Closed mccartney closed 2 years ago
Merging #24 (72bf6dc) into master (2ec992d) will increase coverage by
0.59%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## master #24 +/- ##
============================================
+ Coverage 61.63% 62.22% +0.59%
- Complexity 65 66 +1
============================================
Files 9 9
Lines 318 323 +5
Branches 27 28 +1
============================================
+ Hits 196 201 +5
Misses 108 108
Partials 14 14
Impacted Files | Coverage Δ | |
---|---|---|
...a/com/sumologic/http/queue/BufferWithEviction.java | 66.66% <ø> (ø) |
|
.../sumologic/http/sender/SumoBufferFlushingTask.java | 90.47% <ø> (ø) |
|
...sumologic/http/aggregation/BufferFlushingTask.java | 93.10% <100.00%> (+1.43%) |
:arrow_up: |
.../sumologic/http/aggregation/SumoBufferFlusher.java | 92.30% <100.00%> (ø) |
|
...m/sumologic/http/queue/BufferWithFifoEviction.java | 80.76% <100.00%> (ø) |
|
...mologic/http/queue/CostBoundedConcurrentQueue.java | 88.88% <100.00%> (ø) |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 2ec992d...72bf6dc. Read the comment docs.
Why: Discovered with seemingly unrelated code change.
SumoHttpSenderTest.testRetryRegex()
has failed semi-deterministically (quite deterministic per machine, per JDK version, for some machines) with:It turned out
flushAndSend()
didn't honor themessagesPerRequest
setting. Thus sometimes (depending on how quickly the messages appeared in the input queue) it sent more than 1 message in one go.Which was not what
SumoHttpSenderTest.testRetryRegex()
was expecting. Also I suppose legitimate/production code setting themessagesPerRequest
wouldn't want to see more messages sent.What:
BufferFlushingTask
on flush sends at mostmessagesPerRequest
in one go.