[
https://issues.apache.org/jira/browse/CASSANDRA-6477?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14640602#comment-14640602
]
Benedict commented on CASSANDRA-6477:
-------------------------------------
OK, so we have deadlock in that case. The apply of the batchlog happens on a
separate writeOrder group. The earlier group could be marked as "blocking" in
which case it is permitted to take any necessary action to terminate, but a
later (or even simply different keyspace) group is _not_ marked blocking and we
depend on it, then the system will lock up. In fact, we have this twice: once
for batchlog, once for the MV mutation (if local).
We could also do with some comments around things like:
{code}
if (stage != Stage.MATERIALIZED_VIEW_MUTATION)
{
for (WriteResponseHandlerWrapper wrapper : wrappers)
wrapper.handler.get();
}
{code}
Without any context, it's kind of hard to spot let alone understand the
important semantic difference and why it is there.
In general, I'd prefer more comments around this feature.
> Materialized Views (was: Global Indexes)
> ----------------------------------------
>
> Key: CASSANDRA-6477
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6477
> Project: Cassandra
> Issue Type: New Feature
> Components: API, Core
> Reporter: Jonathan Ellis
> Assignee: Carl Yeksigian
> Labels: cql
> Fix For: 3.0 beta 1
>
> Attachments: test-view-data.sh, users.yaml
>
>
> Local indexes are suitable for low-cardinality data, where spreading the
> index across the cluster is a Good Thing. However, for high-cardinality
> data, local indexes require querying most nodes in the cluster even if only a
> handful of rows is returned.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
