OHDSI Home | Forums | Wiki | Github

Import from AWS yields errors in vocabulary query

Part of our project is to get Atlas running on Kubernetes, and launchable per student (each student will have their own) we has taken to import the ohdsi-in-a-box databases from AWS as an initial approach and have beeen largely successful. with one exception.

Selecting the “Configuration” LHS button, yeilds an error in the webapi.

2022-11-07 21:50:53.979 ERROR http-nio-8080-exec-11 org.ohdsi.webapi.util.GenericExceptionMapper - [] - org.springframework.dao.IncorrectResultSizeDataAccessException: Incorrect result size: expected 1, actual 2
    at org.springframework.dao.support.DataAccessUtils.requiredSingleResult(DataAccessUtils.java:72)
    at org.springframework.jdbc.core.JdbcTemplate.queryForObject(JdbcTemplate.java:799)
    at org.ohdsi.webapi.service.VocabularyService.getInfo(VocabularyService.java:1194)

Which can be found here

this is could be misconfiguration on our part

Hello @Jeff_Waller,

Is it possible that Vocabulary table has two rows with vocabulary_id = 'None'? This query is supposed to always return a single row that contains OMOP Vocabulary package version.

Hi @rookie_crewkie ,
No, it is not, since the vocabulary_id is a primary key in the vocabulary table. I mean, if you have more than one record with the same vocabulary id in the vocabulary table it is not correct.

Can you check this? because @rookie_crewkie is correct that the query that threw the exception is simply:

from @CDM_schema.vocabulary
where VOCABULARY_ID = 'None'

Can you run this query on your database and see what rows you get? Maybe you don’t have the primary keys applied…

I did find that there were multiple entries in the vocabulary table as a result of an incorrect database import. Fixing the import fixed the problem.

1 Like