Traffic Pattern Analyzer

Traffic has a shape — and that shape tells you when things break. A number like "14,302 requests today" tells you almost nothing about what actually happened; a chart of those requests over time immediately shows the morning ramp, the lunchtime dip, the deploy that caused a cliff, or the five minutes where everything fell over. Traffic Pattern Analyzer charts requests over time as a fast ASCII bar chart directly from pasted log lines, with per-minute to per-day zoom and an optional status-class filter, so you get that shape without opening a dashboard or writing a query.

Each line is parsed for a leading timestamp (YYYY-MM-DD HH:MM:SS, with a space or "T" separator) and grouped into a time bucket based on the interval you pick: per-minute for zooming into a short, intense window; per-10-minutes as a middle ground; per-hour for a full day’s shape; per-day for a multi-day trend. Requests in the same bucket are counted, and each bucket is rendered as a row with a proportional bar — the width of every bar is scaled against the busiest bucket in the window, so the relative shape of your traffic is visible at a glance even before you read a single number.

The status-class filter narrows the same chart to only 2xx, 4xx or 5xx responses (or "all" for everything), which turns the same tool into a failure timeline. Charting all requests shows your traffic curve; charting only 5xx responses turns a wall of mostly-successful log lines into an isolated island of bars exactly where the incident happened, with everything else silently excluded rather than diluting the signal.

• Time-series ASCII bar chart — request counts per time bucket rendered as proportional bars, no charting library required.
• Zoomable buckets — per-minute, per-10-minutes, per-hour or per-day, chosen to match the size of the window you pasted.
• Status-class filter — isolate 2xx, 4xx or 5xx responses to turn the same chart into a success or failure timeline.
• Peak + average headline numbers — the busiest bucket (with its count) and the average requests per bucket, computed alongside the chart.
• Copyable chart — the whole ASCII chart as plain text, ready to paste into a ticket, chat message, or postmortem doc.

1. Paste access log lines that begin with a timestamp and include a 3-digit HTTP status code.
2. Choose a bucket interval that fits your window — minutes for a short burst, hours or days for a longer trend — and a status filter if you want to isolate failures.
3. Click Chart. Bars render per bucket along with peak and average stats.
4. Copy the ASCII chart to share the shape of your traffic without a screenshot.

Example 1 — Daily wave. Hourly bars over a full day’s log make your 09:00 ramp-up and 20:00 evening peak visually obvious, which is exactly the information you need to schedule maintenance windows, cron jobs, or deploys around low-traffic hours.

Example 2 — Finding the outage minute. Filter the status class to 5xx and switch to per-minute buckets — the incident stands out as an isolated cluster of bars on an otherwise empty chart, and the bucket labels tell you exactly when it started and how long it lasted.

Example 3 — Proving a cache deployment worked. Chart 2xx-only traffic for a window before a cache change and again for a window after; if the cache is doing its job, the shape (and often the total volume hitting your origin) visibly changes.

Example 4 — Spotting a slow traffic ramp vs. a spike. A gradual, steadily-rising set of bars usually means organic growth or a marketing push; a single bucket that dwarfs its neighbors usually means a bot, a retry storm, or a scraper — the chart makes the difference obvious without doing any math.

• Instant visual — see the shape of your traffic with no charting library or dashboard login.
• Failure focus — status filtering isolates exactly when things went wrong.
• Zoomable — minute-level detail or day-level trend, same tool.
• Headline numbers included — peak bucket and average are computed automatically alongside the chart.
• Private — parsing and charting happen entirely in your browser.