Files
silo-server/internal/notifications/push_sender.go
T
203a18ae83 feat(observability): OpenTelemetry logs+traces with secret redaction and slog standardization (#290)
* feat(observability): OpenTelemetry logs+traces with secret redaction

Part of #265. Adds opt-in OpenTelemetry (logs + traces) alongside the existing
stderr + opslog pipeline, plus secret redaction on all sinks. Default-off: with
no OTEL_* / SILO_OTEL_ENABLED config, behavior is unchanged.

Bootstrap (internal/telemetry):
- Setup() builds one shared resource, a TracerProvider (parent-based trace-id
  ratio sampler), a LoggerProvider, and the W3C TraceContext+Baggage propagator
  from env. It installs NO MeterProvider — metrics stay on Prometheus, and the
  built-in no-op global MeterProvider keeps the trace instrumentation libs from
  double-emitting. Shutdown is deferred with a flush timeout.
- Logs are bridged via otelslog fan-out (slog.MultiHandler), level-gated by the
  shared LevelVar and best-effort so a failing collector can't break the console
  or DB branches. stderr + opslog stay untouched.

Secret redaction (internal/logredact):
- A slog.Handler masks secret-keyed attributes (password, token, api_key,
  authorization, cookie, ...) — including .With-bound attrs, nested groups,
  secret-keyed group subtrees, and values behind a LogValuer — on the console
  and OTLP sinks, with a no-op fast path when a record has no secret keys.
  opslog.shouldRedact delegates to logredact.SecretKey so all sinks share one
  marker list.

Rotation is infra-managed (no custom file sink): container runtime for stderr,
collector/backend for OTLP, opslog partition-pruning for the DB. Documented in
docs/architecture/observability.md.

Verification: go build ./..., go vet, gofmt -l — clean; go test
./internal/telemetry/ ./internal/logredact/ -race pass.

AI-use disclosure: implemented with AI assistance (Claude Code), including
adversarial reviews that hardened the bootstrap and fixed two redaction leak
paths; reviewed by the author.

* refactor(observability): slog context+component sweep, sloglint gate (phase 3)

Part of #265. Builds on the OTel bootstrap + redaction commit.

Standardizes every log call site onto the context-carrying slog variants so
records correlate with the active OpenTelemetry trace, and locks the standard
in with a machine gate so future code (human- or AI-authored) can't drift back.

- Call-site sweep: converted the remaining slog.<Level>(...) calls to the
  slog.<Level>Context(ctx, ...) form wherever a context.Context is in scope
  (background/init calls with no ctx are left as-is), across 183 files. Applied
  via a type-aware AST codemod. Log levels and message strings are preserved
  verbatim; a component attr (canonical per-package name) is added to direct
  package-level slog calls. Bound-logger calls keep their existing .With
  bindings. The main.go and telemetry package conversions rode with their file
  in the previous commit to keep each file within a single commit.
- Enforcement (.golangci.yml): enable sloglint with context=scope, static-msg,
  key-naming-case=snake, no-mixed-args. After the sweep all four report zero
  violations repo-wide (tests included), so make lint / CI now blocks any
  regression to the non-context form. The gate ships with the sweep because it
  cannot be green until the legacy sites are converted.

Metrics remain on Prometheus; no behavior change to /metrics or Grafana.

Verification: go build ./..., go vet ./..., gofmt -l — clean; sloglint (all 4
rules) 0 violations repo-wide; log levels verified unchanged.

AI-use disclosure: implemented with AI assistance (Claude Code), including the
codemod; reviewed by the author.

* fix(observability): honor per-signal OTLP protocol and secret WithGroup names

Two Codex review findings on PR #290:

- telemetry: OTEL_EXPORTER_OTLP_{TRACES,LOGS}_PROTOCOL now override the
  generic OTEL_EXPORTER_OTLP_PROTOCOL per signal, so mixed collector
  setups (e.g. HTTP logs + gRPC traces) build the right exporter.
- logredact: entering a group whose name is secret-bearing (e.g.
  WithGroup("authorization")) now masks every leaf in that subtree,
  matching how slog.Group("authorization", ...) is masked as a whole.

