Local record: one-item retest preserves other evidence
A seven-item diagnosis of example.com retested only IPv4, checking that the other six results and times were retained and restored after a DNS-tool round trip.
Updated · Egress Observatory
Check whether retesting replaces unrelated results
Previously, a one-item retest cleared the displayed run. After result merging was changed, this acceptance checked that only the retested item updated while other evidence retained its original observation time. It verifies the diagnostic workflow, not a network repair.
Steps actually performed
A symptom diagnosis targeted example.com and produced seven result rows. The IPv4 retest was selected, followed by a comparison of item count, the IPv4 time and the times for other items.
The diagnosis then opened the DNS tool. The target was prefilled without an automatic query, and returning to the diagnosis restored the prior results.
Recorded observations
The displayed IPv4 time changed from 12:36:09 to 12:38:03. The other six results and times remained, with all seven items still present. These are browser-displayed times; the source summary did not record the time zone, so no UTC conversion is claimed.
The DNS page received example.com without automatically querying it. Returning restored all seven results and their times. Retained timestamps identify retained evidence; untested items were not presented as new measurements.
Retained results do not mean every probe succeeded
IPv6, DNS or some routing endpoints did not all succeed in this environment, and failures remained visible. Seven retained items establish evidence retention, not seven successful checks or a single identified cause for any failure.
To review this workflow, keep the target and symptom unchanged, retest one item, then inspect the other timestamps. A changed target requires a different diagnosis instead of reusing evidence for the old target.
Record scope and evidence version
Recorded on 2026-11-02 at accepted revision d802437. Chrome accessed local Caddy and Express. This was local integration acceptance, without production deployment, proxy switching or a physical phone.
The download is a redacted structured version of the recorded acceptance summary, containing options, observations and coverage limits. It is not a complete packet capture and contains no visitor IPs, accounts, secrets or request contents.