Well, this turns out to only work for PG 9.3 testing (which is what I was using when I tested it originally). Maybe 9.2 but I didn't test that either.
For PG 9.1, the function fails to create leading to errors like:
psql:300.schema.staged_search.sql:449: ERROR: column "hits" does not exist
LINE 12: WHERE mmrsm.source IN (SELECT * FROM unnest(hits)) ^
psql:300.schema.staged_search.sql:468: ERROR: current transaction is aborted, commands ignored until end of transaction block
ROLLBACK
This causes the entire search schema to fail during database create. Which is the source of the failure for the live tests lately since this merge occurred.
So, question then, do we try to find a solution that works with PG 9.1, or is it time we declare PG 9.3 the minimum version for Evergreen 2.10?
According to http://www.postgresql.org/support/versioning/, PG 9.1 is end of life in September 2016 anyways, so this might be the last version of Evergreen to work with it regardless.
Well, this turns out to only work for PG 9.3 testing (which is what I was using when I tested it originally). Maybe 9.2 but I didn't test that either.
For PG 9.1, the function fails to create leading to errors like:
psql:300. schema. staged_ search. sql:449: ERROR: column "hits" does not exist
^ schema. staged_ search. sql:468: ERROR: current transaction is aborted, commands ignored until end of transaction block
LINE 12: WHERE mmrsm.source IN (SELECT * FROM unnest(hits))
psql:300.
ROLLBACK
This causes the entire search schema to fail during database create. Which is the source of the failure for the live tests lately since this merge occurred.
So, question then, do we try to find a solution that works with PG 9.1, or is it time we declare PG 9.3 the minimum version for Evergreen 2.10?
According to http:// www.postgresql. org/support/ versioning/, PG 9.1 is end of life in September 2016 anyways, so this might be the last version of Evergreen to work with it regardless.