Is it possible to rerun a pipeline on already ingested logs? #5906
-
I've tried to create a pipeline which has:
But I still see old version of the attribute (tagged as String) whereas my new Grok pattern parses them as Float, and the simulation correctly shows the attribute parsed as a float. I'd like to avoid reingesting all my historical logs, so is there a way to "kick" a pipeline such that it deletes existing computed data and recreates it from the source without reingesting? If not, what's the best approach to re-ingest with |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
No, this is not possible today.
It doesn't deduplicate. One option is to identify those logs, delete them and re-ingest using some script. Please create a feature request here https://github.com/SigNoz/signoz/issues |
Beta Was this translation helpful? Give feedback.
No, this is not possible today.
It doesn't deduplicate. One option is to identify those logs, delete them and re-ingest using some script.
Please create a feature request here https://github.com/SigNoz/signoz/issues