summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Fix sphinx bullet pointsfix-sphinx-docsJeff Widman2018-01-081-1/+1
|
* Fix typo in _try_authenticate_plain (#1333)everpcpc2017-12-281-1/+1
|
* Add security layer negotiation to the GSSAPI authentication. (#1283)Andre Araujo2017-12-261-22/+43
| | | | | | | | | | | | | | | | | | When trying to establish a connection with Kafka using SASL with the GSSAPI authentication mechanism the connection was hanging an timing out after 60 secons. On the Kafka broker side I noticed that the SaslServerAuthenticator was going from the AUTHENTICATE to the FAILED state. The GSSAPI auth implementation was missing the second handshake defined in RFC 2222, which happens after the security context is established. This handshake is used by the client and server to negotiate the security layer (QoP) to be used for the connection. Kafka currently only support the "auth" QoP, so the implementation in this commit doesn't make it configurable, but this can be extended later. With this change I was able to successfully connect to a Kerberos-enabled Kafka broker using the SASL_PLAINTEXT protocol and the GSSAPI mechanism.
* Raise non-API exceptions (#1316)Jeff Widman2017-12-211-5/+1
| | | | | | | | | | | | | The original intent was to catch API exceptions (errors returned by the broker when trying to produce a message) and delegate them to the messages' futures. This is copied from the Java producer. However, we were accidentally catching all exceptions, thereby hiding exceptions from users unless they explicitly check the result of the future. Much better to raise client-side errors directly in the foreground so the user is immediately aware of them and can decide how to handle. Fix #1274
* KAFKA-3888 Use background thread to process consumer heartbeats (#1266)Dana Powers2017-12-2114-738/+977
|
* Pin pylint and lz4 to avoid breaking tests (#1330)Dana Powers2017-12-212-2/+4
|
* Minor Exception cleanupJeff Widman2017-12-127-10/+10
|
* Use non-deprecated exception handlingJeff Widman2017-12-121-1/+1
| | | Pulls in the fix upstream from https://github.com/mhils/backports.socketpair/pull/1
* Raise better struct pack/unpack errors (#1320)Jeff Widman2017-12-121-4/+8
|
* Followup cleanup to #1304Jeff Widman2017-12-081-2/+1
| | | https://github.com/dpkp/kafka-python/pull/1304/files#r155842880
* Refactor dns lookup in BrokerConnection (#1312)Dana Powers2017-12-082-61/+74
|
* optimize util.crc32 (#1304)Ofek Lev2017-12-081-8/+15
|
* use python standard max value (#1303)lukeWx2017-12-074-4/+4
|
* changed for to use enumerate() (#1301)Benn Roth2017-12-061-2/+1
|
* Change fixture default host to localhost (#1305)Andre Araujo2017-12-051-15/+1
|
* Fix two bugs in printing bytes instanceAndre Araujo2017-11-271-1/+1
| | | | | | | | | Bug 1: When `value` is None, trying to call `len(None)` throws an exception. Bug 2: When len(`value`) <= 100, the code currently prints b'' rather than `value`.
* Revert ffc7caef13a120f69788bcdd43ffa01468f575f9 / PR #1239Dana Powers2017-11-162-30/+2
| | | | The change caused a regression documented in issue #1290
* Use correct casing for MBJeff Widman2017-11-152-2/+2
| | | | | | These values refer to Megabytes, not Megabits. Fix #1295
* Replace periods (java) with underscores (python)Jeff Widman2017-11-071-2/+2
|
* Add license to wheelJeff Widman2017-11-031-0/+3
| | | https://wheel.readthedocs.io/en/stable/index.html#including-the-license-in-the-generated-wheel-file
* Use socket timeout of request_timeout_ms to prevent blocking forever on send ↵Dana Powers2017-10-251-6/+7
| | | | (#1281)
* Handle lookup_coordinator send failures (#1279)Dana Powers2017-10-242-5/+19
|
* Add DefaultRecordBatch implementation aka V2 message format parser/builder. ↵Taras Voinarovskyi2017-10-2518-31/+1696
| | | | | (#1185) Added bytecode optimization for varint and append/read_msg functions. Mostly based on avoiding LOAD_GLOBAL calls.
