[
https://issues.apache.org/jira/browse/CASSANDRA-13127?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16089940#comment-16089940
]
Fridtjof Sander commented on CASSANDRA-13127:
---------------------------------------------
I found another bug, that I believe falls into the same ballpark as this issue:
{code:java}
@Test
public void testFilterViewTombstones() throws Throwable {
execute("USE " + keyspace());
executeNet(protocolVersion, "USE " + keyspace());
createTable("CREATE TABLE %s (k int, c int, a int, b int, PRIMARY KEY
(k, c))");
createView("mv", "CREATE MATERIALIZED VIEW %s AS SELECT k,c FROM %%s
WHERE k IS NOT NULL AND c IS NOT NULL PRIMARY KEY (k,c)");
updateView("UPDATE %s USING TIMESTAMP 10 SET b=1 WHERE k=1 AND c=1");
assertRows(execute("SELECT * from %s"), row(1, 1, null, 1));
assertRows(execute("SELECT * from mv"), row(1, 1));
updateView("DELETE b FROM %s USING TIMESTAMP 11 WHERE k=1 AND c=1");
assertEmpty(execute("SELECT * from %s"));
assertEmpty(execute("SELECT * from mv"));
updateView("UPDATE %s USING TIMESTAMP 1 SET a=1 WHERE k=1 AND c=1");
assertRows(execute("SELECT * from %s"), row(1, 1, 1, null));
assertRows(execute("SELECT * from mv"), row(1, 1));
}
{code}
A view entry is created based on {{b=1@ts=10}} and deleted with {{ts=11}}.
Later another cell is added with {{a=1@ts=1}}. The view's liveness-info with
{{ts=1}} does not supersede the previously created tombstone with {{ts=11}}.
If you disagree on the similarity, I'll open a new ticket for this. Both issues
are about handling the contribution of base-cells to the view-row's liveness
that are projected out of the view.
On trunk, this doesn't even create an mv-entry (because of the empty
liveness-info), but also fails with [~jasonstack]'s patch. I'm not sure, if
this can be fixed at all without keeping hidden timestamp/liveness-data of
discarded columns.
> Materialized Views: View row expires too soon
> ---------------------------------------------
>
> Key: CASSANDRA-13127
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13127
> Project: Cassandra
> Issue Type: Bug
> Components: Local Write-Read Paths, Materialized Views
> Reporter: Duarte Nunes
> Assignee: ZhaoYang
>
> Consider the following commands, ran against trunk:
> {code}
> echo "DROP MATERIALIZED VIEW ks.mv; DROP TABLE ks.base;" | bin/cqlsh
> echo "CREATE TABLE ks.base (p int, c int, v int, PRIMARY KEY (p, c));" |
> bin/cqlsh
> echo "CREATE MATERIALIZED VIEW ks.mv AS SELECT p, c FROM base WHERE p IS NOT
> NULL AND c IS NOT NULL PRIMARY KEY (c, p);" | bin/cqlsh
> echo "INSERT INTO ks.base (p, c) VALUES (0, 0) USING TTL 10;" | bin/cqlsh
> # wait for row liveness to get closer to expiration
> sleep 6;
> echo "UPDATE ks.base USING TTL 8 SET v = 0 WHERE p = 0 and c = 0;" | bin/cqlsh
> echo "SELECT p, c, ttl(v) FROM ks.base; SELECT * FROM ks.mv;" | bin/cqlsh
> p | c | ttl(v)
> ---+---+--------
> 0 | 0 | 7
> (1 rows)
> c | p
> ---+---
> 0 | 0
> (1 rows)
> # wait for row liveness to expire
> sleep 4;
> echo "SELECT p, c, ttl(v) FROM ks.base; SELECT * FROM ks.mv;" | bin/cqlsh
> p | c | ttl(v)
> ---+---+--------
> 0 | 0 | 3
> (1 rows)
> c | p
> ---+---
> (0 rows)
> {code}
> Notice how the view row is removed even though the base row is still live. I
> would say this is because in ViewUpdateGenerator#computeLivenessInfoForEntry
> the TTLs are compared instead of the expiration times, but I'm not sure I'm
> getting that far ahead in the code when updating a column that's not in the
> view.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
