fix(review): restore Agent Update Required flow, command field, axis alignment

Fixes from Jason's hands-on verification of the branch:

1. Old-agent messaging regression (this branch): a detail-request timeout
   went through the loud poison/reconnect flow, burying the ProcessDetails
   modal's 'Agent Update Required' message under a connection-error modal.
   Old agents IGNORE unknown messages (no late reply, no desync), so the
   optional per-PID endpoints now use quiet_reconnect(): swap the stream
   silently (still safe against merely-slow agents) and let the modal show
   its message. Only a failed reconnect surfaces loudly. Verified against
   a real v1.40.0 agent: message shows, session stays healthy.

2. Draw starvation (this branch): an agent that never answers get_metrics
   put the loop in fetch->timeout->poison->restart cycles that never
   reached the draw call — permanently blank TUI. The iteration now paints
   before fetching, and a second consecutive metrics timeout trips a
   circuit breaker: persistent 'Agent is not responding' error, recovery
   left to the manual/30s retry paths. Verified against a 0.9-era agent.

3. Command & Details pane blank (pre-existing on master): the minimal-
   refresh optimization dropped cmd/exe/cwd from the detail endpoint's
   ProcessRefreshKind, so process.cmd() had nothing to return. Restored
   with UpdateKind::OnlyIfNotSet — immutable values, read once per PID.
   Regression test added; journal E2E re-verified (100 entries render).

4. Scatter-plot axis misalignment: Y labels used a fixed 4-char field from
   the era when CPU times were 1000x too small; honest millisecond values
   (e.g. 136114) blew through it. Labels now right-align to the widest
   value per frame and X labels/titles share the dynamic padding.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

socktop/src/app.rs

@@ -94,6 +94,12 @@ pub struct App {
     last_net_totals: Option<(u64, u64, Instant)>,
     // Agent-side sample timestamp of the previous snapshot (1.51+ agents).
     last_net_sampled_at_ms: Option<u64>,
+
+    // Consecutive metrics-request timeouts. One timeout gets a silent stream
+    // refresh; a second in a row means the agent accepts connections but
+    // never answers, and deserves a persistent error instead of an invisible
+    // reconnect loop that starves the UI.
+    consecutive_request_timeouts: u32,
     rx_hist: VecDeque<u64>,
     tx_hist: VecDeque<u64>,
     rx_peak: u64,
@@ -195,6 +201,7 @@ impl App {
             per_core_hist: PerCoreHistory::new(60),
             last_net_totals: None,
             last_net_sampled_at_ms: None,
+            consecutive_request_timeouts: 0,
             rx_hist: VecDeque::with_capacity(600),
             tx_hist: VecDeque::with_capacity(600),
             rx_peak: 0,
@@ -374,6 +381,30 @@ impl App {
         self.retry_connection().await;
     }
 
+    /// Replace the connection WITHOUT any modal or state churn.
+    ///
+    /// For timeouts on the optional per-process endpoints: an old agent
+    /// ignores those messages entirely (no late reply, so no desync), but a
+    /// merely-slow agent would desync the stream — indistinguishable at
+    /// timeout time, so we still swap to a fresh stream, silently. The
+    /// ProcessDetails modal keeps showing its "Agent Update Required"
+    /// message instead of being buried under a connection-error modal.
+    /// Only a failed reconnect (connection genuinely dead) surfaces loudly.
+    async fn quiet_reconnect(&mut self) {
+        let tls_ca_ref = self.tls_ca.as_deref();
+        match self
+            .try_connect(&self.ws_url, tls_ca_ref, self.verify_hostname)
+            .await
+        {
+            Ok(ws) => {
+                self.replacement_connection = Some(ws);
+            }
+            Err(e) => {
+                self.show_connection_error(format!("Reconnect failed: {e}"));
+            }
+        }
+    }
+
     /// Mark connection as successful and dismiss any error modals
     pub fn mark_connected(&mut self) {
         if self.connection_state != ConnectionState::Connected {
@@ -1053,6 +1084,13 @@ impl App {
                 break;
             }
 
