Expose more information on server process pages (#6278)

This change modifies the Monitor server pages to
expose as much information as possible so that we
can verify the accuracy of the information and
determine which information we want to display vs.
not display. The following modifications are
included:

  1. Metric.java was modified to include the label,
     css class, and description to use in the server
     page table headers.
  2. A new class, ColumnJSGen, which exports the
     information from Metric.java to columns.js at
     build time and is packaged in the monitor jar
     file. The column.js includes the information
     in a javascript Map object which is referenced
     when processing the data from the server.
  3. Changes in AbstractServer to exclude specific
     metrics from being included in the getMetrics
     RPC response.
  4. A new class, ServersView, which replaces the
     class ScanServerView but serves the same
     purpose. This class returns the columns, data,
     and status information for use on the server
     process pages.
  5. A new `servers/view` endpoint which takes a
     path parameter for the server type. Example:
     servers/view;serverType=SCAN_SERVER. This
     endpoint returns a ServersView object. This
     endpoint replaces the `sservers/view` endpoint
     that returned the ScanServerView object.
  6. Updates to the SystemInformation.finish method
     that creates memoized Suppliers for the
     ServersView object for each server type. When
     the endpoint is hit the Supplier will calculate
     the response object the first time, then cache
     it for subsequent requests.
  7. Centralized the javascript code for creating
     the DataTables and HTML table headers into a
     new file called server_process_common.js. The
     methods in this file use the information from
     the ServersView object and columns.js to
     create the necessary objects to render the
     tables in the server process pages.
  8. Updated the server page js and ftl files
     accordingly to use the server_process_common.js
     file and other new items. 



Co-authored-by: Dom G. <domgarguilo@apache.org>

Author: dlmarion

LICENSE

@@ -309,4 +309,12 @@ Files (in server/monitor/src/main/resources/):
     Copyright (c) 2008-2024 SpryMedia Ltd
     Licensed under the MIT license (see above)
 
+## DataTables ColReorder 1.7.0 (https://datatables.net)
+
+Files (in server/monitor/src/main/resources/):
+* org/apache/accumulo/monitor/resources/external/datatables/**/*
+
+    Copyright (c) 2010-2015 SpryMedia Limited
+    Licensed under the MIT license (see above)
+
 **********

core/src/main/java/org/apache/accumulo/core/metrics/Metric.java

@@ -22,6 +22,7 @@
 
 import java.net.UnknownHostException;
 import java.util.OptionalInt;
+import java.util.Set;
 import java.util.concurrent.ScheduledFuture;
 import java.util.concurrent.atomic.AtomicBoolean;
 import java.util.concurrent.atomic.AtomicReference;
@@ -39,6 +40,7 @@
 import org.apache.accumulo.core.conf.SiteConfiguration;
 import org.apache.accumulo.core.data.ResourceGroupId;
 import org.apache.accumulo.core.lock.ServiceLock;
+import org.apache.accumulo.core.metrics.Metric;
 import org.apache.accumulo.core.metrics.MetricsProducer;
 import org.apache.accumulo.core.process.thrift.MetricResponse;
 import org.apache.accumulo.core.process.thrift.MetricSource;
@@ -95,6 +97,7 @@ public static void startServer(AbstractServer server, Logger LOG) throws Excepti
   private final AtomicBoolean shutdownRequested = new AtomicBoolean(false);
   private final AtomicBoolean shutdownComplete = new AtomicBoolean(false);
   private final AtomicBoolean closed = new AtomicBoolean(false);
+  private final Set<String> monitorMetricExclusions;
 
   protected AbstractServer(ServerId.Type serverType, ServerOpts opts,
       BiFunction<SiteConfiguration,ResourceGroupId,ServerContext> serverContextFactory,
@@ -177,6 +180,7 @@ protected AbstractServer(ServerId.Type serverType, ServerOpts opts,
       default:
         throw new IllegalArgumentException("Unhandled server type: " + serverType);
     }
+    monitorMetricExclusions = Metric.getMonitorExclusions(serverType);
   }
 
   /**
@@ -403,10 +407,12 @@ public MetricResponse getMetrics(TInfo tinfo, TCredentials credentials) throws T
     if (context.getMetricsInfo().isMetricsEnabled()) {
       Metrics.globalRegistry.getMeters().forEach(m -> {
         if (m.getId().getName().startsWith("accumulo.")) {
-          m.match(response::writeMeter, response::writeMeter, response::writeTimer,
-              response::writeDistributionSummary, response::writeLongTaskTimer,
-              response::writeMeter, response::writeMeter, response::writeFunctionTimer,
-              response::writeMeter);
+          if (!this.monitorMetricExclusions.contains(m.getId().getName())) {
+            m.match(response::writeMeter, response::writeMeter, response::writeTimer,
+                response::writeDistributionSummary, response::writeLongTaskTimer,
+                response::writeMeter, response::writeMeter, response::writeFunctionTimer,
+                response::writeMeter);
+          }
         }
       });
     }

server/base/src/main/java/org/apache/accumulo/server/AbstractServer.java

@@ -54,6 +54,10 @@
       <groupId>com.github.ben-manes.caffeine</groupId>
       <artifactId>caffeine</artifactId>
     </dependency>
+    <dependency>
+      <groupId>com.google.code.gson</groupId>
+      <artifactId>gson</artifactId>
+    </dependency>
     <dependency>
       <groupId>com.google.guava</groupId>
       <artifactId>guava</artifactId>
@@ -242,6 +246,26 @@
       </plugins>
     </pluginManagement>
     <plugins>
+      <plugin>
+        <groupId>org.codehaus.mojo</groupId>
+        <artifactId>exec-maven-plugin</artifactId>
+        <executions>
+          <execution>
+            <id>metrics-javascript</id>
+            <goals>
+              <goal>java</goal>
+            </goals>
+            <phase>prepare-package</phase>
+            <configuration>
+              <mainClass>org.apache.accumulo.monitor.next.views.ColumnJsGen</mainClass>
+              <classpathScope>compile</classpathScope>
+              <arguments>
+                <argument>${project.build.directory}/classes/org/apache/accumulo/monitor/resources/js/columns.js</argument>
+              </arguments>
+            </configuration>
+          </execution>
+        </executions>
+      </plugin>
       <plugin>
         <groupId>org.apache.maven.plugins</groupId>
         <artifactId>maven-resources-plugin</artifactId>

server/monitor/pom.xml

@@ -33,6 +33,7 @@
 import jakarta.inject.Inject;
 import jakarta.servlet.http.HttpServletRequest;
 import jakarta.ws.rs.GET;
+import jakarta.ws.rs.MatrixParam;
 import jakarta.ws.rs.NotFoundException;
 import jakarta.ws.rs.Path;
 import jakarta.ws.rs.PathParam;
@@ -54,20 +55,25 @@
 import org.apache.accumulo.monitor.next.deployment.DeploymentOverview;
 import org.apache.accumulo.monitor.next.ec.CompactorsSummary;
 import org.apache.accumulo.monitor.next.ec.CoordinatorSummary;
-import org.apache.accumulo.monitor.next.sservers.ScanServerView;
+import org.apache.accumulo.monitor.next.views.ServersView;
 
 import io.micrometer.core.instrument.Meter.Id;
 import io.micrometer.core.instrument.cumulative.CumulativeDistributionSummary;
 
 @Path("/")
 public class Endpoints {
   /**
-   * A {@code String} constant representing supplied resource group in path parameter.
+   * A {@code String} constant representing the supplied resource group in path parameter.
    */
   private static final String GROUP_PARAM_KEY = "group";
 
   /**
-   * A {@code String} constant representing supplied tableId in path parameter.
+   * A {@code String} constant representing the supplied server type in path parameter.
+   */
+  private static final String SERVER_TYPE_KEY = "serverType";
+
+  /**
+   * A {@code String} constant representing the supplied tableId in path parameter.
    */
   private static final String TABLEID_PARAM_KEY = "tableId";
 
@@ -271,11 +277,17 @@ public Map<Id,CumulativeDistributionSummary> getScanServerAllMetricSummary() {
   }
 
   @GET
-  @Path("sservers/view")
+  @Path("servers/view")
   @Produces(MediaType.APPLICATION_JSON)
-  @Description("Returns a UI-ready view model for the Scan Server status page")
-  public ScanServerView getScanServerPageView() {
-    return monitor.getInformationFetcher().getSummaryForEndpoint().getScanServerView();
+  @Description("Returns a UI-ready view model for server processes. Add ';serverType=<ServerId.Type>' to URL")
+  public ServersView getServerProcessView(@MatrixParam(SERVER_TYPE_KEY) ServerId.Type serverType) {
+    ServersView view =
+        monitor.getInformationFetcher().getSummaryForEndpoint().getServerProcessView(serverType);
+    if (view == null) {
+      throw new NotFoundException(
+          "ServersView object for server type " + serverType.name() + " not found");
+    }
+    return view;
   }
 
   @GET

server/monitor/src/main/java/org/apache/accumulo/monitor/next/Endpoints.java

@@ -211,7 +211,7 @@ public void newConnectionEvent() {
   }
 
   // Protect against NPE and wait for initial data gathering
-  public SystemInformation getSummary() throws InterruptedException {
+  private SystemInformation getSummary() throws InterruptedException {
     while (summaryRef.get() == null) {
       Thread.sleep(100);
     }
@@ -254,7 +254,7 @@ public void onRemoval(@Nullable ServerId server, @Nullable MetricResponse respon
   @Override
   public void run() {
 
-    long refreshTime = 0;
+    long lastRunTime = 0;
 
     while (true) {
 
@@ -263,7 +263,7 @@ public void run() {
       // If a connection has not been made in a while, stale data may be displayed.
       // Only refresh every 5s (old monitor logic).
       while (!newConnectionEvent.get() && connectionCount.get() == 0
-          && NanoTime.millisElapsed(refreshTime, NanoTime.now()) > 5000) {
+          && NanoTime.millisElapsed(lastRunTime, NanoTime.now()) > 5000) {
         try {
           Thread.sleep(100);
         } catch (InterruptedException e) {
@@ -275,7 +275,7 @@ public void run() {
       // reset the connection event flag
       newConnectionEvent.compareAndExchange(true, false);
 
-      LOG.info("Fetching metrics from servers");
+      LOG.info("Fetching information from servers");
 
       final List<Future<?>> futures = new ArrayList<>();
       final SystemInformation summary = new SystemInformation(allMetrics, this.ctx);
@@ -315,13 +315,18 @@ public void run() {
         }
       }));
 
-      long monitorFetchTimeout =
+      final long monitorFetchTimeout =
           ctx.getConfiguration().getTimeInMillis(Property.MONITOR_FETCH_TIMEOUT);
-      long allFuturesAdded = NanoTime.now();
+      final long allFuturesAdded = NanoTime.now();
       boolean tookToLong = false;
       while (!futures.isEmpty()) {
 
         if (NanoTime.millisElapsed(allFuturesAdded, NanoTime.now()) > monitorFetchTimeout) {
+          LOG.warn(
+              "Fetching information for Monitor has taken longer {}. Cancelling all"
+                  + " remaining tasks and monitor will display old information. Resolve issue"
+                  + " causing this or increase property {}.",
+              monitorFetchTimeout, Property.MONITOR_FETCH_TIMEOUT.getKey());
           tookToLong = true;
         }
 
@@ -344,26 +349,31 @@ public void run() {
         }
       }
 
-      summary.finish();
-
-      refreshTime = NanoTime.now();
-      LOG.info("Finished fetching metrics from servers");
-      LOG.info(
-          "All: {}, Manager: {}, Garbage Collector: {}, Compactors: {}, Scan Servers: {}, Tablet Servers: {}",
-          allMetrics.estimatedSize(), summary.getManager() != null,
-          summary.getGarbageCollector() != null,
-          summary.getCompactorAllMetricSummary().isEmpty() ? 0
-              : summary.getCompactorAllMetricSummary().entrySet().iterator().next().getValue()
-                  .count(),
-          summary.getSServerAllMetricSummary().isEmpty() ? 0
-              : summary.getSServerAllMetricSummary().entrySet().iterator().next().getValue()
-                  .count(),
-          summary.getTServerAllMetricSummary().isEmpty() ? 0 : summary.getTServerAllMetricSummary()
-              .entrySet().iterator().next().getValue().count());
-
-      SystemInformation oldSummary = summaryRef.getAndSet(summary);
-      if (oldSummary != null) {
-        oldSummary.clear();
+      lastRunTime = NanoTime.now();
+
+      if (tookToLong) {
+        summary.clear();
+      } else {
+        summary.finish();
+
+        LOG.info("Finished fetching metrics from servers");
+        LOG.info(
+            "All: {}, Manager: {}, Garbage Collector: {}, Compactors: {}, Scan Servers: {}, Tablet Servers: {}",
+            allMetrics.estimatedSize(), summary.getManager() != null,
+            summary.getGarbageCollector() != null,
+            summary.getCompactorAllMetricSummary().isEmpty() ? 0
+                : summary.getCompactorAllMetricSummary().entrySet().iterator().next().getValue()
+                    .count(),
+            summary.getSServerAllMetricSummary().isEmpty() ? 0
+                : summary.getSServerAllMetricSummary().entrySet().iterator().next().getValue()
+                    .count(),
+            summary.getTServerAllMetricSummary().isEmpty() ? 0 : summary
+                .getTServerAllMetricSummary().entrySet().iterator().next().getValue().count());
+
+        SystemInformation oldSummary = summaryRef.getAndSet(summary);
+        if (oldSummary != null) {
+          oldSummary.clear();
+        }
       }
     }