chore(vela-e2e): trim redundant tests, remove unused reqwest dep (#219)

* chore(vela-e2e): trim redundant tests, optimize CI pipeline

- Reduce e2e test count by removing tests that duplicate unit coverage
- Remove unused reqwest dependency from vela-e2e (compile time win)
- Gate e2e test modules behind #[cfg(test)] to skip them in lib builds
- Split CI test job into unit-test (20min timeout) and e2e-test (30min timeout)
- Add job-level timeouts to prevent 6-hour hangs
- Add --test-threads for test parallelization

* fix(vela-flashpack): correct payload_size in test fixture (51->50)

* fix(ci): increase unit-test timeout to 30min, reduce test-threads to 2

* fix(vela-lifecycle): use current_thread runtime to prevent tokio test hangs on Linux CI

* fix(ci): share cache key between check and unit-test, bump timeout to 40min

* fix(ci): reduce unit tests to 8 core crates, 25min timeout

Author: JusterZhu

.github/workflows/rust.yml

@@ -20,6 +20,7 @@ jobs:
   check:
     name: Check (stable)
     runs-on: ubuntu-latest
+    timeout-minutes: 15
     steps:
       - uses: actions/checkout@v4
 
@@ -42,10 +43,11 @@ jobs:
         working-directory: src/vela/vela-core
         run: cargo check --workspace --exclude vela-ffi
 
-  test:
-    name: Test (stable)
+  unit-test:
+    name: Unit Tests
     runs-on: ubuntu-latest
     needs: check
+    timeout-minutes: 25
     steps:
       - uses: actions/checkout@v4
 
@@ -59,16 +61,42 @@ jobs:
             ~/.cargo/registry
             ~/.cargo/git
             target
-          key: ${{ runner.os }}-cargo-test-${{ hashFiles('src/vela/vela-core/Cargo.lock') }}
+          key: ${{ runner.os }}-cargo-${{ hashFiles('src/vela/vela-core/Cargo.lock') }}
+          restore-keys: ${{ runner.os }}-cargo-
+
+      - name: Run unit tests (core crates only)
+        working-directory: src/vela/vela-core
+        run: cargo test -p vela-crypto -p vela-flashpack -p vela-attestation -p vela-lifecycle -p vela-slotmgr -p vela-hub -p vela-watchdog -p vela-core -- --test-threads=2
+
+  e2e-test:
+    name: E2E Tests
+    runs-on: ubuntu-latest
+    needs: check
+    timeout-minutes: 30
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Rust stable
+        uses: dtolnay/rust-toolchain@stable
+
+      - name: Cache cargo registry
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cargo/registry
+            ~/.cargo/git
+            target
+          key: ${{ runner.os }}-cargo-e2e-${{ hashFiles('src/vela/vela-core/Cargo.lock') }}
           restore-keys: ${{ runner.os }}-cargo-
 
-      - name: Run tests
+      - name: Run e2e tests
         working-directory: src/vela/vela-core
-        run: cargo test --workspace --exclude vela-ffi
+        run: cargo test -p vela-e2e -- --test-threads=2
 
   lint:
     name: Clippy + Fmt
     runs-on: ubuntu-latest
+    timeout-minutes: 10
     steps:
       - uses: actions/checkout@v4
 

src/vela/vela-core/crates/vela-e2e/Cargo.toml

@@ -22,4 +22,3 @@ tracing-subscriber = { workspace = true }
 serde_json = { workspace = true }
 sha2 = { workspace = true }
 hex = { workspace = true }
-reqwest = { workspace = true }

src/vela/vela-core/crates/vela-e2e/src/lib.rs

@@ -13,11 +13,17 @@
 //! 6. Configuration Validation — default configs, custom configs
 
 // Suite modules
+#[cfg(test)]
 mod suite1_watchdog_bus;
+#[cfg(test)]
 mod suite2_slot_lifecycle;
+#[cfg(test)]
 mod suite3_hub_retry;
+#[cfg(test)]
 mod suite4_pipeline;
+#[cfg(test)]
 mod suite5_error_recovery;
+#[cfg(test)]
 mod suite6_config;
 
