Skip to content

Conversation

@jmckenzie-dev
Copy link
Owner

No description provided.

belliottsmith and others added 30 commits June 3, 2019 11:46
Patch by Benedict Elliot Smith; reviewed by Sam Tunnicliffe for CASSANDRA-15120
Unit test added in CASSANDRA-15126 is broken on 3.11
Patch by Ted Petersson; Reviewed by Michael Shuler for CASSANDRA-15050
Patch by Jeremy Hanna, Reviewed by Michael Shuler for CASSANDRA-15150
patch by Sam Tunnicliffe; reviewed by Michael Shuler
patch by Aleksey Yeschenko; reviewed by Benedict Elliott Smith for
CASSANDRA-15066
patch by Aleksey Yeschenko and Benedict Elliott Smith; reviewed by Alex
Petrov, Aleksey Yeschenko, and Benedict Elliott Smith for CASSANDRA-15066

Co-authored-by: Aleksey Yeschenko <aleksey@apache.org>
Co-authored-by: Benedict Elliott Smith <benedict@apache.org>
…re and message expirations

patch by Alex Petrov; reviewed by Aleksey Yeschenko and Benedict Elliott
Smith for CASSANDRA-15066
Patch by Jeremy Hanna, Reviewed by Michael Shuler for CASSANDRA-15150
Patch by Jeremy Hanna, Reviewed by Michael Shuler for CASSANDRA-15149
patch by Hari Nukala; reviewed by Aleksey Yeschenko for CASSANDRA-15165
Patch by Jeff Jirsa; Reviewed by Chris Lohfink for CASSANDRA-15167
Closes apache#327

Patch by Nathan Jackles; Reviewed by Jeff Jirsa for CASSANDRA-15173
Patch by Per Otterström, Sumanth Pasupuleti; Reviewed by Vinay Chella for CASSANDRA-15105

Co-Authored-By: Per Otterström <per.otterstrom@gmail.com>
Co-Authored-By: Sumanth Pasupuleti <sumanth.pasupuleti.is@gmail.com>
…sion's

patch by Aleksey Yeschenko; reviewed for CASSANDRA-15176 by Sam
Tunnicliffe an Blake Eggleston
Patch by Chris Lohfink; Reviewed by Jon Haddad for CASSANDRA-14670
Patch by Michael Shuler; Reviewed by Jon Haddad for CASSANDRA-15185
 patch by Mick Semb Wever; reviewed by Robert Stupp for CASSANDRA-14757
michaelsembwever and others added 29 commits August 22, 2019 13:34
patch by Dinesh Joshi; reviewed by Jon Meredith for CASSANDRA-15272
Patch by Josh Turner; Reviewed by Jeff Jirsa for CASSANDRA-15186
…and keyspace

Patch by Josh Turner; Reviewed by Jeff Jirsa for CASSANDRA-15287
After building artifacts, a diff is left over, so let's fix that:

diff --git a/conf/cassandra.yaml b/conf/cassandra.yaml
index 28d86fd..e776a5a 100644
--- a/conf/cassandra.yaml
+++ b/conf/cassandra.yaml
@@ -1341,4 +1341,4 @@ enable_sasi_indexes: false

 # Enables creation of transiently replicated keyspaces on this node.
 # Transient replication is experimental and is not recommended for production use.
-enable_transient_replication: false
\ No newline at end of file
+enable_transient_replication: false
…hat doesn't require keyspace knowledge/existence

 patch by Mick Semb Wever; reviewed by Branimir Lambov for CASSANDRA-15260
Closes apache#350

Patch by xuanling.gc, reviewed by Brandon Williams for CASSANDRA-15092
…more) correctly

patch by Aleksey Yeschenko; reviewed by Blake Eggleston for
CASSANDRA-15163
…'s version into the existing release process documentation

  patch by Mick Semb Wever; reviewed by Michael Shuler for CASSANDRA-15333
…otsearch on pom.xml dependency list

Closes apache#358

 patch by Marvin Froeder; reviewed by Mick Semb Wever for CASSANDRA-15321
Patch by Romain Hardouin; Reviewed by Dinesh Joshi for CASSANDRA-15296
…ations

Patch by James Berragan; reviewed by marcuse for CASSANDRA-15286
The original CDC implementation leaned on inheritance on multiple CLSM
with an abstract base. This refactor changes that approach to isntead
use composition with the segment allocation being the axis of
composability. Also tidies up some scope to package private while in
the area.
Adds some javadocs and tightens up package private scope on a couple
things.
Adds a CommitLogTestUtils concept; initial stubbing in of
ResumableCommitLogReader and ResumableCommitLogReaderTest.
Squash this before merge.
Mostly consists of method renames to clarify what we're flushing and on
what conditions in either naming, javadoc, or a combination of both. A
touch of cleaning up unused imports too, but no functional changes.
The original CDC CommitLogReadHandler for test only handled a boolean state
of whether or not a cdc mutation was seen or not. This opens that up to allow
for a tracking of a MutationIdentifier to determine if the specific mutation
we're seeing on replay is a duplicate and if so, how many times we've seen it.
This will play into resumable clreader testing in subsequent commits (some
risk of yagni here, but have a general framework of how I want to test this
in my head. We'll see.)
Had a handful of variables // when /** */ would give us IDE hinting.
Also added a class comment to header of CDC Allocator to give high level
view of why it exists.
Adds a new Single Source of Reading Truth for CommitLogSegments.
This resumable reader has a simpler API than our existing CommitLogReader
and is primarily intended to be externally user-facing.

Primary tradeoffs of this implementation are 1) Requiring init of a new
ResumableCommitLogReader object on each CLS read, 2) taking on ownership of
some complexity in SegmentReader (iterator resumability and RAR reinit)
in order to provide a very simple API to external CDC users, and 3) splitting
out the "Reading a CommitLog" API into both part in CommitLogReader and part
in ResumableCommitLogReader.

For now, for API backwards compatibility, the CommitLogReader APIs are
unchanged and ResumableCommitLogReader's API is purely simplified and
additive.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.