topologyUpdater "invalid hostname" upon pool registration

Staking & Delegation Setup a Stake Pool
,
1

I’m receiving the message “invalid hostname not pointing to client IP” in topologyUpdater_lastresult.json since registering my relays with the pool certificate.

I registered the pool first with my operator cli keys and address with only one relay. I submitted my Ledger account as a pledging owner about 8 hours later, including both relays in the new certificate.

I feel it is the registration that somehow caused this to happen, or at least to visibly manifest, as the topologyUpdater_lastresult.json for each relay reflects the change to “invalid” occured at the time of certificate online propogation for each.

Aside from adding the relays to the certificate and submitting it, nothing about their setup has changed. The “glad you’re staying with us” message reliably came through for hundreds of hours until this.

Both relays are still running great and processing transactions. Block producer shows 2 in and 2 out (both relays) and each relay has outgoing and incoming peer connections.

Adaberry Pi Stake Pool [APSP]

  • pool id: pool1xdcg6kg0g4y3wqf92f8uh44urz06suglhff036k5yak2g0m8zlc
  • hash: 33708d590f4549170125524fcbd6bc189fa8711fba52f8ead4276ca4
  • relay addresses: relay1.adaberrypi.com and relay2.adaberrypi.com
  • both subdomains cname aliased to adaberrypi.3utilities.com (No-IP DDNS)
2

After 24 hourly runs on one relay and 21 on the other of topologyUpdater.sh responding with “invalid hostname not pointing to client IP,” I suddenly started getting “glad you’re staying with us” once more on both relays in the UTC 17:00 hourly update.

No changes were made on the pool’s node configurations. The relays and block producer all ran fine this whole time aside from the topology updater problem on the relays. I assume from the silence within a community that seems exceptionally forward-thinking that this is unusual, but am still open to speculation.