warpdotdev/warp · warning
Cloud follow-up finished before a new session became availab
Error message
Cloud follow-up finished before a new session became available
What it means
Informational terminal error for follow-up polling: the run ended in a terminal, non-failure-like state with no status_message, and no new session ever became joinable (SessionJoinInfo::from_task never matched while InProgress). The stream ends with this message rather than a session event.
Source
Thrown at app/src/ai/ambient_agents/spawn.rs:294
status_message: task.status_message.clone(),
});
}
if task.state.is_terminal() {
if matches!(&mode, RunPollMode::Followup { .. }) {
let exhausted_stale_skips = !seen_working_state
&& skipped_stale_polls >= MAX_STALE_POLLS_BEFORE_FAILURE;
let message = if exhausted_stale_skips {
"Cloud follow-up did not start in time".to_string()
} else {
task.status_message
.as_ref()
.map(|msg| msg.message.clone())
.unwrap_or_else(|| {
if task.state.is_failure_like() {
"Cloud agent failed".to_string()
} else {
"Cloud follow-up finished before a new session became available".to_string()
}
})
};
yield Err(anyhow!(message));
}
return;
}
if task.state == AmbientAgentTaskState::InProgress
&& let Some(session_join_info) = SessionJoinInfo::from_task(&task) {
let has_new_session = match &mode {
RunPollMode::InitialRun
| RunPollMode::Followup {
previous_session_id: None,
} => true,
RunPollMode::Followup {
previous_session_id: Some(previous_session_id),
} => session_join_infoView on GitHub (pinned to e72fd7aacb)
Solutions
- Refresh the agent/task list to pick up any session created near the end of the run
- Treat as benign completion and resync local ambient-agent state instead of surfacing a hard error
- Re-trigger the follow-up if a new session was expected
- If sessions routinely appear after this error, shorten the poll interval so InProgress + join info is observed
Defensive patterns
Strategy: fallback
Try / catch
match event {
Err(e) if e.to_string().contains("before a new session became available") => {
refresh_agent_sessions().await; // benign terminal case; resync instead of erroring
}
other => handle(other),
} Prevention
- Poll fast enough that short InProgress windows with joinable session info are observed
- Treat non-failure-like terminal follow-ups as completions, not hard errors, in UX copy
- Resync the local session list whenever a follow-up stream ends
When it happens
Trigger: Follow-up task reaches a terminal non-failure state (e.g. completed) with task.status_message == None, after never yielding session join info - either it finished before a poll observed InProgress with joinable session data, or it completed without creating a session (spawn.rs:290-295, 303-304).
Common situations: Very short follow-up that completes between two polls; follow-up that ends without spawning a session (e.g. nothing to do); polling interval too coarse for fast runs.
Related errors
- Cloud follow-up did not start in time
- Cloud agent failed
- Schedule not found
- Failed to delete schedule: {}
- Unexpected non-terminal OAuth status returned
AI-assisted analysis of warpdotdev/warp@e72fd7aacb (2026-08-16).
Data as JSON: /api/errors/3b841529ef1158b3.
Report an issue: GitHub.