+            // Paint the current state BEFORE fetching: a request can stall for
+            // the full 5s timeout, and an iteration that ends in a poisoned-
+            // stream restart never reaches the draw at the bottom — without
+            // this, an agent that never answers left the screen permanently
+            // blank. (ratatui diffs make an unchanged repaint nearly free.)
+            terminal.draw(|f| self.draw(f))?;
+
             // Fetch and update. Skipped while disconnected — the retry paths
             // (manual 'r' or the 30s auto-retry) own recovery, and hammering a
             // dead socket with 5s-timeout requests would stall the loop. The
@@ -1061,10 +1099,22 @@ impl App {
             if self.connection_state == ConnectionState::Connected {
                 match timeout(REQUEST_TIMEOUT, ws.request(AgentRequest::Metrics)).await {
                     Err(_) => {
-                        self.poison_connection("Metrics request timed out").await;
+                        self.consecutive_request_timeouts += 1;
+                        if self.consecutive_request_timeouts >= 2 {
+                            // The agent accepts connections but never answers
+                            // (wrong protocol era, or wedged): reconnecting
+                            // can't help, so surface a persistent error and
+                            // leave recovery to the manual/auto retry paths.
+                            self.show_connection_error(
+                                "Agent is not responding to requests".to_string(),
+                            );
+                        } else {
+                            self.poison_connection("Metrics request timed out").await;
+                        }
                     }
                     Ok(Ok(AgentResponse::Metrics(m))) => {
                         self.mark_connected(); // Mark as connected on successful request
+                        self.consecutive_request_timeouts = 0;
                         self.update_with_metrics(m);
 
                         // Only poll processes every 2s
@@ -1209,14 +1259,14 @@ impl App {
                                             self.process_details_unsupported = true;
                                         }
                                         Err(_) => {
-                                            // No reply at all: mark unsupported AND
+                                            // No reply at all: old agents IGNORE
-                                            // reconnect — a late reply would desync
+                                            // this message, so show the "Agent
-                                            // every later request on this stream.
+                                            // Update Required" state and refresh
+                                            // the stream quietly (a merely-slow
+                                            // agent's late reply would otherwise
+                                            // desync it).
                                             self.process_details_unsupported = true;
-                                            self.poison_connection(
+                                            self.quiet_reconnect().await;
-                                                "Process details request timed out",
-                                            )
-                                            .await;
                                         }
                                         Ok(Ok(_)) => {
                                             // Wrong response type
@@ -1244,9 +1294,9 @@ impl App {
                                             self.journal_entries = Some(journal);
                                         }
                                         Err(_) => {
-                                            // No reply: poison the stream (see above).
+                                            // No reply: same quiet stream refresh
-                                            self.poison_connection("Journal request timed out")
+                                            // as the details endpoint above.
-                                                .await;
+                                            self.quiet_reconnect().await;
                                         }
                                         Ok(Err(_)) | Ok(Ok(_)) => {
                                             // Endpoint unsupported or wrong type;

socktop/src/ui/modal_process.rs

@@ -624,16 +624,29 @@ impl ModalManager {
         // labels + axis title + (top) Y-axis title + legend + spacing.
         let mut lines: Vec<Line> = Vec::with_capacity(plot_height + 6);
 
-        // Y-axis labels and plot content
+        // Format a CPU-time value: whole ms once past 100, one decimal below.
-        let mut row_buf = String::with_capacity(plot_width);
+        let fmt_ms = |v: f64| {
-        for y in 0..plot_height {
+            if v >= 100.0 {
-            let y_value = params.max_system * (1.0 - (y as f64 / (plot_height - 1).max(1) as f64));
+                format!("{v:.0}")
-            // 4-char fixed-width label so the axis doesn't shift as digits change.
-            let y_label = if y_value >= 100.0 {
-                format!("{y_value:>4.0}")
             } else {
-                format!("{y_value:>4.1}")
+                format!("{v:.1}")
-            };
+            }
+        };
+
+        // Y-axis labels, right-aligned to the widest value this frame so the
+        // axis stays a straight line. The old fixed 4-char field predates the
+        // CPU-time unit fix; honest millisecond values (e.g. 136114) blew
+        // through it and skewed the whole axis.
+        let y_values: Vec<String> = (0..plot_height)
+            .map(|y| {
+                fmt_ms(params.max_system * (1.0 - (y as f64 / (plot_height - 1).max(1) as f64)))
+            })
+            .collect();
+        let y_label_w = y_values.iter().map(|s| s.len()).max().unwrap_or(4).max(4);
+
+        let mut row_buf = String::with_capacity(plot_width);
+        for (y, y_value) in y_values.iter().enumerate() {
+            let y_label = format!("{y_value:>y_label_w$}");
 
