diff --git a/src/main/java/net/openhft/chronicle/hash/ChronicleHashBuilder.java b/src/main/java/net/openhft/chronicle/hash/ChronicleHashBuilder.java
index 13ddf9980..737863e26 100755
--- a/src/main/java/net/openhft/chronicle/hash/ChronicleHashBuilder.java
+++ b/src/main/java/net/openhft/chronicle/hash/ChronicleHashBuilder.java
@@ -624,6 +624,7 @@ public interface ChronicleHashBuilder<K, H extends ChronicleHash<K, ?, ?, ?>,
      * @see #createOrRecoverPersistedTo(File, boolean)
      * @see #recoverPersistedTo(File, boolean)
      */
+    @Beta
     H createOrRecoverPersistedTo(
             File file, boolean sameLibraryVersion,
             ChronicleHashCorruption.Listener corruptionListener) throws IOException;
@@ -739,6 +740,7 @@ public interface ChronicleHashBuilder<K, H extends ChronicleHash<K, ?, ?, ?>,
      * @see #createOrRecoverPersistedTo(File, boolean, ChronicleHashCorruption.Listener)
      * @see #createPersistedTo(File)
      */
+    @Beta
     H recoverPersistedTo(
             File file, boolean sameBuilderConfigAndLibraryVersion,
             ChronicleHashCorruption.Listener corruptionListener) throws IOException;
diff --git a/src/main/java/net/openhft/chronicle/hash/ChronicleHashCorruption.java b/src/main/java/net/openhft/chronicle/hash/ChronicleHashCorruption.java
index 524783e83..46ff75ee5 100644
--- a/src/main/java/net/openhft/chronicle/hash/ChronicleHashCorruption.java
+++ b/src/main/java/net/openhft/chronicle/hash/ChronicleHashCorruption.java
@@ -38,9 +38,13 @@ import java.io.File;
  * used outside of the {@link Listener#onCorruption(ChronicleHashCorruption)} method body, because
  * {@code ChronicleHashCorruption} objects could be reused during the recovery procedure.
  *
+ * <p>During a recovery procedure, <i>{@link Listener#onCorruption(ChronicleHashCorruption)} might
+ * be called concurrently from multiple threads.</i> If the implementation of this method calls some
+ * methods on some objects, that are not safe for concurrent use from multiple threads, the
+ * implementation must care about synchronization itself.
  *
- * @see ChronicleHashBuilder#recoverPersistedTo(File, boolean, Listener)
- * @see ChronicleHashBuilder#createOrRecoverPersistedTo(File, boolean, Listener)
+ * @see ChronicleHashBuilder#recoverPersistedTo(File, boolean, ChronicleHashCorruption.Listener)
+ * @see ChronicleHashBuilder#createOrRecoverPersistedTo(File, boolean, ChronicleHashCorruption.Listener)
  */
 @Beta
 public interface ChronicleHashCorruption {
@@ -73,9 +77,14 @@ public interface ChronicleHashCorruption {
     @Beta
     interface Listener {
         /**
-         * Called, when <a href="https://github.com/OpenHFT/Chronicle-Map#recovery">recovery</a>
+         * Called when <a href="https://github.com/OpenHFT/Chronicle-Map#recovery">recovery</a>
          * procedure encounters a corruption of a persisted Chronicle Map.
          *
+         * <p>During a recovery procedure, <i>this method might be called concurrently from multiple
+         * threads.</i> If the implementation of this method calls some methods on some objects,
+         * that are not safe for concurrent use from multiple threads, the implementation must
+         * care about synchronization itself.
+         *
          * @param corruption the corruption object, must not be saved and used outside the body of
          *                   the {@code #onCorruption()} method, because during the recovery
          *                   procedure, the same corruption object could be reused.
diff --git a/src/main/java/net/openhft/chronicle/hash/serialization/SizedReader.java b/src/main/java/net/openhft/chronicle/hash/serialization/SizedReader.java
index 36707d505..82e7f8116 100644
--- a/src/main/java/net/openhft/chronicle/hash/serialization/SizedReader.java
+++ b/src/main/java/net/openhft/chronicle/hash/serialization/SizedReader.java
@@ -53,7 +53,7 @@ public interface SizedReader<T> extends Marshallable {
      * given {@code using} object, i. e. to read the deserialized data into the given object. If it
      * is possible, this objects then returned from this method. If it is impossible for any reason,
      * a new object should be created and returned. The given {@code using} object could be {@code
-     * null}, in this case this method, of cause, should create a new object.
+     * null}, in this case read() should always create a new object.
      *
      * <p>This method should increment the position in the given {@code Bytes} by the given {@code
      * size}.