Log Timestamp Converter & Time Delta Annotator

How to use this tool

1. Paste or load your logs. Paste multiline log text into the Log lines box, choose a local .txt, .log, or .csv file with "Load file…", or click "Load sample" to see an example. The selected file is read in your browser.
2. Choose detection mode. Leave "Auto-detect common formats" selected for ISO 8601, Apache/Nginx, RFC 2822, RFC 3164 syslog, and Unix epoch timestamps, or switch to "Custom regex + format" for anything else.
3. Set the unzoned assumption. If your logs contain timestamps with no UTC offset, choose whether to treat them as UTC, your browser’s local time, or a specific IANA timezone.
4. Choose the output timezone and annotate. Choose Browser local time, UTC, or a specific IANA output timezone. Then run the annotator to append a readable time and compact elapsed-time delta to each valid line, and review the summary and diagnostics.
5. Copy or download the result. Use Copy or Download .txt on the output box to save the annotated log. The original line order, unmatched lines, and line endings are preserved.

Turning scattered log timestamps into a readable timeline

Paste a multiline log excerpt or load a local text file. This tool appends a readable time and compact elapsed-time delta to every line where it finds a valid timestamp. Output defaults to your browser's timezone, but you can choose UTC or a specific IANA timezone. Lines without a timestamp, such as stack traces or plain text, stay unchanged and do not break the running comparison. Pick Auto-detect common formats for the grammars below, or switch to Custom regex + format for another format, then click Annotate logs. Parsing, file reading, timezone math, and output formatting all run locally in your browser; you can copy the result or download it as a text file.

Explicit UTC versus an assumed timezone

A timestamp that ends in Z or carries a numeric offset like +02:00 or -0400 is unambiguous: the tool reads it as explicit UTC or an explicit offset and converts it directly. A bare timestamp with no offset, such as 2026-07-21 14:03:20, is inherently ambiguous. No software can recover the author's intended timezone from the digits alone, so the Unzoned timestamps are selector lets you choose how to interpret it: as UTC (the default), your browser's current local time, or a specific IANA timezone such as Europe/Berlin. This source assumption is separate from the Output timezone setting: for example, you can interpret an unzoned server timestamp in America/New_York and display its instant in UTC. The output selector starts with your browser's timezone and is shown just above the input box.

Supported built-in formats

Automatic detection recognizes ISO 8601/RFC 3339 timestamps with a T or space separator and dot- or comma-separated millisecond fractions; Apache/Nginx access-log timestamps like [21/Jul/2026:14:03:20 +0000]; RFC 2822 dates with numeric offsets; RFC 3164 syslog timestamps such as Jan 2 15:04:05; and standalone Unix epoch values, either 10-digit seconds or 13-digit milliseconds. Syslog has no year, so this tool assumes the current calendar year in your browser; check year-boundary entries before relying on their deltas. Both syslog and ISO timestamps without a zone use the selected unzoned assumption. When several patterns could match the same position, the earliest and most specific candidate wins, and only the first timestamp on each line is used. Epoch matching is boundary-guarded so it does not select digits embedded inside longer identifiers.

Custom regex and format tokens

For anything outside those grammars, switch to custom mode. Your regular expression must contain a named capture group, (?<timestamp>...), and the first match on each line is what gets parsed. A companion format pattern then parses that captured text deterministically, using tokens like yyyy, MM, MMM (English month abbreviations), dd, HH or hh with a for a meridiem, mm, ss, fractional-second tokens, and XX/XXX for a compact or colon-separated offset. This deliberately avoids handing arbitrary captured text to the JavaScript engine's built-in date parser, whose behavior for non-standard formats varies by browser.

Relative-time semantics

Every successfully parsed line is compared to the previous successfully parsed line, even when unrelated or invalid lines sit between them. The first valid timestamp is labeled start; an identical timestamp shows same time; later timestamps get a leading + and earlier ones a leading -. Small gaps render as milliseconds or seconds with trailing zeroes trimmed, while gaps of a minute or more show the two largest non-zero units, for example +1m 37s or +2h 4m.

DST ambiguity and nonexistent times

Interpreting an unzoned timestamp against an IANA timezone can hit a daylight-saving transition. During a fall-back overlap, the same wall-clock time occurs twice; this tool picks the earlier of the two instants and counts it as an ambiguous wall time in the summary. During a spring-forward gap, the wall-clock time never occurred at all, so that line is reported as invalid rather than silently shifted to a neighboring hour.

Debugging and incident-response examples

This tool is built for scanning an incident timeline quickly: paste server logs that mix a UTC-stamped upstream service with an unzoned application log, choose the application's known timezone for unzoned lines, and immediately see which request came first and how long each step took, without manually converting every timestamp by hand. It is equally useful for lining up client-reported local times against server UTC logs when triaging a bug report, or for spotting an unusually long gap between two log lines during a performance investigation.