- 6月 17, 2020
- 3月 03, 2020
-
-
由 Confluent Jenkins Bot 创作于
-
- 3月 01, 2020
-
-
-
由 Brian Bushree 创作于
what/why the throttling_test was broken by this PR (#7785) since it depends on the consumer having partitions-assigned before starting the producer this PR provides the ability to wait for partitions to be assigned in the console consumer before considering it started. caveat this does not support starting up the JmxTool inside the console-consumer for custom metrics while using this wait_until_partitions_assigned flag since the code assumes one JmxTool running per node. I think a proper fix for this would be to make JmxTool its own standalone single-node service alternatives we could use the EndToEnd test suite which uses the verifiable producer/consumer under the hood but I found that there were more changes necessary to get this working unfortunately (specifically doesn't seem like this test suite plays nicely with the ProducerPerformanceService) Reviewers: Mathew Wong <mwong@confluent.io>, Bill Bejeck <bbejeck.com>
-
- 2月 28, 2020
-
-
由 Bill Bejeck 创作于
Adding the ZK upgrade information to the Notable Changes section for 2.4.1 Reviewers: Ron Dagostino <rdagostino@confluent.io>, Jim Galasyn <jim.galasyn@confluent.io>
-
-
由 Matthew Wong 创作于
The test_throttled_reassignment test fails because the consumer that is used to validate reassignment does not start on time to consume all messages. This does not seem like an issue with the throttling of the reassignment, since increasing the timeout allowed the test to pass multiple consecutive runs locally. This test seemed to rely on the default JmxTool for the console consumer that was removed in this commit: 179d0d73 The console consumer would check to see if it had partitions assigned to it before beginning to consume. Although the test occasionally failed with the JmxTool, it began to fail much more after the removal. Error messages of failures followed the below format with varying numbers of missed messages. They are the first messages by the producer. 535 acked message did not make it to the Consumer. They are: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19...plus 515 more. Total Acked: 192792, Total Consumed: 192259. We validated that the first 535 of these missing messages correctly made it into Kafka's data files. This suggests they were lost on their way to the consumer. In the scope of the test, this error suggests that the test is falling into the race condition described in produce_consume_validate.py, which has the timeout to prevent the consumer from missing initial messages. This can serve as a temporary fix until the logic of consumer startup is addressed further. Reviewers: Jason Gustafson <jason@confluent.io>, Bill Bejeck <bbejeck@gmail.com>
-
- 2月 27, 2020
-
-
-
由 Chris Egerton 创作于
Author: Chris Egerton <chrise@confluent.io> Reviewer: Randall Hauch <rhauch@gmail.com>
-
-
由 John Roesler 创作于
Adds a couple of extra checks to the test-output-capturing logic in our gradle build. Previously, we were seeing a lot of error logs while attempting to write output for a test whose output file hadn't been initialized. Reviewers: Ewen Cheslack-Postava <ewen@confluent.io>
-
- 2月 26, 2020
-
-
由 Arvind Thirunarayanan 创作于
-
-
- 2月 25, 2020
-
-
由 Chia-Ping Tsai 创作于
The "offset deletion" and "group rebalance" should not be recorded by the same sensor since they are totally different. The code is introduced by #7276. Reviewers: David Jacot <djacot@confluent.io>, Guozhang Wang <wangguoz@gmail.com>
-
- 2月 22, 2020
-
- 2月 21, 2020
-
-
-
由 David Mao 创作于
Previously, checkpointed offsets for a log were only updated if the log was chosen for cleaning once the cleaning job completes. This caused issues in cases where logs with invalid checkpointed offsets would repeatedly emit warnings if the log with an invalid cleaning checkpoint wasn't chosen for cleaning. Proposed fix is to update the checkpointed offset for logs with invalid checkpoints regardless of whether it gets chosen for cleaning. Reviewers: Anna Povzner <anna@confluent.io>, Jun Rao <junrao@gmail.com>
-
由 Michael Viamari 创作于
Fixes a bug where KStream#transformValues would forward null values from the provided ValueTransform#transform operation. A test was added for verification KStreamTransformValuesTest#shouldEmitNoRecordIfTransformReturnsNull. A parallel test for non-key ValueTransformer was not added, as they share the same code path. Reviewers: Bruno Cadonna <bruno@confluent.io>, Bill Bejeck <bbejeck@gmail.com>
-
- 2月 19, 2020
-
-
由 Matthias J. Sax 创作于
Reviewers: Bill Bejeck <bill@confluent.io>
-
- 2月 18, 2020
-
-
由 Arvind Thirunarayanan 创作于
Reviewers: Ismael Juma <ismael@juma.me.uk>
-
-
由 Chia-Ping Tsai 创作于
Change unit tests to make sure the consumer group is in Stable state (i.e. consumers have completed joining the group)
-
由 Ismael Juma 创作于
