summaryrefslogtreecommitdiff
path: root/kafka
Commit message (Collapse)AuthorAgeFilesLines
...
* | | | try to fix uncaught FailedPayloadsErrorreAsOn20102015-06-092-1/+10
| | | |
* | | | Add async_stop_timeout parameter to tune how long to let the producerDana Powers2015-06-061-3/+26
| | | | | | | | | | | | | | | | | | | | keep trying to send messages before timing out. Log an error if async producer was stopped before all messages sent.
* | | | Dont stop async producer until all pending messages have been processedDana Powers2015-06-061-1/+1
| | | |
* | | | Update Producer class docstringDana Powers2015-06-061-4/+25
| | | |
* | | | Deprecate async producer batch_send kwarg -- use 'async' insteadDana Powers2015-06-063-26/+44
| | | |
* | | | Log retries and failed messages in async producer (configurable as full ↵Dana Powers2015-06-061-12/+24
| | | | | | | | | | | | | | | | messages or hash())
* | | | PR 331 fixup: do not attempt to get new messages if there are pending retriesDana Powers2015-06-061-4/+9
| | | |
* | | | PR 331 fixup: Rename reqs dict to request_triesDana Powers2015-06-061-9/+13
| | | |
* | | | PR 331 fixup: Dont need try/except when calling send_produce_requests with ↵Dana Powers2015-06-061-15/+16
| | | | | | | | | | | | | | | | fail_on_error=False
* | | | PR 331 fixup: log warnings on async producer backoff and metadata refreshDana Powers2015-06-061-1/+2
| | | |
* | | | PR 331 fixup: Support unlimited retries with async_retry_limit=NoneDana Powers2015-06-061-6/+8
| | | | | | | | | | | | | | | | Async producer defaults are now retry everything always w/ 100ms backoff
* | | | PR 331 fixup: fix _handle_error closureDana Powers2015-06-061-16/+17
| | | |
* | | | PR 331 fixup: Fix kafka.common imports from kafka.producer.base (one import ↵Dana Powers2015-06-061-3/+2
| | | | | | | | | | | | | | | | block, add ProduceResponse)
* | | | KafkaClient should try/except ConnnectionError when calling _get_connDana Powers2015-06-061-2/+2
| | | |
* | | | FailedPayloadsError now only has a single payload; use .payload attributeDana Powers2015-06-061-2/+2
| | | | | | | | | | | | | | | | .failed_payloads attribute has been removed.
* | | | Raise an error if we attempt to group duplicate topic-partition payloadsDana Powers2015-06-062-0/+5
| | | | | | | | | | | | | | | | - previously this would simply drop one of the payloads
* | | | Use separate module loggers instead of a single 'kafka' loggerDana Powers2015-06-069-9/+17
| |_|/ |/| |
* | | Fix import error in kafka/producer/base.pyDana Powers2015-06-061-1/+1
| | |
* | | Set __version__ string in kafka/version.pyDana Powers2015-06-042-3/+2
| | |
* | | Check response.error for async producerViktor Shlapakov2015-06-031-19/+24
| | |
* | | Async producer: py2.6 backward compatibility fixViktor Shlapakov2015-06-031-2/+2
| | |
* | | Improve async producer code: logic and style fixesViktor Shlapakov2015-06-034-43/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - send_producer_request with fail_on_error=False to retry failed reqs only - using an internal dict with with namedtuple keys for retry counters - refresh metadata on refresh_error irrespective to retries options - removed infinite retries (retry_options.limit=None) as an over-feature - separate producer init args for retries options (limit,backoff,on_timeouts) - AsyncProducerQueueFull returns a list of failed messages - producer tests improved thanks to @rogaha and @toli
* | | Change backoff message log levelViktor Shlapakov2015-06-031-1/+1
| | |
* | | Fix async producer queue put argumentsViktor Shlapakov2015-06-031-2/+4
| | |
* | | Fix names for async retries opts, add timeout for putViktor Shlapakov2015-06-034-20/+34
| | |
* | | Simplification of retry logicViktor Shlapakov2015-06-032-31/+29
| | |
* | | Fix small issues with names/testsViktor Shlapakov2015-06-032-9/+11
| | |
* | | async queue: refactored code; add one more testEduard Iskandarov2015-06-031-2/+7
| | |
* | | add producer send batch queue overfilled testИскандаров Эдуард2015-06-031-1/+7
| | |
* | | Disable retry on timeouts by default (old behaviour)Viktor Shlapakov2015-06-031-1/+1
| | |
* | | Clean and simplify retry logicViktor Shlapakov2015-06-032-60/+49
| | |
* | | Arg fixes for base/keyed producersViktor Shlapakov2015-06-032-4/+2
| | |
* | | Improved retry logicViktor Shlapakov2015-06-034-31/+73
| | |
* | | Fixed tests and other issues after rebaseViktor Shlapakov2015-06-032-2/+1
| | |
* | | Returned default behaviour with no retriesViktor Shlapakov2015-06-031-1/+1
| | |
* | | Fix: check failed reqs to retry only for positive limitViktor Shlapakov2015-06-031-3/+3
| | |
* | | Fixed compatible issues with testsViktor Shlapakov2015-06-031-3/+5
| | |
* | | Producer _send_upstream fixes, added tests for retriesViktor Shlapakov2015-06-031-6/+20
| | |
* | | Fixed base producer importsViktor Shlapakov2015-06-031-1/+2
| | |
* | | Retries for async batchingViktor Shlapakov2015-06-034-15/+50
| |/ |/|
* | Check vs None directly.William Ting2015-05-171-4/+4
|/
* Fix calling of _offsetsArturo Filastò2015-05-141-1/+1
| | | | | | | | Previously you would see this error: ``` self.offsets._fetch[(topic, partition)], AttributeError: 'function' object has no attribute '_fetch' ```
* Merge pull request #367 from dpkp/clean_metadata_refreshDana Powers2015-04-121-13/+14
|\ | | | | Clear local metadata cache before refresh in client.load_metadata_for_topics()
| * Update debug log messages in client.load_metadata_for_topicsDana Powers2015-04-121-2/+2
| |
| * clean local metadata before requesting refresh in load_metadata_for_topics()Dana Powers2015-04-121-2/+7
| |
| * Simplify client.reset_topic_metadataDana Powers2015-04-121-9/+5
| |
* | Merge pull request #366 from dpkp/failed_payloadsDana Powers2015-04-121-79/+77
|\ \ | |/ |/| Refactor client request/response handling to support better retries
| * Improve send_produce_request docstringDana Powers2015-04-061-6/+17
| |
| * Add debug logging for all broker-aware requests/responsesDana Powers2015-04-061-0/+3
| |
| * Refactor KafkaClient._send_broker_aware_request to return a list of responsesDana Powers2015-04-061-73/+57
| | | | | | | | | | | | and include individual (unraised) FailedPayloadsError instances rather than always raising a FailedPayloadsError. This should allow producers to determine which payloads succeeded and which failed, specifically.
View Lorry Controller Status