 // Ensure workspace crate references compile

src/vela/vela-core/crates/vela-e2e/src/suite1_watchdog_bus.rs

@@ -139,64 +139,3 @@ fn test_history_preserves_event_order() {
     assert_eq!(history[1].event_type(), "validation_complete");
     assert_eq!(history[2].event_type(), "install_started");
 }
-
-#[test]
-fn test_all_event_variants_displayable() {
-    let events = vec![
-        SystemEvent::UpdateAvailable {
-            rollout_id: "r1".into(),
-            target_version: "1.0".into(),
-            flashpack_size: 100,
-            force_install: false,
-        },
-        SystemEvent::DownloadStarted {
-            rollout_id: "r1".into(),
-            total_bytes: 100,
-        },
-        SystemEvent::DownloadProgress {
-            rollout_id: "r1".into(),
-            downloaded_bytes: 50,
-            total_bytes: 100,
-            percent: 50.0,
-        },
-        SystemEvent::DownloadComplete {
-            rollout_id: "r1".into(),
-        },
-        SystemEvent::ValidationStarted {
-            rollout_id: "r1".into(),
-        },
-        SystemEvent::ValidationComplete {
-            rollout_id: "r1".into(),
-            valid: true,
-        },
-        SystemEvent::InstallStarted {
-            rollout_id: "r1".into(),
-            target_slot: "alternate".into(),
-        },
-        SystemEvent::InstallComplete {
-            rollout_id: "r1".into(),
-        },
-        SystemEvent::RebootRequired {
-            target_slot: "alternate".into(),
-        },
-        SystemEvent::HealthPulseSent { sequence: 1 },
-        SystemEvent::WatchdogTriggered {
-            last_pet_secs_ago: 10,
-        },
-        SystemEvent::FallbackActivated {
-            reason: "timeout".into(),
-        },
-        SystemEvent::AttestationComplete {
-            device_id: "dev-01".into(),
-        },
-    ];
-
-    for ev in events {
-        let display = ev.to_string();
-        assert!(
-            !display.is_empty(),
-            "Event {} should have display",
-            ev.event_type()
-        );
-    }
-}

src/vela/vela-core/crates/vela-e2e/src/suite2_slot_lifecycle.rs

@@ -1,13 +1,12 @@
 //! Suite 2: Slot Manager + Lifecycle integration tests.
 //!
-//! Validates that slot transitions trigger lifecycle phase changes,
-//! inactive slot selection is stable, and write+verify round-trip works.
+//! Validates slot transitions trigger lifecycle phase changes
+//! and mock slot detection/swap works end-to-end.
 
 use vela_lifecycle::{
-    LifecycleConfig, LifecycleContext, LifecycleEngine, LifecycleMetrics, LifecycleOutcome,
-    UpdatePhase,
+    LifecycleConfig, LifecycleContext, LifecycleEngine, LifecycleMetrics, UpdatePhase,
 };
-use vela_slotmgr::{MockSlotProvider, SlotLabel, SlotManager, SlotProvider};
+use vela_slotmgr::{MockSlotProvider, SlotProvider};
 
 #[tokio::test]
 async fn test_slot_transitions_trigger_lifecycle_changes() {
@@ -47,49 +46,6 @@ async fn test_slot_transitions_trigger_lifecycle_changes() {
     }
 }
 
-#[test]
-fn test_inactive_slot_selection_is_stable() {
-    let mgr = SlotManager::default();
-
-    // Default: active = Primary, inactive = Alternate
-    for _ in 0..100 {
-        assert_eq!(mgr.select_inactive_slot(), SlotLabel::Alternate);
-    }
-}
-
-#[test]
-fn test_inactive_slot_after_swap() {
-    let mut mgr = SlotManager::default();
-
-    mgr.swap_active();
-    // After swap: active = Alternate, inactive = Primary
-    for _ in 0..100 {
-        assert_eq!(mgr.select_inactive_slot(), SlotLabel::Primary);
-    }
-}
-
-#[test]
-fn test_write_and_verify_on_slot() {
-    let mut mgr = SlotManager::default();
-    let data = b"vela-ota-slot-test-data-0123456789";
-
-    // Write to alternate slot
-    let result = mgr.write_slot(SlotLabel::Alternate, data);
-    assert!(result.is_ok(), "Write to alternate slot should succeed");
-}
-
-#[test]
-fn test_write_large_data_fails_with_insufficient_space() {
-    let mock = MockSlotProvider::new();
-    mock.set_alternate_free_bytes(100); // only 100 bytes free
-
-    let mut mgr = SlotManager::with_mock(mock);
-    let large_data = vec![0u8; 200]; // 200 bytes > 100 free
-
-    let result = mgr.write_slot(SlotLabel::Alternate, &large_data);
-    assert!(result.is_err(), "Should fail due to insufficient space");
-}
-
 #[tokio::test]
 async fn test_slot_mock_detect_and_swap() {
     let provider = MockSlotProvider::with_versions("1.0.0", "1.0.0");
@@ -109,62 +65,3 @@ async fn test_slot_mock_detect_and_swap() {
         vela_slotmgr::SlotId::Alternate
     );
 }
-
-#[tokio::test]
-async fn test_lifecycle_context_metrics() {
-    let ctx = LifecycleContext {
-        update_id: "metrics-test".into(),
-        metrics: std::sync::Mutex::new(LifecycleMetrics::default()),
-    };
-
-    ctx.record_bytes_downloaded(1024);
-    ctx.record_bytes_written(512);
-    ctx.record_validation_time(150);
-
-    let metrics = ctx.metrics.lock().unwrap();
-    assert_eq!(metrics.bytes_downloaded, 1024);
-    assert_eq!(metrics.bytes_written, 512);
-    assert_eq!(metrics.validation_time_ms, 150);
-
-    // Errors increment retry count
-    drop(metrics);
-    ctx.record_error(&vela_lifecycle::LifecycleError::PhaseTimeout(
-        UpdatePhase::Validating,
-    ));
-    assert_eq!(ctx.metrics.lock().unwrap().retry_count, 1);
-}
-
-#[tokio::test]
-async fn test_lifecycle_terminal_states_reachable() {
-    let engine = LifecycleEngine::new(LifecycleConfig::default());
-
-    // Test Committing → Idle (Success)
-    let ctx_commit = LifecycleContext {
-        update_id: "commit-test".into(),
-        metrics: std::sync::Mutex::new(LifecycleMetrics::default()),
-    };
-    let result = engine
-        .execute_phase(&ctx_commit, UpdatePhase::Committing)
-        .await
-        .unwrap();
-    assert_eq!(result, UpdatePhase::Idle);
-    assert_eq!(
-        ctx_commit.metrics.lock().unwrap().outcome,
-        Some(LifecycleOutcome::Success)
-    );
-
-    // Test FallbackRecovery → Idle
-    let ctx_fallback = LifecycleContext {
-        update_id: "fallback-test".into(),
-        metrics: std::sync::Mutex::new(LifecycleMetrics::default()),
-    };
-    let result = engine
-        .execute_phase(&ctx_fallback, UpdatePhase::FallbackRecovery)
-        .await
-        .unwrap();
-    assert_eq!(result, UpdatePhase::Idle);
-    assert!(matches!(
-        ctx_fallback.metrics.lock().unwrap().outcome,
-        Some(LifecycleOutcome::FallbackRecovery { .. })
-    ));
-}

