Documenso does not come up after update to 1.6.0

[marcusquinn]

Seems to be a problem with other apps, too.

WARNING: don't restart your apps if they are mission critical, until we have an answer here.

[pepicrft]

We are experiencing the same issue. I downgroaded to a 1.5.0 backup, but the issue remains, probably because the attempt to migrate to 1.6.0 put the DB in a weird state.

[james]

Hello @documenso users

I will look into this.

[james]

I just tried to reproduce this error and got the following log:

Jul 29 10:31:49 ==> Create / Migrate db
Jul 29 10:31:50 => Healthcheck error: Error: connect ECONNREFUSED 172.18.19.52:3000
Jul 29 10:31:51 \Prisma schema loaded from packages/prisma/schema.prisma
Jul 29 10:31:51 Datasource "db": PostgreSQL database "db806162892d894b9f84f2eafb933ba9d4", schema "public" at "postgresql:5432"
Jul 29 10:31:51 2025-07-29T08:31:51Z
Jul 29 10:31:51 132 migrations found in prisma/migrations
Jul 29 10:31:51 2025-07-29T08:31:51Z
Jul 29 10:31:51 Applying migration `20250522054049_add_id_generator`
Jul 29 10:31:51 Applying migration `20250522054050_add_organisations`
Jul 29 10:31:51 Applying migration `20250611052525_add_session_revoke_enum`
Jul 29 10:31:51 2025-07-29T08:31:51Z
Jul 29 10:31:51 The following migration(s) have been applied:
Jul 29 10:31:51 2025-07-29T08:31:51Z
Jul 29 10:31:51 migrations/
Jul 29 10:31:51 └─ 20250522054049_add_id_generator/
Jul 29 10:31:51 └─ migration.sql
Jul 29 10:31:51 └─ 20250522054050_add_organisations/
Jul 29 10:31:51 └─ migration.sql
Jul 29 10:31:51 └─ 20250611052525_add_session_revoke_enum/
Jul 29 10:31:51 └─ migration.sql
Jul 29 10:31:51 2025-07-29T08:31:51Z
Jul 29 10:31:51 All migrations have been successfully applied.

The app update was successful, and the app is in a healthy state.

---

@pepicrft a restore to the previous backup did not work? Did I get this right?

---

@marcusquinn said in Documenso does not come up after update to 1.6.0:

Seems to be a problem with other apps, too.

What other apps are you experiencing this issue with?

[james]

Please follow these steps:

While the app is installed and in an error state for version 1.6.0:

• enable the recovery mode in the app settings
• open the web terminal
• insert the following

   export DATABASE_URL=postgres://${CLOUDRON_POSTGRESQL_USERNAME}:${CLOUDRON_POSTGRESQL_PASSWORD}@${CLOUDRON_POSTGRESQL_HOST}:${CLOUDRON_POSTGRESQL_PORT}/${CLOUDRON_POSTGRESQL_DATABASE}
   export NEXT_PRIVATE_DATABASE_URL=postgres://${CLOUDRON_POSTGRESQL_USERNAME}:${CLOUDRON_POSTGRESQL_PASSWORD}@${CLOUDRON_POSTGRESQL_HOST}:${CLOUDRON_POSTGRESQL_PORT}/${CLOUDRON_POSTGRESQL_DATABASE}
   export NEXT_PRIVATE_DIRECT_DATABASE_URL=postgres://${CLOUDRON_POSTGRESQL_USERNAME}:${CLOUDRON_POSTGRESQL_PASSWORD}@${CLOUDRON_POSTGRESQL_HOST}:${CLOUDRON_POSTGRESQL_PORT}/${CLOUDRON_POSTGRESQL_DATABASE}
   npx prisma migrate status --schema ./packages/prisma/schema.prisma
  

Post the output of this command here.

