chore: build-time quick wins — rsp DAG rebuild fix, CI queue hygiene, symbols retention, LF scripts

.gitattributes

@@ -29,3 +29,13 @@ Explorer/Assets/StreamingAssets/AssetBundles/**/*.meta -filter=lfs -diff=lfs -me
 *.asset binary
 *.controller binary
 *.asset text diff=yaml merge=yaml
+
+# Root-level compiler rsp files must check out byte-identical to what
+# scripts/generate-ignore-warnings.sh writes (LF), or its drift check rewrites
+# them on CRLF-converting builders and re-triggers a Bee DAG rebuild.
+Explorer/Assets/*.rsp text eol=lf
+
+# Shell scripts must check out LF everywhere: CRLF-converting Windows builders
+# fail them with "$'\r': command not found" (seen on every Windows cloud build
+# running the UBA preBuildScript).
+*.sh text eol=lf

.github/workflows/build-release-main-page.yml

@@ -56,7 +56,9 @@ jobs:
         uses: dawidd6/action-download-artifact@v6
         with:
           run_id: ${{ github.event.workflow_run.id }}
-          name: Decentraland_.*
+          # Anchored to the four player zips: Decentraland_.* also matched the multi-GB
+          # *_debug_symbols artifacts, downloading them only to discard them.
+          name: Decentraland_(windows64|macos)(_epic)?$
           name_is_regexp: true
           skip_unpack: true
 

.github/workflows/build-unitycloud.yml

@@ -573,6 +573,7 @@ jobs:
           POLL_TIME: 60
           QUEUE_POLL_TIME: 120
           STALE_POLL_THRESHOLD: 600
+          LOG_STALL_THRESHOLD: 900
           QUEUE_TIMEOUT: 14400
           BUILD_TIMEOUT: 10800
           TARGET: t_${{ matrix.target }}
@@ -599,6 +600,19 @@ jobs:
           PARAM_IS_RELEASE_BUILD: ${{ inputs.is_release_build }}
           PARAM_UNITY_EXTRA_PARAMS: '-disable-assembly-updater'
 
+      # Must run before any always() log/report steps: on cancellation GitHub grants a 5-minute
+      # grace window, and the DELETE that frees the Unity-side queue slot has to win that race
+      # (a large-log upload ahead of it can eat the whole window). Also runs on failure() so an
+      # outer-step timeout doesn't leave a Unity-side build holding a slot.
+      - name: Cancel Unity Cloud build
+        if: ${{ cancelled() || failure() }}
+        timeout-minutes: 2
+        env:
+          API_KEY: ${{ secrets.UNITY_CLOUD_API_KEY }}
+          ORG_ID: ${{ secrets.UNITY_CLOUD_ORG_ID }}
+          PROJECT_ID: ${{ secrets.UNITY_CLOUD_PROJECT_ID }}
+        run: python -u scripts/cloudbuild/build.py --cancel || true
+
       - name: Locate Windows main executable
         if: matrix.target == 'windows64' && (inputs.is_release_build == true || github.event.inputs.force_sign == 'true')
         run: |
@@ -877,14 +891,19 @@ jobs:
             echo "[${ARTIFACT_URL}](${ARTIFACT_URL})"
           } >> "$GITHUB_STEP_SUMMARY"
 
+      # Symbol artifacts have no downstream consumer (Sentry uploads from the on-disk build/
+      # folder below; release drafting attaches only the player zips) — keep them only where a
+      # human might need post-hoc symbolication, and never at default retention.
       - name: Upload debug symbols
+        if: ${{ inputs.is_release_build == true }}
         uses: actions/upload-artifact@v6
         with:
           name: ${{ env.artifact_name }}_debug_symbols
           path: |
             build/**/*_BackUpThisFolder_ButDontShipItWithYourGame
             build/**/*_BurstDebugInformation_DoNotShip
           if-no-files-found: error
+          retention-days: 7
 
       - name: Upload debug symbols to Sentry
         if: ${{ needs.prebuild.outputs.sentry_enabled == 'true' }}
@@ -899,22 +918,22 @@ jobs:
             --project "$SENTRY_PROJECT" \
             "build"
 
