| 17 Collaborator |
|
Marcus Eriksson , Sam Tunnicliffe , Alex Petrov , Mick Semb Wever , Ekaterina Dimitrova , Benedict Elliott Smith , Berenguer Blasi , Blake Eggleston , Ariel Weisberg , Jason Brown , Dinesh Joshi , Stefan Podkowinski , Jon Haddad , Vinay Chella , Venkata Harikrishna Nukala , Alexander Dejanovski , Gregory Ramsperger |
| 13 Patch |
7 Review |
|
674b6cc1a5e905a9c234c649adaad2de79cfa560,
9c1bbf3ac913f9bdf7a0e0922106804af42d2c1e,
b30c8c98a594a5682f6ea1f0b5511463b700b6e8,
27b35799a46dd5b649c4a172f4f8316b48615304,
8ad4a9a522aef7243b2612d084bbae69ea8f1be6,
d5005627b02b4e716947fa05a40473368017c0f9,
1f19d5f7a243cc4227da923459f5eb2f66066778,
801cb70ee811c956e987718a00695638d5bec1b6,
2adfa92044381aa9093104f3a105f3dbd7dda94c,
bfbc5274f2b3a5af2cbbe9679f0e78f1066ef638,
a0f5523910dcc96f204220e3c5ba24f7be15f37b,
da7fcefb16d16af8924cda35c0a6a63ad553694f,
4df308dfe7bbc5fcf5b649fc1e4a3d6bf6daa7ee |
dd974b4b5f7770b293fdd8e4e76d7043508abec4,
a7d8ba7b10a441f9710724e65a939a46add0ae78,
1911a887e8316d343c9bfe3aca3f9d143e9f4a61,
6fc8920889e8537a1f56f45e6c966b3d18325fbb,
a7d8ba7b10a441f9710724e65a939a46add0ae78,
fd44a69fce135c6ddf7ebcba25d032a36a1d9064,
a86780687b3ebe7d9883d81c44eb5f8cd228b2cb |
1911a887e8316d343c9bfe3aca3f9d143e9f4a61 | Author: Venkata Harikrishna Nukala <n.v.harikrishna.apache@gmail.com>
| 2021-10-23 00:03:45+05:30
Fix slow keycache load which blocks startup for tables with many sstables.
Patch by Venkata Harikrishna Nukala; reviewed by Marcus Eriksson and Joseph Lynch for CASSANDRA-14898
6fc8920889e8537a1f56f45e6c966b3d18325fbb | Author: Alexander Dejanovski <alex@thelastpickle.com>
| 2020-06-26 11:08:50+02:00
Fix upgrade failures when using the EC2Snitch in legacy mode
Legacy naming conventions for AWS regions were allowing datacenter names such as us-west and us-west-2, which in the new standard mode become us-west-1 and us-west-2.
As us-west-2 can match both the standard and legacy mode, checking the DC name cannot be reliably used to detect mixed modes being used as it would fail in case of an upgrade from pre-4.0 clusters in a region such as us-west-2.
The rack check should be enough as rack names in the standard mode now includes the region name, which can be easily identified.
patch by Alexander Dejanovski; reviewed by Mick Semb Wever, Joey Lynch for CASSANDRA-15878
da7fcefb16d16af8924cda35c0a6a63ad553694f | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2020-05-04 10:59:35-07:00
Fix compression_test.TestCompression regression
CASSANDRA-15379 changed behavior of how we flush for deflate, namely we
flush LZ4 and then compact to deflate off the hot path. Updated this
test to take this into account.
Patch by Joseph Lynch; Reviewed by Berenguer Blasi for CASSANDRA-15782
9c1bbf3ac913f9bdf7a0e0922106804af42d2c1e | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2019-11-03 16:37:18-08:00
Flush with fast compressors by default
Previously Zstd, Deflate and LZ4HC could hold up the flush thread while
flushing due to their slow compression rates. With this patch we always
default to flush with a fast compressor (default to LZ4) and then
compress into the high compression strategy during normal compactions
that follow. Note that if the existing table compressor is fast (e.g.
LZ4 or Snappy) we use that.
