aokolnychyi commented on code in PR #53143:
URL: https://github.com/apache/spark/pull/53143#discussion_r2547510370
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/CacheManager.scala:
##########
@@ -352,22 +353,35 @@ class CacheManager extends Logging with
AdaptiveSparkPlanHelper {
}
needToRecache.foreach { cd =>
cd.cachedRepresentation.cacheBuilder.clearCache()
+ tryRebuildCacheEntry(spark, cd).foreach { entry =>
+ this.synchronized {
+ if (lookupCachedDataInternal(entry.plan).nonEmpty) {
+ logWarning("While recaching, data was already added to cache.")
+ } else {
+ cachedData = entry +: cachedData
+ CacheManager.logCacheOperation(log"Re-cached Dataframe cache
entry:" +
+ log"${MDC(DATAFRAME_CACHE_ENTRY, entry)}")
+ }
+ }
+ }
+ }
+ }
+
+ private def tryRebuildCacheEntry(
+ spark: SparkSession,
+ cd: CachedData): Option[CachedData] = {
+ try {
val sessionWithConfigsOff = getOrCloneSessionWithConfigsOff(spark)
val (newKey, newCache) = sessionWithConfigsOff.withActive {
val refreshedPlan = V2TableRefreshUtil.refresh(cd.plan)
Review Comment:
It does seem like all our current invocations treat cache refresh as
opportunistic. In other words, it is usually the last step where it is critical
to remove the old entry but reaching may or may not succeed. Is that the same
understanding that everyone has? Any cases I missed?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
