Settings

Settings

Everything on the Settings page is stored in the data volume (/app/backend/data) and survives updates. Most cards have their own Save button that appears when something changed; the top Save / Discard buttons cover General, Update checks, Notifications, Health and Image cleanup.

General

Option	Description
Time zone	Used by every schedule (auto-update and cleanup). The initial value comes from TZ

Update checks

Background checks only mark containers with Update available; nothing is installed unless Automatic updates are set to do it.

Option	Description
Check when DockerUpdates starts	Run a check a few minutes after the app starts (for example after it updated itself), instead of waiting for the next interval
Check interval	How often every container is checked: from every 15 minutes to every week, or Never (startup only)

A new or edited container is also checked 3 seconds after it starts, so it does not stay Not checked.

Notifications

Which events are sent and to which channels. See Notifications.

LAN network (macvlan)

Gives containers their own IP on your LAN. See LAN network.

Health

Option	Description
Stop containers in a restart loop	If a container crashes within a minute of starting and Docker restarts it this many times in a row (2 to 50, default 5), it is stopped and you get a notification

Image cleanup

Option	Description
Remove old image after an update	Deletes the previous image right after a container is updated, if nothing else uses it
What to remove	Dangling images only (untagged leftovers, recommended) or Unused images (every image not used by a container, including tagged ones you may want to keep)
Scheduled cleanup	Run the cleanup on a schedule
Clean up now	Shows how many images and how much space would be freed, and runs it immediately

Account

The username and password used to sign in. Both start as ADMIN_USER / ADMIN_PASSWORD from .env.

• Username: type the new one and click Apply. A popup asks for your current password.
• Password: type the new password twice; the page tells you live whether they match and whether it is long enough (8+ characters, 12+ recommended). Change password opens the same popup asking for the current password.
• A wrong current password counts as a failed login (same lockout as the sign-in page).
• After a change you stay signed in; every other session is signed out.
• The password is stored as a scrypt hash, never in clear text.

Forgot the login you set here? Use RESET_LOGIN_CONFIG.

Server & access

Values from .env are shown filled in and can be overridden here. What you save takes precedence over .env and applies right away. Each field has a badge: Set here, From .env, Auto-detected or Default.

Option	Variable	Description
Host IP	HOST_IP	Pre-filled with the detected IP. Type another one to override it, Auto goes back to detection. Shown in orange with "This IP is not correct" when it is not an address of the server
Server name	HOST_NAME	Pre-filled with the name from Docker; type to override, Auto to go back
Trust a reverse proxy	TRUST_PROXY	Switch plus the proxy IPs (comma separated, ranges allowed). The field is greyed out while the switch is off
HTTPS-only session cookie	COOKIE_SECURE	Warns you when you turn it on while using plain HTTP
Extra allowed origins	ALLOWED_ORIGINS	Normally empty
Console keep-alive	CONSOLE_WS_KEEPALIVE	Seconds between pings on an open console so the reverse proxy does not close it when idle. 0 = off; 25 for Nginx Proxy Manager
Session secret	SESSION_SECRET	Shows Configured or Not set. Change generates a random secret (you can also type one, 32+ characters); it is only visible until you save. Saving signs out every other session
PORT, Docker connection	PORT, DOCKER_HOST	Read-only: only in .env

Registry credentials

Rows of registry / username / token for private registries (REGISTRY_AUTH). Tokens are never sent back to the browser: a saved token shows as ******** and is kept when you save other changes. See Private registries.

Login & sessions

Option	Default	Range	Description
Session lifetime	12 h	1 to 720 h	Sign in again after this long, even when active
Idle timeout	120 min	5 to 10080 min	Sign out after this long without activity
Failed logins per IP	5	3 to 100	Failures from one address before it is locked out
Failed logins in total	30	10 to 1000	Failures from all addresses before sign-in is locked for everyone (stops IP rotation)
Lockout window	15 min	1 to 1440 min	Failures are counted over this window, and a lockout lasts this long

Below the limits, a security checklist shows whether a reverse proxy is pinned, the cookie is HTTPS-only, a session secret is set and the password is long enough. Lockouts can be lifted in Admin Panel > IP access.