When a new release lands on your tenant, the real decision is not « what should I read? » but « who has to act this week, and on what? ». Dynatrace SaaS 1.350 began rolling out on 4 November 2026, according to the official release notes page (last updated 6 November 2026). It lists 32 feature updates, three breaking changes and 26 fixes.
This article starts from an on-call question: which dashboards, queries and alerts on your team might behave differently on Monday? It applies to Dynatrace SaaS; Managed follows another schedule and is not covered. The previous release is covered in the 1.349 article.
A note on honesty: span.observer has not tested 1.350 on a tenant. The facts come from the release notes; the checks below are suggestions, not measured results.
Dynatrace SaaS 1.350: three breaking changes, three owners
Give each breaking change a named owner and a date. The summary below restates what the page says.
The first two touch your queries. Billing Usage Events change their version format: event.version becomes a three-part number such as 1.0.0. Any DQL query that filters on the old value must drop that filter, or it will stop returning what you expect.
The security events change is the most distant but the most useful to prepare for. It is announced for SaaS 1.353, planned for December 2026. These events become configurable in OpenPipeline and move to a new default_security bucket with a new default retention period. Act if you query default_securityevents or default_securityevents_builtin, or need to keep data beyond 35 days.
Dashboards: stricter validation
The third breaking change shows up on screen. Invalid tiles.*.querySettings (query limits) and tiles.*.davis (analyse and alert) settings are removed. The affected dashboards are mainly those created or changed through the API or an external tool.
If you manage dashboards as code, with Terraform for example, review them before someone notices a tile without its limit. To recover a lost setting, the page points to Version history, then Edit JSON. For query examples inside tiles, see our SaaS dashboard DQL examples.
A second trap comes from the fixes: time series tiles with two metrics now assign the left and right axes by field order. An existing chart may therefore show swapped axes. It is not an outage, but a dashboard on the wall during an incident must stay readable.
How to check it works
Here are two checks you can do in ten minutes. I ran neither; adapt them to your environment.
- Billing: list the
event.versionvalues and find saved queries that still filter on them. - Security: search your dashboards, notebooks and workflows for the two old bucket names. Each hit is a December task.
- Dashboards: open the API-fed dashboards and compare query limits before and after.
- OTLP metrics: if you enabled Advanced OTLP Metrics Dimensions, note that
dt.-prefixed metrics sent over OTLP are now dropped, according to the fixes list.
What the release brings to operations
Among the 32 feature updates, a few change an on-call engineer’s work more than the others.
- Maintenance windows: they can suppress problem creation for their scope. Handy against migration noise, as long as the scope is narrow. Window start and end events are also ingested into Grail.
- Dashboards: base queries shared across tiles, named collapsible rows (collapsed rows skip their queries) and PDF export through Print view, in Chromium-based browsers only. See also DQL and Smartscape, since
K8S_SERVICEnodes gain anipattribute. - Workflows: actions to send an event to PagerDuty and an ITOM event to ServiceNow, with templates, plus create, read, update and delete actions for documents. Script limits and pitfalls: Workflow examples.
- Go: static monitoring no longer needs ptrace on Linux; it will replace the opt-in « Go improved static monitoring » feature.
- Permissions:
storage:smartscape:readcan be scoped bystorage:type. - OpenPipeline: forwarding to object storage covers more event types (security events, bizevents, Davis events and problems).
One change touches the API: the OneAgent on-a-host auto-update API returns 400 if the requested version is not higher than the installed one. A rollback script aiming at an older version will therefore fail.
Best practices
- Separate what breaks (three changes), what looks different (axes, tiles) and what is optional (the rest).
- Test the new Workflow actions on a low-stakes case before any PagerDuty or ServiceNow send in production.
- Re-read the release notes before each rollout wave: the page is updated after publication.
- Plan the move to
default_securitynow rather than the night before 1.353.
Frequently asked questions
When was Dynatrace SaaS 1.350 released?
The page states a rollout start on 4 November 2026. It gives no date for each feature or each tenant.
Is the security events change already active?
No, according to the page: it is announced for 1.353, planned for December 2026. Release 1.350 is the warning to prepare your queries and retention.
Are my API-created dashboards at risk?
Only if their query limit or Davis analysis settings are invalid. Those are then removed; Version history lets you recover them.
Conclusion
For 1.350, three actions are enough: drop version filters from billing queries, inventory the old security buckets before December, and review API-pushed dashboards. The rest can wait for a test. Sources: Dynatrace SaaS 1.350 release notes and the vendor’s blog post.
span.observer is an independent community blog, not affiliated with Dynatrace LLC. Dynatrace is a trademark of Dynatrace LLC. Versions and dates change: check them in the official documentation for your environment.
Originally published on span.observer. French version: https://span.observer/fr/dynatrace-saas-1-350-quoi-verifier/.


Top comments (0)