermediate

Objectives:

  • Define a simple tagging convention and apply it at the source.
  • Create automatic tagging rules based on entity properties.
  • Build management zones that split the environment by team or by application.
  • Link user groups, permissions and management zones to give everyone exactly the access they need.

Prerequisites: modules 1 to 5, a test Managed environment with several hosts and services, administrator access to the environment and, for the access part, to the Cluster Management Console (CMC).

An environment that monitors three servers can be read at a glance. An environment that monitors three thousand, shared between twenty teams, becomes unreadable without organisation. Module 6 of the Dynatrace Managed course covers tags, Dynatrace management zones and access in Managed: how to tag entities automatically, how to split the environment into coherent scopes, and how to use those scopes to manage who sees and who changes what.

These three topics are linked: tags feed management zones, which serve both as a filter for the alerting profiles seen in module 5 and as a scope for permissions. The course overview puts this module in context.

Why set up Dynatrace management zones, tags and access from the start?

Adding organisation after the fact is expensive: you have to rename things, track down owners, and rebuild dashboards and alerting profiles. Conversely, a convention set early makes every new application “self-filing” as soon as OneAgent is installed. The table below summarises the division of roles:

Building block Nature Question it answers
Tag Label (key or key:value) applied to an entity “What is it, and who does it belong to?”
Management zone Scope defined by rules “Which slice of the environment are we looking at?”
Group and permissions Rights granted to users “Who is allowed to see or change this slice?”

Defining a tagging convention

Stick to a few stable keys that everyone knows. A common foundation:

env   : prod | preprod | test
app   : payment | catalogue | crm ...
team  : name of the owning team
tier  : front | back | data

Avoid free-form values typed by hand in the interface: they quickly diverge (“Payment”, “payments”, “pay”). Prefer values from a controlled source (CMDB, deployment variables, Kubernetes labels).

Setting metadata at the source

Several mechanisms carry the information through to Dynatrace without manual intervention:

  • Process environment variables, such as DT_TAGS, read by OneAgent when the process starts.
  • Host tags and properties defined when OneAgent is installed or afterwards with the oneagentctl command-line tool.
  • Kubernetes labels and annotations, usable in rules when the cluster is monitored through Dynatrace Operator.
  • Cloud metadata (AWS, Azure tags, etc.) when the corresponding integration is enabled.
DT_TAGS="app=payment tier=back"

Creating automatic tagging rules

An automatic tagging rule applies a tag to every entity that meets a condition: process group name, host property, host group, Kubernetes label, detected technology… It can also take a value dynamically (for example the value of a property) to build the tag. Two good habits:

  • Propagate the tag to related entities when the option exists (from the host to its processes, from the process to its services), so that the whole stack carries the same label.
  • Check coverage: regularly filter the entities that have no team tag to spot orphans.

Building management zones

A management zone is also defined by rules, often based on tags: “all entities carrying app:payment and env:prod“. It is then used everywhere: global interface filter, dashboards, alerting profiles and, above all, permissions. Name them predictably, for example {env}-{app} such as prod-payment.

Watch out for two pitfalls: a zone that is too broad (and ends up containing everything) adds nothing, and a proliferation of very narrow zones becomes impossible to maintain. Start from the real teams and applications. The article Management zones and automatic tagging works through complete rule examples.

Managing access: groups, permissions and zones

In Managed, users and groups are managed at cluster level, in the CMC, either as local accounts or synchronised with an LDAP directory or a SAML provider. Each group receives permissions per environment, and some of these permissions can be restricted to one or more management zones. The principle to apply is least privilege:

Group (example) Scope Typical rights
dt-cluster-admins Cluster CMC administration, reserved for a few people
dt-prod-admins Prod environment Read and change settings
dt-prod-payment Zone prod-payment Read the zone’s data
dt-prod-viewers Prod environment Read-only, without sensitive data

The exact permission names vary between versions: refer to the documentation and to the group management screen of your CMC. For the directory, see Connecting Dynatrace Managed to LDAP / Active Directory and SAML SSO.

Hands-on exercise

Step 1 — Set a tag at the source

On a test host, add the variable DT_TAGS="app=training tier=back" to an application process (systemd unit file or start-up script), then restart that process.

Expected result: the corresponding process group shows the tags app:training and tier:back after a few minutes.

Step 2 — Create an automatic tagging rule

Create a rule that applies the tag env:test to all hosts in a given host group, propagating to processes and services.

Expected result: when you filter services on env:test, you find the ones running on those hosts.

Step 3 — Build a management zone

Create the zone test-training with a rule combining env:test and app:training.

Expected result: when you select this zone in the global filter, only the relevant entities appear.

Step 4 — Restrict a group to the zone

In the CMC, create a group dt-test-training with read access limited to the test-training zone, then add a test user to it.

Expected result: signed in with this account, the user only sees the zone’s entities and cannot change the settings.

Step 5 — Connect the zone to alerting

Edit the alerting profile created in module 5 so that it filters on the test-training zone.

Expected result: only problems in this zone trigger a notification.

Quiz

  1. What is the difference between a tag and a management zone?
    A. None, they are two names for the same thing · B. A tag labels an entity, a zone defines a scope through rules · C. Zones are reserved for hosts
  2. Why prefer DT_TAGS or an automatic rule to a tag typed by hand?
    A. So that values stay consistent and follow new entities · B. Because manual tags cost extra · C. Because manual tags expire
  3. Where are user groups managed in Managed?
    A. In each OneAgent · B. In the Cluster Management Console, optionally synchronised with LDAP or SAML · C. Only through a configuration file on the nodes
  4. What principle guides the assignment of permissions?
    A. Give everyone administration rights to save time · B. Least privilege, limited to a management zone where possible · C. A single group for everyone

Answers

  1. B. The tag describes the entity; the zone groups entities according to rules, often based on tags.
  2. A. An automated source avoids spelling variants and also covers entities created later.
  3. B. Users and groups are managed at cluster level, in the CMC.
  4. B. Everyone receives only the rights they need, on the scope that concerns them.

Further reading

← Module 5: Davis, problems and alerting · Course overview · Module 7: Operating the cluster →

Frequently asked questions

How many tag keys should I plan for?

Four or five stable keys are enough in most organisations. Beyond that, the convention becomes hard to follow and the rules multiply.

Can an entity belong to several management zones?

Yes. A shared service can appear in the zone of every application that uses it, which lets several teams see it.

Should I create one environment per team rather than zones?

Not necessarily. Environments strongly separate data and licences; zones split a shared environment. Separate environments are generally kept for strong isolation needs, such as prod and non-prod.

With a tagging convention, clear zones and well-fitted rights, your environment can welcome new teams without losing readability. The next module changes perspective: it is aimed at those who operate the cluster itself.

span.observer is an independent community blog, not affiliated with Dynatrace LLC. This course does not replace the official Dynatrace University training and certifications. Check features in the official documentation for your version.

Total
0
Shares
1 comment
Leave a Reply

Your email address will not be published. Required fields are marked *

Previous Post
Module 5 — Davis : alerting en Managed, anomalies, problèmes et profils d'alerte

Module 5 — Dynatrace Davis alerting in Managed: anomalies, problems and alerting profiles

Next Post
Module 7 — Exploitation : cluster Managed, santé, sauvegardes, mises à jour et stockage

Module 7 — Dynatrace Managed cluster operations: health, backups, updates and storage

Related Posts