From 4be2279485fc76ab604f5ce1d52bd995eb0f3b6b Mon Sep 17 00:00:00 2001
From: damocles <damocles@hyperhive.local>
Date: Mon, 13 Jul 2026 16:12:40 +0200
Subject: [PATCH] chore(#2400): skip the prebuild warm-build when the container
 is down

---
 hive-c0re/src/job_queue/exec.rs  | 21 +++++++++++++++++----
 hive-c0re/src/job_queue/model.rs |  5 ++++-
 2 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/hive-c0re/src/job_queue/exec.rs b/hive-c0re/src/job_queue/exec.rs
index 7d5dd6d4..09d8c812 100644
--- a/hive-c0re/src/job_queue/exec.rs
+++ b/hive-c0re/src/job_queue/exec.rs
@@ -104,7 +104,11 @@ pub(super) async fn run_node(coord: &Arc<Coordinator>, claim: &Claim) -> Result<
 /// Out-of-band toplevel build while the container keeps serving: meta
 /// sync + optional per-agent relock, then warm
 /// `system.build.toplevel` so the later `Swap` hits cache and skips
-/// straight to the profile-swap.
+/// straight to the profile-swap. The warm build is skipped when the
+/// container is already down: its only purpose is to shrink the swap's
+/// downtime window, so a stopped agent (no uptime to preserve) doesn't
+/// pay the double eval — `Swap` builds inline instead. The meta sync +
+/// dir prep still run unconditionally (the `Swap` depends on them).
 async fn run_prebuild(
     coord: &Arc<Coordinator>,
     claim: &Claim,
@@ -134,9 +138,18 @@ async fn run_prebuild(
             crate::meta::lock_update_for_rebuild(name).await?;
         }
     }
-    ctx.step("nix build");
-    let flake_ref = format!("{}#{name}", crate::paths::meta_root().display());
-    crate::lifecycle::prebuild_toplevel(name, &flake_ref, &|log_id| ctx.build_log(log_id)).await?;
+    // Warm the toplevel build only when the container is up — the whole
+    // point of prebuild is to shrink the swap's downtime window. A
+    // stopped agent has no uptime to preserve, so skip the (expensive)
+    // eval and let the downstream `Swap` build inline.
+    if crate::lifecycle::is_running(name).await {
+        ctx.step("nix build");
+        let flake_ref = format!("{}#{name}", crate::paths::meta_root().display());
+        crate::lifecycle::prebuild_toplevel(name, &flake_ref, &|log_id| ctx.build_log(log_id))
+            .await?;
+    } else {
+        ctx.step("skipped prebuild (container down)");
+    }
     Ok(NodeOutput::default())
 }
 
diff --git a/hive-c0re/src/job_queue/model.rs b/hive-c0re/src/job_queue/model.rs
index 684da72e..7e474785 100644
--- a/hive-c0re/src/job_queue/model.rs
+++ b/hive-c0re/src/job_queue/model.rs
@@ -68,7 +68,10 @@ pub enum NodeKind {
     /// meta `sync_agents`, optional per-agent relock, then
     /// `lifecycle::prebuild_toplevel`. `relock = false` only for
     /// meta-update cascade rebuilds (re-locking would revert the bump
-    /// the cascade just committed).
+    /// the cascade just committed). The `prebuild_toplevel` warm is
+    /// skipped when the container is already down — it only exists to
+    /// shrink the swap's downtime, which a stopped agent doesn't need
+    /// (the sync + dir prep still run; `Swap` builds inline).
     Prebuild { relock: bool },
     /// `nixos-container update` profile-swap (requires the container
     /// stopped). Re-applies nspawn flags + resource limits first —