kursk
kursk
> @liangzh-404 > 这里 牺牲一致性、牺牲可用性是不是写反了， 因为如果你把 cookie 复制到所有服务器，预期来说是可用性更高，但是一致性更难保证；而如果做一致性哈希，那么是一致的（因为只有一份数据），但是可用性上就不行（因为重启等就会丢数据？） 我也觉得是的，我读这一段总觉得有哪里不对，根据[CAP原则](https://en.wikipedia.org/wiki/CAP_theorem)的解释 一致性：每次读都能读到最近一次的写入 可用性：每次请求都能获得一个非报错的响应，但不一定是最近一次的写入 而把每个用户分发到不同的节点上，当这个节点shutdown时，不能保证的应该是可用性，而不是一致性。 采用master-worker方式，同步数据时，如果master shutdown时，可以保证可用性，但是同步数据的速率不一定能保证是最新的写入，所以损失的是一致性
同求
1)改两个文件 vi /usr/local/nginx/conf.d/nextcloud.conf #因为我用nextcloud伪装，所以我是这个，在相同目录下，自己找 vi /usr/local/etc/xray/config.json 如果你没有改过，默认是443端口 将这两个文件的443端口，改为其它端口 比如，nextcloud.conf 文件中 `fastcgi_param SERVER_PORT 443` 2)然后把nginx和 xray重启服务 service nginx restart service xray restart 3)把客户端的端口也改成服务端相同，这不用说了吧
As a final addendum, cp01 is one of the control planes in the K8S cluster, and again I accessed it from a machine external to the K8S cluster, and of...
ok, logs of Easegress ingress cm pod as follow: ``` 2024-02-04T12:26:36.503Z ERROR ingresscontroller/ingresscontroller.go:142 failed to create kubernetes client: error loading config file "apiVersion: v1 clusters: - cluster: certificate-authority-data: ..... server:...
ok, I left the kubeConfig and masterURL values unset and restarted the deployment easegress ConfigMap as follow: ``` apiVersion: v1 kind: ConfigMap metadata: name: easegress-cm namespace: default data: easegress-server.yaml: |...
2. path / is configured in ingress and should be fine. ``` root@cp01:~# kubectl describe ingress ingress-example Name: ingress-example Labels: Namespace: default Address: Ingress Class: easegress Default backend: Rules: Host...
Thanks, I've been on vacation since the Lunar New Year is coming up and I can't access the server, I'll try again after fixing my vacation. Happy New Year!
I just returned to work yesterday and have been very busy at work and haven't had a chance to look into it yet, give me a moment.....
I'm sorry, I've been so busy lately that I haven't had the time to study the methods you've provided, I'll definitely test them out as soon as I'm free after...