diff --git a/README.md b/README.md
index 4f936d0..b5de30a 100644
--- a/README.md
+++ b/README.md
@@ -288,7 +288,7 @@ A possible solution to remove duplicates when reading the written data could be
 </tr>
 
 <tr>
-  <td>`poolTimeoutMs`</td>
+  <td>`pollTimeoutMs`</td>
   <td>A number string in unit of milliseconds </td>
   <td>No</td>
   <td>"120000"</td>
diff --git a/src/main/scala/org/apache/spark/sql/pulsar/PulsarSources.scala b/src/main/scala/org/apache/spark/sql/pulsar/PulsarSources.scala
index 990578d..20d2122 100644
--- a/src/main/scala/org/apache/spark/sql/pulsar/PulsarSources.scala
+++ b/src/main/scala/org/apache/spark/sql/pulsar/PulsarSources.scala
@@ -190,12 +190,12 @@ class PulsarSourceInitialOffsetWriter(sparkSession: SparkSession, metadataPath:
   def getInitialOffset(
                         pulsarHelper: PulsarHelper,
                         startingOffsets: PerTopicOffset,
-                        poolTimeoutMs: Int,
+                        pollTimeoutMs: Int,
                         reportDataLoss: String => Unit): SpecificPulsarOffset = {
     val deserializedOffset = get(0).map(markOffsetUserProvided(_))
     deserializedOffset.getOrElse {
       val actualOffsets = SpecificPulsarOffset(
-        pulsarHelper.actualOffsets(startingOffsets, poolTimeoutMs, reportDataLoss))
+        pulsarHelper.actualOffsets(startingOffsets, pollTimeoutMs, reportDataLoss))
       add(0, actualOffsets)
       logInfo(s"Initial Offsets: $actualOffsets")
       actualOffsets