k-raina commented on code in PR #20144:
URL: https://github.com/apache/kafka/pull/20144#discussion_r2254421825
##########
clients/src/main/java/org/apache/kafka/common/telemetry/internals/ClientTelemetryReporter.java:
##########
@@ -713,14 +716,23 @@ private Optional<Builder<?>>
createPushRequest(ClientTelemetrySubscription local
return Optional.empty();
}
- CompressionType compressionType =
ClientTelemetryUtils.preferredCompressionType(localSubscription.acceptedCompressionTypes());
+ CompressionType compressionType =
ClientTelemetryUtils.preferredCompressionType(localSubscription.acceptedCompressionTypes(),
unsupportedCompressionTypes);
ByteBuffer compressedPayload;
try {
compressedPayload = ClientTelemetryUtils.compress(payload,
compressionType);
- } catch (Throwable e) {
- log.debug("Failed to compress telemetry payload for
compression: {}, sending uncompressed data", compressionType);
+ } catch (IOException e) {
+ log.debug("Failed to compress telemetry payload for
compression: {}, sending uncompressed data", compressionType, e);
compressedPayload = ByteBuffer.wrap(payload.toByteArray());
compressionType = CompressionType.NONE;
+ } catch (Throwable e) {
+ if (e instanceof NoClassDefFoundError || e.getCause()
instanceof NoClassDefFoundError) {
+ log.debug("Compression library {} not found, sending
uncompressed data", compressionType, e);
+ unsupportedCompressionTypes.add(compressionType);
Review Comment:
Below are couple of edge case scenarios i can think of where exception will
happen:
1. Libraries set by broker missing in client application -> Handled by
NoClassDefFoundError
2. Stream Compression failure or protobuf failues -> Handled by
IOException
3. Memory overflow like OutOfMemoryError, BufferOverflowException caused due
to maybe large metrics data -> Should be thrown as Runtime Exception
4. System or JVM Exceptions like UnsatisfiedLinkError, StackOverflowError
-> Should be thrown as Runtime Exception
For cases like 3 & 4 Errors seems to be non recoverable, so telemetry will
be blocked next time as mentioned in
[comment](https://github.com/apache/kafka/pull/20144#discussion_r2254399569).
Please let me know your opinion on this!
--
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: jira-unsubscr...@kafka.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
