Daily Operations
After deployment, day-to-day work happens mostly in the "Ops Management" mode at the top of the panel. It does not impose a fixed order and suits service start/stop, configuration tuning, log viewing, version upgrades, and problem localization.
Overview
The overview page gathers:
- CPU, memory, disk, and average load, refreshed every 5 seconds by default;
- the health of base environment and business services;
- gateway and business service port conflicts;
- the drift dot on service cards (configuration changed but not yet applied); the current running version is shown on the Business Environment page.
Clicking a status card jumps to the matching management page with a focus parameter and locates the specific service.
Task Center
The task entry at the bottom right opens the task center: a right-side drawer shows running tasks (pinned to top) and history; each task opens a floating log window with its output (draggable, resizable, with copy and download).
- Long-running operations — building source, pulling source, deploying, applying gateway configuration — all run as tasks: switching pages after starting one does not interrupt it, and you can come back any time to check progress;
- tasks run concurrently by type: tasks of different types can run in parallel (e.g. apply gateway configuration while images are building); when same-type or exclusive tasks conflict, the panel reports which task holds the resource;
- when some services fail to build, the task ends as "partially successful" — the failure list is in the task log, and the failed services can be rebuilt individually;
- failures and partial successes raise a global notification; unhandled failures show a red badge on the task entry; history supports single delete and clear-all.
Base Environment
The base environment covers PostgreSQL, Redis, and Artemis. Each service card supports start, stop, restart, redeploy, container deletion, log viewing, details, and configuration.
- "Start All", "Stop All", and "Restart All" suit batch operations in a maintenance window.
- "Redeploy" rebuilds the container; whether data survives depends on whether the data volume is deleted — in production, read the confirmation prompt carefully first.
- "View Passwords" in the card's "Maintain" area shows the auto-generated middleware credentials; to change a password, use "Change Password" in the service card's "Config" drawer, then follow the prompt to sync and restart the business services.
- Ports are not mapped to the host by default; for temporary debugging, prefer binding to 127.0.0.1 only.
Business Environment
The business environment has two parts: the running state and version building.
Running State
The main app, channel sub-apps, the admin portal, the merchant portal, and the H5 app support grouped start/stop, restart, single-service deploy, redeploy, container deletion, and log viewing.
The app configuration drawer adjusts runtime parameters, port mappings, and the deployment mode; key management and channel info live in their own drawers, so you don't accidentally edit another service's configuration.
Versions and Builds
Supports downloading or updating source code, checking for updates, switching branch/tag, preparing the database/IP files, and building images.
- Source pulling covers all repositories (including extension channel repos); if the matching credentials are not configured, that repo shows as skipped — this is expected, not a hang;
- "One-Click Build All" is a parallel pipeline (backend Maven groups, both web ends, and H5 all progress at once); extension channel images join automatically once their credentials are configured (their version follows the business source; no separate tag selection);
- the build concurrency tier (smart / low-memory single-job / balanced / parallel) is set on the Global Configuration page; on low-spec machines, pin it to single-job;
- after a version switch or a global configuration change, the page shows a drift banner naming the services that need redeployment.
When versions refresh, extension channel images refresh along with them. For the complete coverage of versions, builds, and upgrades see the Deployment Guide.
Gateway Management
The gateway is the single entry point for external traffic. Day to day you can:
- configure domains for the admin portal, merchant portal, H5, the backend API, and Artemis;
- configure the CDN / fronting mode and the origin-pull token (the origin runs zero certificates; HTTPS is terminated by the outer CDN / gateway);
- toggle external endpoints such as /unipay, /epay, and /unipay/callback;
- deploy, redeploy, start, stop, restart, and uninstall the gateway in one click;
- view nginx access and error logs.
After saving domains, access mode, and endpoint toggles, click "Apply Gateway Configuration" to reload; only changes that require rebuilding the nginx container need "Redeploy". For the complete reference on access modes, origin-pull tokens, and the CDN onboarding checklist see Gateway Management - Domain Binding and CDN Fronting.
Extension Source
The Extension Source page maintains the code-source credentials for commercial extensions (a Yunxiao Codeup account with its HTTPS clone password) and checks access to the extension repositories. Configure the credentials once; they are shared by source pulls, extension-channel image builds, and mini-program builds. See Extension Products for how to configure them.
Mobile Management
The Mobile Management page provides version selection, source sync, and build delivery for the three mini-program clients (admin, merchant, cashier); download the artifact and import it into the platform's developer tools to publish. How to fill in the AppID and API endpoint at build time is covered in Extension Products.
Docker Management
Docker Management provides engine configuration, container, image, and network views:
- the "Config" tab shows Docker engine info, switches the registry mirror, and restarts Docker as prompted;
- the containers page searches, starts/stops containers, and shows logs, details, and disk usage;
- the images page deletes specific images or cleans up dangling images;
- the "Networks" tab lists the Docker networks (both panel-created and system built-ins).
Before deleting an image, confirm no service depends on it; cleaning dangling images is not a substitute for rolling back a business version.
Data Storage
The data storage page gives a read-only breakdown of the host disk and the data/ directory — usage categories, sizes, and file counts. Use it to judge disk pressure; it does not perform risky cleanup. Clean up containers, images, and volumes in Docker Management or in the terminal with explicit commands.
Operation Audit
Write operations — deploys, builds, start/stop, configuration updates, and logins — are recorded in the operation audit. You can filter by functional area, action, object type, and result, and inspect operation details. When investigating "who changed which configuration and when", check the audit trail first, then confirm the outcome with the task terminal and the service logs.
Global Configuration
Global Configuration hosts day-to-day entries such as the memory tier, run mode, and build mirror sources; it is the same configuration used by step 2 of the deployment guide.
After changing it:
- save the configuration;
- check the drift banner for the affected services;
- redeploy or restart as prompted;
- confirm on the overview page and in the logs that the new configuration is in effect.
Secure Entrance
The secure entrance places the login page and every panel API under a random path. After installation, open the full URL that includes the secure entrance, for example http://192.168.1.10:8060/k3x9m2qp4w7nbd5e; the random path is part of the full URL and must be kept. Do not open just http://192.168.1.10:8060 — the bare IP and port return 404, which is the intended protection.
Top-right user menu → "Security Settings" opens the Panel Settings page (Access Control tab), which supports:
- viewing the current entrance;
- rotating to a new random entrance;
- setting a custom entrance;
- disabling the entrance.
The secure entrance, domain restriction, the IP allowlist, and two-step verification together form the panel's access security stack — see Panel Security. After a change, the old URL stops working immediately and the page redirects to the new one. For automation, or when the web panel cannot be opened, use the command line:
daxpay entrance show
daxpay entrance rotate --yesrotate, set, and disable ask for confirmation by default; scripts and CI must pass --yes / -y explicitly.
When the Panel Is Unreachable
Start by running on the server:
daxpay status
daxpay logs --installer
daxpay entrance showYou can also run daxpay (no arguments) to enter the terminal management UI — its "Access Recovery" group walks you through the three defense lines: secure entrance / domain restriction / IP allowlist.
For a login lockout run daxpay unlock; for a forgotten password run daxpay reset-password. For more troubleshooting organized by stage — installation, login, runtime — see Configuration and Troubleshooting.