This project currently forwards decoded GELF events to Seq via seqcli ingest. This was a good way to bootstrap the project, but the .NET dependency introduces unstable runtime behavior. It's also a big additional moving part to manage.
Now that Seq supports native OTLP ingestion, we can remove the seqcli dependency and replace it with a Rust-based OTLP emitter. The quickest way to bootstrap this would probably be via tonic, but we could also roll our own gRPC if we need to for custom header support.
nblumhardt
commented
1 week ago
This project currently forwards decoded GELF events to Seq via
seqcli ingest. This was a good way to bootstrap the project, but the .NET dependency introduces unstable runtime behavior. It's also a big additional moving part to manage.Now that Seq supports native OTLP ingestion, we can remove the
seqclidependency and replace it with a Rust-based OTLP emitter. The quickest way to bootstrap this would probably be viatonic, but we could also roll our own gRPC if we need to for custom header support.