Cannot resolve all.mainnet.ethdisco.net

[TheAviationDoctor]

Description

• Besu is reporting DNS resolution errors to all.mainnet.ethdisco.net every 10 minutes in the log.
• Besu behaves otherwise as expected (attesting and peering are fine - even a sync committee happened successfully), so the issue does not appear to affect node and staking operations in any way.
• This is on a staking NUC running Besu+Nimbus with dedicated 1Gbs fiber and ports 9000 & 30303 open.

Steps to Reproduce (Bug)

As far as I can tell, simply run Besu v23.10.0 on the config described below.

Expected behavior: Besu should be able to resolve all.mainnet.ethdisco.net (if the server exists) or not attempt to resolve it (if the server does not exist).

Actual behavior: Besu attempts and is unable to resolve all.mainnet.ethdisco.net. I also tried resolving/pinging/tracerouting all.mainnet.eth.disco myself, and am unable to (ping: all.mainnet.ethdisco.net: No address associated with hostname). I can ping other known servers just fine.

Frequency: Every 10 minutes exactly.

Logs (if a bug)

Oct 22 14:54:48 intel-nuc besu[906]: 2023-10-22 14:54:48.204+08:00 | vert.x-eventloop-thread-1 | WARN  | DNSResolver | I/O exception contacting remote DNS server when resolving all.mainnet.ethdisco.net
Oct 22 14:54:48 intel-nuc besu[906]: io.netty.channel.unix.Errors$NativeIoException: sendToAddress(..) failed: Network is unreachable
[...]
Oct 22 15:04:48 intel-nuc besu[906]: 2023-10-22 15:04:48.204+08:00 | vert.x-eventloop-thread-1 | WARN  | DNSResolver | I/O exception contacting remote DNS server when resolving all.mainnet.ethdisco.net
Oct 22 15:04:48 intel-nuc besu[906]: io.netty.channel.unix.Errors$NativeIoException: sendToAddress(..) failed: Network is unreachable

Versions (Add all that apply)

• Software version: besu v23.10.0
• Java version: 17.0.8.1 2023-08-24
• OS Name & Version: Ubuntu 22.04.3 LTS
• Kernel Version: Linux 5.15.0-87-generic 97
• Virtual Machine software & version: N/A
• Docker Version: N/A
• Cloud VM, type, size: N/A
• Consensus Client & Version if using Proof of Stake: Nimbus beacon node v23.10.0-8b07f4-stateofus

Additional Information (Add any of the following or anything else that may be relevant)

• Besu setup info: see below
• System info: 32Gb, 11th Gen Intel(R) Core(TM) i7-1165G7 @ 2.80GHz
• DNS was switched from the ISP's to 8.8.8.8 in an attempt to rule out resolution issues, to no effect

[Unit]
Description = Besu Execution Client (Mainnet)
Wants       = network-online.target
After       = network-online.target

[Service]
User        = besu
Group       = besu
Type        = simple
Restart     = always
RestartSec  = 5
Environment = "JAVA_OPTS=-Xmx8g"
ExecStart   = /usr/local/bin/besu/bin/besu \
  --network=mainnet \
  --sync-mode=X_CHECKPOINT \
  --data-path=/mnt/ssd/besu \
  --data-storage-format=BONSAI \
  --rpc-http-enabled=true \
  --engine-host-allowlist=localhost,127.0.0.1 \
  --engine-jwt-secret=/mnt/ssd/secrets/jwtsecret \
  --max-peers=32 \
  --Xplugin-rocksdb-high-spec-enabled

[Install]
WantedBy    = multi-user.target

[shemnon]

DNS info is not currently published

ethereum/discv4-dns-lists#9

The bug to besu would be to fail in a quieter fashion.

[non-fungible-nelson]

Referencing #5725 - Needs Tuweni logging change.

[zrock23]

I immediately joined in I personally work with a dns server: nessi.cloudns.cl

[siladu]

Hi @TheAviationDoctor is this still a problem for you?

[TheAviationDoctor]

Hi @TheAviationDoctor is this still a problem for you?

Hi @siladu, yes, I am still seeing this DNSResolver issue every 10 minutes in the Besu log, even after upgrading Besu to v23.10.1.

Not a big deal, though, as it does not seem to impact peering nor attestations in any meaningful way.