A production-grade Rust toolkit for Solana infrastructure diagnostics, validator optimization, RPC health checks, caching, metrics, and production-readiness workflows.
solana-validator-optimizer is a modular enhancement suite for Solana validator infrastructure. It helps operators by:
- Running validator preflight readiness checks before deployment or restart
- Validating RPC health, cluster genesis hash, current slot, and blockhash availability
- Checking host CPU, memory, and disk readiness
- Prefetching ledger snapshots from trusted mirrors
- Validating snapshot integrity via SHA-256
- Auto-tuning validator configuration based on hardware resources
- Adding an in-memory LRU cache for frequently used RPC calls
- Exposing Prometheus metrics via a
/metricsendpoint
The project currently builds against the Solana 3.x Rust crate line.
- Solana crate line: 3.x
- RPC diagnostics have been exercised against Mainnet-Beta RPC endpoints reporting Agave/Solana 4.x
- Tested against:
- Mainnet RPC: https://api.mainnet-beta.solana.com
solana-test-validator(local)
RPC-level interoperability with a newer Agave endpoint does not imply full build-time compatibility with the Agave 4.x Rust crate line.
- Solana Validator Operators
- RPC Infrastructure Maintainers
- Performance-Focused Mainnet/Devnet Deployments
| Feature | Description |
|---|---|
| Validator Preflight | Aggregates RPC, cluster identity, CPU, memory, disk, and readiness checks into PASS/WARN/FAIL results |
| Genesis Validation | Retrieves the cluster genesis hash and optionally validates it against an operator-supplied expected hash |
| Snapshot Prefetching | Downloads or copies snapshots locally with optional SHA-256 validation |
| RPC LRU Cache | Reduces redundant RPC calls like getBalance, getEpochInfo, etc. |
| Prometheus Metrics | /metrics endpoint for cache hits, misses, request counts, and latency |
| Config Auto-Tuner | Adjusts validator configuration based on CPU, RAM, disk, and network |
| RPC Health Checker | Checks RPC health, Solana/Agave version, genesis hash, current slot, latest blockhash availability, and RPC latency |
git clone https://github.com/0rlych1kk4/solana-validator-optimizer.git
cd solana-validator-optimizerCreate a Config.toml file in the project root:
# Config.toml
# Snapshot settings
snapshot_url = "" # e.g. "https://snapshots.myvalidator.com/latest.tar.zst"
snapshot_sha256 = "" # optional SHA-256 checksum for validation
# RPC and Metrics
rpc_url = "https://api.mainnet-beta.solana.com"
metrics_port = 9090 # Prometheus scrapes at http://<host>:9090/metrics
# Cache settings
cache_size = 128 # Number of entries in the RPC LRU cacheRun all modules (snapshot, RPC cache, auto-tuner, metrics):
cargo run -p solana-validator-optimizer-cli --releaseAdd this to your Cargo.toml:
[dependencies]
solana-validator-optimizer = "2.3.0"
tokio = { version = "1", features = ["full"] }
anyhow = "1"Example usage:
use solana_validator_optimizer::{
config::AppConfig,
snapshot_prefetcher,
rpc_cache_layer,
metrics,
config_autotuner,
};
#[tokio::main]
async fn main() -> anyhow::Result<()> {
// Load configuration from Config.toml and/or OPTIMIZER_* env vars
let cfg = AppConfig::load()?;
// Prefetch snapshot (no-op if snapshot_url is empty)
snapshot_prefetcher::run(&cfg).await?;
// Auto-tune configuration (currently informational)
config_autotuner::autotune_config(&cfg).await?;
// Start background RPC cache workers
rpc_cache_layer::start_rpc_cache(&cfg).await?;
// Start Prometheus metrics server (blocks)
metrics::start_metrics_server(&cfg).await?;
Ok(())
}Check the configured Solana RPC endpoint from Config.toml:
cargo run -p solana-validator-optimizer-cli -- rpc-healthExample output:
Solana RPC Health Report
Endpoint: https://api.mainnet-beta.solana.com
Healthy: true
Health Status: ok
Solana Version: 4.2.0-rc.1
Genesis Hash: 5eykt4UsFv8P8NJdTREpY1vzqKqZKvdpKuc147dw2N9d
Current Slot: 438204077
Latest Blockhash: 77L8S6DcWGDsAVQJr1S4n4n7EPn7wP365DgHiMNFJbMc
getHealth Latency: 956 ms
getVersion Latency: 328 ms
getGenesisHash Latency: 308 ms
getSlot Latency: 295 ms
getLatestBlockhash Latency: 626 ms
Use a custom endpoint:
cargo run -p solana-validator-optimizer-cli -- rpc-health --endpoint https://api.mainnet-beta.solana.comOutput as JSON:
cargo run -p solana-validator-optimizer-cli -- rpc-health --jsonThe RPC Health Checker helps Solana infrastructure teams validate endpoint health, cluster identity, blockhash availability, and latency before transaction submission or production deployment.
