[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16004483#comment-16004483
]
Andrés de la Peña commented on CASSANDRA-13412:
---
Committed as
[b0db519b79701cecac92a7a2c93101cf17fb928d|https://github.com/apache/cassandra/commit/b0db519b79701cecac92a7a2c93101cf17fb928d],
thanks for reviewing.
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16002688#comment-16002688
]
Alex Petrov commented on CASSANDRA-13412:
-
Sorry for the delay! Thank you for the patch,
+1 from my side, LGTM.
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15984859#comment-15984859
]
Enrique Bautista Barahona commented on CASSANDRA-13412:
---
I've tried to identify a precise sequence of steps to reproduce this in 3.0.x
but have had no luck so far. You can close this issue if you want. If I manage
to reproduce it in 3.0.x consistently I can open another issue (or re-open this
one, whatever you prefer).
Thanks!
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15976935#comment-15976935
]
Andrés de la Peña commented on CASSANDRA-13412:
---
Totally agree. I have added a new test for an index on a regular column. Here
is the new patch, where the unaffected 3.0+ branches have only the unit tests:
||[2.1|https://github.com/apache/cassandra/compare/cassandra-2.1...adelapena:13412-2.1]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.1-testall/]|[dtests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.1-dtest/]|
||[2.2|https://github.com/apache/cassandra/compare/cassandra-2.2...adelapena:13412-2.2]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.2-testall/]|[dtests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.2-dtest/]|
||[3.0|https://github.com/apache/cassandra/compare/cassandra-3.0...adelapena:13412-3.0]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-3.0-testall/]|
|
||[3.11|https://github.com/apache/cassandra/compare/cassandra-3.11...adelapena:13412-3.11]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-3.11-testall/]|
|
||[3.0|https://github.com/apache/cassandra/compare/cassandra-trunk...adelapena:13412-trunk]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-trunk-testall/]|
|
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15976293#comment-15976293
]
Alex Petrov commented on CASSANDRA-13412:
-
+1, the patch looks good.
Minor remark: we might want to add a test for a regular column expiry, too
(e.g. that the new value isn't queryable) and possibly add same tests to 3.0+,
as the behaviour is important (although it does work on the later branches).
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15970976#comment-15970976
]
Andrés de la Peña commented on CASSANDRA-13412:
---
You are very welcome. At first, I've reproduced the problem in a three nodes
ccm cluster. Then, since the cause of the problem seems to be local, I've
reproduced it in a single instance and I've added the two unit tests that can
be found in the suggested patch.
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15970877#comment-15970877
]
Enrique Bautista Barahona commented on CASSANDRA-13412:
---
[~adelapena], thanks a lot for taking care of this. I'll try to reproduce it
again on 3.0.11 and write back with more information.
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
> Fix For: 2.1.x, 2.2.x
>
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15967286#comment-15967286
]
Andrés de la Peña commented on CASSANDRA-13412:
---
The problem is also affecting to indexes on partition key components:
{code}
CREATE TABLE k.t (
pk1 int,
pk2 int,
a int,
b int,
PRIMARY KEY ((pk1, pk2))
);
CREATE INDEX ON k.t(pk1);
INSERT INTO k.t (pk1, pk2, a, b) VALUES (1, 2, 3, 4);
UPDATE k.t USING TTL 10 SET b = 10 WHERE pk1 = 1 AND pk2 = 2;
-- Wait 10 seconds
SELECT * FROM k.t WHERE pk1 = 1 AND pk2 = 2; -- 1 row
SELECT * FROM k.t WHERE pk1 = 1; -- 0 rows
{code}
Index entries inherit the TTL of the indexed cell, and this is right with
regular columns. However, indexes of primary key columns index every column,
meaning that their index entries will always get the TTL of the last updated
row column.