* fix(observability): address review feedback on telemetry bootstrap

- Telemetry setup failure no longer kills boot: Setup returns usable
  no-op providers alongside the error and main logs and continues with
  telemetry disabled, honoring the best-effort contract.
- Honor OTEL_TRACES_SAMPLER (always_on/off, traceidratio, parentbased_*
  variants); unsupported values fall back to parentbased_traceidratio.
- Attach node identity as semconv service.instance.id instead of the
  non-semconv node.name.
- Rename opslog retention-scope log attrs to target_component/target_level
  so they no longer collide with the canonical component routing key, and
  tag those lines with component=opslog.
- Fix stale levelGated comment casing; use WarnContext in the telemetry
  shutdown defer; document the LogValuer double-resolve on the redaction
  slow path.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

---------

Co-authored-by: Quick <31828688+Quick104@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-07-09 08:53:52 -04:00

344 lines
11 KiB
Go

package notifications
import (
"bytes"
"context"
"encoding/json"
"fmt"
"io"
"log/slog"
"net/http"
"strings"
"time"
"github.com/Silo-Server/silo-server/internal/secret"
)
var pushRetrySchedule = []time.Duration{
0,
30 * time.Second,
2 * time.Minute,
10 * time.Minute,
30 * time.Minute,
}
const (
pushMaxAttempts = 5
pushDispatchQueue = 512
pushRetryClaimLimit = 100
relayAppleSendPath = "/v1/apple/send"
)
func pushRetryDelay(completedAttempt int) (time.Duration, bool) {
if completedAttempt < 1 || completedAttempt >= pushMaxAttempts {
return 0, false
}
return pushRetrySchedule[completedAttempt] - pushRetrySchedule[completedAttempt-1], true
}
type pushRelayAppleRequest struct {
Token string `json:"token"`
Environment string `json:"environment"`
Topic string `json:"topic"`
Mode string `json:"mode"`
ServerDeviceID string `json:"server_device_id"`
DeliveryID string `json:"delivery_id"`
CollapseID *string `json:"collapse_id,omitempty"`
}
type pushRelayAppleResponse struct {
RequestID string `json:"request_id"`
APNsID string `json:"apns_id"`
Status string `json:"status"`
}
type pushRelayErrorResponse struct {
Error struct {
Code string `json:"code"`
Message string `json:"message"`
RequestID string `json:"request_id"`
} `json:"error"`
}
type pushSendResult struct {
OK bool
HTTPStatus int
RetryAfter time.Duration
RelayRequestID string
UpstreamReason string
Message string
TerminalDevice bool
}
type pushSender struct {
devices *PushDeviceRepository
deliveries *DeliveryRepository
cipher *secret.Cipher
settings *Settings
client *http.Client
logger *slog.Logger
}
func newPushSender(devices *PushDeviceRepository, deliveries *DeliveryRepository, cipher *secret.Cipher, settings *Settings) *pushSender {
return &pushSender{
devices: devices,
deliveries: deliveries,
cipher: cipher,
settings: settings,
client: newWebhookHTTPClient(nil),
logger: slog.Default().With("component", "notifications.apple_push"),
}
}
func (s *pushSender) processAttempt(ctx context.Context, attempt PushDeliveryAttempt) *PushDeliveryAttempt {
device, err := s.devices.getPushDeviceByID(ctx, attempt.PushDeviceID)
if err != nil || device == nil {
if err == nil {
return s.finalize(ctx, attempt, PushOutcomeFailed, "", "push device deleted", nil, "", nil)
}
if ctx.Err() == nil {
s.logger.WarnContext(ctx, "push device lookup failed", "attempt_id", attempt.ID, "error", err)
}
return nil
}
if !device.Enabled || device.PushMode != PushModePrivatePush || !s.settings.ApplePushDeliveryEnabled(ctx) {
return s.finalize(ctx, attempt, PushOutcomeFailed, "delivery_disabled", "apple push delivery disabled", nil, "", nil)
}
if attempt.NotificationDeliveryID != nil {
row, err := s.deliveries.GetRowByID(ctx, *attempt.NotificationDeliveryID)
if err != nil {
if ctx.Err() == nil {
s.logger.WarnContext(ctx, "push delivery lookup failed",
