Skip to content

fix: exempt non-autonomous agents from heartbeat inactivity timeout #708

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Fail-Safe wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: exempt non-autonomous agents from heartbeat inactivity timeout #708

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 10 additions & 7 deletions crates/openfang-kernel/src/heartbeat.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -130,7 +130,7 @@ impl Default for RecoveryTracker {
///
/// This is a pure function — it doesn't start a background task.
/// The caller (kernel) can run this periodically or in a background task.
pub fn check_agents(registry: &AgentRegistry, config: &HeartbeatConfig) -> Vec<HeartbeatStatus> {
pub fn check_agents(registry: &AgentRegistry, _config: &HeartbeatConfig) -> Vec<HeartbeatStatus> {
let now = Utc::now();
let mut statuses = Vec::new();

Expand All @@ -143,16 +143,19 @@ pub fn check_agents(registry: &AgentRegistry, config: &HeartbeatConfig) -> Vec<H

let inactive_secs = (now - entry_ref.last_active).num_seconds();

// Determine timeout: use agent's autonomous config if set, else default
let timeout_secs = entry_ref
// Non-autonomous (reactive) agents have no inactivity timeout: they wait
// indefinitely for incoming messages and are never unresponsive due to
// idleness. Only autonomous agents need the inactivity check.
let timeout_secs: Option<i64> = entry_ref
.manifest
.autonomous
.as_ref()
.map(|a| a.heartbeat_interval_secs * UNRESPONSIVE_MULTIPLIER)
.unwrap_or(config.default_timeout_secs) as i64;
.map(|a| (a.heartbeat_interval_secs * UNRESPONSIVE_MULTIPLIER) as i64);

// Crashed agents are always considered unresponsive
let unresponsive = entry_ref.state == AgentState::Crashed || inactive_secs > timeout_secs;
let unresponsive = match timeout_secs {
Some(t) => entry_ref.state == AgentState::Crashed || inactive_secs > t,
None => entry_ref.state == AgentState::Crashed,
};

if unresponsive && entry_ref.state == AgentState::Running {
warn!(
Expand Down
Loading