You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Keith Wansbrough (JIRA)" <ji...@apache.org> on 2016/08/23 10:54:20 UTC
[jira] [Commented] (CASSANDRA-12418) sstabledump JSON fails after
row tombstone
[ https://issues.apache.org/jira/browse/CASSANDRA-12418?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15432610#comment-15432610 ]
Keith Wansbrough commented on CASSANDRA-12418:
----------------------------------------------
It looks to me like JsonTransformer#serializePartition needs to call json.writeEndObject() after serializeDeletion() around line 191. Before this point there are two json.writeStartObject() calls, but after it in this branch there is only one json.writeEndObject() (on line 221), whereas if there is no partition-level deletion there are two (on lines 195 and 221).
> sstabledump JSON fails after row tombstone
> ------------------------------------------
>
> Key: CASSANDRA-12418
> URL: https://issues.apache.org/jira/browse/CASSANDRA-12418
> Project: Cassandra
> Issue Type: Bug
> Components: Tools
> Reporter: Keith Wansbrough
>
> sstabledump fails in JSON generation on an sstable containing a row deletion, using Cassandra 3.10-SNAPSHOT accf7a4724e244d6f1ba921cb11d2554dbb54a76 from 2016-07-26.
> There are two exceptions displayed:
> * Fatal error parsing partition: aye org.codehaus.jackson.JsonGenerationException: Can not start an object, expecting field name
> * org.codehaus.jackson.JsonGenerationException: Current context not an ARRAY but OBJECT
> Steps to reproduce:
> {code}
> cqlsh> create KEYSPACE foo WITH replication = {'class': 'SimpleStrategy', 'replication_factor': 1};
> cqlsh> create TABLE foo.bar (id text, str text, primary key (id));
> cqlsh> insert into foo.bar (id, str) values ('aye', 'alpha');
> cqlsh> insert into foo.bar (id, str) values ('bee', 'beta');
> cqlsh> delete from foo.bar where id = 'bee';
> cqlsh> insert into foo.bar (id, str) values ('bee', 'beth');
> cqlsh> select * from foo.bar;
> id | str
> -----+-------
> bee | beth
> aye | alpha
> (2 rows)
> cqlsh>
> {code}
> Now find the sstable:
> {code}
> $ cassandra/bin/nodetool flush
> $ cassandra/bin/sstableutil foo bar
> [..]
> Listing files...
> [..]
> /home/kw217/cassandra/data/data/foo/bar-407c56f05e1a11e6835def64bf5c656e/mb-1-big-Data.db
> [..]
> {code}
> Now check with sstabledump \-d. This works just fine.
> {code}
> $ cassandra/tools/bin/sstabledump -d /home/kw217/cassandra/data/data/foo/bar-407c56f05e1a11e6835def64bf5c656e/mb-1-big-Data.db
> [bee]@0 deletedAt=1470737827008101, localDeletion=1470737827
> [bee]@0 Row[info=[ts=1470737832405510] ]: | [str=beth ts=1470737832405510]
> [aye]@31 Row[info=[ts=1470737784401778] ]: | [str=alpha ts=1470737784401778]
> {code}
> Now run sstabledump. This should work as well, but it fails as follows:
> {code}
> $ cassandra/tools/bin/sstabledump /home/kw217/cassandra/data/data/foo/bar-407c56f05e1a11e6835def64bf5c656e/mb-1-big-Data.db
> ERROR 10:26:07 Fatal error parsing partition: aye
> org.codehaus.jackson.JsonGenerationException: Can not start an object, expecting field name
> at org.codehaus.jackson.impl.JsonGeneratorBase._reportError(JsonGeneratorBase.java:480) ~[jackson-core-asl-1.9.2.jar:1.9.2]
> at org.codehaus.jackson.impl.WriterBasedGenerator._verifyValueWrite(WriterBasedGenerator.java:836) ~[jackson-core-asl-1.9.2.jar:1.9.2]
> at org.codehaus.jackson.impl.WriterBasedGenerator.writeStartObject(WriterBasedGenerator.java:273) ~[jackson-core-asl-1.9.2.jar:1.9.2]
> at org.apache.cassandra.tools.JsonTransformer.serializePartition(JsonTransformer.java:181) ~[main/:na]
> at java.util.stream.ForEachOps$ForEachOp$OfRef.accept(ForEachOps.java:184) ~[na:1.8.0_77]
> at java.util.stream.ReferencePipeline$2$1.accept(ReferencePipeline.java:175) ~[na:1.8.0_77]
> at java.util.Iterator.forEachRemaining(Iterator.java:116) ~[na:1.8.0_77]
> at java.util.Spliterators$IteratorSpliterator.forEachRemaining(Spliterators.java:1801) ~[na:1.8.0_77]
> at java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:481) ~[na:1.8.0_77]
> at java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:471) ~[na:1.8.0_77]
> at java.util.stream.ForEachOps$ForEachOp.evaluateSequential(ForEachOps.java:151) ~[na:1.8.0_77]
> at java.util.stream.ForEachOps$ForEachOp$OfRef.evaluateSequential(ForEachOps.java:174) ~[na:1.8.0_77]
> at java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234) ~[na:1.8.0_77]
> at java.util.stream.ReferencePipeline.forEach(ReferencePipeline.java:418) ~[na:1.8.0_77]
> at org.apache.cassandra.tools.JsonTransformer.toJson(JsonTransformer.java:99) ~[main/:na]
> at org.apache.cassandra.tools.SSTableExport.main(SSTableExport.java:237) ~[main/:na]
> [
> {
> "partition" : {
> "key" : [ "bee" ],
> "position" : 0,
> "deletion_info" : { "marked_deleted" : "2016-08-09T10:17:07.008101Z", "local_delete_time" : "2016-08-09T10:17:07Z" }
> }
> }
> ]org.codehaus.jackson.JsonGenerationException: Current context not an ARRAY but OBJECT
> at org.codehaus.jackson.impl.JsonGeneratorBase._reportError(JsonGeneratorBase.java:480)
> at org.codehaus.jackson.impl.WriterBasedGenerator.writeEndArray(WriterBasedGenerator.java:257)
> at org.apache.cassandra.tools.JsonTransformer.toJson(JsonTransformer.java:100)
> at org.apache.cassandra.tools.SSTableExport.main(SSTableExport.java:237)
> {code}
> If possible, please can this be fixed in the 3.0.x stream as well as trunk?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)