"attempt_id", attempt.ID,
"delivery_id", *attempt.NotificationDeliveryID,
"error", err)
}
return nil
}
if row == nil {
return s.finalize(ctx, attempt, PushOutcomeFailed, "delivery_missing", "delivery row missing", nil, "", nil)
}
if row.ProfileID != device.ProfileID {
return s.finalize(ctx, attempt, PushOutcomeFailed, "device_reassigned", "push device reassigned", nil, "", nil)
}
}
token, err := s.cipher.Decrypt(device.APNsTokenCiphertext, pushDeviceAPNsTokenAAD(device.ID))
if err != nil {
return s.finalize(ctx, attempt, PushOutcomeFailed, "decrypt_failed", "APNs token decrypt failed", nil, "", nil)
}
result := s.send(ctx, attempt, device, token)
attemptNumber := attempt.AttemptNumber + 1
if result.OK {
updated, _ := s.devices.FinalizePushAttempt(ctx, attempt.ID, PushOutcomeDelivered, attemptNumber,
result.RelayRequestID, &result.HTTPStatus, result.UpstreamReason, "", nil)
_ = s.devices.RecordPushSuccess(ctx, device.ID)
return updated
}
statusPtr := (*int)(nil)
if result.HTTPStatus > 0 {
statusPtr = &result.HTTPStatus
}
code := result.UpstreamReason
if code == "" {
code = result.Message
}
if code == "" {
code = "push_delivery_failed"
}
_ = s.devices.RecordPushFailure(ctx, device.ID, code, result.TerminalDevice)
delay, more := pushRetryDelay(attemptNumber)
if result.RetryAfter > 0 {
delay = result.RetryAfter
}
if more && !result.TerminalDevice && retryableHTTPStatus(result.HTTPStatus) {
nextRetry := time.Now().Add(delay)
return s.finalize(ctx, attempt, PushOutcomeRetrying, result.UpstreamReason, result.Message, statusPtr, result.RelayRequestID, &nextRetry)
}
return s.finalize(ctx, attempt, PushOutcomeFailed, result.UpstreamReason, result.Message, statusPtr, result.RelayRequestID, nil)
}
func (s *pushSender) finalize(ctx context.Context, attempt PushDeliveryAttempt, outcome string, reason, message string, statusPtr *int, relayRequestID string, nextRetryAt *time.Time) *PushDeliveryAttempt {
attemptNumber := attempt.AttemptNumber + 1
updated, err := s.devices.FinalizePushAttempt(ctx, attempt.ID, outcome, attemptNumber, relayRequestID, statusPtr, reason, message, nextRetryAt)
if err != nil && ctx.Err() == nil {
s.logger.WarnContext(ctx, "finalize push attempt failed", "attempt_id", attempt.ID, "error", err)
}
return updated
}
func (s *pushSender) send(ctx context.Context, attempt PushDeliveryAttempt, device *PushDevice, token string) pushSendResult {
apiKey := s.settings.PushRelayAPIKey(ctx)
if apiKey == "" {
return pushSendResult{Message: "push relay API key not configured", UpstreamReason: "relay_api_key_missing"}
}
relayURL := s.settings.PushRelayURL(ctx)
deliveryID := attempt.ID
if attempt.NotificationDeliveryID != nil {
deliveryID = *attempt.NotificationDeliveryID
}
collapseID := deliveryID
body, err := json.Marshal(pushRelayAppleRequest{
Token: token,
Environment: device.APNsEnvironment,
Topic: device.APNsTopic,
Mode: "private_alert",
ServerDeviceID: device.ServerDeviceID,
DeliveryID: deliveryID,
CollapseID: &collapseID,
})
if err != nil {
return pushSendResult{Message: "relay payload build failed", UpstreamReason: "payload_build_failed"}
}
req, err := http.NewRequestWithContext(ctx, http.MethodPost, relayURL+relayAppleSendPath, bytes.NewReader(body))
if err != nil {
return pushSendResult{Message: "invalid push relay URL", UpstreamReason: "invalid_relay_url"}
}
if req.URL.Scheme != schemeHTTPS {
return pushSendResult{Message: "invalid push relay URL", UpstreamReason: "invalid_relay_url"}
}
req.Header.Set("Authorization", "Bearer "+apiKey)
req.Header.Set("Content-Type", "application/json")
req.Header.Set("User-Agent", "Silo-Push/1.0")
req.Header.Set("Idempotency-Key", fmt.Sprintf("%s:%d", attempt.ID, attempt.AttemptNumber+1))
resp, err := s.client.Do(req)
if err != nil {
return pushSendResult{Message: classifyWebhookError(err), UpstreamReason: "relay_unreachable"}
}
defer func() { _ = resp.Body.Close() }()
