Any ideas as to where I am going wrong here.. .Between , observed that by
changing hive txn manager , some other things are breaking in my existing
setup so I have reverted to the old configuration for now to keep the
cluster in a workable mode.
regards
Devopam
On Wed, Jan 14, 2015 at 4:25 PM, Devopam Mittra <devo...@gmail.com> wrote:
> Env: Hive 0.14 , HDP2.2, Centos 6.6
>
> I created a simple table (ref syntax below)
> DROP TABLE IF EXISTS twitter_user_info;
>
> CREATE TABLE twitter_user_info
> ( userid BIGINT COMMENT 'User Id',
> username STRING COMMENT 'User Name',
> is_latest BOOLEAN COMMENT 'Currently active',
> last_modified_ts TIMESTAMP COMMENT 'Last modified timestamp'
> )
> COMMENT 'Twitter User Info Table'
> CLUSTERED BY (userid,is_latest) SORTED BY (userid) INTO 30 BUCKETS
> STORED AS ORC
> TBLPROPERTIES('transactional'='true');
>
>
> set the parameters in hive conf via ambari:
>
> set hive.support.concurrency=true;
> set hive.enforce.bucketing=true;
> set hive.exec.dynamic.partition.mode=nonstrict;
> set hive.txn.manager=org.apache.hadoop.hive.ql.lockmgr.DbTxnManager;
> set hive.compactor.initiator.on=true;
> set hive.compactor.worker.threads=100;
>
> I inserted few rows via the following query (not-relevant , yet pasting):
> INSERT INTO TABLE twitter_user_info
> SELECT created_by_id, created_by_name,'Y',MAX(created_at)
> FROM twitter_data B
> WHERE b.created_by_id IS NOT NULL
> AND b.created_by_id NOT IN (
> SELECT created_by_id FROM
> (SELECT created_by_id,COUNT(DISTINCT
> created_by_name)
> FROM twitter_data
> WHERE created_by_id IS NOT NULL
> GROUP BY created_by_id
> HAVING COUNT(DISTINCT created_by_name) >1) A
> )
> GROUP BY b.created_by_id,b.created_by_name;
>
> For each operation below I am hitting the same error, and am clueless at
> this point .
>
> hive> UPDATE twitter_user_info
> > SET is_latest = false
> > WHERE userid = 298779403
> > AND last_modified_ts = '2015-01-13 14:25:14';
> *FAILED: NoMatchingMethodException No matching method for class
> org.apache.hadoop.hive.ql.udf.UDFToLong with
> (struct<transactionid:bigint,bucketid:int,rowid:bigint>). Possible choices:
> _FUNC_(bigint) _FUNC_(boolean) _FUNC_(decimal(38,18)) _FUNC_(double)
> _FUNC_(float) _FUNC_(int) _FUNC_(smallint) _FUNC_(string)
> _FUNC_(timestamp) _FUNC_(tinyint) _FUNC_(void) *
> hive> DELETE FROM twitter_user_info WHERE userid=100;
>
> *FAILED: NoMatchingMethodException No matching method for class
> org.apache.hadoop.hive.ql.udf.UDFToLong with
> (struct<transactionid:bigint,bucketid:int,rowid:bigint>). Possible choices:
> _FUNC_(bigint) _FUNC_(boolean) _FUNC_(decimal(38,18)) _FUNC_(double)
> _FUNC_(float) _FUNC_(int) _FUNC_(smallint) _FUNC_(string)
> _FUNC_(timestamp) _FUNC_(tinyint) _FUNC_(void) *
>
> Any pointers are welcome
>
> --
> Devopam Mittra
> Life and Relations are not binary
>
--
Devopam Mittra
Life and Relations are not binary