-      # Will run always (even if failing)
+      # Runs even on failure, but only once a log actually exists (a run cancelled in-queue
+      # never writes one, and a guaranteed-failing upload would burn the cancellation grace window).
       - name: Upload cloud logs
-        if: always()
+        if: ${{ always() && hashFiles('unity_cloud_log.log') != '' }}
         uses: actions/upload-artifact@v6
         with:
           name: ${{ matrix.target }}_${{ needs.prebuild.outputs.install_source }}_unity_log
           path: unity_cloud_log.log
           if-no-files-found: error
 
-      # Will run always (even if failing)
       - name: Print cloud logs
-        if: always()
+        if: ${{ always() && hashFiles('unity_cloud_log.log') != '' }}
         run: cat unity_cloud_log.log
 
       - name: Extract and display errors
-        if: always()
+        if: ${{ always() && hashFiles('unity_cloud_log.log') != '' }}
         run: |
           echo "=== Extracted Errors for ${{ matrix.target }} ${{ needs.prebuild.outputs.install_source }} ==="
           grep -iE "error c|fatal" unity_cloud_log.log | sed 's/^/\x1b[31m/' | sed 's/$/\x1b[0m/' || echo "No 'error c' or 'fatal' errors found in ${{ matrix.target }} log."
@@ -931,15 +950,6 @@ jobs:
           path: shader_compilation_report.log
           if-no-files-found: warn
 
-      # Also runs on failure() so an outer-step timeout doesn't leave a Unity-side build holding a slot.
-      - name: Cancel Unity Cloud build
-        if: ${{ cancelled() || failure() }}
-        env:
-          API_KEY: ${{ secrets.UNITY_CLOUD_API_KEY }}
-          ORG_ID: ${{ secrets.UNITY_CLOUD_ORG_ID }}
-          PROJECT_ID: ${{ secrets.UNITY_CLOUD_PROJECT_ID }}
-        run: python -u scripts/cloudbuild/build.py --cancel || true
-
   build-gate:
     name: Build Gate (Windows + macOS)
     runs-on: ubuntu-latest

Explorer/Assets/csc.rsp

@@ -0,0 +1,14 @@
+-nowarn:8618
+-nowarn:8625
+-nowarn:8602
+-nowarn:8604
+-nowarn:8619
+-nowarn:8620
+-nowarn:8603
+-nowarn:8600
+-nowarn:8601
+-nowarn:0649
+-nowarn:0414
+-nowarn:0168
+-nowarn:0219
+-nowarn:8632

Explorer/Assets/gmcs.rsp

@@ -0,0 +1,14 @@
+-nowarn:8618
+-nowarn:8625
+-nowarn:8602
+-nowarn:8604
+-nowarn:8619
+-nowarn:8620
+-nowarn:8603
+-nowarn:8600
+-nowarn:8601
+-nowarn:0649
+-nowarn:0414
+-nowarn:0168
+-nowarn:0219
+-nowarn:8632

Explorer/Assets/mcs.rsp

@@ -0,0 +1,14 @@
+-nowarn:8618
+-nowarn:8625
+-nowarn:8602
+-nowarn:8604
+-nowarn:8619
+-nowarn:8620
+-nowarn:8603
+-nowarn:8600
+-nowarn:8601
+-nowarn:0649
+-nowarn:0414
+-nowarn:0168
+-nowarn:0219
+-nowarn:8632

Explorer/Assets/smcs.rsp

@@ -0,0 +1,14 @@
+-nowarn:8618
+-nowarn:8625
+-nowarn:8602
+-nowarn:8604
+-nowarn:8619
+-nowarn:8620
+-nowarn:8603
+-nowarn:8600
+-nowarn:8601
+-nowarn:0649
+-nowarn:0414
+-nowarn:0168
+-nowarn:0219
+-nowarn:8632

