General gateway fixes from the Atlas integration branch

bin/lloom.mjs

@@ -67,6 +67,12 @@ import { loadRecipeById, loadRecipes, planRecipe } from '../src/recipes.mjs';
 import { RuntimeManager } from '../src/runtime-manager.mjs';
 import { runCommand } from '../src/process-control.mjs';
 import { applyRuntimePolicyPlan, createRuntimePolicyPlan } from '../src/runtime-policy.mjs';
+import {
+  applyRuntimePolicyConfig,
+  createRuntimePolicyConfigReport,
+  parseNumericMemoryFlags,
+  NUMERIC_MEMORY_FLAGS
+} from '../src/runtime-policy-config.mjs';
 import { createLloomServer } from '../src/server.mjs';
 import { applySetup, createSetupPlan, syncClusterSetupMembers } from '../src/setup.mjs';
 import { createSetupStatus } from '../src/setup-status.mjs';
@@ -110,7 +116,8 @@ const COMMAND_REGISTRY = [
   { name: 'backend-install', aliases: [], tier: 'advanced', needsInstalledConfig: false },
   { name: 'runtimes', aliases: ['runtime-status'], tier: 'advanced', needsInstalledConfig: true },
   { name: 'cluster', aliases: ['cluster-status'], tier: 'advanced', needsInstalledConfig: true },
-  { name: 'runtime-plan', aliases: ['runtime-policy'], tier: 'advanced', needsInstalledConfig: true },
+  { name: 'runtime-plan', aliases: [], tier: 'advanced', needsInstalledConfig: true },
+  { name: 'runtime-policy', aliases: [], tier: 'advanced', needsInstalledConfig: true },
   { name: 'runtime-admit', aliases: [], tier: 'advanced', needsInstalledConfig: true },
   { name: 'runtime-start', aliases: [], tier: 'advanced', needsInstalledConfig: true },
   { name: 'runtime-warmup', aliases: [], tier: 'advanced', needsInstalledConfig: true },
@@ -205,6 +212,7 @@ Backends and runtimes:
   lloom backend-install <backend-id> [--apply --yes] [--step step-id]
   lloom runtimes [runtime-id|all]
   lloom runtime-plan <runtime-id>
+  lloom runtime-policy [--max-memory-utilization f] [--reserve-memory-gb n] [--apply --yes]
   lloom suspend <model-or-alias> [--apply --yes] [--drain-timeout-ms 300000]
   lloom resume <model-or-alias> [--apply --yes]
   lloom runtime-admit <runtime-id> [--apply --yes]
@@ -2632,6 +2640,16 @@ async function main() {
       console.log(JSON.stringify(response, null, 2));
     },
     'runtime-plan': async ({ args, config, command: _command }) => {
+      if (args.some((arg) => NUMERIC_MEMORY_FLAGS.some((flag) => arg === flag || arg.startsWith(`${flag}=`)))) {
+        throw new Error(
+          'runtime-plan is read-only and does not accept numeric memory flags; use `lloom runtime-policy --max-memory-utilization <f> --reserve-memory-gb <n> --apply --yes`'
+        );
+      }
+      if (hasFlag(args, '--apply') || hasFlag(args, '--yes')) {
+        throw new Error(
+          'runtime-plan is read-only and does not accept --apply/--yes; use `lloom runtime-policy --apply --yes` to write memory-policy changes'
+        );
+      }
       const requestedRuntimeId = positional(args)[1];
       const livePlan = clusterConfigured(config)
         ? await gatewayRequest(
@@ -2850,7 +2868,30 @@ async function main() {
   handlers['routing'] = handlers['route'];
   handlers['runtime-status'] = handlers['runtimes'];
   handlers['cluster-status'] = handlers['cluster'];
-  handlers['runtime-policy'] = handlers['runtime-plan'];
+  handlers['runtime-policy'] = async ({ args, config, command: _command }) => {
+    const flags = parseNumericMemoryFlags(args);
+    const apply = hasFlag(args, '--apply');
+    const yes = hasFlag(args, '--yes');
+    if (!flags) {
+      if (apply || yes) {
+        throw new Error('runtime-policy --apply/--yes requires --max-memory-utilization and/or --reserve-memory-gb');
+      }
+      // Compatibility: a bare `runtime-policy` with no numeric flags keeps the
+      // historical read-only alias behavior and delegates to the runtime plan.
+      return handlers['runtime-plan']({ args, config, command: 'runtime-plan' });
+    }
+    if (yes && !apply) {
+      throw new Error('runtime-policy --yes requires --apply');
+    }
+    if (!apply) {
+      console.log(JSON.stringify(createRuntimePolicyConfigReport(config, flags), null, 2));
+      return;
+    }
+    if (!yes) {
+      throw new Error('applying a memory-policy change requires both --apply and --yes');
+    }
+    console.log(JSON.stringify(await applyRuntimePolicyConfig(config, flags, { apply, yes }), null, 2));
+  };
   handlers['voice-list'] = handlers['voices'];
   handlers['install-voice'] = handlers['voice-install'];
   handlers['voice-rm'] = handlers['voice-remove'];

