You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Deploy telegraf into AWS EKS as a POD with a config (specified)
Send only 1 request to telegraf with the metrics in JSON (input = http_listener_v2)
Get duplicated records in the output source TimescaleDB (PostgreSQL)
See that the Buffer is not being flushed (emptied) at all
Expected behavior
Metrics are sent to input and processed to output without duplication.
Actual behavior
The Telegraf is deployed into Kubernetes (AWS EKS)
We tried to use buffer_strategy = "disk" and started receiving tons of duplicated records into TimescaleDB (PostgreSQL)
It might be a bug or our config issue, though we've tried changing different properties at [agent] level.
So far, looks like we will stick with memory strategy until this is figured out.
Additional info
No response
The text was updated successfully, but these errors were encountered:
Thanks for trying out the feature! As far as I can tell, this doesn't seem like a configuration issue on your end. I'm looking into it now and will get back to you once I've found a solution
Relevant telegraf.conf
Logs from Telegraf
System info
1.32-alpine from dockerHub (https://hub.docker.com/_/telegraf)
Docker
No response
Steps to reproduce
Expected behavior
Metrics are sent to input and processed to output without duplication.
Actual behavior
The Telegraf is deployed into Kubernetes (AWS EKS)
We tried to use
buffer_strategy = "disk"
and started receiving tons of duplicated records into TimescaleDB (PostgreSQL)It might be a bug or our config issue, though we've tried changing different properties at [agent] level.
So far, looks like we will stick with
memory
strategy until this is figured out.Additional info
No response
The text was updated successfully, but these errors were encountered: