Starting refactor

Author: jonathan-buttner

server/src/main/java/org/elasticsearch/TransportVersions.java

@@ -365,6 +365,7 @@ static TransportVersion def(int id) {
     public static final TransportVersion ESQL_LOOKUP_JOIN_ON_MANY_FIELDS = def(9_139_0_00);
     public static final TransportVersion SIMULATE_INGEST_EFFECTIVE_MAPPING = def(9_140_0_00);
     public static final TransportVersion RESOLVE_INDEX_MODE_ADDED = def(9_141_0_00);
+    public static final TransportVersion INFERENCE_DISABLE_EIS_DISABLE_RATE_LIMITING_ADDED = def(9_142_0_00);
 
     /*
      * STOP! READ THIS FIRST! No, really,

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/external/http/sender/RequestExecutorService.java

@@ -252,16 +252,20 @@ private void cleanup() {
 
     private void handleTasks() {
         try {
-            if (shutdown.get()) {
-                logger.debug("Shutdown requested while handling tasks, cleaning up");
-                cleanup();
-                return;
-            }
+            TimeValue timeToWait;
 
-            var timeToWait = settings.getTaskPollFrequency();
-            for (var endpoint : rateLimitGroupings.values()) {
-                timeToWait = TimeValue.min(endpoint.executeEnqueuedTask(), timeToWait);
-            }
+            do {
+                if (shutdown.get()) {
+                    logger.debug("Shutdown requested while handling tasks, cleaning up");
+                    cleanup();
+                    return;
+                }
+
+                timeToWait = settings.getTaskPollFrequency();
+                for (var endpoint : rateLimitGroupings.values()) {
+                    timeToWait = TimeValue.min(endpoint.executeEnqueuedTask(), timeToWait);
+                }
+            } while (timeToWait.compareTo(TimeValue.ZERO) <= 0);
 
             scheduleNextHandleTasks(timeToWait);
         } catch (Exception e) {

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/external/http/sender/RequestExecutorServiceSettings.java

@@ -40,6 +40,7 @@ public class RequestExecutorServiceSettings {
     static final Setting<TimeValue> TASK_POLL_FREQUENCY_SETTING = Setting.timeSetting(
         "xpack.inference.http.request_executor.task_poll_frequency",
         DEFAULT_TASK_POLL_FREQUENCY_TIME,
+        TimeValue.timeValueNanos(1),
         Setting.Property.NodeScope,
         Setting.Property.Dynamic
     );
@@ -51,6 +52,7 @@ public class RequestExecutorServiceSettings {
     static final Setting<TimeValue> RATE_LIMIT_GROUP_CLEANUP_INTERVAL_SETTING = Setting.timeSetting(
         "xpack.inference.http.request_executor.rate_limit_group_cleanup_interval",
         DEFAULT_RATE_LIMIT_GROUP_CLEANUP_INTERVAL,
+        TimeValue.timeValueNanos(1),
         Setting.Property.NodeScope,
         Setting.Property.Dynamic
     );
@@ -63,6 +65,7 @@ public class RequestExecutorServiceSettings {
     static final Setting<TimeValue> RATE_LIMIT_GROUP_STALE_DURATION_SETTING = Setting.timeSetting(
         "xpack.inference.http.request_executor.rate_limit_group_stale_duration",
         DEFAULT_RATE_LIMIT_GROUP_STALE_DURATION,
+        TimeValue.timeValueNanos(1),
         Setting.Property.NodeScope,
         Setting.Property.Dynamic
     );

x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/settings/RateLimitSettings.java

@@ -53,6 +53,10 @@ public static RateLimitSettings of(
         return requestsPerMinute == null ? defaultValue : new RateLimitSettings(requestsPerMinute);
     }
 
+    // public static RateLimitSettings disabledRateLimiting() {
+    //
+    // }
+
     public static Map<String, SettingsConfiguration> toSettingsConfigurationWithDescription(
         String description,
         EnumSet<TaskType> supportedTaskTypes
@@ -84,6 +88,8 @@ public RateLimitSettings(long requestsPerMinute) {
     }
 
     /**
+     * This should only be used for testing.
+     *
      * Defines the settings in requests per the time unit provided
      * @param requestsPerTimeUnit number of requests
      * @param timeUnit _
@@ -98,6 +104,12 @@ public RateLimitSettings(long requestsPerTimeUnit, TimeUnit timeUnit) {
         this.timeUnit = Objects.requireNonNull(timeUnit);
     }
 
+    private RateLimitSettings(long requestsPerTimeUnit, TimeUnit timeUnit, boolean disableRateLimiting) {
+        this.requestsPerTimeUnit = 0;
+        this.timeUnit = Objects.requireNonNull(timeUnit);
+
+    }
+
     public RateLimitSettings(StreamInput in) throws IOException {
         requestsPerTimeUnit = in.readVLong();
         timeUnit = TimeUnit.MINUTES;