Scaffold Phase 0: agent -> Redpanda -> ingest -> ClickHouse -> api -> web

End-to-end log pipeline for Linux hosts, per /docs/architecture.md:

- proto: shared gRPC contract (agent <-> ingest), Go bindings checked in
- agent: Rust, musl-targeted, journald/file sourcing, RFC5424 parser,
  mTLS gRPC client, no required config for the common case
- ingest: Go, single binary with --mode server|consumer|all; gRPC front
  end forwards to Redpanda unchanged, consumer normalizes and
  batch-writes to ClickHouse with at-least-once delivery
- storage: ClickHouse schema + a plain SQL-file migration runner
- api: minimal SELECT-only query endpoint, plain REST (not gRPC+gateway
  yet -- see api/README.md)
- web: SvelteKit static SPA, one query page
- transport: Redpanda compose + topic provisioning
- cli: sentryctl ping stub
- hack/dev-certs: throwaway CA + cert generation for local mTLS
- root docker-compose.yml + docs/phase-0-runbook.md tie it together

Not yet run end-to-end against real Docker/ClickHouse/Redpanda -- see the
runbook's caveats section before relying on this working as-is.
This commit is contained in:
2026-08-13 08:25:19 -07:00
commit b6b092c912
92 changed files with 7796 additions and 0 deletions
+58
View File
@@ -0,0 +1,58 @@
// Package normalize maps the wire-format LogRecord (as agents send it)
// into the ClickHouse row shape defined in /storage. This is the "OTel-log-
// like schema" normalization step called for in the ingest design — Phase
// 0 keeps it to the minimal column set; full OTel field mapping (separate
// SeverityNumber/SeverityText, resource attributes, etc.) is deferred, see
// the open questions in /docs/architecture.md.
package normalize
import (
"time"
logsv1 "github.com/sentry/sentry/proto/sentry/logs/v1"
)
type Row struct {
Timestamp time.Time
Host string
Service string
Severity string
Message string
Attributes map[string]string
}
func ToRow(rec *logsv1.LogRecord) Row {
attrs := rec.GetAttributes()
if attrs == nil {
attrs = map[string]string{}
}
return Row{
Timestamp: time.Unix(0, rec.GetTimestampUnixNano()).UTC(),
Host: rec.GetHost(),
Service: rec.GetService(),
Severity: severityText(rec.GetSeverity()),
Message: rec.GetMessage(),
Attributes: attrs,
}
}
// severityText maps the proto Severity enum to short OTel-style severity
// names, stored as the `severity` column's value.
func severityText(sev logsv1.Severity) string {
switch sev {
case logsv1.Severity_SEVERITY_TRACE:
return "TRACE"
case logsv1.Severity_SEVERITY_DEBUG:
return "DEBUG"
case logsv1.Severity_SEVERITY_INFO:
return "INFO"
case logsv1.Severity_SEVERITY_WARN:
return "WARN"
case logsv1.Severity_SEVERITY_ERROR:
return "ERROR"
case logsv1.Severity_SEVERITY_FATAL:
return "FATAL"
default:
return "UNSPECIFIED"
}
}
@@ -0,0 +1,68 @@
package normalize
import (
"testing"
"time"
logsv1 "github.com/sentry/sentry/proto/sentry/logs/v1"
)
func TestToRowMapsFieldsAndSeverity(t *testing.T) {
rec := &logsv1.LogRecord{
TimestampUnixNano: time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC).UnixNano(),
Host: "host-1",
Service: "svc-a",
Severity: logsv1.Severity_SEVERITY_ERROR,
Message: "boom",
Attributes: map[string]string{"k": "v"},
}
row := ToRow(rec)
if row.Host != "host-1" || row.Service != "svc-a" || row.Message != "boom" {
t.Fatalf("unexpected row: %+v", row)
}
if row.Severity != "ERROR" {
t.Fatalf("expected severity ERROR, got %s", row.Severity)
}
if row.Attributes["k"] != "v" {
t.Fatalf("expected attribute k=v, got %+v", row.Attributes)
}
if !row.Timestamp.Equal(time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)) {
t.Fatalf("unexpected timestamp: %v", row.Timestamp)
}
}
func TestToRowNilAttributesBecomesEmptyMap(t *testing.T) {
rec := &logsv1.LogRecord{Host: "h", Service: "s", Message: "m"}
row := ToRow(rec)
if row.Attributes == nil {
t.Fatal("expected non-nil empty map, got nil")
}
if len(row.Attributes) != 0 {
t.Fatalf("expected empty map, got %+v", row.Attributes)
}
}
func TestSeverityTextCoversAllEnumValues(t *testing.T) {
cases := map[logsv1.Severity]string{
logsv1.Severity_SEVERITY_UNSPECIFIED: "UNSPECIFIED",
logsv1.Severity_SEVERITY_TRACE: "TRACE",
logsv1.Severity_SEVERITY_DEBUG: "DEBUG",
logsv1.Severity_SEVERITY_INFO: "INFO",
logsv1.Severity_SEVERITY_WARN: "WARN",
logsv1.Severity_SEVERITY_ERROR: "ERROR",
logsv1.Severity_SEVERITY_FATAL: "FATAL",
}
for sev, want := range cases {
if got := severityText(sev); got != want {
t.Errorf("severityText(%v) = %q, want %q", sev, got, want)
}
}
}
func TestSeverityTextUnknownValueFallsBackToUnspecified(t *testing.T) {
if got := severityText(logsv1.Severity(99)); got != "UNSPECIFIED" {
t.Fatalf("expected UNSPECIFIED for unknown severity, got %q", got)
}
}