Unknown column 'location' in 'INSERT INTO' when adding new monitor (HTTP(s))

[epifeny]

When adding a new HTTP(s) monitor with only the friendly name and URL filled in, clicking Save fails with a database error: Unknown column 'location' in 'INSERT INTO'.

🛡️ Security Policy

• I have read and agree to Uptime Kuma's Security Policy.

📝 Description

👟 Reproduction steps

1. Open Uptime Kuma (Add New Monitor).
2. Leave Monitor Type as HTTP(s).
3. Enter a Friendly Name and URL only (all other fields left at defaults).
4. Click Save.

👀 Expected behavior

The monitor is created and saved successfully.

😓 Actual Behavior

A red error overlay appears with a long SQL INSERT statement and the message:

insert into `monitor` (`accepted_statuscodes_json`, `auth_method`, `cache_bust`, `conditions`, `dns_resolve_server`, `dns_resolve_type`, `docker_container`, `docker_host`, `domain_expiry_notification`, `expiry_notification`, `gamedig_given_port_only`, `http_body_encoding`, `ignore_tls`, `interval`, `ip_family`, `json_path`, `json_path_operator`, `kafka_producer_allow_auto_topic_creation`, `kafka_producer_brokers`, `kafka_producer_sasl_options`, `kafka_producer_ssl`, `location`, `maxredirects`, `maxretries`, `method`, `mqtt_check_type`, `mqtt_password`, `mqtt_success_message`, `mqtt_topic`, `mqtt_username`, `mqtt_websocket_path`, `name`, `oauth_auth_method`, `packet_size`, `parent`, `ping_count`, `ping_numeric`, `ping_per_request_timeout`, `protocol`, `proxy_id`, `rabbitmq_nodes`, `rabbitmq_password`, `rabbitmq_username`, `remote_browser`, `resend_interval`, `response_max_length`, `retry_interval`, `retry_only_on_status_code_failure`, `save_error_response`, `save_response`, `screenshot_delay`, `snmp_version`, `system_service_name`, `timeout`, `type`, `upside_down`, `url`, `user_id`, `ws_subprotocol`) values ('[\"200-299\"]', NULL, false, '[]', '', 'A', '', NULL, true, false, true, 'json', false, 60, NULL, '$', '==', false, '[]', '{\"mechanism\":\"None\"}', false, 'world', 10, 0, 'GET', 'keyword', '', '', '', '', '', 'REDUCTED', 'client_secret_basic', 56, NULL, 3, true, 2, NULL, NULL, '[]', '', '', NULL, 0, 1024, 60, false, true, false, 0, '2c', '', 48, 'http', false, 'https://REDUCTED', 1, '') - Unknown column 'location' in 'INSERT INTO'

The monitor is not created.

🐻 Uptime-Kuma Version

louislam/uptime-kuma:nightly2

💻 Operating System and Arch

Linux 5.4.0-216-generic #236-Ubuntu SMP Fri Apr 11 19:53:21 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux

🌐 Browser

Firefox 147.0.2 (64-bit)

🖥️ Deployment Environment

• Deployment: Docker
• Image: louislam/uptime-kuma:nightly2 (nightly)
• Database: Embedded MariaDB. Data directory has not been restored from backup; it has been in use since first setup (~1 week ago).
• Current container: Recreated 2026-01-30 (today); same data volume mounted.

📝 Relevant log output

2026-01-30T23:30:44+02:00 [MONITOR] ERROR: Error adding Monitor: undefined User ID: 1

[epifeny]

What I did to help validate

To confirm this is a migration/schema issue and not the image:

1. New nightly container from scratch – Same image (louislam/uptime-kuma:nightly2), new data directory, first-time setup. Called it uptime-kuma-nightly-test on port 3003.
2. Add HTTP(s) monitor on the new install – Same steps (friendly name + URL, Save). It works – monitor saves without error.
3. Checked both DBs in MariaDB – Ran the same queries on both containers. Output below.

same image, different outcome. The failing DB never had migration 2025-12-17-0000-add-globalping-monitor.js applied, so monitor is missing location, subtype, and protocol. The fresh one ran all migrations and has the full schema. Hope this helps narrow it down.

Failing install (original nightly – port 3002):

docker exec uptime-kuma-nightly mariadb -S /app/data/run/mariadb.sock -u root -e \
  "SELECT id, name FROM kuma.knex_migrations WHERE name LIKE '%2025-12-17%' OR name LIKE '%globalping%';"
(empty – no rows)

docker exec uptime-kuma-nightly mariadb -S /app/data/run/mariadb.sock -u root -e \
  "SELECT COLUMN_NAME FROM information_schema.COLUMNS WHERE TABLE_SCHEMA='kuma' AND TABLE_NAME='monitor' AND COLUMN_NAME IN ('location','subtype','protocol');"
(empty – no rows)

---

docker exec uptime-kuma-nightly-test mariadb -S /app/data/run/mariadb.sock -u root -e \
  "SELECT id, name FROM kuma.knex_migrations WHERE name LIKE '%2025-12-17%' OR name LIKE '%globalping%';"
id	name
36	2025-12-17-0000-add-globalping-monitor.js

docker exec uptime-kuma-nightly-test mariadb -S /app/data/run/mariadb.sock -u root -e \
  "SELECT COLUMN_NAME FROM information_schema.COLUMNS WHERE TABLE_SCHEMA='kuma' AND TABLE_NAME='monitor' AND COLUMN_NAME IN ('location','subtype','protocol');"
COLUMN_NAME
subtype
location
protocol

[CommanderStorm]

Actually, just going from nightly to nightly does not have any stability guarnntees.

Don't think this is a bug.
If any of our stable versions have this, it is.
Migrating nightly -> nightly?
No

[louislam]

When starting the server, there should be some db migration logs. See if any error messages.

[epifeny]

When starting the server, there should be some db migration logs. See if any error messages.

I've opted to not invest in this, since it is the nightly build. I've just added the missing tables by hand, and that was it. If this breaks, I'll just start a brand new nightly. This isn't "production" so I both don't expect it to break anything worthwhile investing too much effort in.