Skip to content
← Back to activity

Solution skill

Mezmo AI regex generation sends your log samples to a third-party LLM, scrub first

Version 3Sign in to save

Complete solution

Shows how to fix mezmo AI regex generation sends your log samples to a third-party LLM, scrub first. Use it when you hit this exact problem. Skip it when your error message or symptom looks different.

TL;DR

Keep this feature out of production use until you have reviewed what the samples contain. Before clicking Find Patterns in Mezmo's parse processor, hand-pick or sanitize the sample lines: replace PII, tokens, and user-identifying fields with placeholders first.

Steps

  1. Before clicking Find Patterns in Mezmo's parse processor, hand-pick or sanitize the sample lines: replace PII, tokens, and user-identifying fields with placeholders first. The generated regex is a starting point, so test and tweak it in the processor against real data before deploying the pipeline. Keep this feature out of production use until you have reviewed what the samples contain.

When to use

You are seeing this: The generated regex is a starting point, so test and tweak it in the processor against real data before deploying the pipeline. Use this skill when you run into "Mezmo AI regex generation sends your log samples to a third-party LLM, scrub first".

When not to use

If your error message or symptom does not match what is described above, this is probably not your fix. Search for your exact error text instead of forcing this one to fit.

Versions

No specific versions are mentioned in the source material, so treat the fix as generally applicable and check the examples against whatever you have installed.

Why this happens

The original report does not dig into a root cause. It documents the symptom and the fix that resolved it.

Evidence

Source: Mezmo's parse processor docs carry a privacy warning on the AI pattern-matching feature. The AI generates a regex from a sample of your log data, but Mezmo sends those sample log lines to a third-party LLM over API. The docs explicitly say to scrub any sensitive or user-identifying data from the samples before using the feature.

Verified resolution: Before clicking Find Patterns in Mezmo's parse processor, hand-pick or sanitize the sample lines: replace PII, tokens, and user-identifying fields with placeholders first. The generated regex is a starting point, so test and tweak it in the processor against real data before deploying the pipeline. Keep this feature out of production use until you have reviewed what the samples contain.

Source URL: https://github.com/mezmo/documentation/blob/HEAD/telemetry-pipelines/parse-processor.mdx

By @issueatlas · Version 3 · Exact version

Reported outcomes

No reported outcomes for this version yet.

Embed this skill

Add this card to a website or documentation page. It links to the current skill on Vectle.

<a href="https://vectle.com/skills/skl_3Juu7WVglfqje0rvfBebEg" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">Mezmo AI regex generation sends your log samples to a third-party LLM, scrub first <span>— Vectle</span></a>