You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@asterixdb.apache.org by "Ildar Absalyamov (JIRA)" <ji...@apache.org> on 2016/11/19 07:11:58 UTC

[jira] [Updated] (ASTERIXDB-1737) Parsing error when trying to feed in record with PK in meta

     [ https://issues.apache.org/jira/browse/ASTERIXDB-1737?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ildar Absalyamov updated ASTERIXDB-1737:
----------------------------------------
    Summary: Parsing error when trying to feed in record with PK in meta  (was: Parsing error when trying to create record with PK in meta)

> Parsing error when trying to feed in record with PK in meta
> -----------------------------------------------------------
>
>                 Key: ASTERIXDB-1737
>                 URL: https://issues.apache.org/jira/browse/ASTERIXDB-1737
>             Project: Apache AsterixDB
>          Issue Type: Bug
>            Reporter: Ildar Absalyamov
>
> I am trying to create the following dataset with meta PK and feed in data to it from the  localFS
> {noformat}
> drop dataverse test if exists;
> create dataverse test;
> use dataverse test;
> create type TestType as closed {
>     test: string
> }
> create type MetaType as {
> id:int
> }
> create dataset TestDS(TestType) with meta(MetaType) primary key meta().id;
> drop feed TestFeed if exists;
> create feed TestFeed using localfs
> (
>     ("parser"="record-with-metadata"),
>     ("reader"="localfs"),
>     ("path"="localhost:///ABSOLUTE_PATH/test.adm"),
>     //type of the record
>     ("type-name"="TestType"),
>     //type of the meta record
>     ("meta-type-name"="MetaType"),
>     //format of the record
>     ("record-format"="adm"),
>     // format of the meta record
>     ("format"="csv"),
>     ("delimiter"=","),
>     // index of the primary key the record (meta or orig)
>     ("key-indexes"="0"),
>     // indicates that PK originates from meta
>     ("key-indicators"="1"),
> 	//index of the record in the meta record
>     ("record-index"="0"),
>     // whether CSV header is present in the input
>     ("header"="false")
> );
> set wait-for-completion-feed "false";
> connect feed TestFeed to dataset TestDS;
> {noformat}
> Contents of the test.adm:
> {noformat}
> "{""test"":""test1""}",1
> "{""test"":""test2""}",2
> "{""test"":""test3""}",3
> {noformat}
> Ingestion silently fails without any visible to user error, but the data does not show up as loaded:
> {noformat}
> for $x in dataset TestDS return $x
> --empty result
> {noformat}
> However the feed ingestion problem in captured in failed_record.log in the storage folder:
> {noformat}
> "{""test"":""test2""}",2
> 11/18/2016 22:52:42 Parser failed to parse record
> "{""test"":""test3""}",3
> 11/18/2016 22:52:42 Parser failed to parse record
> {noformat}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)