vvcephei commented on a change in pull request #9098:
URL: https://github.com/apache/kafka/pull/9098#discussion_r469327500
##########
File path:
streams/src/main/java/org/apache/kafka/streams/state/internals/metrics/RocksDBMetricsRecorder.java
##########
@@ -187,37 +206,39 @@ public void record(final long now) {
long bytesReadDuringCompaction = 0;
long numberOfOpenFiles = 0;
long numberOfFileErrors = 0;
- for (final DbAndCacheAndStatistics valueProviders :
storeToValueProviders.values()) {
- bytesWrittenToDatabase +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BYTES_WRITTEN);
- bytesReadFromDatabase +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BYTES_READ);
- memtableBytesFlushed +=
valueProviders.statistics.getAndResetTickerCount(TickerType.FLUSH_WRITE_BYTES);
- memtableHits +=
valueProviders.statistics.getAndResetTickerCount(TickerType.MEMTABLE_HIT);
- memtableMisses +=
valueProviders.statistics.getAndResetTickerCount(TickerType.MEMTABLE_MISS);
- blockCacheDataHits +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BLOCK_CACHE_DATA_HIT);
- blockCacheDataMisses +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BLOCK_CACHE_DATA_MISS);
- blockCacheIndexHits +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BLOCK_CACHE_INDEX_HIT);
- blockCacheIndexMisses +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BLOCK_CACHE_INDEX_MISS);
- blockCacheFilterHits +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BLOCK_CACHE_FILTER_HIT);
- blockCacheFilterMisses +=
valueProviders.statistics.getAndResetTickerCount(TickerType.BLOCK_CACHE_FILTER_MISS);
- writeStallDuration +=
valueProviders.statistics.getAndResetTickerCount(TickerType.STALL_MICROS);
- bytesWrittenDuringCompaction +=
valueProviders.statistics.getAndResetTickerCount(TickerType.COMPACT_WRITE_BYTES);
- bytesReadDuringCompaction +=
valueProviders.statistics.getAndResetTickerCount(TickerType.COMPACT_READ_BYTES);
- numberOfOpenFiles +=
valueProviders.statistics.getAndResetTickerCount(TickerType.NO_FILE_OPENS)
- -
valueProviders.statistics.getAndResetTickerCount(TickerType.NO_FILE_CLOSES);
- numberOfFileErrors +=
valueProviders.statistics.getAndResetTickerCount(TickerType.NO_FILE_ERRORS);
+ if (storeToValueProviders.values().stream().anyMatch(valueProviders ->
valueProviders.statistics != null)) {
Review comment:
I think it's implied in `addValueProviders` that if any of the
valueProviders' statistics are non-null, then they are all non-null, in which
case, this makes sense as a guard. Still, it's kind of subtle.
Why not just put a guard inside the loop instead to `continue` or `break` if
it turns out that `valueProviders.statistics == null`?
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
