kubernetes: dial tcp i/o timeout error when getting logs of pod

12/4/2015

I'm working on OpenShift Origin 1.1 (which is using kubernetes as its orchestration tool for docker containers). I'm creating pods, but I'm unable to see the build-logs.

[user@ip master]# oc get pods
NAME           READY     STATUS      RESTARTS   AGE
test-1-build   0/1       Completed   0          14m
test-1-iok8n   1/1       Running     0          12m
[user@ip master]# oc logs test-1-iok8n
Error from server: Get https://ip-10-0-x-x.compute.internal:10250/containerLogs/test/test-1-iok8n/test: dial tcp 10.0.x.x:10250: i/o timeout

my /var/logs/messages shows:

Dec  4 13:28:24 ip-10-0-x-x origin-master: E1204 13:28:24.579794   32518 apiserver.go:440] apiserver was unable to write a JSON response: Get https://ip-10-0-x-x.compute.internal:10250/containerLogs/test/test-1-iok8n/test: dial tcp 10.0.x.x:10250: i/o timeout
Dec  4 13:28:24 ip-10-0-x-x origin-master: E1204 13:28:24.579822   32518 errors.go:62] apiserver received an error that is not an unversioned.Status: Get https://ip-10-0-x-x.compute.internal:10250/containerLogs/test/test-1-iok8n/test: dial tcp 10.0.x.x:10250: i/o timeout

My versions are:

origin v1.1.0.1-1-g2c6ff4b
kubernetes v1.1.0-origin-1107-g4c8e6f4
etcd 2.1.2
-- lvthillo
kubernetes
openshift-origin
timeout

1 Answer

12/4/2015

I forgot to open port 10250 (tcp) (in my aws security group). This was the only issue for me.

-- lvthillo
Source: StackOverflow