Files

Log File Summarizer

Turns a wall of log lines into level counts, a time span and the messages that repeat.

Loading the tool…

Processing happens locally in your browser. What you paste or load is processed by this page and is not uploaded to a server. Nothing is stored unless you use a control that says it stores something, and you can clear anything this site has kept from the privacy page.

How to use this tool

  1. Load the log file, or paste the section you care about.
  2. Select Summarise the log.
  3. Read the level counts and the time span first — they tell you whether this is one incident or a bad day.
  4. Then work down the repeated messages. The one at the top with a large count is almost always the thing to fix.

What log file summarizer does

The useful thing about a log is almost never one line. It is that the same error appeared four hundred times in ninety seconds, or that everything before ten past nine was fine. Reading top to bottom finds neither, because every retry of one failure looks like a different message: the timestamps differ, the durations differ, the request identifiers differ, and grep counts them as four hundred distinct problems.

This page strips the variable parts out of each line — numbers, timestamps, UUIDs, IP addresses, hex values and quoted strings — and groups what is left, so those four hundred lines collapse into one row with a count of four hundred. Alongside that it reports the counts by level, the percentage of the file each level accounts for, and the first and last timestamps it recognised. It works on ISO 8601 and Apache-style stamps, and everything runs in your browser.

Frequently asked questions

By stripping the parts that vary. Numbers, timestamps, durations, UUIDs, IP addresses, hex values and quoted strings are replaced with placeholders, and lines that are then identical are counted together. That is why four hundred retries of one failure collapse into a single row instead of four hundred distinct messages.

It does not require a format. Levels are found by looking for the usual words anywhere in a line — fatal, error, warn, info, debug and their variants — and timestamps by matching ISO 8601 and Apache-style stamps. Lines with no recognisable level are counted as unlabelled rather than dropped.

It reports the first and last timestamps in the order they appear in the file, not the earliest and latest by value. That is the right answer for a normal log and the wrong one for a file that has been sorted or concatenated out of order, which is worth knowing about in itself.

The processing is entirely local — nothing is uploaded and no request leaves the page — so the tool itself adds no exposure. Whether the content should be on that machine at all is a separate question and one only you can answer.