Struct ServerLimits 

pub struct ServerLimits { /* private fields */ }

Wire-level resource limits for the native protocol adapters.

These limits are enforced before an operation is dispatched. They therefore protect every application independently of its extractors or business code. Header/body limits apply to both protocols; chunk and keep-alive request counts are HTTP/1-specific, while concurrent streams are HTTP/2-specific. The socket deadlines bound a peer that opens a connection and then stops making progress.

Implementations

impl ServerLimits

pub const fn new() -> Self

pub const fn with_max_header_bytes(self, bytes: usize) -> Self

pub const fn with_max_headers(self, count: usize) -> Self

Sets the maximum header count.

The native parser keeps its header slots on the stack, so this value is capped at 128 instead of allocating once per request.

pub const fn with_max_body_bytes(self, bytes: usize) -> Self

pub const fn with_max_chunks(self, count: usize) -> Self

Limits the number of chunks in one chunked HTTP/1 request.

pub const fn with_max_pipeline_batch(self, count: NonZeroUsize) -> Self

Bounds the number of pipelined HTTP/1 responses coalesced into one socket write. A value of one disables response coalescing.

pub const fn with_max_requests_per_connection( self, count: Option<NonZeroUsize>, ) -> Self

Bounds keep-alive reuse. None allows requests until either peer closes.

pub const fn with_stream_read_bytes(self, bytes: usize) -> Self

Sets the read window for streaming request bodies.

A streamed upload travels as owned chunks of at most this many bytes, and each chunk costs one read, one pass through the body channel, and one wakeup of the consuming handler. The window therefore trades a bounded amount of per-upload memory — at most a handful of windows is ever in flight for one request — against per-chunk overhead. The default of 64 KiB moves a five-megabyte upload in eighty chunks where the 8 KiB connection buffer would take six hundred forty, for tens of kilobytes of look-ahead per in-flight upload.

The body channel’s queue limit scales with this window, so enlarging it never strangles the producer against a fixed queue.

pub const fn with_header_read_timeout(self, timeout: Duration) -> Self

Bounds how long a complete request head may take to arrive once its first byte has been buffered.

pub const fn with_body_read_timeout(self, timeout: Duration) -> Self

Bounds how long one request-body read may stall.

pub const fn with_idle_timeout(self, timeout: Duration) -> Self

Bounds how long an idle keep-alive connection may wait for the first byte of its next request.

pub const fn with_write_timeout(self, timeout: Duration) -> Self

Bounds how long one response write may stall.

pub const fn max_header_bytes(self) -> usize

pub const fn max_headers(self) -> usize

pub const fn max_body_bytes(self) -> usize

pub const fn max_chunks(self) -> usize

pub const fn max_pipeline_batch(self) -> usize

pub const fn max_requests_per_connection(self) -> Option<NonZeroUsize>

pub const fn stream_read_bytes(self) -> usize

pub const fn header_read_timeout(self) -> Duration

pub const fn body_read_timeout(self) -> Duration

pub const fn idle_timeout(self) -> Duration

pub const fn write_timeout(self) -> Duration

Trait Implementations

impl Clone for ServerLimits

fn clone(&self) -> ServerLimits

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for ServerLimits

impl Debug for ServerLimits

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

Formats the value using the given formatter.

impl Default for ServerLimits

fn default() -> Self

Returns the “default value” for a type.

impl Eq for ServerLimits

impl PartialEq for ServerLimits

fn eq(&self, other: &ServerLimits) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for ServerLimits