adamzhoul

> crds app noodpool is `apiextensions.k8s.io/v1beta1`， need change to `apiextensions.k8s.io/v1` this is indeed a problem for v1.22 [changeLog](https://kubernetes.io/blog/2021/07/14/upcoming-changes-in-kubernetes-1-22/#api-changes) - The beta CustomResourceDefinition API (apiextensions.k8s.io/v1beta1) projects : - https://github.com/openyurtio/yurt-app-manager/blob/master/config/setup/all_in_one.yaml - https://github.com/openyurtio/yurtcluster-operator...

if yurtcluster-operator will be deprecated how can we install openyurt to an existed cluster?

有点疑惑，求大佬指点： 如果是这里说的 os的bug。那导致的结果也就是cgroup的泄漏。 进而导致新pod无法被创建。最终宿主机被卡死。 然而，这里的问题是cache持续不释放，导致pod重启。 可能的case，就是最后流量高峰时刻。释放的速度赶不上申请的速度。导致最终OOM。 这个bug和cache自身的管理会有什么关系么？

Organization/Company: [lixiang](https://www.lixiang.com/#li) Location: Beijing, China Contact: [adamzhoul](mailto:[email protected]) Purpose/Scenario: Using OpenYurt for managing edge nodes, deploying edge applications

反馈一个问题: 这里wrap了一下OnReceive 实际上，会先调用w.stream.DestroyStream()。 这里会将client（conn) 放回到connPool 然后在调用w.streamReceiver.OnReceive ，将数据返回。 https://github.com/mosn/mosn/blob/cafbd5c6b9fe677845d43531e2d7c2c6f231b2bb/pkg/stream/client.go#L177-L180 这就导致一个问题。当有新的请求进来的时候。由于stream返回还没完成。所以会卡在 serve上一次for循环中 ``` func (conn *clientStreamConnection) serve() { for { sendStreamResponse(s) // sendStreamResponse } } ``` 导致，新的请求，从连接池获取到conn，发送完数据。但是迟迟没有开始处理response。

same issue. +1

check the pilot, discovery log get : warn unable to retrieve region label for pod: productpage-v1-777c468697-4c7pq is that the reason ? why

thank you @linsun @liamawhite each pod runs fine , only they can't communicate with the pilot. and, what else should i check? one more thing, after using tcpflow , i...