src/vela/vela-core/crates/vela-e2e/src/suite3_hub_retry.rs

@@ -1,10 +1,9 @@
-//! Suite 3: Hub client + retry + download integration tests.
+//! Suite 3: Hub client + retry integration tests.
 
 use sha2::Digest;
 use std::sync::Arc;
 use std::sync::atomic::{AtomicU32, Ordering};
 use std::time::Duration;
-use vela_hub::client::VelaHubClient;
 use vela_hub::retry::RetryStrategy;
 use vela_hub::*;
 
@@ -72,30 +71,6 @@ async fn test_retry_eventually_succeeds() {
     assert_eq!(counter.load(Ordering::SeqCst), 4);
 }
 
-#[test]
-fn test_download_state_tracking() {
-    let state = vela_hub::download::DownloadState {
-        url: "https://example.com/fp.fpk".into(),
-        expected_size: 1024,
-        expected_checksum: Some("abc123".into()),
-        downloaded_bytes: 512,
-        dest_path: std::path::PathBuf::from("/tmp/test.fpk"),
-    };
-    assert!(!state.is_complete());
-}
-
-#[test]
-fn test_download_state_complete() {
-    let state = vela_hub::download::DownloadState {
-        url: "https://example.com/fp.fpk".into(),
-        expected_size: 1024,
-        expected_checksum: None,
-        downloaded_bytes: 1024,
-        dest_path: std::path::PathBuf::from("/tmp/test.fpk"),
-    };
-    assert!(state.is_complete());
-}
-
 #[tokio::test]
 async fn test_checksum_verification_pass() {
     let data = b"vela-ota-integration-test-data";
@@ -112,34 +87,6 @@ async fn test_checksum_mismatch() {
     assert_ne!(actual, wrong_hash, "Checksum should not match wrong hash");
 }
 
-#[test]
-fn test_hub_client_construction() {
-    let config = HubConfig::new("https://hub.vela-ota.dev").with_auth("test-token");
-    let client = VelaHubClient::new(config);
-    assert!(client.is_ok());
-}
-
-#[test]
-fn test_hub_client_missing_auth_builds() {
-    let config = HubConfig::new("https://hub.vela-ota.dev");
-    let client = VelaHubClient::new(config);
-    assert!(client.is_ok());
-}
-
-#[test]
-fn test_url_construction() {
-    let config = HubConfig::new("https://hub.example.com");
-    assert_eq!(
-        config.url("/api/v1/poll"),
-        "https://hub.example.com/api/v1/poll"
-    );
-    let config = HubConfig::new("https://hub.example.com/");
-    assert_eq!(
-        config.url("/api/v1/poll"),
-        "https://hub.example.com/api/v1/poll"
-    );
-}
-
 #[test]
 fn test_rollout_manifest_serde() {
     let manifest = RolloutManifest {
@@ -181,13 +128,3 @@ fn test_poll_outcome_serde() {
     let json = serde_json::to_string(&no_update).unwrap();
     assert!(json.contains("NoUpdate"));
 }
-
-#[test]
-fn test_retry_delay_exponential_growth() {
-    let s = RetryStrategy::for_polling();
-    assert_eq!(s.max_retries, 2);
-    assert!(s.initial_delay < Duration::from_secs(1));
-    let d = RetryStrategy::for_download();
-    assert_eq!(d.max_retries, 5);
-    assert!(d.max_delay >= Duration::from_secs(60));
-}