docs/runtime-policy.md

@@ -21,6 +21,35 @@ an alias, or the config root.
 The legacy `evictable` field (top-level or `policy.evictable`) still normalizes
 to `keepWarm: true` when set to `false`, unchanged from earlier releases.
 
+## Explicit memory limits
+
+Preview a numeric memory-policy change before applying it:
+
+```bash
+lloom runtime-policy --max-memory-utilization 0.90 --reserve-memory-gb 8
+lloom runtime-policy --max-memory-utilization 0.90 --reserve-memory-gb 8 --apply --yes
+```
+
+These example values are not a model-specific capacity recommendation. The
+utilization must be greater than zero and less than one; the reserve must be
+positive. Either option can be supplied alone. The command reports the selected
+before/after fields and any per-node overrides, without printing the config.
+
+Utilization updates both predictive admission and the memory guard's utilization
+limit. Reserve updates both admission's reserve and the guard's minimum available
+memory. The existing guard mode is preserved, including an explicitly selected
+`yolo` mode; this command does not turn enforcement on or off. Percentage and
+absolute-reserve limits both apply, so the stricter limit can determine admission.
+
+Changes affect the resolved local config file. Configure each managed member's
+local policy as needed. Cluster node resource overrides are preserved and shown
+in the preview; they can take precedence over global values. A global absolute
+`runtimePolicy.memoryBudgetGb` must be resolved explicitly before using these
+numeric options. Unrelated settings, routes and credentials are preserved.
+
+`runtime-plan` remains read-only. Without numeric options, `runtime-policy`
+retains its historical read-only runtime-plan behavior.
+
 ## Idle restoration
 
