Log Merge Desk

Log Merge Desk

Merge logs into one timeline, trace requests across systems and export a shareable incident timeline.

  • Local-first
    The app runs on your own device and keeps its data there. Its main features work offline. Review checks that it does not go online on its own to send your data anywhere; features that need the internet, such as connecting to an AI, only connect after you set them up yourself.
  • No ads
  • No subscription
  • AI-modifiable
    The app comes with its source code. If something does not fit, tell an AI coding tool what you want and it changes the app for you. You can also publish your version for others to use. See how to change an app
  • Windows
  • macOS
ID
top1211306251
Creator
投如墨 (36 works)
Category
Tools
Age rating
12+
Price
$0.99
Released
2026-11-19
Built with
Codex · GPT-6
Package size
89.1 KB
Interface language
中文, English
Install this app

Installs in TopArrow. If you do not have it yet, setup opens this app.

Details

Overview

Log Merge Desk merges the logs from one incident into a single time-ordered timeline, links records by request or trace ID, and shows what failed first and where the time went. It exports a timeline you can paste into a ticket. Everything runs offline; your logs never leave your computer.

Getting started

  1. Choose Import log files and pick your application, gateway and other logs. Before importing, each file shows its encoding, detected timestamp format and parse ratio.
  2. Choose Merge timeline to browse every source in time order, and narrow it with the time, level, source and keyword filters.
  3. Select an error record and choose Trace this request to see its order and duration across systems, then choose Export for Markdown or CSV.

Features

  • Many logs together: import up to 20 TXT/LOG/JSON/JSONL files, detect encodings and timestamp formats, then merge and sort them. Records without a detectable timestamp go to the Untimed group instead of disappearing.
  • Timeline browsing: filter by time range, level, source and keyword; select a record to see its raw line, parsed fields and file line number.
  • Request tracing: common fields such as request_id and traceId are detected automatically; you can also type a request ID or a regular expression to see the path across systems, total duration and first error.
  • Marks and notes: flag key records and write down your judgement.
  • Deliverable exports: Markdown timeline with the first error, marks and request summary, or CSV with an option to include raw text; existing files are never overwritten.
  • Projects that reopen: each analysis is a local project; if a source file changed, the app offers to re-parse or keep the previous result.
  • Chinese and English interface: switch in the top bar; your choice is remembered.

Who it is for

  • Backend and operations engineers tracing cause and effect across logs after an incident.
  • Support engineers who need the sequence clear before sharing a fault report.
  • Testers and implementation engineers proving the order in which things happened.
  • Technical leads who need a clear timeline for a post-incident review.

Inputs, outputs and limits

  • Inputs: TXT, LOG, JSON and JSONL; up to 200 MB per file, 20 files per import and 1,000,000 records per project; lines over 64 KiB are truncated with a notice.
  • Encodings: UTF-8 (with or without BOM), UTF-16 and GB18030 are detected; if detection is wrong, pick the encoding and re-read.
  • Time: ISO 8601 with offsets, YYYY-MM-DD HH:MM:SS, MM/DD/YYYY, syslog without a year, and epoch seconds or milliseconds. Each source can have its own time zone.
  • Outputs: Markdown timeline and CSV (UTF-8 with BOM) in a folder you choose; existing files are never overwritten; raw-text export is capped at 50,000 records.
  • Not supported: live tailing, remote collection, proprietary binary log formats, log editing or redaction, charts and alerting, AI root-cause analysis, collaboration or cloud sync. The app shows order, paths and error positions; it does not decide the root cause for you.

Data and privacy

  • Everything runs locally: no account, no network, no telemetry. Logs are never uploaded.
  • Projects, parsed results and marks live in the TopArrow app data folder as plain text. Original log files are only read — never copied or modified.
  • Exports go where you choose; when an export includes raw log text, the app asks you to confirm there is no sensitive data first.
  • If you must share logs outside your team, redact credentials and personal data with the Ticket Log Redactor first; this app does not redact.

Troubleshooting

  • A timestamp was not detected: pick the timestamp format or the source time zone in the import preview; lines that still fail go to Untimed where you can inspect them.
  • The merged order looks wrong: check that source's time zone. Year-less syslog times are inferred from the file's modification year.
  • My request is not traced: make sure the logs share a request or trace field; you can type an ID or a regular expression on the trace view.
  • Large files import slowly: imports show progress and can be cancelled; import just the incident time window if the file is huge.
  • Continue later: reopen the project; if a source file changed, the app offers to re-parse or keep the last result.
  • Export failed: check the destination folder permissions and free space, then retry with another folder; project data is kept.
  • There is sensitive data in the logs: this app does not redact; check before exporting, and use the Ticket Log Redactor for anything leaving your team.

Change it if it does not fit

Add a field, change a page, add a chart — one sentence to an AI coding tool is enough.

The Age of AI Coding —
Change Your Own Software Yourself.