What causes "daemon sent a lag notice as a history event" in atuinsh/atuin?

Restart the daemon so client and server protocol versions match

daemon sent a lag notice as a history event

Error message

daemon sent a lag notice as a history event

What it means

TailHistoryReply::from_proto converts gRPC tail events into typed history events. A Lagged event is a valid tail-stream signal but must be handled as its own kind, not as a history event; if the daemon (or a desynced client) delivers it through the history-event path, this invariant is violated and the conversion bails.

Source

Thrown at crates/atuin/src/command/client/history.rs:702

    #[serde(skip_serializing_if = "Option::is_none")]
    duration_ns: Option<i64>,
    #[serde(skip_serializing_if = "Option::is_none")]
    duration: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    success: Option<bool>,
    #[serde(skip_serializing_if = "Option::is_none")]
    finished_at: Option<String>,
}

#[cfg(feature = "daemon")]
impl TailEvent {
    fn from_proto(reply: TailHistoryReply) -> Result<Self> {
        let (kind, history) = match reply.event {
            Some(TailEventProto::Started(history)) => (TailKind::Started, history),
            Some(TailEventProto::Ended(history)) => (TailKind::Ended, history),
            Some(TailEventProto::Cancelled(history)) => (TailKind::Cancelled, history),
            Some(TailEventProto::Lagged(_)) => {
                bail!("daemon sent a lag notice as a history event")
            }
            None => bail!("daemon sent an unspecified history tail event"),
        };
        let timestamp = OffsetDateTime::from_unix_nanos_i64(history.timestamp);
        let author_kind = history.author_kind();

        Ok(Self {
            kind,
            history: History {
                id: history
                    .id
                    .ok_or_else(|| eyre::eyre!("daemon tail event is missing the history id"))?
                    .try_into()?,
                timestamp,
                duration: history.duration,
                exit: history.exit,
                command: history.command,
                cwd: history.cwd,

View on GitHub (pinned to c0c717ab04)

Solutions

Restart the daemon so client and server protocol versions match
Upgrade both CLI and daemon to the same atuin version
Report as a bug if it persists with matched versions

Defensive patterns

Strategy: try-catch

Try / catch

match TailHistoryReply::from_proto(reply) {
    Err(e) if e.to_string().contains("lag notice as a history event") => {
        // handle Lagged events on the lag path instead; restart daemon to fix skew
    }
    other => other?,
}

Prevention

Run matching CLI/daemon versions
Handle Lagged as its own event kind in stream consumers
Keep the tail-stream decode paths synchronized with the proto contract

When it happens

Trigger: A daemon sends TailEventProto::Lagged through the history-event decode path in the tail stream, i.e. a malformed or out-of-contract stream message.

Common situations: CLI/daemon protocol skew where the daemon packs lag notices differently; bugs in a newer/older daemon; custom or patched daemon builds.

Related errors

AI-assisted analysis of atuinsh/atuin@c0c717ab04 (2026-09-12). Data as JSON: /api/errors/ec506f6fa0896b53. Report an issue: GitHub.