[ https://issues.jboss.org/browse/TEIIDDES-2481?page=com.atlassian.jira.plu... ] Barry LaFond commented on TEIIDDES-2481: ---------------------------------------- [~shawkins] I looked at that test class and followed the methods into TeiidProcessor but I don't see where any query resolving is happening (i.e. valid groups/symbols). The error we're getting is around the "internal" group name (Emp_CTE above). Matus verified that a Dynamic VDB deploys OK with this type of SQL, so I'm assuming the example we're using is valid? I'm having the same issue with one of your tests : TestWithClauseProcessing.testRecursive() {code} WITH t(n) AS ( VALUES (1) UNION ALL SELECT n+1 FROM t WHERE n < 64 ) SELECT sum(n) FROM t {code} ERROR : *Group does not exist: t* Since *t* is not an external table/view, shouldn't that internal "metadata" be taken into account during process? -- This message was sent by Atlassian JIRA (v6.4.11#64026)