Explorer/Assets/us.rsp

@@ -0,0 +1,14 @@
+-nowarn:8618
+-nowarn:8625
+-nowarn:8602
+-nowarn:8604
+-nowarn:8619
+-nowarn:8620
+-nowarn:8603
+-nowarn:8600
+-nowarn:8601
+-nowarn:0649
+-nowarn:0414
+-nowarn:0168
+-nowarn:0219
+-nowarn:8632

scripts/cloudbuild/build.py

@@ -44,6 +44,10 @@ def _extract_member(self, member, targetpath, pwd):
 POLL_TIME = int(os.getenv('POLL_TIME', '60'))
 QUEUE_POLL_TIME = int(os.getenv('QUEUE_POLL_TIME', '120'))
 STALE_THRESHOLD = int(os.getenv('STALE_POLL_THRESHOLD', '600'))
+# If the build log has not grown for this many seconds while the build is active,
+# the build is presumed deadlocked and is cancelled so the retry lands on a fresh
+# builder VM.  15 min is generous enough to survive silent IL2CPP / shader phases.
+LOG_STALL_THRESHOLD = int(os.getenv('LOG_STALL_THRESHOLD', '900'))
 
 # Queue time and active build time use separate budgets so a long Unity Cloud
 # queue does not eat into the actual build window.
@@ -300,7 +304,7 @@ def cancel_build(id):
     except requests.exceptions.RequestException as e:
         print(f'Pre-cancel status check failed ({e}); attempting cancel anyway.')
 
-    response = requests.delete(f'{URL}/buildtargets/{os.getenv('TARGET')}/builds/{id}', headers=HEADERS)
+    response = requests.delete(f'{URL}/buildtargets/{os.getenv('TARGET')}/builds/{id}', headers=HEADERS, timeout=30)
 
     if response.status_code == 204:
         print('Build canceled successfully')
@@ -467,6 +471,39 @@ def download_log(id):
 
     print('Build log ready!')
 
+def get_log_byte_count(id):
+    """Return the current byte length of the build log, or None on any error.
+
+    Uses a HEAD request first (zero-body, cheapest).  If the server does not
+    honour HEAD, falls back to a single-byte Range request and reads the total
+    from the Content-Range response header.  Never downloads the full log.
+    """
+    url = f'{URL}/buildtargets/{os.getenv("TARGET")}/builds/{id}/log'
+    try:
+        resp = requests.head(url, headers=HEADERS, timeout=30)
+        if resp.status_code == 200 and 'Content-Length' in resp.headers:
+            return int(resp.headers['Content-Length'])
+        # Range fallback: fetch exactly one byte; read the total from Content-Range.
+        resp = requests.get(
+            url,
+            headers={**HEADERS, 'Range': 'bytes=0-0'},
+            timeout=30,
+            stream=True,
+        )
+        resp.close()
+        if resp.status_code in (200, 206):
+            content_range = resp.headers.get('Content-Range', '')
+            m = re.search(r'/(\d+)$', content_range)
+            if m:
+                return int(m.group(1))
+            # Server returned 200 without Content-Range → use Content-Length.
+            if 'Content-Length' in resp.headers:
+                return int(resp.headers['Content-Length'])
+    except requests.exceptions.RequestException as e:
+        print(f'Warning: log size probe failed ({e})')
+    return None
+
+
 def delete_build(id):
     response = requests.delete(f'{URL}/buildtargets/{os.getenv('TARGET')}/builds/{id}/artifacts', headers=HEADERS)
 
@@ -618,6 +655,10 @@ def run_poll_loop(id, build_already_active=False, resumed_build_elapsed=0):
     attempt can reattach via try_resume_build and keep the queue position.
     BUILD_TIMEOUT still cancels — a runaway active build should not keep
     holding a Unity Cloud slot.
+    LOG_STALL_THRESHOLD: if the build log has not grown for this many seconds
+    while the build is active, the build is cancelled and retried on a fresh
+    builder VM (exit 99).  This catches deadlocked builders that keep reporting
+    status=started while producing no output.
 
     Returns (final_outcome, phase_durations, queue_reasons, queue_elapsed, build_elapsed).
     """
@@ -630,6 +671,8 @@ def run_poll_loop(id, build_already_active=False, resumed_build_elapsed=0):
     last_status = None
     last_status_change = now
     last_poll = now
+    last_log_byte_count = None  # most recently observed log size in bytes
+    last_log_growth = None      # wall-clock time of the last log-size increase
 
     while True:
         now = time.time()
@@ -663,8 +706,32 @@ def run_poll_loop(id, build_already_active=False, resumed_build_elapsed=0):
 
         if build_start is None and status in ACTIVE_STATUSES:
             build_start = now
+            last_log_growth = now  # start the stall clock from when the build went active
             print(f'Build picked up by builder after {datetime.timedelta(seconds=int(now - queue_start))} in queue.')
 
