fishjam

@lni I have checked the code and try to add OnBecameUnreachable/OnBecameAvailable functions. I found the leader will send `Heartbeat` message to follower , but there is not other follower's active...

I know you dont' want add it in dragonboat's raft. example: - Availability statuses [#235](https://github.com/lni/dragonboat/issues/235) - How can I know the status of group [#212](https://github.com/lni/dragonboat/issues/212) - [Help]Listening for cluster changes...

Sorry for causing you misunderstanding. We do not want to know the server status , but just the ***raft node status in dragonboat***.

@lni I already know your thoughts about it. But I'm a little confused, what is the purpose of creating this issue?

we have same issue... dragonboat is a good opensource libraray, but seems the [dragonboat-example](https://github.com/lni/dragonboat-example) is a little simple. example: all the examples have fixed cluster, and all the cluster created...

after debug , I think maybe can get remote node's address from `node.clusterInfo.Nodes` , if the propose node is not in the cluster, can dragboat forward the propose to the...

> @fishjam @JohnMcPMS could you please take a look? @AtariDreams sorry, I'm not the owner. And if you want upgrade the toolset, I think you need provide the reason.

我加入了 SortField, TruncateOutFile, SkipFields 等配置信息, 可以通过导出到本地文件的方式进行比较. 而且似乎代码中以前使用 fasthttp 等 http 库似乎有问题, 我简单的想通过 `--regenerate_id --repeat_times=10` 来生成测试数据也不行. 而更改 http 使用方式后自己就好了(我没有改这部分的实现). 这些一起体现在 https://github.com/medcl/esm/pull/84 里. 示例: ``` esm --sort=_id --source=http://localhost:9200 --src_indexes=bank --truncate_output --skip=_index --output_file=src.json...

个人建议: 1. 尝试使用 --output_file + --fields 的功能, 将所有 records 的 id 等特殊字段导出到文件, 然后通过 diff/Beyond Compare 等工具比较到底是哪些记录没有被同步. 2. 然后通过 --source_proxy 和 --dest_proxy= 指定本地的 fiddler 等工具抓包; 通过 --query 指定没有同步的数据 , 然后通过 fiddler,...

看起来是类似 https://stackoverflow.com/questions/45427887/how-to-get-the-maximum-id-value-in-elasticsearch 的原因. 因为 --sync 功能需要排序, 所以默认采用了 _id(我只通过 docker 尝试并验证了 6.x 及以后的版本, 确实没有测试 5.x 版本) , 你可以尝试加入 `--sort=` ( 表示不指定排序字段 ), 应该可以规避这个问题. 如果想使用 `--sync` 功能的话, 必须指定排序字段. 如果你的 src/dst 数据中有唯一可以排序的字段(比如你们自己的业务 id),...