Remote management
AI collects logs and finds the root cause
At a glance · AI Remote Log Analysis
LUMA AI remote log analysis is a technical support service that remotely collects system logs from a store’s LUMA box, with AI organizing severity, root cause, supporting log lines, and recommended steps. Log collection and remote app updates are handled by the LUMA technical support team, while customers view diagnosis results and history in the console. Personal information in logs is masked before the AI reads it.
Who it's for
If this sounds familiar
Key features
The LUMA support team receives the device's system logs, kernel logs, device info, and app logs as a single bundle without visiting the store.
Pressing a button runs log collection, upload, and AI diagnosis in sequence automatically. The device usually uploads logs within 1 minute.
Customer engineers can paste raw log text they already have and get a diagnosis report the same way.
If a new collection is requested from the same device within 1 minute, it's blocked to avoid overloading the device and network.
Decisive signals such as app force-stops, not responding, native errors, reboots, and out-of-memory kills are surfaced first, then read by AI.
Service impact is classified into four levels—info, warning, error, critical—along with how confident the AI is.
The most likely cause is summarized in one line, with up to 4 actual log lines attached as evidence along with the reasoning. If there's no evidence, nothing is invented.
Up to 5 steps of actions that store staff can follow are provided, and steps requiring remote support are marked as such.
When a device firmware fix seems needed, it suggests what to change and the risk level.
Download the diagnosis report as a file to share, and get the raw logs for engineer review. Diagnosis history accumulates per device to compare recurrence.
The server masks credentials, emails, IP/MAC addresses, and phone numbers in logs before passing them to the AI.
Sensitive, personally identifiable lines and large data blobs are removed or masked by the device itself before logs are sent.
Collected log bundles are automatically deleted after 30 days. Diagnosis result records are kept for history comparison.
The device counts app force-stops and not-responding events every minute, and uploads those from offline periods once reconnected. Shown in the console device row as, e.g., “24h force-stops: N”.
No matter which app stops, error dialogs are prevented from appearing over the customer screen, and instead logged as stop events so nothing is quietly buried.
If the cause is in the LUMA Box app, the support team remotely installs the fix on each device one by one. If installation fails, it stops there and the device keeps running the previous version.
From the same device screen, you can view diagnosis results and move straight into remote screen control or restart.
Turn “it freezes sometimes” into evidence-based diagnosis
Logs are collected, and AI sorts out the cause and the fix · AI-generated videoEquipment
The target that collects and sends logs — the device with the LUMA Box app installed
RequiredLog upload and diagnosis result delivery — freeze counts build up on the device while disconnected and upload once reconnected
RequiredView diagnosis results and history in the LUMA console, and paste logs for diagnosis — web browser
RequiredHow it works
Use cases by industry
24-hour Signage store
By watching the freeze count and running a remote log diagnosis when it rises
Even hard-to-reproduce issues get narrowed down based on supporting logs
Franchise kiosks
By comparing diagnosis history per store and fixing app issues with remote updates one device at a time
You can respond before the same issue spreads to multiple stores
Installation and maintenance companies
By getting the cause and action order first through remote collection and AI reports
You know the needed actions and parts before the site visit
Overseas and remote stores
By having HQ receive logs, diagnose, and act through Remote Support
Issues are handled with the same process regardless of distance or time zone
Exhibitions & pop-up events
By watching freeze counts during the event and diagnosing logs immediately when something looks off
Quickly identify issues during the event and respond remotely
Get started
Tell us your store size and the equipment you use, and we'll recommend the right setup and quote. We'll also check your existing screens and printers.
Specs
Works even better together
Frequently asked questions
The LUMA support team receives logs remotely, reviews them alongside AI Diagnosis results, and takes action.
Diagnostic logs are automatically deleted after 30 days.
It finds problems that leave traces in the logs, such as forced app closures, unresponsive apps, repeated reboots and out-of-memory shutdowns. The report shows the severity and confidence level, a one-line cause, the actual log lines used as evidence, and the steps to follow to fix it. If there is no evidence, it does not make anything up.
Phone numbers, emails, IP and MAC addresses, and credentials are masked by the server before the AI reads them. Sensitive lines that could identify a person are removed or masked by the device first, before it sends the logs.
If the cause is in the LUMA box app, the technical support team remotely installs a fixed version one device at a time. If an installation fails, it stops there and the device keeps running the previous version. The steps for the store to take are listed in order in the report, and from the same screen you can go straight on to remote control or a restart.
Our team will help you work out the setup and quote.