if resp.StatusCode >= 200 && resp.StatusCode < 300 {
var parsed pushRelayAppleResponse
_ = json.NewDecoder(io.LimitReader(resp.Body, 16<<10)).Decode(&parsed)
if parsed.RequestID == "" {
parsed.RequestID = resp.Header.Get("X-Request-ID")
}
return pushSendResult{
OK: true,
HTTPStatus: resp.StatusCode,
RelayRequestID: parsed.RequestID,
}
}
var parsed pushRelayErrorResponse
data, _ := io.ReadAll(io.LimitReader(resp.Body, 16<<10))
_ = json.Unmarshal(data, &parsed)
code := parsed.Error.Code
if code == "" {
code = fmt.Sprintf("http_%d", resp.StatusCode)
}
message := parsed.Error.Message
if message == "" {
message = http.StatusText(resp.StatusCode)
}
if message == "" {
message = fmt.Sprintf("HTTP %d", resp.StatusCode)
}
return pushSendResult{
HTTPStatus: resp.StatusCode,
RetryAfter: parseRetryAfter(resp.Header.Get("Retry-After"), time.Now()),
RelayRequestID: parsed.Error.RequestID,
UpstreamReason: code,
Message: strings.TrimSpace(message),
TerminalDevice: resp.StatusCode == http.StatusUnprocessableEntity && code == "apns_rejected",
}
}
// PushDispatcher implements the channel Dispatcher interface for Apple push
// on top of the shared channelDispatcher core, with the retry/recovery sweep
// integrated.
type PushDispatcher struct {
core channelDispatcher[PushDeliveryAttempt]
}
func newPushDispatcher(sender *pushSender) *PushDispatcher {
return &PushDispatcher{core: channelDispatcher[PushDeliveryAttempt]{
channel: "apple push",
queue: make(chan string, pushDispatchQueue),
logger: slog.Default().With("component", "notifications.apple_push.dispatch"),
claimPending: sender.devices.ClaimPendingPushForDelivery,
process: func(ctx context.Context, attempt PushDeliveryAttempt) {
sender.processAttempt(ctx, attempt)
},
enabled: sender.settings.ApplePushDeliveryEnabled,
claimDue: sender.devices.ClaimDuePushAttempts,
claimLimit: pushRetryClaimLimit,
}}
}
// Dispatch queues the delivery's Apple push attempts for immediate send.
func (d *PushDispatcher) Dispatch(_ context.Context, delivery DeliveryRow) error {
if d == nil {
return nil
}
d.core.dispatch(delivery.ID)
return nil
}
// Run consumes the dispatch queue and the retry/recovery sweep until ctx is
// canceled.
func (d *PushDispatcher) Run(ctx context.Context) {
d.core.run(ctx)
}
type ApplePushTestResult struct {
AttemptID string
PushDeviceID string
ServerDeviceID string
Outcome string
RelayRequestID string
UpstreamStatus *int
UpstreamReason string
FailureMessage string
}
func (s *System) SendApplePushTest(ctx context.Context, profileID, serverDeviceID string) (*ApplePushTestResult, error) {
if s == nil || s.pushDeviceRepo == nil || s.pushSender == nil {
return nil, ErrPushDeliveryUnavailable
}
if !s.Settings.ApplePushDeliveryEnabled(ctx) {
return nil, ErrPushDeliveryUnavailable
}
if s.Settings.PushRelayAPIKey(ctx) == "" {
return nil, ErrPushDeliveryUnavailable
}
attempt, device, err := s.pushDeviceRepo.EnqueueAppleTestAttempt(ctx, profileID, serverDeviceID)
if err != nil {
return nil, err
}
claimed, err := s.pushDeviceRepo.ClaimPushAttemptByID(ctx, attempt.ID)
if err != nil {
return nil, err
}
if len(claimed) != 1 {
return nil, fmt.Errorf("push test attempt was not claimable")
}
updated := s.pushSender.processAttempt(ctx, claimed[0])
if updated == nil {
updated, _ = s.pushDeviceRepo.GetPushAttempt(ctx, attempt.ID)
}
if updated == nil {
return nil, fmt.Errorf("push test attempt disappeared")
}
result := &ApplePushTestResult{
AttemptID: updated.ID,
PushDeviceID: updated.PushDeviceID,
ServerDeviceID: device.ServerDeviceID,
Outcome: updated.Outcome,
UpstreamStatus: updated.UpstreamStatus,
}
if updated.RelayRequestID != nil {
result.RelayRequestID = *updated.RelayRequestID
}
if updated.UpstreamReason != nil {
result.UpstreamReason = *updated.UpstreamReason
}
if updated.FailureMessage != nil {
result.FailureMessage = *updated.FailureMessage
}
return result, nil
}