Opened 5 years ago

Closed 5 years ago

#4303 closed defect (fixed)

Getting this error on many bots -ALTER EXTENSION failing

Reported by: robe Owned by: Algunenano
Priority: blocker Milestone: PostGIS 3.0.0
Component: postgis Version: master
Keywords: Cc:

Description

17:30:49                                                               ^
17:30:49 HINT:  Perhaps you meant to reference the column "proc.probin".
17:30:49 QUERY:  SELECT 'ALTER EXTENSION ' || e.extname || ' DROP ' || $3 || ' ' || COALESCE(proc.proname || '(' || oidvectortypes(proc.proargtypes) || ')' ,typ.typname, cd.relname, op.oprname,
17:30:49 					cs.typname || ' AS ' || ct.typname || ') ', opcname, opfname) || ';' AS remove_command
17:30:49 			FROM pg_depend As d INNER JOIN pg_extension As e
17:30:49 				ON d.refobjid = e.oid INNER JOIN pg_class As c ON
17:30:49 					c.oid = d.classid
17:30:49 					LEFT JOIN pg_proc AS proc ON proc.oid = d.objid
17:30:49 					LEFT JOIN pg_type AS typ ON typ.oid = d.objid
17:30:49 					LEFT JOIN pg_class As cd ON cd.oid = d.objid
17:30:49 					LEFT JOIN pg_operator As op ON op.oid = d.objid
17:30:49 					LEFT JOIN pg_cast AS ca ON ca.oid = d.objid
17:30:49 					LEFT JOIN pg_type AS cs ON ca.castsource = cs.oid
17:30:49 					LEFT JOIN pg_type AS ct ON ca.casttarget = ct.oid
17:30:49 					LEFT JOIN pg_opclass As oc ON oc.oid = d.objid
17:30:49 					LEFT JOIN pg_opfamily As ofa ON ofa.oid = d.objid
17:30:49 			WHERE d.deptype = 'e' and e.extname = $1 and c.relname = $2 AND (proc.prokind = 'a')  = $4;
17:30:49 CONTEXT:  PL/pgSQL function postgis_extension_remove_objects(text,text) line 48 at FOR over EXECUTE statement

Change History (8)

comment:1 by robe, 5 years ago

The two bots I spot checked winnie and bessie were running 10.5 when they failed on this.

comment:2 by Algunenano, 5 years ago

Owner: changed from pramsey to Algunenano

Totally my fault in #4296. current_setting('server_version_num') returns a 5 digit integer, not 3.

comment:3 by Raul Marin, 5 years ago

In 17204:

Revert r17178

It changes the behaviour of _postgis_pgsql_version

References #4303

comment:4 by Raul Marin, 5 years ago

In 17205:

Partially revert r17179

It changed the behaviour of _postgis_pgsql_version
Now uses the appropiate 5-6 digit version from server_version_num

References #4296
References #4303

comment:5 by Raul Marin, 5 years ago

In 17206:

Partially revert r17180

It changed the behaviour of _postgis_pgsql_version
Now uses the appropiate 5-6 digit version from server_version_num

References #4296
References #4303

comment:6 by Raul Marin, 5 years ago

In 17207:

Partially revert r17181

It changed the behaviour of _postgis_pgsql_version
Now uses the appropiate 5-6 digit version from server_version_num
References #4296
References #4303

comment:7 by Algunenano, 5 years ago

I've reverted the invalid changes and changed postgis_extension_remove_objects to use the 5-6 digits integer. I'll wait until the tests pass in the bots to close the issue.

comment:8 by Algunenano, 5 years ago

Resolution: fixed
Status: newclosed

Fixed now.

Note: See TracTickets for help on using tickets.