Fix heartbeat detection to properly detect offline hosts
All checks were successful
Build and Release / build-and-release (push) Successful in 2m34s

- Add independent heartbeat checking timer (1 second interval) separate from metric reception
- Move get_connected_hosts() call outside metric receive condition to run periodically
- Remove duplicate update_hosts() call from metric processing to avoid redundancy
- Ensure offline host detection works even when no new metrics are received
- Fix issue where hosts going offline were never detected due to conditional heartbeat check
- Heartbeat timeouts now properly detected within configured timeout + 1 second
- Bump version to 0.1.62
This commit is contained in:
Christoffer Martinsson 2025-11-07 14:27:03 +01:00
parent 3e3d3f0c2b
commit c980346d05
5 changed files with 21 additions and 13 deletions

6
Cargo.lock generated
View File

@ -270,7 +270,7 @@ checksum = "a1d728cc89cf3aee9ff92b05e62b19ee65a02b5702cff7d5a377e32c6ae29d8d"
[[package]] [[package]]
name = "cm-dashboard" name = "cm-dashboard"
version = "0.1.60" version = "0.1.61"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"chrono", "chrono",
@ -292,7 +292,7 @@ dependencies = [
[[package]] [[package]]
name = "cm-dashboard-agent" name = "cm-dashboard-agent"
version = "0.1.60" version = "0.1.61"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"async-trait", "async-trait",
@ -315,7 +315,7 @@ dependencies = [
[[package]] [[package]]
name = "cm-dashboard-shared" name = "cm-dashboard-shared"
version = "0.1.60" version = "0.1.61"
dependencies = [ dependencies = [
"chrono", "chrono",
"serde", "serde",

View File

@ -1,6 +1,6 @@
[package] [package]
name = "cm-dashboard-agent" name = "cm-dashboard-agent"
version = "0.1.61" version = "0.1.62"
edition = "2021" edition = "2021"
[dependencies] [dependencies]

View File

@ -1,6 +1,6 @@
[package] [package]
name = "cm-dashboard" name = "cm-dashboard"
version = "0.1.61" version = "0.1.62"
edition = "2021" edition = "2021"
[dependencies] [dependencies]

View File

@ -149,6 +149,8 @@ impl Dashboard {
let mut last_metrics_check = Instant::now(); let mut last_metrics_check = Instant::now();
let metrics_check_interval = Duration::from_millis(100); // Check for metrics every 100ms let metrics_check_interval = Duration::from_millis(100); // Check for metrics every 100ms
let mut last_heartbeat_check = Instant::now();
let heartbeat_check_interval = Duration::from_secs(1); // Check for host connectivity every 1 second
loop { loop {
// Handle terminal events (keyboard input) only if not headless // Handle terminal events (keyboard input) only if not headless
@ -254,14 +256,8 @@ impl Dashboard {
} }
} }
// Update TUI with new hosts and metrics (only if not headless) // Update TUI with new metrics (only if not headless)
if let Some(ref mut tui_app) = self.tui_app { if let Some(ref mut tui_app) = self.tui_app {
let connected_hosts = self
.metric_store
.get_connected_hosts(Duration::from_secs(self.config.zmq.heartbeat_timeout_seconds));
tui_app.update_hosts(connected_hosts);
tui_app.update_metrics(&self.metric_store); tui_app.update_metrics(&self.metric_store);
} }
} }
@ -280,6 +276,18 @@ impl Dashboard {
last_metrics_check = Instant::now(); last_metrics_check = Instant::now();
} }
// Check for host connectivity changes (heartbeat timeouts) periodically
if last_heartbeat_check.elapsed() >= heartbeat_check_interval {
if let Some(ref mut tui_app) = self.tui_app {
let connected_hosts = self
.metric_store
.get_connected_hosts(Duration::from_secs(self.config.zmq.heartbeat_timeout_seconds));
tui_app.update_hosts(connected_hosts);
}
last_heartbeat_check = Instant::now();
}
// Render TUI (only if not headless) // Render TUI (only if not headless)
if !self.headless { if !self.headless {
if let Some(ref mut terminal) = self.terminal { if let Some(ref mut terminal) = self.terminal {

View File

@ -1,6 +1,6 @@
[package] [package]
name = "cm-dashboard-shared" name = "cm-dashboard-shared"
version = "0.1.61" version = "0.1.62"
edition = "2021" edition = "2021"
[dependencies] [dependencies]