[
https://issues.apache.org/jira/browse/CASSANDRA-15625?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sam Tunnicliffe updated CASSANDRA-15625:
----------------------------------------
Status: Awaiting Feedback (was: Triage Needed)
[~james1] The stacktrace suggests that a read or write touched a 1 mb key in
the app300 table during the sampling period, which shouldn't be possible with
the schema you defined. Is this reproducible in a completely clean db?
I'm not sure if it's coincidental that the final argument you've passed to the
nodetool command matches the partition key from the insert statement, but that
argument is the duration in millis for which to sample reads/writes (i.e. 500s
in this case).
> Nodetool toppartitions error
> ----------------------------
>
> Key: CASSANDRA-15625
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15625
> Project: Cassandra
> Issue Type: Bug
> Reporter: Antonio
> Assignee: Alex Lumpov
> Priority: Normal
>
> c* version :3.0.15
> here's my test table：
> CREATE TABLE app300.test (
> a bigint PRIMARY KEY,
> b text,
> c text
> )
> INSERT INTO app300.test(a ,b, c ) VALUES (500000, 'test1', 'test1');
> when i use topartition ：nodetool toppartitions app300 test 500000，get error
> error: Expected 8 or 0 byte long (1048576)
> -- StackTrace --
> org.apache.cassandra.serializers.MarshalException: Expected 8 or 0 byte long
> (1048576)
> at
> org.apache.cassandra.serializers.LongSerializer.validate(LongSerializer.java:42)
> at
> org.apache.cassandra.db.marshal.AbstractType.getString(AbstractType.java:128)
> at
> org.apache.cassandra.db.ColumnFamilyStore.finishLocalSampling(ColumnFamilyStore.java:1579)
> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> but when i flush this table， topartition can work
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
