MJ12bot — "not IP-identifiable" is wrong: 90 FCrDNS nodes
by shop owners for shop owners.
Majestic calls MJ12bot a distributed crawler with no IP list. Empirically wrong: the live core self-identifies via FCrDNS crawl-*.mj12bot.com, 72% on OVH. Triangulated from logs, multilingual research and 1.7M blacklist IPs → 90 nodes.
MJ12bot is the backlink crawler of Majestic (UK). Majestic states it's a "community based distributed crawler" and "cannot be blocked by IP". Empirically that's wrong: the current core self-identifies cleanly via FCrDNS crawl-<random>.mj12bot.com and sits 72% on OVH (AS16276) plus Veloxserv. Part of our PARC Security feed.
mj12bot.json — view feed mj12bot.com →
Verification — every IP individually
The PTR pattern crawl-<random>.mj12bot.com is FCrDNS-confirmed (reverse + forward A record back to the same IP) — only Majestic can set it. We triangulated the fleet from three independent sources:
access logs (own sites) 250 observed → 72 FCrDNS
+ multilingual research forums/crawler DBs (RU/ZH/DE/EN) → 90
+ 1.7M blacklist IPs AbuseIPDB-500k, ipsum, CINS, FireHOL 2-4,
(counter-check) blocklist.de, HaGeZi TIF etc. → +0 new
The blacklists yielded nothing — which is the real insight: MJ12 is too “polite” for abuse lists (respects robots.txt, rate-limits), barely appears there. Identification runs purely on the FCrDNS pattern. Stale legacy nodes (Netify’s ~144 includes historical Hetzner/Contabo/residential IPs that no longer reverse to mj12bot.com) are excluded — otherwise you’d hit foreign/customer IPs. Rolling re-verification (OVH rotates).