Struct RocksSpoolParams 

pub struct RocksSpoolParams {

    pub increase_parallelism: Option<i32>,
    pub optimize_level_style_compaction: Option<usize>,
    pub optimize_universal_style_compaction: Option<usize>,
    pub paranoid_checks: bool,
    pub compression_type: DBCompressionTypeDef,
    pub compaction_readahead_size: Option<usize>,
    pub level_compaction_dynamic_level_bytes: bool,
    pub max_open_files: Option<usize>,
    pub write_buffer_size: Option<usize>,
    pub level0_stop_writes_trigger: Option<i32>,
    pub log_level: LogLevelDef,
    pub memtable_huge_page_size: Option<usize>,
    pub log_file_time_to_roll: Duration,
    pub obsolete_files_period: Duration,
    pub limit_concurrent_stores: Option<usize>,
    pub limit_concurrent_loads: Option<usize>,
    pub limit_concurrent_removes: Option<usize>,
    pub store_deadline: Duration,
    pub error_latch_duration: Duration,
    pub error_unlatch_duration: Duration,
    pub allow_error_unlatch: bool,
}

Fields

increase_parallelism: Option<i32>

optimize_level_style_compaction: Option<usize>

optimize_universal_style_compaction: Option<usize>

paranoid_checks: bool

compression_type: DBCompressionTypeDef

compaction_readahead_size: Option<usize>

If non-zero, we perform bigger reads when doing compaction. If you’re running RocksDB on spinning disks, you should set this to at least 2MB. That way RocksDB’s compaction is doing sequential instead of random reads

level_compaction_dynamic_level_bytes: bool

max_open_files: Option<usize>

write_buffer_size: Option<usize>

Size in bytes of the rocksdb memtable that buffers writes before being flushed to disk as a new SST file.

Smaller values produce smaller, more frequent SST files and trigger compactions sooner – useful in test setups that need to force the storage through its full write/compact lifecycle quickly. Larger values amortize compaction overhead but increase memory use and recovery time after restart. Leave unset to use the rocksdb default.

level0_stop_writes_trigger: Option<i32>

Number of level-0 SST files at which rocksdb will stop accepting writes. Lower values transition the database into the write-stopped state more quickly when background compaction cannot keep up, which is useful for tests that need to deterministically observe that condition. Leave unset to use the rocksdb default.

log_level: LogLevelDef

memtable_huge_page_size: Option<usize>

See: https://docs.rs/rocksdb/latest/rocksdb/struct.Options.html#method.set_memtable_huge_page_size

log_file_time_to_roll: Duration

obsolete_files_period: Duration

limit_concurrent_stores: Option<usize>

limit_concurrent_loads: Option<usize>

limit_concurrent_removes: Option<usize>

store_deadline: Duration

Upper bound on the wait that store() and remove() will tolerate when rocksdb is applying backpressure. Callers may provide a shorter deadline (typically derived from an SMTP client’s idle timeout); the effective deadline is the minimum of the two. Going longer than the caller-provided value risks the client timing out and retrying, which would produce duplicate deliveries – this option therefore only narrows the effective deadline, it never extends it.

error_latch_duration: Duration

How long the composite “this database is wedged” signal must hold continuously before the load-shedding gate latches. The signal goes high whenever the rocksdb background-errors counter has grown above the value observed at process start, or any foreground spool operation has returned a rocksdb error since process start. Brief blips that recover within this window do not latch the gate.

error_unlatch_duration: Duration

How long the healthy state must hold continuously before the load-shedding gate auto-unlatches. Only consulted when allow_error_unlatch is true. A relatively long value (minutes) gives operators time to inspect the database after a brief failure window before the daemon starts accepting writes again on its own.

allow_error_unlatch: bool

When true (the default), the load-shedding gate clears itself after error_unlatch_duration of observed recovery. Set to false to require an operator restart to clear the gate, which is appropriate when you want a human to confirm the underlying cause is resolved before accepting traffic again.

Trait Implementations

impl Debug for RocksSpoolParams

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for RocksSpoolParams

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for RocksSpoolParams

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for RocksSpoolParams

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.