 The preferred-restore pass starts after all owned, enabled keep-warm pins are

package.json

@@ -64,7 +64,7 @@
     "check": "npm run check:cluster && npm run check:js && npm run check:entity-stagger && npm run check:python && npm run check:shell && npm run check:spark-deploy && npm run check:workers",
     "check:cluster": "node --check src/cluster.mjs && node --check test/cluster.test.mjs",
     "check:entity-stagger": "node --check scripts/entity-stagger-benchmark.mjs && node --check test/entity-stagger-benchmark.test.mjs",
-    "check:js": "node --check bin/lloom.mjs && node --check bin/lloom-host.mjs && node --check src/backend-catalog.mjs && node --check src/benchmarks.mjs && node --check src/bootstrap.mjs && node --check src/client-integrations.mjs && node --check src/community-client.mjs && node --check src/community-site.mjs && node --check src/config.mjs && node --check src/dashboard.mjs && node --check src/doctor.mjs && node --check src/host-memory.mjs && node --check src/host-server.mjs && node --check src/init.mjs && node --check src/installer.mjs && node --check src/interchange.mjs && node --check src/machine-profile.mjs && node --check src/managed-environment.mjs && node --check src/model-acquisition.mjs && node --check src/model-intake.mjs && node --check src/model-removal.mjs && node --check src/onboarding.mjs && node --check src/recipe-index.mjs && node --check src/recipe-pack-export.mjs && node --check src/recipe-pack.mjs && node --check src/registry.mjs && node --check src/route-control.mjs && node --check src/recipes.mjs && node --check src/resource-fit.mjs && node --check src/runtime-manager.mjs && node --check src/runtime-policy.mjs && node --check src/server.mjs && node --check src/setup.mjs && node --check src/setup-status.mjs && node --check src/process-control.mjs && node --check src/security.mjs && node --check src/tts-catalog.mjs && node --check src/voice-profiles.mjs && node --check src/protocol/text.mjs && node --check src/protocol/reasoning-normalize.mjs && node --check src/protocol/reasoning-effort.mjs && node --check src/protocol/responses.mjs && node --check src/protocol/anthropic.mjs && node --check src/protocol/sse.mjs && node --check src/protocol/stream-anthropic.mjs && node --check src/protocol/stream-responses.mjs && node --check src/protocol/index.mjs && node --check scripts/apply-spark-route-catalog.mjs && node --check scripts/capture-live-metrics.mjs && node --check scripts/merge-live-metrics.mjs && node --check scripts/run-qualitative-chat-benchmark.mjs && node --check scripts/run-local-decision-benchmark.mjs && node --check scripts/check-community-deploy.mjs && node --check scripts/generate-client-configs.mjs && node --check scripts/check-interchange.mjs && node --check scripts/check-package.mjs && node --check scripts/release-lib.mjs && node --check scripts/build-release.mjs && node --check scripts/deploy-spark.mjs && node --check test/audio-generations.test.mjs && node --check test/community-host.test.mjs && node --check test/dashboard-status.test.mjs && node --check test/ds4fv-recipe.test.mjs && node --check test/dspark-hotfixes.test.mjs && node --check test/glm53-exl3-recipe.test.mjs && node --check test/host-memory.test.mjs && node --check test/metrics-persistence.test.mjs && node --check test/model-acquisition.test.mjs && node --check test/model-failover.test.mjs && node --check test/qwen38-vllm-recipe.test.mjs && node --check test/reasoning-normalize.test.mjs && node --check test/reasoning-effort.test.mjs && node --check test/resource-fit.test.mjs && node --check test/route-control.test.mjs && node --check test/spark-route-catalog.test.mjs && node --check test/runtime-watchdog.test.mjs && node --check test/tts-catalog.test.mjs && node --check test/voice-profiles.test.mjs && node --check test/chatterbox-recipe.test.mjs && node --check test/hear-recipe.test.mjs && node --check test/smoke.mjs && node --check src/runtime-supervisor.mjs && node --check test/runtime-supervisor.test.mjs && node --check test/comfyui-media.test.mjs && node --check src/runtime-residency.mjs && node --check src/dashboard-scene.mjs && node --check src/dashboard-presence.mjs && node --check src/dashboard-presence-client.mjs && node --check src/first-run.mjs && node --check src/first-run-page.mjs && node --check src/browser-setup.mjs && node --check src/runtime-preferences.mjs && node --check src/installation-jobs.mjs && node --check src/model-installation.mjs && node --check src/dashboard-installation.mjs && node --check src/runtime-memory-safety.mjs && node --check src/dashboard-memory.mjs && node --check src/runtime-memory-usage.mjs && node --check src/config-profiles.mjs && node --check test/config-profiles.test.mjs",
+    "check:js": "node --check bin/lloom.mjs && node --check bin/lloom-host.mjs && node --check src/backend-catalog.mjs && node --check src/benchmarks.mjs && node --check src/bootstrap.mjs && node --check src/client-integrations.mjs && node --check src/community-client.mjs && node --check src/community-site.mjs && node --check src/config.mjs && node --check src/dashboard.mjs && node --check src/doctor.mjs && node --check src/host-memory.mjs && node --check src/host-server.mjs && node --check src/init.mjs && node --check src/installer.mjs && node --check src/interchange.mjs && node --check src/machine-profile.mjs && node --check src/managed-environment.mjs && node --check src/model-acquisition.mjs && node --check src/model-intake.mjs && node --check src/model-removal.mjs && node --check src/onboarding.mjs && node --check src/recipe-index.mjs && node --check