Run production-readiness checks using the RPC endpoint configured in Config.toml:
cargo run -p solana-validator-optimizer-cli -- preflightThe preflight command evaluates:
- RPC health
- Solana/Agave version
- Cluster genesis hash
- Current slot
- Latest blockhash availability
- RPC latency
- Logical CPU count
- Total memory
- Available memory
- Available disk space
Example output:
Solana Validator Preflight
Endpoint: https://api.mainnet-beta.solana.com
[PASS] rpc_health RPC health status: ok
[PASS] solana_version Detected Solana/Agave version: 4.2.0-rc.1
[PASS] genesis_hash Genesis hash: 5eykt4UsFv8P8NJdTREpY1vzqKqZKvdpKuc147dw2N9d
[PASS] current_slot Current slot: 438203915
[PASS] latest_blockhash Latest blockhash available
[PASS] rpc_latency RPC latency is within the 1000 ms threshold
[PASS] cpu Logical CPU count: 8
[WARN] memory_total Total memory: 8.0 GiB
[WARN] memory_available Available memory: 0.34 GiB
[PASS] disk_available Disk space: 57.8 GiB available / 228.3 GiB total
Overall readiness: READY
Use an expected genesis hash to detect a wrong-cluster configuration:
cargo run -p solana-validator-optimizer-cli -- preflight --expected-genesis-hash 5eykt4UsFv8P8NJdTREpY1vzqKqZKvdpKuc147dw2N9dA matching hash produces:
[PASS] genesis_hash Genesis hash matches expected value: 5eykt4UsFv8P8NJdTREpY1vzqKqZKvdpKuc147dw2N9d
Overall readiness: READY
A mismatch is treated as a hard readiness failure:
[FAIL] genesis_hash Genesis hash mismatch: expected 11111111111111111111111111111111, observed 5eykt4UsFv8P8NJdTREpY1vzqKqZKvdpKuc147dw2N9d
Overall readiness: NOT READY
Warnings such as elevated RPC latency or limited available memory do not automatically fail readiness. Hard failures such as an unhealthy RPC endpoint, missing required cluster data, or a genesis-hash mismatch result in NOT READY.
Machine-readable output is also available:
cargo run -p solana-validator-optimizer-cli -- preflight --jsonAny Config.toml value can be overridden using the OPTIMIZER_ prefix.
| Config Key | Environment Variable |
|---|---|
| snapshot_url | OPTIMIZER_SNAPSHOT_URL |
| snapshot_sha256 | OPTIMIZER_SNAPSHOT_SHA256 |
| rpc_url | OPTIMIZER_RPC_URL |
| metrics_port | OPTIMIZER_METRICS_PORT |
| cache_size | OPTIMIZER_CACHE_SIZE |
Example:
OPTIMIZER_RPC_URL=https://api.mainnet-beta.solana.com \
OPTIMIZER_CACHE_SIZE=256 \
cargo run --releasePrometheus endpoint:
http://<host>:<metrics_port>/metricsrpc_requests_totalrpc_cache_hits_totalrpc_cache_misses_total
Enable Prometheus metrics endpoint:
cargo run -p solana-validator-optimizer-cli --release --features metrics -- --config Config.toml run
curl -s http://127.0.0.1:9090/metrics | grep rpc_Contributions are welcome!
Feel free to open issues, feature requests, or pull requests.
Licensed under the MIT License.
