Twelve Dollars for Google.com

Easy

· Backend

Incident

SEV-1

BUG-GDOMAINS

Priority: Critical · reported by registrar review

Time to fix: 10 min

Failing checks on the shipped build

  • a locked domain is not available

    expected false but got true

How to work this incident

  1. Read the report: write down the expected behaviour and the actual one in a sentence each.
  2. Reproduce it: run the visible tests and find the failing check that matches the report.
  3. Read the logs and the failing checks for the first value that is wrong — the earliest wrong value is nearer the cause than the loudest error.
  4. Follow the request through the handler, the validation and the service it calls; find the first step whose output disagrees with the report.
  5. Fix the cause rather than the symptom, run the visible tests again, and submit against the hidden ones — then write the root cause down.

Inspired by

Google Domains · 2015

Modelled on google.com being bought through Google Domains in September 2015. Sanmay Ved, a former Google employee, found google.com listed as available in Google Domains and bought it for $12; the purchase went through, and he briefly received the domain's webmaster messages before Google cancelled the order about a minute later and rewarded him through its bug bounty programme. Google did not publish the root cause.

This reconstruction plants one plausible mechanism — not Google's actual code: the availability check reads the registry record's EPP statuses and treats a domain as taken only when it carries the ok status. A heavily protected domain carries lock statuses instead of ok, so it shows as available.

Fix is_available so any registered domain is reported as taken.

Code map

def normalise(name):
    name = name.strip().lower()
    if name.endswith("."):
        name = name[:-1]
    return name


def is_available(name, registry):
    record = registry.lookup(normalise(name))
    if record is None:
        return True
    return "ok" not in record["statuses"]

Sign in to edit the file, run the tests and submit a fix

A free account — no card, no trial clock. You come straight back here.

Twelve Dollars for Google.com — Python Bug Hunt

Modelled on google.com being bought through Google Domains in September 2015.

  • Language: Python
  • Layer: Backend
  • Difficulty: Easy
  • Concepts: Validation, Parsing
  • Modelled on: Google Domains · 2015
  • Visible tests: unregistered is available, a plain registration is not; a locked domain is not available
  • Reward: 50 XP for a complete fix

Incident

  • Severity: SEV-1 (priority: Critical)
  • Ticket: BUG-GDOMAINS
  • Reported by: registrar review
  • Time to fix: 10 minutes, the target for an easy incident

Failing checks on the shipped build

  • a locked domain is not available — expected false but got true

How to work this incident

  1. Read the report: write down the expected behaviour and the actual one in a sentence each.
  2. Reproduce it: run the visible tests and find the failing check that matches the report.
  3. Read the logs and the failing checks for the first value that is wrong — the earliest wrong value is nearer the cause than the loudest error.
  4. Follow the request through the handler, the validation and the service it calls; find the first step whose output disagrees with the report.
  5. Fix the cause rather than the symptom, run the visible tests again, and submit against the hidden ones — then write the root cause down.

Briefing

Modelled on google.com being bought through Google Domains in September 2015. Sanmay Ved, a former Google employee, found google.com listed as available in Google Domains and bought it for $12; the purchase went through, and he briefly received the domain's webmaster messages before Google cancelled the order about a minute later and rewarded him through its bug bounty programme. Google did not publish the root cause.

This reconstruction plants one plausible mechanism — not Google's actual code: the availability check reads the registry record's EPP statuses and treats a domain as taken only when it carries the ok status. A heavily protected domain carries lock statuses instead of ok, so it shows as available.

Fix is_available so any registered domain is reported as taken.

Bug report

BUG-GDOMAINS · Priority: Critical · Reported by: registrar review

is_available(name, registry):

  • the name is normalised first: surrounding whitespace stripped, lowercased, one trailing "." removed; the registry is queried with the normalised name
  • True ONLY when registry.lookup(...) returns None (no record)
  • any record means the domain is registered and NOT available, whatever its statuses: "ok", lock statuses such as "clientTransferProhibited", "pendingDelete", or an empty list

Observed: google.com (statuses clientDeleteProhibited, clientTransferProhibited, clientUpdateProhibited, serverDeleteProhibited, serverTransferProhibited, serverUpdateProhibited) was offered for $12.

Logs

[search] google.com statuses=[clientDeleteProhibited, clientTransferProhibited, ...] -> available
[cart] google.com added, price=12.00 USD
[order] completed

The code as shipped

src/registrar/availability.py (editable)

def normalise(name):
    name = name.strip().lower()
    if name.endswith("."):
        name = name[:-1]
    return name


def is_available(name, registry):
    record = registry.lookup(normalise(name))
    if record is None:
        return True
    return "ok" not in record["statuses"]

Read-only context: src/registrar/registry.py.

Open the hunt to start the incident clock, edit the files, run the visible tests and submit against the hidden ones, then write the root cause down. More Python bug hunts.