EOS研究系列【怎么同步公网节点】

这两天简单运行了一下cleos连接EOS主网，但是连接网上节点经常会断，返回数据也比较慢，如果想用得开心点还是得自己跑节点了！

在网上看到很多都是运行本地测试节点，其实我更需要的是同步公共节点数据，找到一两篇文章按其方式完成了了同步节点的配置，记录一下。

如果还没有nodeos、cleos参考：EOS研究系列【mac下编译运行】

创建genesis.json

下载区块链创世文件genesis.json，位置为：https://github.com/EOS-Mainnet/eos/blob/launch-rc-1.0.2/mainnet-genesis.json ，看内容不多直接粘贴下来，方便使用，如下：

{
  "initial_timestamp": "2018-06-08T08:08:08.888",
  "initial_key": "EOS7EarnUhcyYqmdnPon8rm7mBCTnBoot6o7fE2WzjvEX2TdggbL3",
  "initial_configuration": {
    "max_block_net_usage": 1048576,
    "target_block_net_usage_pct": 1000,
    "max_transaction_net_usage": 524288,
    "base_per_transaction_net_usage": 12,
    "net_usage_leeway": 500,
    "context_free_discount_net_usage_num": 20,
    "context_free_discount_net_usage_den": 100,
    "max_block_cpu_usage": 200000,
    "target_block_cpu_usage_pct": 1000,
    "max_transaction_cpu_usage": 150000,
    "min_transaction_cpu_usage": 100,
    "max_transaction_lifetime": 3600,
    "deferred_trx_expiration_window": 600,
    "max_transaction_delay": 3888000,
    "max_inline_action_size": 4096,
    "max_inline_action_depth": 4,
    "max_authority_depth": 6
  }
}

看起来这个文件和Bitshares差别还是挺大的，把这个文件拷贝到自己放eos数据目录下：

(env3) Chaim:~ Chaim$ mkdir eosdata
(env3) Chaim:~ Chaim$ vi genesis.json

genesis.json放上刚下载的内容，我们把目录建在用户根目录下，没有在代码目录下，因为代码目录经常清或重编容易弄丢。

生成config.ini

先执行下面命令在用户目录/eosdata下生成config.ini，执行几秒“Ctrl+C”即可：

Chaim:nodeos Chaim$ ./nodeos -d ~/eosdata/generator_node --config-dir ~/eosdata/generator_node --plugin eosio::wallet_api_plugin --plugin eosio::chain_api_plugin --p2p-listen-endpoint localhost:5555

增加p2p节点

编辑~/eosdata/generator_node/config.ini文件，增加p2p节点，节点可以在这取到，太多不全列出，如下节点亲测可用：

p2p-peer-address = 54.38.250.15:19878
p2p-peer-address = mainnet.eospay.host:19876
p2p-peer-address = peer.eosn.io:9876
p2p-peer-address = node.eosflare.io:1883
p2p-peer-address = eos.staked.us:9870

可能很多节点连不上，这还有一份列表，看日志把连接出错的给删除吧！

增加plugins

同样的config.ini，增加以下plugins：

plugin = eosio::producer_plugin 
plugin = eosio::wallet_api_plugin 
plugin = eosio::chain_api_plugin 
plugin = eosio::http_plugin 
plugin = eosio::history_api_plugin

http-server-address

这一项改不改都行，如果keosd没有占用的话就不用改：

# The local IP and port to listen for incoming http connections; set blank to disable. (eosio::http_plugin)
http-server-address = 127.0.0.1:8899

第一次启动

./nodeos -d ~/eosdata/generator_node --config-dir ~/eosdata/generator_node --genesis-json ~/eosdata/genesis.json --delete-all-blocks

以后启动就不需要“--genesis-json”和“--delete-all-blocks”参数了！

直到出现类似以下提示表明已经接收到数据包：

2515041ms thread-0   producer_plugin.cpp:290       on_incoming_block    ] Received block ddc03486114e6a1c... #1000 @ 2018-06-09T12:06:33.000 signed by eosio [trxs: 1, lib: 999, conf: 0, latency: 2345722041 ms]

不过这些节点在国内连起来很多都有问题，不知道是网络问题呢还是节点问题！同步了一天数据才同步完成！

轻松访问EOS数据

(env3) Chaim:cleos Chaim$ ./cleos -u http://127.0.0.1:8899 get info

早两天用网上节点取块数据经常一会就卡住，现在用本地节点应该不会有这些问题了吧！

踩个坑

同步得好好的，突然退出了，需要“replay”，replay就是区块数据重新跑一遍，可千万不要重新下载区块啊，都跑一天了！

libc++abi.dylib: terminating with uncaught exception of type boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<std::logic_error> >: Could not restore object, most likely a uniqueness constraint was violated
./pub.sh: line 1: 78500 Abort trap: 6           ./nodeos -d ~/eosdata/generator_node --config-dir ~/eosdata/generator_node

Chaim:nodeos Chaim$ ./pub.sh 
2433929ms thread-0   chain_plugin.cpp:209          plugin_initialize    ] initializing chain plugin
2434000ms thread-0   main.cpp:123                  main                 ] database dirty flag set (likely due to unclean shutdown): replay required

加上“--replay-blockchain”参数还是出错：

Chaim:nodeos Chaim$ ./nodeos -d ~/eosdata/generator_node --config-dir ~/eosdata/generator_node --replay-blockchain
3086942ms thread-0   chain_plugin.cpp:209          plugin_initialize    ] initializing chain plugin
3086942ms thread-0   chain_plugin.cpp:341          plugin_initialize    ] Replay requested: deleting state database
3087788ms thread-0   main.cpp:123                  main                 ] database dirty flag set (likely due to unclean shutdown): replay required

还有一招“--hard-replay-blockchain”先试试：

Chaim:nodeos Chaim$ ./nodeos -d ~/eosdata/generator_node --config-dir ~/eosdata/generator_node --replay-blockchain --hard-replay-blockchain
3178561ms thread-0   chain_plugin.cpp:209          plugin_initialize    ] initializing chain plugin
3178562ms thread-0   chain_plugin.cpp:322          plugin_initialize    ] Hard replay requested: deleting state database
3178646ms thread-0   block_log.cpp:323             repair_log           ] Recovering Block Log...
3178652ms thread-0   block_log.cpp:343             repair_log           ] Moved existing blocks directory to backup location: '/Users/Chaim/eosdata/generator_node/blocks-2018-07-07T11:52:58.646'
3178653ms thread-0   block_log.cpp:348             repair_log           ] Reconstructing '/Users/Chaim/eosdata/generator_node/blocks/blocks.log' from backed up block log
3231752ms thread-0   block_log.cpp:458             repair_log           ] Existing block log was undamaged. Recovered all irreversible blocks up to block number 4419270.
3231891ms thread-0   chain_plugin.cpp:534          recover_reversible_b ] Reconstructing '/Users/Chaim/eosdata/generator_node/blocks/reversible' from backed up reversible directory
3232222ms thread-0   chain_plugin.cpp:577          recover_reversible_b ] Recovered 2890 blocks from reversible block database: blocks 4419271 to 4422160

结果很慢的执行还是出错了，最终没有办法重新走第一次启动的过程，删除全部区块重来！

参考

https://blog.csdn.net/lizhe_dashuju/article/details/80746377

https://eosio.stackexchange.com/questions/725/how-to-vote-using-cleos/726#726

---

感谢您阅读 @chaimyu 的帖子，期待您能留言交流！