Files
coder/agent/boundarylogproxy/codec/boundary.proto
Zach 5b7377c375 feat: add Prometheus metrics for boundary log drop reporting (#22521)
Add Prometheus metrics to the boundary log proxy for observability:
- batches_dropped_total (reason: buffer_full, forward_failed)
- logs_dropped_total (reason: buffer_full, forward_failed,
  boundary_channel_full, boundary_batch_full)
- batches_forwarded_total

Also add BoundaryStatus to the BoundaryMessage envelope so boundary
can report dropped log counts as a separate wire message. The agent
records these as Prometheus metrics, making boundary-side data loss
visible. Backwards compatibility for older versions of boundary is maintained.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-03 12:42:34 -07:00

30 lines
1.0 KiB
Protocol Buffer

syntax = "proto3";
option go_package = "github.com/coder/coder/v2/agent/boundarylogproxy/codec";
package coder.boundarylogproxy.codec.v1;
import "agent/proto/agent.proto";
// BoundaryMessage is the envelope for all TagV2 messages sent over the
// boundary <-> agent unix socket. TagV1 carries a bare
// ReportBoundaryLogsRequest for backwards compatibility; TagV2 wraps
// everything in this envelope so the protocol can be extended with new
// message types without adding more tags.
message BoundaryMessage {
oneof msg {
coder.agent.v2.ReportBoundaryLogsRequest logs = 1;
BoundaryStatus status = 2;
}
}
// BoundaryStatus carries operational metadata from boundary to the agent.
// The agent records these values as Prometheus metrics. This message is
// never forwarded to coderd.
message BoundaryStatus {
// Logs dropped because boundary's internal channel buffer was full.
int64 dropped_channel_full = 1;
// Logs dropped because boundary's batch buffer was full after a
// failed flush attempt.
int64 dropped_batch_full = 2;
}