Since each migration in the _prisma_migrations table has a logs column that stores the error we can also inspect that with:

PGPASSWORD=${CLOUDRON_POSTGRESQL_PASSWORD} psql -h ${CLOUDRON_POSTGRESQL_HOST} -p ${CLOUDRON_POSTGRESQL_PORT} -U ${CLOUDRON_POSTGRESQL_USERNAME} -d ${CLOUDRON_POSTGRESQL_DATABASE} -c "SELECT id,migration_name,started_at,rolled_back_at,logs FROM _prisma_migrations;"

and see what migration failed and has logs for further analysis.

---

For @CaeruleusAqua since the log provided the migration failed is 20250522054050_add_organisations we can attempt the following while the app is in recovery mode:

export DATABASE_URL=postgres://${CLOUDRON_POSTGRESQL_USERNAME}:${CLOUDRON_POSTGRESQL_PASSWORD}@${CLOUDRON_POSTGRESQL_HOST}:${CLOUDRON_POSTGRESQL_PORT}/${CLOUDRON_POSTGRESQL_DATABASE}
export NEXT_PRIVATE_DATABASE_URL=postgres://${CLOUDRON_POSTGRESQL_USERNAME}:${CLOUDRON_POSTGRESQL_PASSWORD}@${CLOUDRON_POSTGRESQL_HOST}:${CLOUDRON_POSTGRESQL_PORT}/${CLOUDRON_POSTGRESQL_DATABASE}
export NEXT_PRIVATE_DIRECT_DATABASE_URL=postgres://${CLOUDRON_POSTGRESQL_USERNAME}:${CLOUDRON_POSTGRESQL_PASSWORD}@${CLOUDRON_POSTGRESQL_HOST}:${CLOUDRON_POSTGRESQL_PORT}/${CLOUDRON_POSTGRESQL_DATABASE}

# Mark the migration as rolled back - this updates the migration record in the _prisma_migrations table to register it as rolled back, allowing it to be applied again:
npx prisma migrate resolve --rolled-back 20250522054050_add_organisations --schema ./packages/prisma/schema.prisma
# Try to re-deploy the migration
npx prisma migrate deploy --schema ./packages/prisma/schema.prisma

[joseph]

Seems to work fine for me in my test server. Can you give postgres more memory ? https://docs.cloudron.io/services/#configure

[marcusquinn]

@james All apps that have updated recently now experiencing this issue:

• Docuseal
• Librechat
• n8n
• Postiz
• Privatebin
• StirlingPDF
• Vaultwarden
• Ghost

I just increased the PostGreSQL DB Memory from 1GB to 2GB, and still no joy on app restarts.

[james]

Good Morning @marcusquinn
This sounds something else is astray with PostgreSQL on your instance.
Can you please review and/or send the PostgreSQL log file?

[pepicrft]

@james in my case the same migration failed, and when I tried to migrate, I got the following migration error:

Error: P3018

A migration failed to apply. New migrations cannot be applied before the error is recovered from. Read more about how to resolve migration issues in a production database: https://pris.ly/d/migrate-resolve

Migration name: 20250522054050_add_organisations

Database error code: 23502

Database error:
ERROR: null value in column "organisationId" of relation "OrganisationGroup" violates not-null constraint
DETAIL: Failing row contains (org_group_wxiansskyyehhuuo, null, INTERNAL_TEAM, MEMBER, null, 1, ADMIN).

DbError { severity: "ERROR", parsed_severity: Some(Error), code: SqlState(E23502), message: "null value in column \"organisationId\" of relation \"OrganisationGroup\" violates not-null constraint", detail: Some("Failing row contains (org_group_wxiansskyyehhuuo, null, INTERNAL_TEAM, MEMBER, null, 1, ADMIN)."), hint: None, position: None, where_: None, schema: Some("public"), table: Some("OrganisationGroup"), column: Some("organisationId"), datatype: None, constraint: None, file: Some("execMain.c"), line: Some(2006), routine: Some("ExecConstraints") }

So it seems the migration is making assumptions on the state of some tables.

[pepicrft]

For anyone running into this, it seems the migration assumes that all teams have a subscription. Until the migration is fixed, you can workaround it by faking a subscription.