src/recipe-pack-export.mjs && node --check src/recipe-pack.mjs && node --check src/registry.mjs && node --check src/route-control.mjs && node --check src/recipes.mjs && node --check src/resource-fit.mjs && node --check src/runtime-manager.mjs && node --check src/runtime-policy.mjs && node --check src/server.mjs && node --check src/setup.mjs && node --check src/setup-status.mjs && node --check src/process-control.mjs && node --check src/security.mjs && node --check src/tts-catalog.mjs && node --check src/voice-profiles.mjs && node --check src/protocol/text.mjs && node --check src/protocol/reasoning-normalize.mjs && node --check src/protocol/reasoning-effort.mjs && node --check src/protocol/responses.mjs && node --check src/protocol/anthropic.mjs && node --check src/protocol/sse.mjs && node --check src/protocol/stream-anthropic.mjs && node --check src/protocol/stream-responses.mjs && node --check src/protocol/index.mjs && node --check scripts/apply-spark-route-catalog.mjs && node --check scripts/capture-live-metrics.mjs && node --check scripts/merge-live-metrics.mjs && node --check scripts/run-qualitative-chat-benchmark.mjs && node --check scripts/run-local-decision-benchmark.mjs && node --check scripts/check-community-deploy.mjs && node --check scripts/generate-client-configs.mjs && node --check scripts/check-interchange.mjs && node --check scripts/check-package.mjs && node --check scripts/release-lib.mjs && node --check scripts/build-release.mjs && node --check scripts/deploy-spark.mjs && node --check test/audio-generations.test.mjs && node --check test/community-host.test.mjs && node --check test/dashboard-status.test.mjs && node --check test/ds4fv-recipe.test.mjs && node --check test/dspark-hotfixes.test.mjs && node --check test/glm53-exl3-recipe.test.mjs && node --check test/host-memory.test.mjs && node --check test/metrics-persistence.test.mjs && node --check test/model-acquisition.test.mjs && node --check test/model-failover.test.mjs && node --check test/qwen38-vllm-recipe.test.mjs && node --check test/reasoning-normalize.test.mjs && node --check test/reasoning-effort.test.mjs && node --check test/resource-fit.test.mjs && node --check test/route-control.test.mjs && node --check test/spark-route-catalog.test.mjs && node --check test/runtime-watchdog.test.mjs && node --check test/tts-catalog.test.mjs && node --check test/voice-profiles.test.mjs && node --check test/chatterbox-recipe.test.mjs && node --check test/hear-recipe.test.mjs && node --check test/smoke.mjs && node --check src/runtime-supervisor.mjs && node --check test/runtime-supervisor.test.mjs && node --check test/comfyui-media.test.mjs && node --check src/runtime-residency.mjs && node --check src/dashboard-scene.mjs && node --check src/dashboard-presence.mjs && node --check src/dashboard-presence-client.mjs && node --check src/first-run.mjs && node --check src/first-run-page.mjs && node --check src/browser-setup.mjs && node --check src/runtime-preferences.mjs && node --check src/installation-jobs.mjs && node --check src/model-installation.mjs && node --check src/dashboard-installation.mjs && node --check src/runtime-memory-safety.mjs && node --check src/dashboard-memory.mjs && node --check src/runtime-memory-usage.mjs && node --check src/config-profiles.mjs && node --check test/config-profiles.test.mjs && node --check src/runtime-policy-config.mjs && node --check test/runtime-policy-config.test.mjs",
     "check:python": "python3 -m py_compile backends/dspark-vllm/apply-patch-pack.py backends/dspark-vllm/packs/miaai-dsv4flash-d1b76251-defaults/patches/*.py backends/dspark-vllm/packs/miaai-ds4fv-f5665e8/patches/*.py backends/dspark-vllm/packs/miaai-ds4fv-f5665e8/patches/vision_exp/*.py backends/qwen38-vllm/apply-ple-fp8-patch.py backends/qwen38-sglang/*.py backends/glm53-exl3/*.py backends/mlx-audio/lloom_audio_server.py backends/hear/lloom_hear_server.py backends/chatterbox/lloom_chatterbox_server.py patches/apply_mtplx_longctx_fix.py backends/comfyui-media/bridge/*.py backends/comfyui-media/graphs/*.py backends/comfyui-media/build/*.py backends/comfyui-media/install.py backends/qwen-image-diffusers/*.py backends/hear/test/test_hear_security.py src/darwin-memory-usage.py",
     "check:shell": "bash -n backends/dspark-vllm/entrypoint.sh backends/dspark-vllm/runtime/build-hardened-image.sh backends/dspark-vllm/packs/miaai-dsv4flash-d1b76251-defaults/patches/*.sh backends/dspark-vllm/packs/miaai-ds4fv-f5665e8/apply-runtime.sh backends/dspark-vllm/packs/miaai-ds4fv-f5665e8/patches/*.sh backends/qwen38-vllm/entrypoint.sh backends/qwen38-sglang/entrypoint.sh backends/glm53-exl3/entrypoint.sh backends/mlx-audio/install.sh backends/hear/install.sh backends/chatterbox/install.sh clients/examples/lloom-claude clients/examples/lloom-codex clients/examples/lloom-hermes clients/examples/lloom-opencode clients/examples/lloom-zero scripts/remote-install-spark.sh scripts/remote-stage-spark-worker.sh",
     "check:spark-deploy": "node --check test/spark-deploy.test.mjs && node test/spark-deploy.test.mjs",
@@ -86,7 +86,7 @@
     "test:comfyui": "python3 scripts/test-comfyui-media.py",
     "test:hear": "python3 -m unittest discover -s backends/hear/test -v",
     "test:ux": "node --test test/first-run.test.mjs test/runtime-preferences.test.mjs test/admin-browser-security.test.mjs test/installation-jobs.test.mjs",
-    "test:memory-safety": "node --test test/runtime-memory-safety.test.mjs"
+    "test:memory-safety": "node --test test/runtime-memory-safety.test.mjs test/runtime-policy-config.test.mjs"
   },
   "engines": {
     "node": ">=22.19.0"