The proposed solution is to not generate expiring cells when the indexed column
is part of the primary key. Here is the patch for 2.1 and 2.2:
||[2.1|https://github.com/apache/cassandra/compare/cassandra-2.1...adelapena:13412-2.1]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.1-testall/]|[dtests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.1-dtest/]|
||[2.2|https://github.com/apache/cassandra/compare/cassandra-2.2...adelapena:13412-2.2]|[utests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.2-testall/]|[dtests|http://cassci.datastax.com/view/Dev/view/adelapena/job/adelapena-13412-2.2-dtest/]|
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
[jira] [Commented] (CASSANDRA-13412) Update of column with TTL results in secondary index not returning row
[
https://issues.apache.org/jira/browse/CASSANDRA-13412?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15965740#comment-15965740
]
Andrés de la Peña commented on CASSANDRA-13412:
---
[~ebautistabar], I can reproduce the problem in 2.1 and 2.2 this way:
{code}
CREATE KEYSPACE k WITH replication = {'class': 'SimpleStrategy',
'replication_factor': 3};
CREATE TABLE k.t (
a text,
b text,
c text,
d set,
e float,
f text,
g int,
h double,
j set,
k float,
m set,
PRIMARY KEY (a, b, c)
);
CREATE INDEX ON k.t(c);
INSERT INTO k.t (a, b, c, h) VALUES ('test_a', 'test_b', 'test_c', 1);
UPDATE k.t USING TTL 10 SET h = 10 WHERE a = 'test_a' AND b = 'test_b' AND c =
'test_c';
-- Wait 10 seconds
SELECT * FROM k.t WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c'; -- 1 row
SELECT * FROM k.t WHERE c = 'test_c'; -- 0 rows
{code}
Or, in a simpler way:
{code}
CREATE KEYSPACE k WITH replication = {'class': 'SimpleStrategy',
'replication_factor': 1};
CREATE TABLE k.t (
pk int,
ck int,
a int,
b int,
PRIMARY KEY (pk, ck)
);
CREATE INDEX ON k.t(ck);
INSERT INTO k.t (pk, ck, a, b) VALUES (1, 2, 3, 4);
UPDATE k.t USING TTL 10 SET b = 10 WHERE pk = 1 AND ck = 2;
-- Wait 10 seconds
SELECT * FROM k.t WHERE pk = 1 AND ck = 2; -- 1 row
SELECT * FROM k.t WHERE ck = 2; -- 0 rows
{code}
However, I can't reproduce the problem in 3.0.11. Could you please provide a
sequence of insertions producing the failure in 3.0.11?
> Update of column with TTL results in secondary index not returning row
> --
>
> Key: CASSANDRA-13412
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13412
> Project: Cassandra
> Issue Type: Bug
>Reporter: Enrique Bautista Barahona
>Assignee: Andrés de la Peña
>
> Cassandra versions: 2.2.3, 3.0.11
> 1 datacenter, keyspace has RF 3. Default consistency level.
> Steps:
> 1. I create these table and index.
> {code}
> CREATE TABLE my_table (
> a text,
> b text,
> c text,
> d set,
> e float,
> f text,
> g int,
> h double,
> j set,
> k float,
> m set,
> PRIMARY KEY (a, b, c)
> ) WITH read_repair_chance = 0.0
>AND dclocal_read_repair_chance = 0.1
>AND gc_grace_seconds = 864000
>AND bloom_filter_fp_chance = 0.01
>AND caching = { 'keys' : 'ALL', 'rows_per_partition' : 'NONE' }
>AND comment = ''
>AND compaction = { 'class' :
> 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy' }
>AND compression = { 'sstable_compression' :
> 'org.apache.cassandra.io.compress.LZ4Compressor' }
>AND default_time_to_live = 0
>AND speculative_retry = '99.0PERCENTILE'
>AND min_index_interval = 128
>AND max_index_interval = 2048;
> CREATE INDEX my_index ON my_table (c);
> {code}
> 2. I have 9951 INSERT statements in a file and I run the following command to
> execute them. The INSERT statements have no TTL and no consistency level is
> specified.
> {code}
> cqlsh -u -f
> {code}
> 3. I update a column filtering by the whole primary key, and setting a TTL.
> For example:
> {code}
> UPDATE my_table USING TTL 30 SET h = 10 WHERE a = 'test_a' AND b = 'test_b'
> AND c = 'test_c';
> {code}
> 4. After the time specified in the TTL I run the following queries:
> {code}
> SELECT * FROM my_table WHERE a = 'test_a' AND b = 'test_b' AND c = 'test_c';
> SELECT * FROM my_table WHERE c = 'test_c';
> {code}
> The first one returns the correct row with an empty h column (as it has
> expired). However, the second query (which uses the secondary index on column
> c) returns nothing.
> I've done the query through my app which uses the Java driver v3.0.4 and
> reads with CL local_one, from the cql shell and from DBeaver 3.8.5. All
> display the same behaviour. The queries are performed minutes after the
> writes and the servers don't have a high load, so I think it's unlikely to be
> a consistency issue.
> I've tried to reproduce the issue in ccm and cqlsh by creating a new keyspace
> and table, and inserting just 1 row, and the bug doesn't manifest. This leads
> me to think that it's an issue only present with not trivially small amounts
> of data, or maybe present only after Cassandra compacts or performs whatever
> maintenance it needs to do.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