A couple of critical fixes: ZOOKEEPER-3644: Data loss after upgrading standalone ZK server 3.4.14 to 3.5.6 with snapshot.trust.empty=true ZOOKEEPER-3701: Split brain on log disk full (3.5) Full release notes: https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310801&version=12346098 Also fixed an existing unused import issue that was causing checkstyle to fail in the cherry-pick to the 2.4 branch. Reviewers: Bill Bejeck <bbejeck@gmail.com>
-
- 2月 17, 2020
-
-
由 Boyang Chen 创作于
Today if we attempt to list offsets with a fenced leader epoch, consumer will retry without updating the metadata until the timeout is reached. This affects synchronous APIs such as `offsetsForTimes`, `beginningOffsets`, and `endOffsets`. The fix in this patch is to trigger the metadata update call whenever we see a retriable error before additional attempts. Reviewers: Jason Gustafson <jason@confluent.io>
-
- 2月 14, 2020
-
-
由 Guozhang Wang 创作于
-
由 Bruno Cadonna 创作于
* Added init() method to RocksDBMetricsRecorder * Added call to init() of RocksDBMetricsRecorder to init() of RocksDB store * Added call to init() of RocksDBMetricsRecorder to openExisting() of segmented state stores * Adapted unit tests * Added integration test that reproduces the situation in which the bug occurred Reviewers: Guozhang Wang <wangguoz@gmail.com>
-
- 2月 13, 2020
-
-
由 Vikas Singh 创作于
-
由 Randall Hauch 创作于
Author: Randall Hauch <rhauch@gmail.com> Reviewer: Jason Gustafson <jason@confluent.io>
-
由 Vikas Singh 创作于
-
由 Konstantine Karantasis 创作于
Author: Konstantine Karantasis <konstantine@confluent.io> Reviewer: Randall Hauch <rhauch@gmail.com>
-
由 Lev Zemlyanov 创作于
Signed-off-by: Lev Zemlyanov <lev@confluent.io>
-
由 Lev Zemlyanov 创作于
Author: Lev Zemlyanov <lev@confluent.io> Reviewers: Greg Harris <gregh@confluent.io>, Randall Hauch <rhauch@gmail.com>
-
由 Jason Gustafson 创作于
The recent increase in the flakiness of one of the offset reset tests (KAFKA-9538) traces back to https://github.com/apache/kafka/pull/7941. After investigation, we found that following this patch, the consumer was sending an additional metadata request prior to performing the group assignment. This slight timing difference was enough to trigger the test failures. The problem turned out to be due to a bug in `SubscriptionState.groupSubscribe`, which no longer counted the local subscription when determining if there were new topics to fetch metadata for. Hence the extra metadata update. This patch restores the old logic. Without the fix, we saw 30-50% test failures locally. With it, I could no longer reproduce the failure. However, #6561 is probably still needed to improve the resilience of this test. Reviewers: Rajini Sivaram <rajinisivaram@googlemail.com>
-
由 Rajini Sivaram 创作于
KAFKA-9181; Maintain clean separation between local and group subscriptions in consumer's SubscriptionState (#7941) Reviewers: Jason Gustafson <jason@confluent.io>, Guozhang Wang <wangguoz@gmail.com>
-
- 2月 12, 2020
-
-
由 John Roesler 创作于
Corrects a flaw leading to an exception while building topologies that include both: * A foreign-key join with the result not explicitly materialized * An operation after the join that requires source materialization Also corrects a flaw in TopologyTestDriver leading to output records being enqueued in the wrong order under some (presumably rare) circumstances. Cherry-pick of 1681c78f from trunk Cherry-pick of 1679839c from 2.5 Reviewers: Matthias J. Sax <matthias@confluent.io>, Guozhang Wang <wangguoz@gmail.com>
-
由 John Roesler 创作于
Migrates TopologyTestDriver processing to be closer to the same processing/ordering semantics as KafkaStreams. This corrects the output order for recursive topologies as reported in KAFKA-9503, and also works similarly in the case of task idling. Cherry-pick of 998f1520 from trunk Cherry-pick of 7b71cb92 from 2.5 Reviewers: Matthias J. Sax <matthias@confluent.io>
-
由 John Roesler 创作于
During the discussion for KIP-213, we decided to pass "pseudo-topics" to the internal serdes we use to construct the wrapper serdes for CombinedKey and hashing the left-hand-side value. However, during the implementation, this strategy wasn't fully implemented, and we wound up using the same topic name for a few different data types. Cherry-pick of e16859dc from trunk Cherry-pick of 557ad551 from 2.5 Reviewers: Guozhang Wang <guozhang@confluent.io>
-
由 Konstantine Karantasis 创作于
Changed `EmbeddedConnectCluster` to add utility methods that return `Response`, throw `ConnectException` instead of `IOException` for failures, and deprecate the old methods that returned primitive types rather than `Response`. Also introduce common assertions for embedded clusters under `EmbeddedConnectClusterAssertions`. Author: Konstantine Karantasis <konstantine@confluent.io> Reviewer: Randall Hauch <rhauch@gmail.com>, Chia-Ping Tsai <chia7712@gmail.com>
-
由 A. Sophie Blee-Goldman 创作于
Reviewers: Guozhang Wang <wangguoz@gmail.com>
-