Replay and SLI Analyzer FAQ

Reading time: 0 minute(s) (0 words)

How long does Replay load historical data?

Normally, loading historical data in Replay does not exceed two hours.

Replay time for standard SLOs depends on the following:

Length of the replayed period. Longer retrieval periods result in more data to load.
Number of objectives in the SLO. Additional objectives increase complexity and processing demands.
Number of unique queries in the SLO. Each query requires additional resources and processing.

Replay also relies on integration between Nobl9 and your data source, the following external factors can also have an impact:

Data source performance and current load. High load on the data source may reduce response speeds.
API rate limits. Throttling by data sources can impact query execution time.
Retrieved data density. High-frequency data or large datasets take longer to load.
Data compression or downsampling. The format of retrieved data can affect loading time.

If your Replay process takes longer than two hours, and none of the listed factors is contributing, contact Nobl9 support.

How does Replay handle Datadog API rate limits?

If you exceed Datadog's API rate limit during Replay, the system will attempt to retrieve data for up to 20 minutes. If all attempts fail, the Replay will be unsuccessful, and the SLO will not be replayed.

For SLOs being created with Replay, if the process fails, the SLO will still be created, but without any historical data.