Proxy tier benchmark Targets

Why does the Nebraska corporate search time out, and does a proxy fix it?

State business entity registries · minimum viable tier: no tier sufficient · measured 2026-08-18

A proxy changes the failure but does not fix it. All ten unproxied requests hung and timed out after 30 seconds without a response; through datacenter, anti-detect and residential exits the same host answered in about a second with a 26,240-byte reCAPTCHA page. Every arm scored 0%. Measured 2026-08-18.

What was measured

URL https://www.nebraska.gov/sos/corp/corpsearch.cgi
Content assertion Corporate|Businessnever verified against a live page
Arms 6
Rounds per arm 10
Requests 60
Measured 2026-08-18, 15:38–18:09 UTC
Benchmark host DigitalOcean droplet, New York (nyc3)

A request counts as a success only if the expected content is present. An HTTP 200 carrying a challenge page, an empty body or the wrong page is a failure, because it is not a page you can use.

This measures whether a tier works, not long-run stability under sustained load. Ten requests per arm, at a two-second floor, never approach a per-IP rate limit — which is the thing a rotating pool exists to defeat.

Results by arm

Arm Success p50 p95 Bytes per success Outcomes
direct (no proxy) 0% 30,166 ms 30,176 ms n/a error 10
datacenter, one pinned exit 0% 15,386 ms 16,065 ms n/a challenge 2, error 8
datacenter, rotating exit 0% 15,311 ms 15,777 ms n/a challenge 4, error 6
anti-detect, one pinned exit 0% 15,728 ms 19,147 ms n/a challenge 3, server error 7
anti-detect, rotating exit 0% 15,439 ms 17,736 ms n/a challenge 5, server error 5
residential, rotating exit 0% 2,480 ms 19,051 ms n/a challenge 9, server error 1

What protects this target

A tarpit for direct traffic, a captcha for everything else.

The unproxied arm's ten requests each ran to the harness's 30-second ceiling with a ReadTimeout and a zero-byte response: the connection was accepted and never answered. PROTECTION-CENSUS.md flags this signature on five state hosts — Nebraska, Missouri, Tennessee, Wisconsin and Ohio — and makes the point that matters most here: a status-code-only measurement records nothing at all. There is no 403 to count, no challenge page to fingerprint, and a scraper without a timeout simply stops.

Through a proxy the host behaves differently. Twenty-three of 60 requests returned a byte-identical 26,240-byte reCAPTCHA page, in a median of around a second on the residential arm, and the remaining failures were split between transport timeouts on the datacenter arms and HTTP 502s on the anti-detect and residential arms. So the tarpit is applied to the requesting address, not to the request: change the exit and the same URL answers promptly.

That identifies the mechanism without offering a route through it. Converting a 30-second hang into a reCAPTCHA is diagnostic progress and nothing more; no HTTP client solves a reCAPTCHA at any tier.

robots.txt status

The URL measured is allowed by the host's robots.txt.

Status allowed
File https://www.nebraska.gov/robots.txt
Crawl-delay not declared
Fetched as the benchmark's own User-Agent
Note robots.txt 97 bytes via benchmark_ua

Limits of this result

  • The content assertion for this target was never verified against a live page. The 26,240-byte body was classified by its reCAPTCHA signature before the assertion ran, so it is not established whether that body is a challenge page or the search form with a reCAPTCHA widget on it — the same ambiguity documented for Iowa and Louisiana.
  • The search page was measured, not a submitted query.
  • Ten rounds per arm in a single three-hour window on 2026-08-18. Anti-bot behaviour drifts; a result older than a few months is a historical record, not a current one.
  • One exit geography. Every request left from a US datacenter host or a US exit, so nothing here describes how this target answers traffic from elsewhere.
  • The unproxied arm ran from a clean, previously unused cloud IP with no accumulated reputation. It is a best case a production scraper will not sustain.

Run date: 2026-08-18 · Full benchmark report · Harness and raw data · All 32 targets

The 60 rows behind this page are in results/six-arm/raw.csv; filter them with grep '^[^,]*,us_ne_corpsearch,' raw.csv.


Test your own target

The result above describes one URL on one day. Paste your own URL into the live tester and see the real response before creating an account.

Test your target See plans