Ingesting BizEvents data after connectivity issues

[michiel_otten]

Hi community,

I've got a question about using the OpenPipeline whilst ingesting either events, metrics or BizEvents.
When using the API for BizEvents, the timestamp can be a maximum of 24 hours in the past. 
This causes a risk from a data consistency level. If communication is not possible from our ingest source towards Dynatrace, the start times can't be accurate anymore. This causes data to be in Dynatrace which is not representative nor usable for the business analysis.

Has anyone find a solution for this? Can we change metrics and timestamps after they've been ingested or add data in the platform that is older than (f.e.) the 24 hours that is referenced for BizEvents?

Many thanks in advance,

Michiel

[sujit_k_singh]

From what I’ve seen, the 24‑hour window on BizEvents in OpenPipeline is intentional to keep analytics reliable and near real‑time, and once data lands in Dynatrace the timestamps aren’t editable or backdatable. When connectivity hiccups happen, the safest route is to buffer/retry upstream so you can replay within that window; if you still need the original occurrence time for business reporting, include it as a separate attribute  so analysts can reference it while the platform maintains a clean ingestion timeline. In situations where strict historical accuracy is a must (compliance/audit), I’d store the raw feed in a secondary system and let Dynatrace focus on operational insights—this keeps dashboards and anomaly detection consistent while you preserve a definitive record elsewhere.

[michiel_otten]

Wouldnt this make the bizevents a bit unusable for now? Seems to me that these events need to be very accurate. Also the Business Flow app uses the timestamp for calculations. Do we need an external system populating and caching? What is a smart way of doing this, you think? 

[sujit_k_singh]

You’re absolutely right that BizEvents need accurate timestamps for Business Flow calculations, and the 24‑hour limit can feel restrictive. At the moment, Dynatrace enforces this to keep analytics near real-time and consistent, so backdating isn’t supported.

A smart way is to use a queue or caching layer to buffer events and replay them within the 24-hour window. Include the original timestamp as a custom attribute for reporting, and optionally archive the raw feed in a secondary system for compliance or deep historical analysis. I have created one diagram based on my analysis.