Struct FailoverResetEvent 

#[non_exhaustive]
pub struct FailoverResetEvent {
    pub failed_addr: Endpoint,
    pub new_addr: Endpoint,
    pub new_server_info: Option<ServerInfo>,
    pub new_request_id: i64,
    pub attempts: u32,
    pub trigger: Error,
    pub elapsed: Duration,
}

Notification delivered to the ReaderQuery::on_failover_reset callback right before replayed batches start arriving on a new connection. Mirrors the Java onFailoverReset(newNode) contract: the user-side handler is responsible for discarding any rows it had accumulated from the previous (now-dead) connection, since the query restarts from batch_seq=0 against the new endpoint.

Marked #[non_exhaustive] so we can add fields without breaking downstream pattern matches.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

failed_addr: Endpoint

Endpoint that just failed. Use failed_addr.host / failed_addr.port directly; the Endpoint struct replaces the older (String, u16) tuple.

The address-list index is deliberately not exposed: indices are brittle if the connect string is reordered between runs, and the endpoint host/port is stable.

new_addr: Endpoint

Endpoint of the new connection.

new_server_info: Option<ServerInfo>

SERVER_INFO of the new endpoint (None only if the server omitted it).

new_request_id: i64

Newly-allocated request_id the cursor will receive frames for from now on. Different from Cursor::request_id before the failover.

attempts: u32

Count of reconnect dials the current failover cycle burned before this success. 1 means the first reconnect dial succeeded and its replay write went through cleanly. Larger values mean earlier dials in this cycle missed (rotating through endpoints) before one landed. Pairs with elapsed — both measure the same failover event.

trigger: Error

The error that triggered this failover (the failure of the previous connection). The full error — code + message — is preserved so callers can both route on the ErrorCode (for metrics / categorization) and log the raw message (for diagnostics: errno text on SocketError, peer info on TlsError, decode-site detail on ProtocolError, etc.). Use Error::code to extract just the category.

Without this, the cause-of-death of the previous connection is lost forever once failover succeeds — it’s not re-surfaced as Err anywhere else in the cursor’s API.

elapsed: Duration

Wall-clock time spent reconnecting (sleep + dial + handshake + SERVER_INFO read). Excludes the time from the cursor’s last successful read until the failure was observed.

Trait Implementations

impl Clone for FailoverResetEvent

fn clone(&self) -> FailoverResetEvent

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for FailoverResetEvent

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

Formats the value using the given formatter.