Pratik Mankawde
f66a53cfc9
Merge branch 'pratik/otel-phase1b-telemetry-infra' into pratik/otel-phase1c-rpc-integration
...
Signed-off-by: Pratik Mankawde <3397372+pratikmankawde@users.noreply.github.com >
2026-05-29 14:51:12 +01:00
Pratik Mankawde
8b790ebac9
bumped otel version to 1.26.0
...
Signed-off-by: Pratik Mankawde <3397372+pratikmankawde@users.noreply.github.com >
2026-05-28 12:18:20 +01:00
Pratik Mankawde
824f63216a
Merge branch 'pratik/otel-phase1b-telemetry-infra' into pratik/otel-phase1c-rpc-integration
...
Signed-off-by: Pratik Mankawde <3397372+pratikmankawde@users.noreply.github.com >
2026-05-27 16:57:08 +01:00
Pratik Mankawde
a104140a51
addressing code review comments
...
Signed-off-by: Pratik Mankawde <3397372+pratikmankawde@users.noreply.github.com >
2026-05-27 16:46:35 +01:00
Pratik Mankawde
7a854ccad2
refactor(telemetry): simplify attr naming on phase-1c — drop xrpl.<domain>. prefix
...
- Drop xrpl.rpc.* prefix from per-span attrs (command, version).
- Qualify collision-prone fields: role -> rpc_role/grpc_role,
status -> rpc_status/grpc_status.
- Rename payload_size -> request_payload_size for cross-domain clarity.
- Simplify link.type -> link_type (bare name, no join).
- Update convention doc in SpanNames.h to reflect new naming rules.
- Update telemetry.md doc with renamed attr keys.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-05-13 15:54:13 +01:00
Pratik Mankawde
5e8277f36a
docs(telemetry): fix doc references to match pimpl architecture
...
Replace references to non-existent TracingInstrumentation.h with
SpanGuard.cpp pimpl implementation that actually exists on this branch.
Update conditional compilation section to describe the pimpl approach.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-28 14:26:05 +01:00
Pratik Mankawde
4bb2030315
feat(telemetry): add FilteringSpanProcessor and SpanGuard::discard()
...
Add span discard mechanism that drops unwanted spans before they enter
the batch export queue, saving both network bandwidth and storage.
FilteringSpanProcessor is a custom SpanProcessor decorator that wraps
BatchSpanProcessor. SpanGuard::discard() sets a thread-local flag
(tl_discardCurrentSpan) before calling Span::End(). The OTel SDK calls
OnEnd() synchronously on the same thread, where the flag is checked and
cleared to drop the span.
New file: DiscardFlag.h — zero-dependency header for the thread-local
flag, avoiding transitive include bloat from Telemetry.h.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-04-28 14:25:31 +01:00
Pratik Mankawde
3852b5ae4b
fix(telemetry): address review findings and PR #6437 comments
...
Critical fixes:
- Restore accidentally removed mallocTrim call and MallocTrim.h include
- Add missing shouldTraceLedger() to interface and all implementations
- Derive networkId/networkType from config_->NETWORK_ID (0=mainnet,
1=testnet, 2=devnet) instead of leaving defaults unpopulated
- Clamp sampling_ratio to [0.0, 1.0] in config parser
PR comment fixes:
- Rename rippled -> xrpld in service name defaults, getTracer() calls,
Docker network, comments, and docs/build/telemetry.md
- Remove exporter config option (only otlp_http supported)
- Add trace_ledger and service_name to example config
- Clarify head-based sampling semantics in config comments
- Add filter descriptions for span intrinsic filters in Grafana datasource
- Add inline comments to Docker Compose services
Docker/config improvements:
- Remove deprecated version: "3.8" from docker-compose.yml
- Pin images: collector 0.121.0, grafana 11.5.2
- Add health_check extension to otel-collector-config.yaml
- Comment out Tempo metrics_generator remote_write (no Prometheus service)
- Add Prometheus datasource caveat in Grafana datasource config
Other:
- Revert unrelated formatting changes in ServiceRegistry.h
- Change Conan telemetry default to False (matches CMake OFF)
- Add CLAUDE.md-required docs (ASCII diagrams, usage examples,
@note thread-safety) to Telemetry.h and SpanGuard.h
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-04-28 14:25:31 +01:00
Pratik Mankawde
ca2d616277
refactor(telemetry): remove Jaeger service, exporter, and datasource
...
Tempo is now the sole trace backend. Remove Jaeger all-in-one service
from docker-compose, otlp/jaeger exporter from OTel Collector config,
and Jaeger Grafana datasource provisioning file.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-28 14:25:31 +01:00
Pratik Mankawde
88686af850
Phase 1b: Telemetry core infrastructure - CMake, Conan, SpanGuard, config
...
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-04-28 14:25:31 +01:00
