qixiaobo
qixiaobo
> Curator 5.0 Breaking Changes > Curator 5.0 contains a few non-backward compatible/breaking changes from previous versions. > > ZooKeeper 3.4.x is no longer supported (the associated Compatibility classes/methods have...
> I also met this before, that's may caused by the net issue. If you want to avoid the endless log, you may can also set log level for specific...
> Hi @qixiaobo , the curator version has been updated to 4.3.0, you can pull the newest master branch and try again. You can refer to this PR #3569 ....
The same problem
使用多层cache，目前我们是放在redis中 springcache
其实目前没有好的办法，我们之前使用memcache作为监听也出现了不能序列化的问题 因此需要手动实现Serializable接口，但是对于具体的client其实此时已经可能在反序列化已经失败。 private transient final MemcachedClient client; ``` public class ApiConfigChangeHandler extends AbstractApiConfigChangeHandle implements Serializable { private static final long serialVersionUID = 4683537583713221627L; private transient final MemcachedClient client; private final...
对于集群部署的话acessToken的存储？多台服务器对于同一个appId的话？
继承AbstractApiConfigChangeHandle
很多ops应用都有此类需求 甚至包含创建app
> 重新部署可能换一个ip？为啥部署就要换ip？ > 目前美团内部大部分服务已实现容器化部署，leaf服务运行在固定十几台机器上，机器ip也都是固定的，除非新增机器或下掉机器才会出现新消耗workerId的情况。leaf服务开始运行到现在几年时间了，workerId消耗的个数还不到100个。 他的意思docker的虚拟ip吧 重启都会更换的