cassandra-pr mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From aweisberg <...@git.apache.org>
Subject [GitHub] cassandra pull request #262: Replicalayout followup
Date Mon, 10 Sep 2018 16:23:34 GMT
Github user aweisberg commented on a diff in the pull request:

    https://github.com/apache/cassandra/pull/262#discussion_r216386124
  
    --- Diff: src/java/org/apache/cassandra/batchlog/BatchlogManager.java ---
    @@ -459,36 +458,31 @@ private void writeHintsForUndeliveredEndpoints(int startFrom, Set<InetAddressAnd
                 Keyspace keyspace = Keyspace.open(ks);
                 Token tk = mutation.key().getToken();
     
    +
                 ReplicaLayout.ForTokenWrite liveAndDown = ReplicaLayout.forTokenWriteLiveAndDown(keyspace,
tk);
                 Replicas.temporaryAssertFull(liveAndDown.all()); // TODO in CASSANDRA-14549
     
    -            EndpointsForToken.Builder liveReplicasBuilder = EndpointsForToken.builder(tk);
    +            Replica selfReplica = liveAndDown.all().selfIfPresent();
    +            if (selfReplica != null)
    --- End diff --
    
    Looked at mutateAtomically and sendToHintedEndpoints, I guess I can see that we do have
the batch log attempt to write to the coordinator. I guess that sort of makes sense in the
batch log doesn't really make sense sort of way. It's possible the coordinator didn't hand
the write off to itself.


---

---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscribe@cassandra.apache.org
For additional commands, e-mail: pr-help@cassandra.apache.org


Mime
View raw message