kubernetesClient does not support setting connection pool size.

[happyzzz1997]

I want to set the size of the connection pool, but the kubernetesClient does not support it. DefaultKubernetesClient(Config config) and DefaultKubernetesClient(HttpClient httpClient, Config config) create objects that look very different in implementation, how should I set the connection pool size?

[shawkins]

This depends on the underlying http client implementation. More than likely you'll need to override the specific http client factory to do this. Can you elaborate on which http client you are using and why you need to set the pool size?

[happyzzz1997]

This depends on the underlying http client implementation. More than likely you'll need to override the specific http client factory to do this. Can you elaborate on which http client you are using and why you need to set the pool size?

I use the DefaultKubernetesClient(Config config) method to create a k8s client. I found that the client may continuously timeout when disk performance is poor or CPU frequency is unstable. I actively catch the exceptions that occur when using the client and close the connection, but it does not work. The exception continues until it recovers on its own, and the HTTP connection remains open during the fault period and is not closed as expected. I am not sure if the exception is related to this connection issue, so I want to set the connection pool size to 0 to create a new connection for each request.

[happyzzz1997]

kubernetes version: 1.21.14
fabric8 version: 4.9.0

io.fabric8.kubernetes.client.KubernetesClientException: An error has occurred.
at io.fabric8.kubernetes.client.KubernetesClientException.launderThrowable(KubernetesClientException.java:64) ~[?:?]
at io.fabric8.kubernetes.client.KubernetesClientException.launderThrowable(KubernetesClientException.java:53) ~[?:?]
at io.fabric8.kubernetes.client.utils.Serialization.unmarshal(Serialization.java:248) ~[?:?]
at io.fabric8.kubernetes.client.utils.Serialization.unmarshal(Serialization.java:199) ~[?:?]
at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse(OperationSupport.java:474) ~[?:?]
at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse(OperationSupport.java:430) ~[?:?]
at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse(OperationSupport.java:412) ~[?:?]
at io.fabric8.kubernetes.client.dsl.base.BaseOperation.listRequestHelper(BaseOperation.java:151) ~[?:?]
at io.fabric8.kubernetes.client.dsl.base.BaseOperation.list(BaseOperation.java:621) ~[?:?]
at io.fabric8.kubernetes.client.dsl.base.BaseOperation.list(BaseOperation.java:70) ~[?:?]
at com.h3c.matrix.util.K8sClientHelper.getAllPods(K8sClientHelper.java:177) ~[?:?]
at com.h3c.matrix.monitor.util.MonitorUtil.getContainerResourceFromMetrics(MonitorUtil.java:1099) ~[?:?]
at com.h3c.matrix.monitor.main.LeaderOnlyMonitor.AlertManager.doMonitor(AlertManager.java:241) ~[?:?]
at com.h3c.matrix.monitor.main.LeaderOnlyMonitor.AlertManager.runOnce(AlertManager.java:213) ~[?:?]
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [?:1.8.0_322]
at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308) [?:1.8.0_322]
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180) [?:1.8.0_322]
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294) [?:1.8.0_322]
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) [?:1.8.0_322]
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) [?:1.8.0_322]
at java.lang.Thread.run(Thread.java:750) [?:1.8.0_322]
Caused by: java.net.SocketTimeoutException: timeout
at okhttp3.internal.http2.Http2Stream$StreamTimeout.newTimeoutException(Http2Stream.java:593) ~[?:?]
at okhttp3.internal.http2.Http2Stream$StreamTimeout.exitAndThrowIfTimedOut(Http2Stream.java:601) ~[?:?]
at okhttp3.internal.http2.Http2Stream$FramingSource.waitUntilReadable(Http2Stream.java:374) ~[?:?]
at okhttp3.internal.http2.Http2Stream$FramingSource.read(Http2Stream.java:337) ~[?:?]
at okhttp3.internal.http2.Http2Codec$StreamFinishingSource.read(Http2Codec.java:210) ~[?:?]
at okio.RealBufferedSource.read(RealBufferedSource.java:46) ~[?:?]
at okio.RealBufferedSource.request(RealBufferedSource.java:67) ~[?:?]
at okio.RealBufferedSource.require(RealBufferedSource.java:60) ~[?:?]
at okio.GzipSource.consumeHeader(GzipSource.java:114) ~[?:?]
at okio.GzipSource.read(GzipSource.java:73) ~[?:?]
at okio.RealBufferedSource$1.read(RealBufferedSource.java:430) ~[?:?]
at java.io.BufferedInputStream.fill(BufferedInputStream.java:246) ~[?:1.8.0_322]
at java.io.BufferedInputStream.read(BufferedInputStream.java:265) ~[?:1.8.0_322]
at io.fabric8.kubernetes.client.utils.Serialization.unmarshal(Serialization.java:238) ~[?:?]
... 18 more

E0927 21:46:41.993869 1 wrap.go:54] timeout or abort while handling: GET "/api/v1/pods"
E0927 21:46:41.994160 1 writers.go:117] apiserver was unable to write a JSON response: http2: stream closed
{"level":"warn","ts":"2024-09-27T21:46:41.994+0900","caller":"clientv3/retry_interceptor.go:62","msg":"retrying of unary invoker failed","target":"endpoint://client-3f2a8b4a-85dc-430c-9e53-edad3c8b818d/matrix-node1:2379","attempt":0,"error":"rpc error: code = Canceled desc = context canceled"}
E0927 21:46:41.994256 1 status.go:71] apiserver received an error that is not an metav1.Status: &errors.errorString{s:"context canceled"}: context canceled
E0927 21:46:41.995375 1 status.go:71] apiserver received an error that is not an metav1.Status: &errors.errorString{s:"http2: stream closed"}: http2: stream closed
E0927 21:46:41.996591 1 writers.go:130] apiserver was unable to write a fallback JSON response: http2: stream closed
E0927 21:46:41.997747 1 writers.go:117] apiserver was unable to write a JSON response: http: Handler timeout
I0927 21:46:41.998940 1 trace.go:205] Trace[1245557671]: "SerializeObject" method:GET,url:/api/v1/pods,protocol:HTTP/2.0,mediaType:application/json,encoder:{"encodeGV":"v1","encoder":"{"name":"json","pretty":"false","yaml":"false"}","name":"versioning"} (27-Sep-2024 21:46:31.959) (total time: 10039ms):
Trace[1245557671]: ---"Write call finished" writer:*gzip.Writer,size:619863,firstWrite:true 10029ms (21:46:41.994)
Trace[1245557671]: [10.03949272s] [10.03949272s] END
I0927 21:46:41.998986 1 trace.go:205] Trace[1383485295]: "List" url:/api/v1/pods,user-agent:okhttp/3.9.1,client:127.0.0.1,accept:,protocol:HTTP/2.0 (27-Sep-2024 21:46:31.955) (total time: 10043ms):
Trace[1383485295]: ---"Writing http response done" count:66 10039ms (21:46:41.998)
Trace[1383485295]: [10.04351861s] [10.04351861s] END

[shawkins]

Do you mean version 4.9? If so please update to a much later version.

As for the timeout exception, I can't really say at first glance what that is related to, and especially if changing the connection pool size would matter. Please reproduce whatever behavior you are seeing the latest version of the client as possible and re-open or re-create an issue.

[happyzzz1997]

Okay, I will upgrade to the latest version . Thank you for your support.