googleapis · mutianf · Sep 8, 2026 · Sep 8, 2026
@@ -167,32 +167,15 @@ workers accordingly.
 
 ### Snappy compressed Snapshots
 
-1. Set the environment variables.
-    ```
-    PROJECT_ID=your-project-id
-    INSTANCE_ID=your-instance-id
-    TABLE_NAME=your-table-name
-    REGION=us-central1
-
-    SNAPSHOT_GCS_PATH="$BUCKET_NAME/hbase-migration-snap"
-    SNAPSHOT_NAME=your-snapshot-name
-    ```
-
-1. Run the import.
-    ```
-    java -jar bigtable-beam-import-2.3.0.jar importsnapshot \
-        --runner=DataflowRunner \
-        --project=$PROJECT_ID \
-        --bigtableInstanceId=$INSTANCE_ID \
-        --bigtableTableId=$TABLE_NAME \
-        --hbaseSnapshotSourceDir=$SNAPSHOT_GCS_PATH/data \
-        --snapshotName=$SNAPSHOT_NAME \
-        --stagingLocation=$SNAPSHOT_GCS_PATH/staging \
-        --gcpTempLocation=$SNAPSHOT_GCS_PATH/temp \
-        --maxNumWorkers=$(expr 3 \* $CLUSTER_NUM_NODES) \
-        --region=$REGION \
-        --enableSnappy=true
-    ```
+Snappy compressed snapshots are read natively, so they are imported with the
+exact same command as above, with no extra configuration.
+
+> [!WARNING]
+> The `--enableSnappy` flag is **deprecated**. Since
+> [#4338](https://github.com/googleapis/java-bigtable-hbase/pull/4338) (released
+> in 2.14.0) the import job runs on HBase 2.x / Hadoop 3, which decompresses
+> Snappy without the custom worker container the flag used to install. The flag
+> is accepted but ignored, and it will be removed in a future release.
 
 ### Sequence Files
 

@@ -26,7 +26,7 @@ The script relies on the following environment variables. You should set them be
 | `DISK_SIZE_GB` | *Optional* | Worker disk size in Gigabytes. | `500` (Default) |
 | `MAX_NUM_WORKERS` | *Optional* | Maximum number of active Dataflow workers per job. | `10` (Default) |
 | `USE_PUBLIC_IPS` | *Optional* | Whether Dataflow workers should have public IPs. | `false` (Default) |
-| `ENABLE_SNAPPY` | *Optional* | Whether to enable Snappy compression for transit files. | `true` (Default) |
+| `ENABLE_SNAPPY` | **Deprecated** | Ignored. Snappy compressed snapshots are read natively, so no flag is needed. Setting it only prints a warning. | N/A |
 
 ## Understanding Sharding
 

@@ -114,7 +114,11 @@ WORKER_MACHINE_TYPE="${WORKER_MACHINE_TYPE:-n1-highmem-4}"
 DISK_SIZE_GB="${DISK_SIZE_GB:-500}"
 MAX_NUM_WORKERS="${MAX_NUM_WORKERS:-10}"
 USE_PUBLIC_IPS="${USE_PUBLIC_IPS:-false}"
-ENABLE_SNAPPY="${ENABLE_SNAPPY:-true}"
+
+# Deprecated: Snappy compressed snapshots are read natively, this variable is ignored.
+if [ -n "${ENABLE_SNAPPY:-}" ]; then
+    echo "⚠️  Warning: ENABLE_SNAPPY is deprecated and ignored. Snappy compressed snapshots are supported natively."
+fi
 
 # Generate a safe, unique job name prefix to prevent collisions
 SAFE_TABLE_NAME=$(echo "${TABLE_NAME}" | tr '[:upper:]' '[:lower:]' | tr '_' '-' | tr -cd '[:alnum:]-')
@@ -284,7 +288,6 @@ for (( i=START_SHARD; i<=END_SHARD; i++ )); do
     --region="${REGION}" \
     "${SERVICE_ACCOUNT_ARGS[@]}" \
     --usePublicIps="${USE_PUBLIC_IPS}" \
-    --enableSnappy="${ENABLE_SNAPPY}" \
     --skipRestoreStep="${SKIP_RESTORE}" \
     --deleteRestoredSnapshots=false \
     --restorePath="${RESTORE_DIR}" \

@@ -120,10 +120,23 @@ public interface ImportOptions extends ImportJob.ImportOptions {
     @SuppressWarnings("unused")
     void setSnapshotName(String snapshotName);
 
-    @Description("Is importing Snappy compressed snapshot.")
+    /**
+     * @deprecated This option is a no-op since the job moved to HBase 2.x / Hadoop 3. Snappy
+     *     compressed snapshots are read natively, so the flag no longer needs to be set and is
+     *     ignored. It will be removed in a future release.
+     */
+    @Deprecated
+    @Description(
+        "Deprecated: this option is ignored. Snappy compressed snapshots are supported natively"
+            + " and no longer require a flag.")
     @Default.Boolean(false)
     Boolean getEnableSnappy();
 
+    /**
+     * @deprecated This option is a no-op since the job moved to HBase 2.x / Hadoop 3. See {@link
+     *     #getEnableSnappy()}.
+     */
+    @Deprecated
     @SuppressWarnings("unused")
     void setEnableSnappy(Boolean enableSnappy);
 
@@ -260,6 +273,8 @@ public static void main(String[] args) throws Exception {
     // To determine the Google Cloud Storage file scheme (gs://)
     FileSystems.setDefaultPipelineOptions(options);
 
+    warnIfSnappyFlagSet(options);
+
     LOG.info("Building Pipeline");
     Pipeline pipeline = null;
     ImportConfig importConfig = null;
@@ -286,6 +301,16 @@ public static void main(String[] args) throws Exception {
     }
   }
 
+  @SuppressWarnings("deprecation")
+  private static void warnIfSnappyFlagSet(ImportOptions options) {
+    if (Boolean.TRUE.equals(options.getEnableSnappy())) {
+      LOG.warn(
+          "--enableSnappy is deprecated and ignored. Snappy compressed snapshots are read natively"
+              + " since the job moved to HBase 2.x / Hadoop 3; the flag will be removed in a future"
+              + " release.");
+    }
+  }
+
   @VisibleForTesting
   static ImportConfig buildImportConfigFromConfigFile(String configFilePath) throws Exception {
     Gson gson = new GsonBuilder().create();

@@ -419,8 +419,8 @@ public void testHBaseSnapshotImportWithCorruptions() throws Exception {
   @Test
   public void testSnappyCompressedHBaseSnapshotImport() throws Exception {
     // Start import
+    // Snappy compressed snapshots are read natively, no flag needed.
     ImportOptions importOpts = createImportOptions();
-    importOpts.setEnableSnappy(true);
     importOpts.setSnapshotName(TEST_SNAPPY_SNAPSHOT_NAME);
 
     // run pipeline