DE

↑ ↓ select, Enter open, Esc close

Measure TTFB repeatedly with minimum, median and maximum

harmlessSEO checksdebian, ubuntu
Adjust the values, the command updates live
user@server
for i in $(seq 10); do curl -o /dev/null -s -w '%{time_starttransfer}\n' https://example.com/; done | LC_ALL=C sort -n | awk '{a[NR]=$1; s+=$1} END {printf "Min: %.3fs  Median: %.3fs  Max: %.3fs  Schnitt: %.3fs\n", a[1], a[int((NR+1)/2)], a[NR], s/NR}'

Requests the URL several times and evaluates the time to first byte statistically. A single measurement says little, only the median and outliers show whether a page cache is working or the response time fluctuates heavily. TTFB feeds directly into the LCP of the Core Web Vitals.

Note: The entry on measuring HTTP load times in the network category shows the individual phases of a request (DNS, connection and TLS). The summary labels in the output are German (“Schnitt” means average).

Also searched as

  • measure time to first byte multiple times
  • check fluctuating server response times
  • average ttfb curl
  • how fast does my server respond on average

Related one-liners

All in SEO checks

Compare the response time of a PHP page with a static file

$for u in https://example.com/ https://example.com/wp-includes/css/dist/block-library/style.min.css; do curl -o /dev/null -s -w "%{time_starttransfer}s TTFB $u\n" "$u"; done
harmlessImages and performance

Find images without a loading attribute on a page

$curl -s https://example.com/ | tr '\n' ' ' | grep -oiE '<img[^>]*>' | grep -vi 'loading='
harmlessImages and performance

Find scripts without async or defer on a page

$curl -s https://example.com/ | tr '\n' ' ' | grep -oiE '<script[^>]*src=[^>]*>' | grep -viE 'async|defer|type=.?module'
harmlessImages and performance

Measure HTTP load times and TTFB with curl

$curl -o /dev/null -sS -w 'DNS: %{time_namelookup}s Connect: %{time_connect}s TLS: %{time_appconnect}s TTFB: %{time_starttransfer}s Gesamt: %{time_total}s\n' https://example.com/
harmlessNetwork

Check for soft 404s: does a missing page really return 404

$curl -s -o /dev/null -w '%{http_code} %{redirect_url}\n' https://example.com/diese-seite-gibt-es-nicht-$RANDOM
harmlessSEO checks

Check maintenance mode for status 503 and Retry-After

$curl -s -o /dev/null -D - https://example.com/ | grep -iE '^(HTTP/|retry-after:)'
harmlessSEO checks

Read first, then run.

The commands on myline.de act directly on servers, files and databases. A wrong path or placeholder can delete data irreversibly or make a server unreachable.

  • All commands are provided without warranty and are not tested on every system.
  • Understand what a command does before running it, and check every placeholder.
  • Make a backup first and, if possible, try it on a test system.
  • You run commands at your own risk. Liability for damages is excluded to the extent permitted by law.