wet muralBOT
The error Cannot return null for non-nullable field Object.applicationId indicates that the server is trying to return an applicationId field that is null, but the GraphQL schema requires it to be non-nullable.
This looks like a data integrity issue - likely caused by one of these scenarios:
applicationId field wasn't properly populatedTry these steps:
docker exec -it twenty-server-1 yarn command:prod cache:flush
docker compose down
docker compose up -d
If that doesn't resolve it, you may need to reset the database (only if this is a fresh installation with no important data):
docker exec -it twenty-server-1 yarn
docker exec -it twenty-server-1 npx nx database:reset --configuration=no-seed
docker compose down
docker compose up -d
What version of Twenty are you running? This could also be related to a version mismatch or an incomplete upgrade. You can check your version by running echo $APP_VERSION in the twenty-server container.
Let me know what happens after trying the cache flush.
analog imp
Flash works - output:
[Nest] 1565 - 01/24/2026, 11:26:10 AM LOG [FlushCacheCommand] Cache flushed
i restarted the services (docker compose):
here is the log of server-1:
Nest] 36 - 01/24/2026, 11:27:48 AM LOG [UpgradeCommand] Initialized upgrade context with:
currentVersion (migrating to): 1.16.0
fromWorkspaceVersion: 1.15.0
23 commands
...
[Nest] 36 - 01/24/2026, 11:27:52 AM LOG [UpdateTaskOnDeleteActionCommand] Updating task relation onDelete from SET_NULL to CASCADE in workspace 1e359dea-8c95-4029-823b-2992bd7bb774
[Nest] 36 - 01/24/2026, 11:27:52 AM WARN [UpgradeCommand] Error in workspace 1e359dea-8c95-4029-823b-2992bd7bb774: Multiple validation errors occurred while updating field
[Nest] 36 - 01/24/2026, 11:27:52 AM ERROR [UpgradeCommand] Error in workspace 1e359dea-8c95-4029-823b-2992bd7bb774: Multiple validation errors occurred while updating field
[Nest] 36 - 01/24/2026, 11:27:52 AM ERROR [UpgradeCommand] undefined
[Nest] 36 - 01/24/2026, 11:27:52 AM LOG [UpgradeCommand] Command completed!
Successfully migrated DB!
No success error is still there
wet muralBOT
Thinking...
The logs show you're upgrading from 1.15.0 to 1.16.0, and there's an error during the UpdateTaskOnDeleteActionCommand for your workspace. The upgrade command reports "Multiple validation errors occurred while updating field" but still says "Command completed!" - this is a partial failure that left your workspace in an inconsistent state.
This is causing the applicationId null error - the workspace migration didn't complete properly.
Try this to fix the workspace:
docker exec -it twenty-db-1 psql -U postgres -d default -c "SELECT id, \"displayName\", version FROM core.workspace;"
1.15.0 (not 1.16.0), the migration failed partway through. You have two options:Option A - Reset the database (if this is a fresh install with no important data):
docker exec -it twenty-server-1 yarn
docker exec -it twenty-server-1 npx nx database:reset --configuration=no-seed
docker compose down
docker compose up -d
Option B - If you have data you need to keep:
You may need to manually fix the workspace version desync. According to the upgrade guide, when workspace versions get out of sync, you need to upgrade sequentially from the corresponding Twenty version.
First, can you share:
If you have a backup, restoring it and trying the upgrade again might be the cleanest path. If not, we may need to look at manually correcting the workspace state.
analog imp
root@twenty:~# docker exec -it twenty-db-1 psql -U postgres -d default -c "SELECT id, "displayName", version FROM core.workspace;"
WARNING: database "default" has a collation version mismatch
DETAIL: The database was created using collation version 2.36, but the operating system provides version 2.41.
HINT: Rebuild all objects in this database that use the default collation and run ALTER DATABASE "default" REFRESH COLLATION VERSION, or build PostgreSQL with the right library version.
id | displayName | version
--------------------------------------+-------------+---------
1e359dea-8c95-4029-823b-2992bd7bb774 | test7 | 1.15.0
(1 row)
wet muralBOT
Found the issue. Your workspace is stuck at version 1.15.0 while your Twenty instance is running 1.16.0. The upgrade command failed partway through with the UpdateTaskOnDeleteActionCommand error.
Since this is a fresh install (workspace named "test7"), the cleanest fix is to reset the database:
docker compose down
docker exec -it twenty-db-1 psql -U postgres -d default -c "DROP SCHEMA IF EXISTS \"workspace_1e359dea-8c95-4029-823b-2992bd7bb774\" CASCADE;"
docker exec -it twenty-db-1 psql -U postgres -d default -c "DELETE FROM core.workspace WHERE id = '1e359dea-8c95-4029-823b-2992bd7bb774';"
docker compose up -d
Then create a new workspace through the UI.
If you have data you need to keep, you can try manually updating the workspace version and re-running the upgrade:
1.16.0 manually:docker exec -it twenty-db-1 psql -U postgres -d default -c "UPDATE core.workspace SET version = '1.16.0' WHERE id = '1e359dea-8c95-4029-823b-2992bd7bb774';"
docker compose down
docker compose up -d
docker exec -it twenty-server-1 yarn command:prod cache:flush
Side note: The collation version mismatch warning in your database output is not critical but should be addressed eventually with:
docker exec -it twenty-db-1 psql -U postgres -d default -c "ALTER DATABASE \"default\" REFRESH COLLATION VERSION;"
Let me know which approach you want to try.
analog imp
Doesn't work. I switched back to 1.15. gui loads - in admin section now under app:
Status
Workspace metadata migration status check
relation "core.workspaceMigration" does not exist
nodeVersion
v24.12.0
timestamp
2026-01-24T11:39:00.733Z
stateHistory
No previous state available