ERROR: column core_user.google_auth does not exist Position: 204

2023-08-18 05:50:14,085 ERROR middleware.log :: GET /api/user/current 500 8.4 ms (1 DB calls)
{:via
[{:type clojure.lang.ExceptionInfo,
:message "ERROR: column core_user.google_auth does not exist\n Position: 204",
:data
{:toucan2/context-trace
[["execute SQL with class com.mchange.v2.c3p0.impl.NewProxyConnection"
{:toucan2.jdbc.query/sql-args
["SELECT "core_user"."id", "core_user"."email", "core_user"."date_joined", "core_user"."first_name", "core_user"."last_name", "core_user"."last_login", "core_user"."is_superuser", "core_user"."is_qbnewb", "core_user"."google_auth", "core_user"."ldap_auth", "core_user"."sso_source", "core_user"."is_active", "core_user"."updated_at", "core_user"."login_attributes", "core_user"."locale" FROM "core_user" WHERE "id" = ?"
9]}]
["resolve connection" {:toucan2.connection/connectable metabase.db.connection.ApplicationDB}]
["resolve connection" {:toucan2.connection/connectable :default}]
["resolve connection" {:toucan2.connection/connectable nil}]
{:toucan2.pipeline/rf
#object[clojure.core$map$fn__5931$fn__5932 0x2a27b3a "clojure.core$map$fn__5931$fn__5932@2a27b3a"]}
["with compiled query"
{:toucan2.pipeline/compiled-query
["SELECT "core_user"."id", "core_user"."email", "core_user"."date_joined", "core_user"."first_name", "core_user"."last_name", "core_user"."last_login", "core_user"."is_superuser", "core_user"."is_qbnewb", "core_user"."google_auth", "core_user"."ldap_auth", "core_user"."sso_source", "core_user"."is_active", "core_user"."updated_at", "core_user"."login_attributes", "core_user"."locale" FROM "core_user" WHERE "id" = ?"
9]}]
["with built query"
{:toucan2.pipeline/built-query
{:select

You upgraded and downgraded without restoring a backup

I'm having the same problem after the upgrade. We only have RDS instance backup. Would this help to restore the users that can't login?

You either do this or use the migrate command to take your app db to a lower version

I'm also running into this issue. We upgraded from 0.46.6.1 to 0.47.1 and had issues with many questions (similar to this issue Editor sorting and custom column editing may result in an error - #9 by Yuya )

We ran the migrate down command on the jar and reverted both metabase.jar and the metabase.db.mv.db but still get this error. Is this not the correct way of downgrading?

Thanks for any help

Please move away from H2 as soon as you can. Right now migrate to the latest version and see if it works

0.47.2 didn't resolve the issue. I was able to get back to 0.46.6.1 by restoring the whole VM to a previous backup. I'm going to migrate from H2 to postgreSQL following this Migrating to a production application database . I'll wait some time before upgrading to 0.47 to see if others are reporting similar issues or a fix for our issue gets implemented. Thanks Luiggi.

This issue hasn’t been reported in 47 but rather on instances that didn’t revert correctly to older versions