src/config.mjs

@@ -370,6 +370,41 @@ function validateConfig(config, sourcePath, env) {
     }
   }
 
+  // Top-level numeric memory-policy fields. These mirror the ranges used by the
+  // runtime admission and memory-safety layers so a hand-edited or CLI-written
+  // config fails loudly instead of silently falling through to defaults.
+  const topLevelMaxMemoryUtilization = config.runtimePolicy?.maxMemoryUtilization;
+  if (topLevelMaxMemoryUtilization != null) {
+    if (
+      typeof topLevelMaxMemoryUtilization !== 'number' ||
+      !Number.isFinite(topLevelMaxMemoryUtilization) ||
+      topLevelMaxMemoryUtilization <= 0 ||
+      topLevelMaxMemoryUtilization >= 1
+    ) {
+      errors.push('runtimePolicy.maxMemoryUtilization must be a finite number greater than 0 and less than 1');
+    }
+  }
+  const topLevelReserveMemoryGb = config.runtimePolicy?.reserveMemoryGb;
+  if (topLevelReserveMemoryGb != null) {
+    if (
+      typeof topLevelReserveMemoryGb !== 'number' ||
+      !Number.isFinite(topLevelReserveMemoryGb) ||
+      topLevelReserveMemoryGb <= 0
+    ) {
+      errors.push('runtimePolicy.reserveMemoryGb must be a positive finite number');
+    }
+  }
+  const topLevelMemoryBudgetGb = config.runtimePolicy?.memoryBudgetGb;
+  if (topLevelMemoryBudgetGb != null) {
+    if (
+      typeof topLevelMemoryBudgetGb !== 'number' ||
+      !Number.isFinite(topLevelMemoryBudgetGb) ||
+      topLevelMemoryBudgetGb <= 0
+    ) {
+      errors.push('runtimePolicy.memoryBudgetGb must be a positive finite number');
+    }
+  }
+
   const memorySafety = config.runtimePolicy?.memorySafety;
   if (memorySafety != null) {
     if (typeof memorySafety !== 'object' || Array.isArray(memorySafety)) {