 Nils Andreas Svee
|
e7cbdbb265
|
docs(observability): document rationale for unbatched UDP metric flushing
Explain that individual metric delivery prevents data truncation caused by small UDP receive buffers on the remote host.
|
2026-04-13 17:36:30 +07:00 |
|
|
Nils Andreas Svee
|
699c8782d9
|
fix(observability): send individual metrics over UDP to avoid packet size limits
Differentiate flushing logic between transport types to prevent UDP payload truncation while maintaining batched delivery for TCP.
|
2026-04-13 17:21:14 +07:00 |
|
|
Nils Andreas Svee
|
0f902e27a3
|
perf(observability): optimize metric flushing to reduce memory allocations
Use a pre-allocated buffer for metric serialization to improve efficiency when processing large batches.
|
2026-04-13 16:54:33 +07:00 |
|
|
Nils Andreas Svee
|
72b4bc8dab
|
fix(observability): ensure all metric outputs are newline-terminated
Append trailing newlines to metric string formats to ensure consistent parsing and record separation by monitoring backends.
|
2026-04-13 15:43:09 +07:00 |
|
|
Nils Andreas Svee
|
757a715a54
|
chore: cleanup unused imports
|
2026-04-13 15:29:04 +07:00 |
|
|
Nils Andreas Svee
|
671e3f2671
|
refactor: standardize channel naming with tx and rx suffixes
Rename internal channel senders and receivers to follow standard Rust conventions for improved code consistency and readability.
|
2026-04-13 15:07:00 +07:00 |
|
|
Nils Andreas Svee
|
44f551d38c
|
feat(observability): Add InfluxDB Line Protocol metrics module and config
|
2026-04-13 15:01:45 +07:00 |
|
|
Nils Andreas Svee
|
853f807e2c
|
feat: add metrics module for monitoring
Initialize the Metrics structure to receive and process ping statistics as part of the monitoring subsystem.
|
2026-04-13 12:11:05 +07:00 |
|