Skip to content

MINOR: AbstractIndex.close should unmap #5757

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ijuma merged 2 commits into from
Oct 11, 2018
Merged

MINOR: AbstractIndex.close should unmap #5757

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a335181
MINOR: AbstractIndex.close should unmap
ijuma Oct 8, 2018
a37aabd
Call `closeHandler` from `deleteIfExists`
ijuma Oct 8, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 6 additions & 7 deletions core/src/main/scala/kafka/log/AbstractIndex.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,13 +223,7 @@ abstract class AbstractIndex[K, V](@volatile var file: File, val baseOffset: Lon
* not exist
*/
def deleteIfExists(): Boolean = {
inLock(lock) {
// On JVM, a memory mapping is typically unmapped by garbage collector.
// However, in some cases it can pause application threads(STW) for a long moment reading metadata from a physical disk.
// To prevent this, we forcefully cleanup memory mapping within proper execution which never affects API responsiveness.
// See https://issues.apache.org/jira/browse/KAFKA-4614 for the details.
safeForceUnmap()
}
closeHandler()
Files.deleteIfExists(file.toPath)
}

Expand All @@ -251,9 +245,14 @@ abstract class AbstractIndex[K, V](@volatile var file: File, val baseOffset: Lon
/** Close the index */
def close() {
trimToValidSize()
closeHandler()
}

def closeHandler(): Unit = {
// On JVM, a memory mapping is typically unmapped by garbage collector.
// However, in some cases it can pause application threads(STW) for a long moment reading metadata from a physical disk.
// To prevent this, we forcefully cleanup memory mapping within proper execution which never affects API responsiveness.
// See https://issues.apache.org/jira/browse/KAFKA-4614 for the details.
inLock(lock) {
safeForceUnmap()
}
Expand Down