[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14989298#comment-14989298
]
Sylvain Lebresne commented on CASSANDRA-10608:
--
Alright, last dtest run looks good (we have too much flapping dtests right now
so comparison to the 3.0 branch are a bit difficult, but I did ran this once
more [here|http://cassci.datastax.com/job/pcmanus-10608-dtest/2/] and it does
look good). Committed, thanks.
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985404#comment-14985404
]
Sylvain Lebresne commented on CASSANDRA-10608:
--
That last unit tests looks good. The
{{CQLMetricsTest.testPreparedStatementsExecuted}} is definitively not related
to this (I've seen that test flapping on other branches). Let's wait on the new
dtest run to finish just to be sure and I'll commit (but please do ping me once
the dtests finish since I won't get an email :)).
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985343#comment-14985343
]
Mike Adamson commented on CASSANDRA-10608:
--
[~slebresne] Apologies, I thought I'd rebased before the last run but obviously
not. I've kicked off the test builds again and am seeing 6 failures on the 3.0
branch. 4 of these are failing on the current 3.0 test build and the other 2
are successful here but seem to be failing on cassci.
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985040#comment-14985040
]
Sylvain Lebresne commented on CASSANDRA-10608:
--
The patch lgtm, but the test results aren't all that good: the unit tests have
a few failures in {{ReadCommandTest}} that aren't in the cassandra-3.0 branch,
and the dtests have basically all failed. I doubt any of this is related to
this patch, but it would be nice to have slightly cleaner runs. Can you make
sure your 3.0 branch is up to date and re-submit the tests?
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14982338#comment-14982338
]
Sylvain Lebresne commented on CASSANDRA-10608:
--
Some tests are failing with a NPE because the 2nd branch doesn't account for
{{def}} being {{null}} (nit: I'd fact I'd change that 2nd branch to test for
{{def == null}}).
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14982765#comment-14982765
]
Mike Adamson commented on CASSANDRA-10608:
--
Good point, missed that. I've changed the branch logic in the inner branch to
your suggestion. Here is my branch with the fixes:
||3.0||trunk||
|[branch|https://github.com/mike-tr-adamson/cassandra/tree/10608-3.0]|[branch|https://github.com/mike-tr-adamson/cassandra/tree/10608-trunk]|
|[testall|http://cassci.datastax.com/view/Dev/view/madamson/job/mike-tr-adamson-10608-3.0-testall/]|[testall|http://cassci.datastax.com/view/Dev/view/madamson/job/mike-tr-adamson-10608-trunk-testall/]|
|[dtests|http://cassci.datastax.com/view/Dev/view/madamson/job/mike-tr-adamson-10608-3.0-dtest/]|[dtests|http://cassci.datastax.com/view/Dev/view/madamson/job/mike-tr-adamson-10608-trunk-dtest/]|
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14980486#comment-14980486
]
Sam Tunnicliffe commented on CASSANDRA-10608:
-
while [~mike_tr_adamson] gets set on with cassci I've pushed his branches to my
fork to get a CI run going:
||3.0||trunk||
|[branch|https://github.com/beobal/cassandra/tree/10608-3.0]|[branch|https://github.com/beobal/cassandra/tree/10608-trunk]|
|[testall|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-10608-3.0-testall/]|[testall|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-10608-trunk-testall/]|
|[dtests|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-10608-3.0-dtest/]|[dtests|http://cassci.datastax.com/view/Dev/view/beobal/job/beobal-10608-trunk-dtest/]|
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14980489#comment-14980489
]
Mike Adamson commented on CASSANDRA-10608:
--
[~slebresne] I have incorporated your suggestions in [~beobal]'s branch.
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14980053#comment-14980053
]
Sylvain Lebresne commented on CASSANDRA-10608:
--
I would strongly suspect that clustering columns are also affected so I think
the patch should use {{def.isPrimaryKeyColumn()}} rather than
{{def.isPartitionKey()}}.
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (CASSANDRA-10608) Adding a dynamic column to a compact storage table with the same name as the partition key causes a memtable flush deadlock
[
https://issues.apache.org/jira/browse/CASSANDRA-10608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14978995#comment-14978995
]
T Jake Luciani commented on CASSANDRA-10608:
The change itself looks fine. Do you have a dtest/utest run on cassci you can
link?
> Adding a dynamic column to a compact storage table with the same name as the
> partition key causes a memtable flush deadlock
> ---
>
> Key: CASSANDRA-10608
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10608
> Project: Cassandra
> Issue Type: Bug
>Reporter: Mike Adamson
>Assignee: Mike Adamson
>Priority: Minor
> Fix For: 3.0.0
>
> Attachments: 10608.txt
>
>
> The reproduction steps for this are as follows:
> # Create the following schema:
> {noformat}
> CREATE KEYSPACE ks WITH replication = { 'class': 'SimpleStrategy',
> 'replication_factor': '1'};
> CREATE TABLE ks.cf (k int, v int, PRIMARY KEY(k)) WITH COMPACT STORAGE;
> {noformat}
> # Using the thrift client execute the following:
> {noformat}
> Column column = new Column(ByteBufferUtil.bytes("k"));
> column.setValue(ByteBufferUtil.bytes(1));
> column.setTimestamp(1);
> client.insert(key, new ColumnParent(compactCf), column,
> ConsistencyLevel.ONE);
> {noformat}
> This causes an invalid {{PartitionUpdate}} to be added to {{Memtable}}
> containing a {{PartitionColumns}} containing the partition key
> {{ColumnDefinition}}.
> This happens because {{LegacyLayout.decodeCellName}} does not check whether
> the {{ColumnDefinition}} is a partition key
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
