基于Geth的Clique共识PoA私有以太坊分叉：eth.blockNumber停滞在1

环境与部署 #

• 基于Geth搭建的权威证明（PoA）私有以太坊分叉，自定义chainId为1698369，采用Clique共识算法
• 4台服务器部署架构：
  • 服务器1：部署在NGINX反向代理后的公共RPC节点
  • 服务器2：私有RPC节点 + 3个独立端口的验证节点
  • 服务器3：私有RPC节点 + 3个独立端口的验证节点
  • 服务器4：私有RPC节点 + 3个独立端口的验证节点
• 所有13个服务（1个公共RPC、3个私有RPC、9个验证节点）均能通过genesis.json正常启动，systemctl状态显示为Active，但所有节点的eth.blockNumber始终停留在1，运行数小时无任何增长

排查操作 #

执行以下命令查看验证节点日志：

journalctl -u geth-mainnet-validator1.service -n 50000 | grep -E "Block|seal|mine|Commit|unlock"

日志示例 #

Mar 21 01:10:53 node2 test-geth-mainnet-validator1[139532]: INFO [03-21|01:10:53.493] Commit new sealing work                  number=1 sealhash=d264b8..b8970e txs=0 gas=0 fees=0 elapsed="148.258µs"
Mar 21 01:10:53 node2 test-geth-mainnet-validator1[139532]: WARN [03-21|01:10:53.493] Block sealing failed                     err="unauthorized signer"
Mar 21 01:13:28 node2 test-geth-mainnet-validator1[139532]: INFO [03-21|01:13:28.162] Commit new sealing work                  number=2 sealhash=f12875..f24402 txs=0 gas=0 fees=0 elapsed="115.696µs"
Mar 21 01:13:28 node2 test-geth-mainnet-validator1[139532]: WARN [03-21|01:13:28.162] Block sealing failed                     err="unauthorized signer"
Mar 21 01:11:03 node2 test-geth-mainnet-validator3[139562]: INFO [03-21|01:11:03.659] Successfully sealed new block            number=1 sealhash=bfc8a4..647af4 hash=77a147..1771e5 elapsed=7.713ms
Mar 21 01:11:03 node2 test-geth-mainnet-validator3[139562]: INFO [03-21|01:11:03.660] Commit new sealing work                  number=2 sealhash=1310ab..2ea0d8 txs=0 gas=0 fees=0 elapsed="543.651µs"
Mar 21 01:11:03 node2 test-geth-mainnet-validator3[139562]: WARN [03-21|01:11:03.660] Block sealing failed                     err="signed recently, must wait for others"

日志分析 #

• 多数验证节点尝试出块时失败，错误信息为unauthorized signer
• 仅validator3成功打包了区块1，之后无法生成新区块，错误信息为signed recently, must wait for others

genesis.json配置 #

{
  "config": {
    "chainId": 1698369,
    "homesteadBlock": 0,
    "eip150Block": 0,
    "eip155Block": 0,
    "eip158Block": 0,
    "clique": {
      "period": 5,
      "epoch": 30000
    }
  },
  "nonce": "0x0",
  "timestamp": "0x67d1ddfd",
  "extraData": "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",
  "gasLimit": "0x989680",
  "difficulty": "0x1",
  "mixHash": "0x0000000000000000000000000000000000000000000000000000000000000000",
  "coinbase": "0x0000000000000000000000000000000000000000",
  "alloc": {
    "8582102B6e433AEb76B4baA2247c8BB673054056": {"balance": "180000000000000000000000"},
    "44C41862AFe35E7ffA5d46D106E78e56282106D2": {"balance": "180000000000000000000000"},
    "4C0F541D9e0b6026dcee2532F778dB15E52AA716": {"balance": "180000000000000000000000"},
    "C17CfBaa87Ec82a26c2Ed9e0206aBF8c74d39d21": {"balance": "180000000000000000000000"},
    "eDDC1aD264D782A598DeB424d284DA751b0237eE": {"balance": "180000000000000000000000"},
    "44E3bA47fB9c036e3f9441F8c817d58f0d714c7F": {"balance": "180000000000000000000000"},
    "C79DE6A1eefAA4325B71590585B4b056B0750e97": {"balance": "180000000000000000000000"},
    "CEb07760b2b9797b7e31cfd648F7302925c28d58": {"balance": "180000000000000000000000"},
    "2C1EB859B739829ea7D3B99f4445710EfBED2017": {"balance": "180000000000000000000000"},
    "F7bCeae4a6B59e451f97C6D91BA7115C8ed0c00d": {"balance": "1361000000000000000000000"},
    "d2DEBaecF0591Ab97a4e1e8214fd357390f1879C": {"balance": "0"},
    "68a87bA8cb51aC05422138f639171d36f831F27c": {"balance": "99997019000000000000000000000"}
  }
}

内容的提问来源于stack exchange，提问作者Primea