Patch by Joey Lynch; reviewed by Dinesh Joshi for CASSANDRA-15379
674b6cc1a5e905a9c234c649adaad2de79cfa560 | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2019-08-16 18:15:55-07:00
Update defaults for server and client TLS settings
This fixes backwards compatibility with the 3.11 server_encryption_options
while exposing the correct defaults after the 4.0 Netty refactor.
Patch by Joseph Lynch; Reviewed by Ekaterina Dimitrova for CASSANDRA-15262
a7d8ba7b10a441f9710724e65a939a46add0ae78 | Author: Alex Petrov <oleksandr.petrov@gmail.com>
| 2019-02-11 15:37:07+01:00
Fix memory-retention problem arising from thread locals from schema change in in-jvm tests
patch by Alex Petrov; reviewed by Joseph Lynch for CASSANDRA-15014
a7d8ba7b10a441f9710724e65a939a46add0ae78 | Author: Alex Petrov <oleksandr.petrov@gmail.com>
| 2019-02-11 15:37:07+01:00
Fix memory-retention problem arising from thread locals from schema change in in-jvm tests
patch by Alex Petrov; reviewed by Joseph Lynch for CASSANDRA-15014
4df308dfe7bbc5fcf5b649fc1e4a3d6bf6daa7ee | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2019-02-05 12:11:53-08:00
Fix the token_generator_test for Cassandra 2.2
Ref: Make tools/bin/token-generator py2/3 compatible
Patch by Joseph Lynch; reviewed by Stefan Podkowinski for CASSANDRA-15012
8ad4a9a522aef7243b2612d084bbae69ea8f1be6 | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2018-12-07 18:29:08-08:00
In JVM dtests need to clean up after instance shutdown
Adds additional cleanup logic to ensure we don't leak classloaders and
their associated objects when running the in JVM dtests.
Patch by Joseph Lynch; reviewed by Alex Petrov for CASSANDRA-14922
d5005627b02b4e716947fa05a40473368017c0f9 | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2018-12-07 18:29:08-08:00
In JVM dtests need to clean up after instance shutdown
Adds additional cleanup logic to ensure we don't leak classloaders and
their associated objects when running the in JVM dtests.
Patch by Joseph Lynch; reviewed by Alex Petrov for CASSANDRA-14922
2adfa92044381aa9093104f3a105f3dbd7dda94c | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2018-10-31 20:51:34-07:00
Fix incorrect sorting of replicas in SimpleStrategy.calculateNaturalReplicas
patch by Joseph Lynch; reviewed by Benedict Elliott Smith for
CASSANDRA-14862
801cb70ee811c956e987718a00695638d5bec1b6 | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2018-08-23 15:19:20-07:00
Startup checker should wait for count rather than percentage
This improves on the wait for healthy work from CASSANDRA-13993 to
solve CASSANDRA-14297. In particular now the connectivity checker waits
for all but a single node in either the local datacenter or every
datacenter (defaults to just local, but the user can configure it to
wait for every datacenter). This way users can use this feature to ensure
availability of their application during restarts of Cassandra. The default
behavior waits for all but a single local datacenter node.
Patch by Joseph Lynch; Reviewed by Ariel Weisberg for CASSANDRA-14297
1f19d5f7a243cc4227da923459f5eb2f66066778 | Author: Joseph Lynch <joe.e.lynch@gmail.com>
| 2018-03-12 20:37:30-07:00
Auto-expand replication_factor for NetworkTopologyStrategy
This re-defines the 'replication_factor' configuration option for
NetworkTopologyStrategy to auto-expand out to the datacenters that the
cluster knows about. This allows users to not worry about their cluster
layout when creating keyspaces. On ALTER this keyword will only ever add
datacenters (it keeps all previous datacenters unless explicitly overridden)
Patch by Joseph Lynch; reviewed by Jon Haddad for CASSANDRA-14303
fd44a69fce135c6ddf7ebcba25d032a36a1d9064 | Author: Gregory Ramsperger <ramsperger@users.noreply.github.com>
| 2016-04-16 00:42:03-05:00
Use standard Amazon naming for datacenter and rack in Ec2Snitch.
patch by Gregory Ramsperger and jasobrown; reviewed by Joey Lynch for CASSANDRA-7839