+        # Log-stall watchdog: probe log size on every poll tick while the build
+        # is active.  A deadlocked builder keeps status=started but its log stops
+        # growing.  Cancel and retry (exit 99 → fresh builder VM) after the
+        # configured threshold.  If the probe itself fails we skip rather than
+        # false-positive cancel.
+        if status in ACTIVE_STATUSES:
+            log_bytes = get_log_byte_count(id)
+            if log_bytes is not None:
+                if last_log_byte_count is None or log_bytes > last_log_byte_count:
+                    last_log_byte_count = log_bytes
+                    last_log_growth = now
+                elif last_log_growth is not None and (now - last_log_growth) > LOG_STALL_THRESHOLD:
+                    stall_duration = datetime.timedelta(seconds=int(now - last_log_growth))
+                    print(
+                        f'Build log has not grown for {stall_duration} '
+                        f'(threshold {datetime.timedelta(seconds=LOG_STALL_THRESHOLD)}). '
+                        f'Builder appears deadlocked — cancelling and retrying on a fresh VM.'
+                    )
+                    cancel_build(id)
+                    queue_elapsed = (build_start or now) - queue_start
+                    build_elapsed = now - (build_start or now)
+                    return 'log_stall', phase_durations, queue_reasons, queue_elapsed, build_elapsed
+
         if status != last_status:
             queue_elapsed = (build_start or now) - queue_start
             build_elapsed = (now - build_start) if build_start else 0
@@ -707,6 +774,17 @@ def run_poll_loop(id, build_already_active=False, resumed_build_elapsed=0):
     id = build_info["id"]
 
     if args.cancel:
+        if id is None:
+            # The runner died between the build POST and the id write; the queued build is
+            # findable only as the target's latest non-terminal build.
+            latest = get_latest_build(os.getenv('TARGET'))
+            if latest and latest.get('buildStatus') not in TERMINAL_STATUSES:
+                id = latest['build']
+                print(f'No build id persisted; cancelling latest non-terminal build #{id} on {os.getenv("TARGET")}')
+            else:
+                print('No build id persisted and no non-terminal build found; nothing to cancel.')
+                utils.delete_build_info()
+                sys.exit(0)
         cancel_build(id)
         utils.delete_build_info()
         sys.exit(0)
@@ -741,6 +819,9 @@ def get_clean_build_bool():
             else:
                 raise ValueError(f"Invalid boolean value for CLEAN_BUILD: {value}")
 
+        # Persist the target before the POST: if the runner dies mid-request, --cancel can still
+        # find the queued build via the target's latest-build lookup.
+        utils.persist_build_info(os.getenv('TARGET'), None)
         id = run_build(os.getenv('BRANCH_NAME'), get_clean_build_bool())
         utils.persist_build_info(os.getenv('TARGET'), id)
         print(f'For more info and live logs, go to https://cloud.unity.com/ and search for target "{os.getenv('TARGET')}" and build ID "{id}"')
@@ -752,14 +833,15 @@ def get_clean_build_bool():
 )
 write_step_summary(os.getenv('TARGET'), id, final_outcome, phase_durations, queue_reasons, queue_elapsed, build_elapsed)
 
-if final_outcome in ('queue_timeout', 'build_timeout'):
-    if final_outcome == 'build_timeout':
+if final_outcome in ('queue_timeout', 'build_timeout', 'log_stall'):
+    if final_outcome in ('build_timeout', 'log_stall'):
         # Build was cancelled; the persisted info points to a dead build.
+        # Delete it so the next retry creates a fresh build on a different VM.
         utils.delete_build_info()
         try:
             download_log(id)
         except Exception as e:
-            print(f'Warning: could not download log after timeout: {e}')
+            print(f'Warning: could not download log after {final_outcome}: {e}')
     sys.exit(RETRYABLE_EXIT_CODE)
 
 utils.delete_build_info()