fix: base36 IDs for kind-compatible cluster names, test --perform-cluster-management

- ids.py: use base36 (lowercase+digits) instead of base62 — kind
  cluster names must match ^[a-z0-9.-]+$
- k8s deploy test: pass --perform-cluster-management on first start
  since 'start' defaults to --skip-cluster-management

Found by running tests/k8s-deploy/run-deploy-test.sh locally.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

stack_orchestrator/ids.py

@@ -1,11 +1,11 @@
 """Sortable timestamp-based ID generation for cluster naming.
 
-Uses base62 encoding with 100ms resolution and a 2024-01-01 epoch
+Uses base36 encoding with 100ms resolution and a 2024-01-01 epoch
-to produce compact, sortable IDs like 'laconic-iqE6Za'.
+to produce compact, sortable IDs like 'laconic-3k7m2ab'.
 
 Format: {prefix}-{timestamp}{random}
-- timestamp: 5 chars (100ms resolution, ~180 years from 2024)
+- timestamp: 7 chars base36 (100ms resolution, ~2500 years from 2024)
-- random: 2 chars (3,844 unique per 100ms slot)
+- random: 2 chars (1,296 unique per 100ms slot)
 """
 # Adapted from exophial/src/exophial/ids.py
 
@@ -15,23 +15,24 @@ import time
 # 2024-01-01 00:00:00 UTC in milliseconds
 EPOCH_2024 = 1704067200000
 
-# Sortable base62 alphabet (0-9, A-Z, a-z)
+# Sortable base36 alphabet (0-9, a-z) — lowercase only to satisfy
-ALPHABET = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
+# kind cluster name validation (^[a-z0-9.-]+$)
+ALPHABET = "0123456789abcdefghijklmnopqrstuvwxyz"
 
 
-def _base62(n: int) -> str:
+def _base36(n: int) -> str:
-    """Encode integer as base62 string."""
+    """Encode integer as base36 string."""
     if n == 0:
         return ALPHABET[0]
     s = ""
     while n:
-        n, r = divmod(n, 62)
+        n, r = divmod(n, len(ALPHABET))
         s = ALPHABET[r] + s
     return s
 
 
 def _random_suffix(length: int = 2) -> str:
-    """Generate random base62 suffix."""
+    """Generate random base36 suffix."""
     return "".join(random.choice(ALPHABET) for _ in range(length))
 
 
@@ -39,9 +40,9 @@ def _timestamp_id() -> str:
     """Generate a sortable timestamp ID (100ms resolution, 2024 epoch) with random suffix."""
     now_ms = int(time.time() * 1000)
     offset = (now_ms - EPOCH_2024) // 100  # 100ms resolution
-    return f"{_base62(offset)}{_random_suffix()}"
+    return f"{_base36(offset)}{_random_suffix()}"
 
 
 def generate_id(prefix: str) -> str:
-    """Generate a sortable ID with an arbitrary prefix like 'laconic-iqE6Za'."""
+    """Generate a sortable ID with an arbitrary prefix like 'laconic-3k7m2ab'."""
     return f"{prefix}-{_timestamp_id()}"

tests/k8s-deploy/run-deploy-test.sh

@@ -115,8 +115,9 @@ sed -i 's/^secrets: {}$/secrets:\n  test-secret:\n    - TEST_SECRET_KEY/' ${depl
 deployment_id=$(cat ${test_deployment_dir}/deployment.yml | cut -d ' ' -f 2)
 
 echo "deploy create output file test: passed"
-# Try to start the deployment
+# Try to start the deployment (--perform-cluster-management needed on first start
-$TEST_TARGET_SO deployment --dir $test_deployment_dir start
+# because 'start' defaults to --skip-cluster-management)
+$TEST_TARGET_SO deployment --dir $test_deployment_dir start --perform-cluster-management
 wait_for_pods_started
 # Check logs command works
 wait_for_log_output