mirror of
https://github.com/XRPLF/rippled.git
synced 2026-10-01 09:18:04 +00:00
ac07e1345f
The collector reads the per-node directory off the log file path and stamps it
as the Loki label service_instance_id, so the directory name has to equal the
node's own [telemetry] service_instance_id or log lines carry a node name that
no trace or metric shares and nothing joins.
Both harness scripts disagreed with themselves: run-full-validation.sh wrote to
node$i while setting validator-${i}, and benchmark.sh wrote to node$i while
setting bench-node-${i}. Rename the directories to match the ids rather than
the reverse, so no existing trace or metric label value moves and no harness
expectation has to be re-checked. Only path references are renamed; the
human-readable "node$i" in log and error messages is left as prose.
The config template is not rendered by any script, so its DATA_DIR
documentation gains a note about the same constraint instead.
Also rename the deprecated otlphttp/filelog collector component names in the
harness scripts and docs.
128 lines
4.1 KiB
Plaintext
128 lines
4.1 KiB
Plaintext
# xrpld validator node configuration template for workload harness.
|
|
#
|
|
# Not consumed by anything today: run-full-validation.sh writes each node's
|
|
# cfg inline. Kept as the reference layout for a validator run as a container,
|
|
# whose entrypoint would substitute the placeholders below.
|
|
#
|
|
# Because nothing reads this file, nothing catches it drifting from the cfg the
|
|
# runner actually generates. Any change to the [telemetry], [insight] or
|
|
# [rpc_startup] stanzas here must be mirrored in run-full-validation.sh, and
|
|
# vice versa.
|
|
#
|
|
# Placeholders:
|
|
# {{NODE_INDEX}} — Node number (1-based)
|
|
# {{RPC_PORT}} — HTTP RPC port
|
|
# {{WS_PORT}} — WebSocket port
|
|
# {{PEER_PORT}} — Peer protocol port
|
|
# {{DATA_DIR}} — Node data directory. Its last path segment must
|
|
# equal service_instance_id below: the collector's
|
|
# file_log receiver reads that segment off the log
|
|
# file path and stamps it as the Loki label
|
|
# service_instance_id, so a mismatch gives log lines
|
|
# a node name no trace or metric shares.
|
|
# {{VALIDATION_SEED}} — Validator seed from key generation
|
|
# {{VALIDATORS_FILE}} — Path to shared validators.txt
|
|
# {{IPS_FIXED}} — Peer addresses (one per line)
|
|
# {{OTEL_ENDPOINT}} — OTel Collector OTLP/HTTP traces endpoint
|
|
# {{OTEL_METRICS_ENDPOINT}} — OTel Collector OTLP/HTTP metrics endpoint
|
|
# {{LOG_LEVEL}} — Log level. Must be at least `info` for the
|
|
# log-trace correlation checks to pass: they need a
|
|
# log line emitted while a span is current on the
|
|
# thread. The guaranteed such line is the consensus
|
|
# accept pair in RCLConsensus.cpp, which is at info
|
|
# and sits inside the activation doAccept makes over
|
|
# its whole body. `warning` and above suppress it and
|
|
# leave the checks with nothing to find. Do not use
|
|
# `debug`: it adds log I/O inside the spans whose
|
|
# latency regression-metrics.json gates.
|
|
|
|
[server]
|
|
port_rpc
|
|
port_ws
|
|
port_peer
|
|
|
|
# RPC and WebSocket stay on loopback, matching the cfg that
|
|
# run-full-validation.sh generates and the rest of the repo's node configs.
|
|
# Only the peer port below needs to listen on all interfaces.
|
|
[port_rpc]
|
|
port = {{RPC_PORT}}
|
|
ip = 127.0.0.1
|
|
admin = 127.0.0.1
|
|
protocol = http
|
|
|
|
[port_ws]
|
|
port = {{WS_PORT}}
|
|
ip = 127.0.0.1
|
|
admin = 127.0.0.1
|
|
protocol = ws
|
|
|
|
[port_peer]
|
|
port = {{PEER_PORT}}
|
|
ip = 0.0.0.0
|
|
protocol = peer
|
|
|
|
[node_db]
|
|
type=NuDB
|
|
path={{DATA_DIR}}/nudb
|
|
online_delete=256
|
|
|
|
[database_path]
|
|
{{DATA_DIR}}/db
|
|
|
|
[debug_logfile]
|
|
{{DATA_DIR}}/debug.log
|
|
|
|
[validation_seed]
|
|
{{VALIDATION_SEED}}
|
|
|
|
[validators_file]
|
|
{{VALIDATORS_FILE}}
|
|
|
|
[ips_fixed]
|
|
{{IPS_FIXED}}
|
|
|
|
[peer_private]
|
|
1
|
|
|
|
# --- OpenTelemetry tracing (all categories enabled) ---
|
|
[telemetry]
|
|
enabled=1
|
|
service_instance_id=validator-{{NODE_INDEX}}
|
|
endpoint={{OTEL_ENDPOINT}}
|
|
exporter=otlp_http
|
|
batch_size=512
|
|
batch_delay_ms=2000
|
|
max_queue_size=2048
|
|
trace_rpc=1
|
|
trace_transactions=1
|
|
trace_consensus=1
|
|
trace_peer=1
|
|
trace_ledger=1
|
|
|
|
# --- Native OTel metrics (beast::insight over OTLP/HTTP) ---
|
|
# The collector has no StatsD receiver (its metrics pipeline is
|
|
# [otlp, spanmetrics]), so beast::insight must export natively over OTLP for
|
|
# system metrics to reach Prometheus at all. `server=otel` is the only
|
|
# load-bearing key here -- it selects the OTel collector.
|
|
#
|
|
# No `prefix` is set on purpose. It would be inert: OTelCollector applies no
|
|
# prefix to instrument names, so exported names are the lowercased raw names
|
|
# (`jobq_job_count`, `rpc_requests_total`, `total_bytes_in`). The service is
|
|
# identified by the OTel resource `service.name`, not by a name prefix.
|
|
[insight]
|
|
server=otel
|
|
endpoint={{OTEL_METRICS_ENDPOINT}}
|
|
|
|
[rpc_startup]
|
|
{ "command": "log_level", "severity": "{{LOG_LEVEL}}" }
|
|
|
|
[ssl_verify]
|
|
0
|
|
|
|
# --- Network tuning for local cluster ---
|
|
[network_id]
|
|
0
|
|
|
|
[sntp_servers]
|
|
time.google.com
|