I'm trying to run a SQL-H from Teradata 14.10 VM Image to a Apache Hadoop 2.2.0 single node cluster but it fails.
The query is being run from Teradata Studio. I read in the SQL Quick Reference manual (B035-1510-112B) that this feature is licenced separately. So wondering if it isn't enabled on the VM images that were download from Teradata Developer Exchange Downloads page.
Would anyone know how I can test this feature without enabling it in my production / test clusters?
Executed as Single statement. Failed [3707 : 42000] Syntax error, expected something like ';' between the word 'load_from_hcatalog' and '('.
Elapsed time = 00:00:00.008
STATEMENT 1: Select Statement failed.
webhcat is up and running
curl -i http://192.168.100.240:50111/templeton/v1/status
HTTP/1.1 200 OK
I'm having the same problem with both the 14.10 VM and the 15.0 VM. Is there a way to enable load_from_hcatalog in the VMs? Is there an additional download?
Same here. After a proper installation of the TDCH 1.3.1 connector on Teradata Studio 15.00.01 I see this exact error. HDP version used to create Hadoop connection profile in studio is HDP 2.1.
Any ideas on this ?
Load_From_HCatalog is a table operator that became available originally with Teradata Aster, to be used together with Teradata DB and Hadoop in Unified Data Architecture (UDA). Aster started off with SQL-H which upgraded to QueryGrid, but it is still in development phase in Teradata 15 as yet. Basically you have to purchase a package for QueryGrid that installs LOAD_FROM_HCATALOG and LOAD_TO_HCATALOG in your Syslib library. You can browse the SYSLIB library in the TD 15 Express VM but it does not have these functions installed by default. As and when you get the package, you can install SQL-H using the instructions here