* Fix more gssapi errors (#1280)Dana Powers2017-10-241-1/+4
|
* Add fixture support for upcoming 1.0.0 broker release (#1275)Dana Powers2017-10-234-2/+194
|
* Fix timestamp not passed to RecordMetadata (#1273)Taras Voinarovskyi2017-10-226-24/+195
| | | | | | | | * Fix timestamp not being passed to RecordMetadata properly * Add more tests for LegacyBatch * Fix producer test for recordmetadata
* Fixup for PR 1264 -- required to propagate configuration to BrokerConnectionDana Powers2017-10-211-0/+3
|
* Fixup for PR 1258 / _try_authenticateDana Powers2017-10-211-1/+4
|
* Added controlled thread shutdown to example.py (#1268)Benn Roth2017-10-211-8/+32
|
* Merge pull request #1258 from dpkp/pending_completionsTaras Voinarovskyi2017-10-214-31/+65
|\ | | | | Move callback processing from BrokerConnection to KafkaClient
| * Move callback processing from BrokerConnection to KafkaClientpending_completionsDana Powers2017-10-154-31/+65
| |
* | Explicitly check for None rather than falseyJeff Widman2017-10-191-3/+3
| | | | | | | | Be pedantic about checking for identity rather than equality to avoid issues like #1237 / 411bc08f214b7afc36f11bde2047096c06467088
* | Fix overriding sasl_kerberos_service_name in KafkaConsumer / KafkaProducer ↵Nathanael Smith2017-10-172-0/+6
| | | | | | | | (#1264)
* | Cleanup gssapi code; fix bug report re AttributeError. (#1262)Dana Powers2017-10-161-10/+9
| |
* | Fix docstringJeff Widman2017-10-161-3/+4
|/
* Fix SASL authentication bugs (#1257)Dana Powers2017-10-151-23/+34
| | | | | | * Use _send_bytes_blocking in BrokerConnection * _try_authenticate should call recv() so that futures are resolved * _sasl_auth_future can be reset if recv() causes disconnect * validate sasl_mechanism against SaslHandShakeResponse enabled_mechanisms
* Merge pull request #1252 from dpkp/legacy_records_refactorTaras Voinarovskyi2017-10-1426-461/+1317
|\ | | | | Refactor MessageSet and Message into LegacyRecordBatch
| * Added limit of 100 bytes for Bytes debug printslegacy_records_refactorTaras Voinarovskiy2017-10-141-0/+4
| |
| * Added minor fixes for PR reviewTaras2017-10-126-27/+20
| |
| * Fix tests and rebase problemsTaras2017-10-122-97/+28
| |
| * Added specific to record/ folder micro benchmarks to get exact speed change ↵Taras2017-10-123-0/+155
| | | | | | | | after updating to V2 message format
| * Remove the check for timestamp None in producer, as it's done in RecordBatch ↵Taras2017-10-123-22/+7
| | | | | | | | | | | | anyway. Minor abc doc fixes.
| * Fix snappy compression on PyPyTaras2017-10-121-4/+5
| |
| * Refactor MessageSet and Message into LegacyRecordBatch to later support v2 ↵Taras2017-10-1121-355/+1142
| | | | | | | | message format
* | Merge pull request #1247 from dpkp/add_makefileTaras Voinarovskyi2017-10-142-0/+72
|\ \ | | | | | | Add a Makefile
| * | Add build_intergation job for makefileadd_makefileTaras Voinarovskiy2017-10-141-6/+11
| | |
| * | Forgot the requirement-dev.txt fileTaras2017-10-141-0/+15
| | |
| * | Add a Makefile, so users with less knowladge can easily build the project, ↵Taras2017-10-141-0/+52
| |/ | | | | | | generate docs and run tests.
* | KAFKA-4034: Avoid unnecessary consumer coordinator lookup (#1254)Dana Powers2017-10-115-26/+78
|/
* Merge pull request #1245 from dpkp/KAFKA_3977_defer_fetch_parsingTaras Voinarovskyi2017-10-112-277/+537
|\ | | | | KAFKA-3977: defer fetch response parsing and raise exceptions to user