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
Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
If you are interested in working on this issue or have submitted a pull request, please leave a comment
Problem
I've recently rewritten our log aggregation pipeline to use Vector, which works very well, but as I was rolling it out I noticed that the vast majority of our logs weren't showing up in our Grafana Cloud Loki instance. After some further investigation it turns out that when sink.otel.protocol.batch.max_events > 1 it only registers the first log event. I created a reproduction for this: https://github.com/FredrikAugust/vector-otel-loss-repro. When turning off batching by setting protocol.batch.max_events = 1 it works perfectly. I suppose this is because to ingest several logs at the same time the log events have to be concatenated in the JSON structure (per protobuf: https://github.com/open-telemetry/opentelemetry-proto/blob/main/opentelemetry/proto/logs/v1/logs.proto#L38).
A note for the community
Problem
I've recently rewritten our log aggregation pipeline to use Vector, which works very well, but as I was rolling it out I noticed that the vast majority of our logs weren't showing up in our Grafana Cloud Loki instance. After some further investigation it turns out that when
sink.otel.protocol.batch.max_events > 1
it only registers the first log event. I created a reproduction for this: https://github.com/FredrikAugust/vector-otel-loss-repro. When turning off batching by settingprotocol.batch.max_events = 1
it works perfectly. I suppose this is because to ingest several logs at the same time the log events have to be concatenated in the JSON structure (per protobuf: https://github.com/open-telemetry/opentelemetry-proto/blob/main/opentelemetry/proto/logs/v1/logs.proto#L38).Configuration
Version
vector 0.44.0 (aarch64-apple-darwin 3cdc7c3 2025-01-13 21:26:04.735691656)
Debug Output
Example Data
https://raw.githubusercontent.com/FredrikAugust/vector-otel-loss-repro/refs/heads/master/logs.log
Additional Context
As mentioned; by disabling batching it works great. I could always see all the correct logs in the stdout sink.
References
#22188
The text was updated successfully, but these errors were encountered: