DRILL-8504: Add Schema Caching to Splunk Plugin (apache#2929)

.github/workflows/ci.yml

@@ -55,7 +55,7 @@ jobs:
         # added to the Runner and memory hungry tests are run separately.
         run: |
           sudo sh -c "
-            fallocate -l 4G /tmp/swapfile
+            fallocate -l 2G /tmp/swapfile
             chmod 0600 /tmp/swapfile
             mkswap /tmp/swapfile
             swapon /tmp/swapfile
@@ -67,6 +67,7 @@ jobs:
       - name: Remove swap space
         run : |
           sudo sh -c "
+            free -h
             swapoff /tmp/swapfile
             rm /tmp/swapfile
           "

contrib/storage-splunk/README.md

@@ -4,7 +4,7 @@ This plugin enables Drill to query Splunk.
 ## Configuration
 
 | Option                | Default   | Description                                                     | Since |
-|-----------------------| --------- | --------------------------------------------------------------- | ----- |
+|-----------------------| --------- | --------------------------------------------------------------- |-------|
 | type                  | (none)    | Set to "splunk" to use this plugin                              | 1.19  |
 | username              | null      | Splunk username to be used by Drill                             | 1.19  |
 | password              | null      | Splunk password to be used by Drill                             | 1.19  |
@@ -13,12 +13,15 @@ This plugin enables Drill to query Splunk.
 | port                  | 8089      | TCP port over which Drill will connect to Splunk                | 1.19  |
 | earliestTime          | null      | Global earliest record timestamp default                        | 1.19  |
 | latestTime            | null      | Global latest record timestamp default                          | 1.19  |
-| app                   | null      | The application context of the service[^1]                      | 2.0   |
-| owner                 | null      | The owner context of the service[^1]                            | 2.0   |
-| token                 | null      | A Splunk authentication token to use for the session[^2]        | 2.0   |
-| cookie                | null      | A valid login cookie                                            | 2.0   |
-| validateCertificates  | true      | Whether the Splunk client will validates the server's SSL cert  | 2.0   |
+| app                   | null      | The application context of the service[^1]                      | 1.21  |
+| owner                 | null      | The owner context of the service[^1]                            | 1.21  |
+| token                 | null      | A Splunk authentication token to use for the session[^2]        | 1.21  |
+| cookie                | null      | A valid login cookie                                            | 1.21  |
+| validateCertificates  | true      | Whether the Splunk client will validates the server's SSL cert  | 1.21  |
 | validateHostname      | true      | Whether the Splunk client will validate the server's host name  | 1.22  |
+| maxColumns            | 1024      | The maximum number of columns Drill will accept from Splunk     | 1.22  |
+| maxCacheSize          | 10000     | The size (in bytes) of Splunk's schema cache.                   | 1.22  |
+| cacheExpiration       | 1024      | The number of minutes for Drill to persist the schema cache.    | 1.22  |
 
 [^1]: See [this Splunk documentation](https://docs.splunk.com/Documentation/Splunk/latest/Admin/Apparchitectureandobjectownership) for more information.
 [^2]: See [this Splunk documentation](https://docs.splunk.com/Documentation/Splunk/latest/Security/CreateAuthTokens) for more information.
@@ -46,6 +49,10 @@ To bypass it by Drill please specify "reconnectRetries": 3. It allows you to ret
 ### User Translation
 The Splunk plugin supports user translation.  Simply set the `authMode` parameter to `USER_TRANSLATION` and use either the plain or vault credential provider for credentials.
 
+## Schema Caching
+For every query that you send to Splunk from Drill, Drill will have to pull schema information from Splunk.  If you have a lot of indexes, this process can cause slow planning time.  To improve planning time, you can configure Drill to cache the index names so that it does not need to make additional calls to Splunk.
+
+There are two configuration parameters for the schema caching: `maxCacheSize` and `cacheExpiration`.  The maxCacheSize defaults to 10k bytes and the `cacheExpiration` defaults to 1024 minutes.  To disable schema caching simply set the `cacheExpiration` parameter to a value less than zero.
 
 ## Understanding Splunk's Data Model
 Splunk's primary use case is analyzing event logs with a timestamp. As such, data is indexed by the timestamp, with the most recent data being indexed first.  By default, Splunk

contrib/storage-splunk/pom.xml

@@ -58,6 +58,12 @@
         </exclusion>
       </exclusions>
     </dependency>
+    <dependency>
+      <groupId>com.github.ben-manes.caffeine</groupId>
+      <artifactId>caffeine</artifactId>
+      <version>${caffeine.version}</version>
+    </dependency>
+
     <!-- Test dependencies -->
     <dependency>
       <groupId>org.apache.drill.exec</groupId>

contrib/storage-splunk/src/main/java/org/apache/drill/exec/store/splunk/SplunkBatchReader.java

@@ -54,9 +54,6 @@ public class SplunkBatchReader implements ManagedReader<SchemaNegotiator> {
   private static final Logger logger = LoggerFactory.getLogger(SplunkBatchReader.class);
   private static final List<String> INT_COLS = new ArrayList<>(Arrays.asList("date_hour", "date_mday", "date_minute", "date_second", "date_year", "linecount"));
   private static final List<String> TS_COLS = new ArrayList<>(Arrays.asList("_indextime", "_time"));
-  private static final String EARLIEST_TIME_COLUMN = "earliestTime";
-  private static final String LATEST_TIME_COLUMN = "latestTime";
-
   private final SplunkPluginConfig config;
   private final SplunkSubScan subScan;
   private final List<SchemaPath> projectedColumns;
@@ -88,6 +85,8 @@ public SplunkBatchReader(SplunkPluginConfig config, SplunkSubScan subScan) {
     RowListProcessor rowProcessor = new RowListProcessor();
     csvSettings.setProcessor(rowProcessor);
     csvSettings.setMaxCharsPerColumn(ValueVector.MAX_BUFFER_SIZE);
+    // Splunk can produce a lot of columns. The SDK default maximum is 512.
+    csvSettings.setMaxColumns(config.getMaxColumns());
   }
 
   @Override
@@ -174,7 +173,7 @@ private TupleMetadata buildSchema() {
         }
       }
     }
-    logger.debug("Time to build schmea: {} milliseconds", timer.elapsed().getNano() / 100000);
+    logger.debug("Time to build schema: {} milliseconds", timer.elapsed().getNano() / 100000);
     return builder.buildSchema();
   }
 
@@ -241,18 +240,18 @@ private String buildQueryString () {
 
     // Splunk searches perform best when they are time bound.  This allows the user to set
     // default time boundaries in the config.  These will be overwritten in filter pushdowns
-    if (filters != null && filters.containsKey(EARLIEST_TIME_COLUMN)) {
-      earliestTime = filters.get(EARLIEST_TIME_COLUMN).value.value.toString();
+    if (filters != null && filters.containsKey(SplunkUtils.EARLIEST_TIME_COLUMN)) {
+      earliestTime = filters.get(SplunkUtils.EARLIEST_TIME_COLUMN).value.value.toString();
 
       // Remove from map
-      filters.remove(EARLIEST_TIME_COLUMN);
+      filters.remove(SplunkUtils.EARLIEST_TIME_COLUMN);
     }
 
-    if (filters != null && filters.containsKey(LATEST_TIME_COLUMN)) {
-      latestTime = filters.get(LATEST_TIME_COLUMN).value.value.toString();
+    if (filters != null && filters.containsKey(SplunkUtils.LATEST_TIME_COLUMN)) {
+      latestTime = filters.get(SplunkUtils.LATEST_TIME_COLUMN).value.value.toString();
 
       // Remove from map so they are not pushed down into the query
-      filters.remove(LATEST_TIME_COLUMN);
+      filters.remove(SplunkUtils.LATEST_TIME_COLUMN);
     }
 
     if (earliestTime == null) {

contrib/storage-splunk/src/main/java/org/apache/drill/exec/store/splunk/SplunkGroupScan.java

@@ -309,6 +309,63 @@ public GroupScan clone(List<SchemaPath> columns) {
     return new SplunkGroupScan(this, columns);
   }
 
+  /**
+   * Generates the query which will be sent to Splunk. This method exists for debugging purposes so
+   * that the actual SPL will be recorded in the query plan.
+   */
+  private String generateQuery() {
+    String earliestTime = null;
+    String latestTime = null;
+
+    // Splunk searches perform best when they are time bound.  This allows the user to set
+    // default time boundaries in the config.  These will be overwritten in filter pushdowns
+    if (filters != null && filters.containsKey(SplunkUtils.EARLIEST_TIME_COLUMN)) {
+      earliestTime = filters.get(SplunkUtils.EARLIEST_TIME_COLUMN).value.value.toString();
+    }
+
+    if (filters != null && filters.containsKey(SplunkUtils.LATEST_TIME_COLUMN)) {
+      latestTime = filters.get(SplunkUtils.LATEST_TIME_COLUMN).value.value.toString();
+    }
+
+    if (earliestTime == null) {
+      earliestTime = config.getEarliestTime();
+    }
+
+    if (latestTime == null) {
+      latestTime = config.getLatestTime();
+    }
+
+    // Special case: If the user wishes to send arbitrary SPL to Splunk, the user can use the "SPL"
+    // Index and spl filter
+    if (splunkScanSpec.getIndexName().equalsIgnoreCase("spl")) {
+      if (filters != null && filters.containsKey("spl")) {
+        return filters.get("spl").value.value.toString();
+      }
+    }
+
+    SplunkQueryBuilder builder = new SplunkQueryBuilder(splunkScanSpec.getIndexName());
+
+    // Set the sourcetype
+    if (filters != null && filters.containsKey("sourcetype")) {
+      String sourcetype = filters.get("sourcetype").value.value.toString();
+      builder.addSourceType(sourcetype);
+    }
+
+    // Add projected columns, skipping star and specials.
+    for (SchemaPath projectedColumn: columns) {
+      builder.addField(projectedColumn.getAsUnescapedPath());
+    }
+
+    // Apply filters
+    builder.addFilters(filters);
+
+    // Apply limits
+    if (maxRecords > 0) {
+      builder.addLimit(maxRecords);
+    }
+    return builder.build();
+  }
+
   @Override
   public int hashCode() {
 
@@ -344,6 +401,7 @@ public String toString() {
       .field("scan spec", splunkScanSpec)
       .field("columns", columns)
       .field("maxRecords", maxRecords)
+      .field("spl", generateQuery())
       .toString();
   }
 }

contrib/storage-splunk/src/main/java/org/apache/drill/exec/store/splunk/SplunkPluginConfig.java

@@ -40,6 +40,9 @@ public class SplunkPluginConfig extends StoragePluginConfig {
   public static final String NAME = "splunk";
   public static final int DISABLED_RECONNECT_RETRIES = 1;
   public static final int DEFAULT_WRITER_BATCH_SIZE = 1000;
+  public static final int DEFAULT_MAX_READER_COLUMNS = 1024;
+  public static final int DEFAULT_MAX_CACHE_SIZE = 10000;
+  public static final int DEFAULT_CACHE_EXPIRATION = 1024;
 
   private final String scheme;
   private final String hostname;
@@ -55,6 +58,9 @@ public class SplunkPluginConfig extends StoragePluginConfig {
   private final Integer reconnectRetries;
   private final boolean writable;
   private final Integer writerBatchSize;
+  private final Integer maxColumns;
+  private final Integer maxCacheSize;
+  private final Integer cacheExpiration;
 
   @JsonCreator
   public SplunkPluginConfig(@JsonProperty("username") String username,
@@ -74,7 +80,11 @@ public SplunkPluginConfig(@JsonProperty("username") String username,
                             @JsonProperty("reconnectRetries") Integer reconnectRetries,
                             @JsonProperty("authMode") String authMode,
                             @JsonProperty("writable") boolean writable,
-                            @JsonProperty("writableBatchSize") Integer writerBatchSize) {
+                            @JsonProperty("writableBatchSize") Integer writerBatchSize,
+                            @JsonProperty("maxColumns") Integer maxColumns,
+                            @JsonProperty("maxCacheSize") Integer maxCacheSize,
+                            @JsonProperty("cacheExpiration") Integer cacheExpiration
+      ) {
     super(CredentialProviderUtils.getCredentialsProvider(username, password, credentialsProvider),
         credentialsProvider == null, AuthMode.parseOrDefault(authMode, AuthMode.SHARED_USER));
     this.scheme = scheme;
@@ -91,6 +101,9 @@ public SplunkPluginConfig(@JsonProperty("username") String username,
     this.latestTime = latestTime == null ? "now" : latestTime;
     this.reconnectRetries = reconnectRetries;
     this.writerBatchSize = writerBatchSize;
+    this.maxColumns = maxColumns;
+    this.maxCacheSize = maxCacheSize;
+    this.cacheExpiration = cacheExpiration;
   }
 
   private SplunkPluginConfig(SplunkPluginConfig that, CredentialsProvider credentialsProvider) {
@@ -109,6 +122,9 @@ private SplunkPluginConfig(SplunkPluginConfig that, CredentialsProvider credenti
     this.latestTime = that.latestTime;
     this.reconnectRetries = that.reconnectRetries;
     this.writerBatchSize = that.writerBatchSize;
+    this.maxColumns = that.maxColumns;
+    this.maxCacheSize = that.maxCacheSize;
+    this.cacheExpiration = that.cacheExpiration;
   }
 
   /**
@@ -225,6 +241,21 @@ public int getWriterBatchSize() {
     return writerBatchSize != null ? writerBatchSize : DEFAULT_WRITER_BATCH_SIZE;
   }
 
+  @JsonProperty("maxColumns")
+  public int getMaxColumns() {
+    return maxColumns != null ? maxColumns : DEFAULT_MAX_READER_COLUMNS;
+  }
+
+  @JsonProperty("maxCacheSize")
+  public int getMaxCacheSize() {
+    return maxCacheSize != null ? maxCacheSize : DEFAULT_MAX_CACHE_SIZE;
+  }
+
+  @JsonProperty("cacheExpiration")
+  public int getCacheExpiration() {
+    return cacheExpiration != null ? cacheExpiration : DEFAULT_CACHE_EXPIRATION;
+  }
+
   private static CredentialsProvider getCredentialsProvider(CredentialsProvider credentialsProvider) {
     return credentialsProvider != null ? credentialsProvider : PlainCredentialsProvider.EMPTY_CREDENTIALS_PROVIDER;
   }
@@ -250,26 +281,32 @@ public boolean equals(Object that) {
       Objects.equals(validateHostname, thatConfig.validateHostname) &&
       Objects.equals(earliestTime, thatConfig.earliestTime) &&
       Objects.equals(latestTime, thatConfig.latestTime) &&
-      Objects.equals(authMode, thatConfig.authMode);
+      Objects.equals(authMode, thatConfig.authMode) &&
+        Objects.equals(maxCacheSize, thatConfig.maxCacheSize) &&
+        Objects.equals(maxColumns, thatConfig.maxColumns) &&
+        Objects.equals(cacheExpiration, thatConfig.cacheExpiration);
   }
 
   @Override
   public int hashCode() {
     return Objects.hash(
-      credentialsProvider,
-      scheme,
-      hostname,
-      port,
-      app,
-      owner,
-      token,
-      cookie,
-      writable,
-      validateCertificates,
-      validateHostname,
-      earliestTime,
-      latestTime,
-      authMode
+        credentialsProvider,
+        scheme,
+        hostname,
+        port,
+        app,
+        owner,
+        token,
+        cookie,
+        writable,
+        validateCertificates,
+        validateHostname,
+        earliestTime,
+        latestTime,
+        authMode,
+        cacheExpiration,
+        maxCacheSize,
+        maxColumns
     );
   }
 
@@ -290,6 +327,9 @@ public String toString() {
       .field("earliestTime", earliestTime)
       .field("latestTime", latestTime)
       .field("Authentication Mode", authMode)
+      .field("maxColumns", maxColumns)
+      .field("maxCacheSize", maxCacheSize)
+      .field("cacheExpiration", cacheExpiration)
       .toString();
   }