             // Build the row's char slice into a reusable String buffer.
             row_buf.clear();
@@ -650,8 +663,8 @@ impl ModalManager {
             ]));
         }
 
-        // Add X-axis
+        // Add X-axis (padding = Y label width + the space before the bar)
-        let x_axis_padding = "     ".to_string(); // Match Y-axis label width
+        let x_axis_padding = " ".repeat(y_label_w + 1);
         let x_axis_line = "─".repeat(plot_width + 1);
         lines.push(Line::from(vec![
             Span::styled(x_axis_padding, Style::default()),
@@ -659,13 +672,14 @@ impl ModalManager {
         ]));
 
         // Add X-axis labels
-        let x_label_start = "0.0".to_string();
+        let x_label_start = fmt_ms(0.0);
-        let x_label_mid = format!("{:.1}", params.max_user / 2.0);
+        let x_label_mid = fmt_ms(params.max_user / 2.0);
-        let x_label_end = format!("{:.1}", params.max_user);
+        let x_label_end = fmt_ms(params.max_user);
 
         let spacing = plot_width / 3;
         let x_labels = format!(
-            "     {}{}{}{}{}",
+            "{}{}{}{}{}{}",
+            " ".repeat(y_label_w + 1),
             x_label_start,
             " ".repeat(spacing.saturating_sub(x_label_start.len())),
             x_label_mid,
@@ -677,7 +691,7 @@ impl ModalManager {
 
         // Add axis titles with better visibility
         lines.push(Line::from(vec![Span::styled(
-            "     User CPU Time (ms) →",
+            format!("{}User CPU Time (ms) →", " ".repeat(y_label_w + 1)),
             Style::default()
                 .fg(Color::Yellow)
                 .add_modifier(Modifier::BOLD),

socktop_agent/src/metrics.rs

@@ -1161,10 +1161,17 @@ pub async fn collect_process_metrics(
     system.refresh_processes_specifics(
         ProcessesToUpdate::Some(&[sysinfo::Pid::from_u32(pid)]),
         false,
+        // cmd/exe/cwd feed the modal's Command & Details pane. They're
+        // immutable per process, so OnlyIfNotSet reads them once per PID and
+        // serves the cache afterwards — the "minimal refresh" optimization
+        // had dropped them entirely, leaving the pane blank.
         ProcessRefreshKind::nothing()
             .with_memory()
             .with_cpu()
-            .with_disk_usage(),
+            .with_disk_usage()
+            .with_cmd(sysinfo::UpdateKind::OnlyIfNotSet)
+            .with_exe(sysinfo::UpdateKind::OnlyIfNotSet)
+            .with_cwd(sysinfo::UpdateKind::OnlyIfNotSet),
     );
 
     let process = system

socktop_agent/tests/process_details.rs

@@ -87,3 +87,19 @@ async fn test_collect_journal_entries_invalid_pid() {
         }
     }
 }
+
+/// The Command & Details pane went blank when the minimal-refresh
+/// optimization dropped cmd from the detail endpoint's refresh kind.
+#[tokio::test]
+async fn test_process_metrics_include_command() {
+    let state = AppState::new();
+    let pid = std::process::id();
+    let resp = collect_process_metrics(pid, &state)
+        .await
+        .expect("collect self");
+    assert!(
+        !resp.process.command.is_empty(),
+        "command should not be empty for self (cmdline is always readable)"
+    );
+    println!("command = {}", resp.process.command);
+}