- 31 12月, 2015 3 次提交
-
-
由 Ufuk Celebi 提交于
-
由 Ufuk Celebi 提交于
Adds a simple tracker of checkpoint statistics.
-
由 Ufuk Celebi 提交于
[FLINK-3131] [contrib, runtime, streaming-java] Add long getStateSize() to StateHandle and KvStateSnapshot In order to report the state sizes, we need to expose them. All state backends currently available backends know the state size. Only the LazyDbKvState does not expose it at the moment, because it serializes the data lazily. This can be changed in a follow-up fix.
-
- 28 12月, 2015 2 次提交
-
-
由 vasia 提交于
This closes #1467
-
由 mjsax 提交于
- removed BoltWrapperTwoInputs - add generic multi-input stream bolt wrapper - updated FlinkTopology to translate multi-input bolts correctly - added multi-input union test wrapper - updated FlinkTopology to translate multi-input bolts correctly - added multi-input union test
-
- 27 12月, 2015 6 次提交
-
-
由 Till Rohrmann 提交于
This closes #1456
-
由 Stephan Ewen 提交于
-
由 Greg Hogan 提交于
[FLINK-3166] [runtime] The first program in ObjectReuseITCase has the wrong expected result, and it succeeds - TestEnvironment now honors configuration of object reuse - Fixed reduce transformations to allow the user to modify and return either input This closes #1464
-
由 Ufuk Celebi 提交于
This closes #1472
-
由 Ufuk Celebi 提交于
-
由 Jun Aoki 提交于
WriteMode has 2 values NO_OVERWRITE and OVERWRITE but it refers CREATE in the javadoc. Probably old code. This closes #1475
-
- 24 12月, 2015 1 次提交
-
-
由 tedyu 提交于
[FLINK-3103] [runtime] Remove unnecessary synchronization in FsStateBackend.FsCheckpointStateOutputStream This closes 1474
-
- 21 12月, 2015 5 次提交
-
-
由 Till Rohrmann 提交于
This closes #1473 This closes #1098 This closes #1459
-
由 Ufuk Celebi 提交于
- The off-heap stats reported by the memory logger didn't include direct memory. This closes #1466
-
由 Jun Aoki 提交于
This closes #1444
-
由 Fabian Hueske 提交于
-
由 chengxiang li 提交于
This closes #1255
-
- 18 12月, 2015 1 次提交
-
-
由 mjsax 提交于
Enalbes submission of Flink programs written in Clojure. This closes #1457
-
- 17 12月, 2015 2 次提交
-
-
由 Maximilian Michels 提交于
- use AMRMClientAsync instead of AMRMClient - handle allocation and startup of containers in callbacks - remove YarnHeartbeat message The AMRMClientAsync uses one thread to communicate with the resource manager and an additional thread to execute the callbacks. This closes #1450.
-
由 Peter Vandenabeele 提交于
This closes #1463.
-
- 16 12月, 2015 6 次提交
-
-
由 Till Rohrmann 提交于
-
由 Ufuk Celebi 提交于
This closes #1439
-
由 Nick Dimiduk 提交于
This closes #1442
-
由 Greg Hogan 提交于
- Memoize string representations of AbstractID. - Use lookup table for byte-to-hex conversion in StringUtils. This closes #1455
-
由 Maximilian Michels 提交于
-
由 Aljoscha Krettek 提交于
Before, when the Kafka Source would emit a final Long.MAX_VALUE watermark to signal the end of elements (i.e. no partition assigned). This would trip up the AutomaticWatermarkContext and the NonWatermarkContext. Now, a Long.MAX_VALUE watermark is allowed through.
-
- 15 12月, 2015 9 次提交
-
-
由 Stephan Ewen 提交于
[FLINK-3171] [misc] Consolidate zoo of wrapper classes for input/output-stream to data-input/output-view.
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
[hotfix] Minor cleanups in classes FileSystem, OperatingSystem, CopyableValue, GenericCsvInputFormat
-
由 Ufuk Celebi 提交于
This was removed in 3b8b4f0f. The issue was fixed by this, but it disabled local testing as well. This change re-introduces the variable and sets it to the empty string by default. This way, we can still use the proxy server for local testing. This closes #1449
-
由 Ufuk Celebi 提交于
Many browsers don't allow cross-origin HTTP requests if the respective HTTP header is not set by the server. Because of this it was not possible to test changes to the web frontend with the local proxy server and a running job manager. See here: https://developer.mozilla.org/en-US/docs/Web/HTTP/Access_control_CORS
-
- 13 12月, 2015 2 次提交
-
-
由 Sachin Goel 提交于
This closes #1418
-
由 Sachin Goel 提交于
This closes #1422
-
- 11 12月, 2015 2 次提交
-
-
由 Aljoscha Krettek 提交于
Kafka sources that don't read from any partition never emit a watermark, thereby blocking the progress of event-time in downstream operations. This changes the Kafka Source to emit a Long.MAX_VALUE watermark if it knows that it will never receive data. This also changes the Timestamp Extraction operator to reacto to a Long.MAX_VALUE watermark by itself emitting a Long.MAX_VALUE watermark.
-
由 Aljoscha Krettek 提交于
Before, would throw a ClassCastException when emitting watermarks with timestamp/watermark multiplexing disabled.
-
- 10 12月, 2015 1 次提交
-
-
由 Robert Metzger 提交于
-