Tasks

Step-by-step instructions for performing operations with Kubernetes.

Edit This Page

Debug Services

An issue that comes up rather frequently for new installations of Kubernetes is that Services are not working properly. You’ve run all your Pods and Deployments, but you get no response when you try to access them. This document will hopefully help you to figure out what’s going wrong.

Conventions

Throughout this doc you will see various commands that you can run. Some commands need to be run within a Pod, others on a Kubernetes Node, and others can run anywhere you have kubectl and credentials for the cluster. To make it clear what is expected, this document will use the following conventions.

If the command “COMMAND” is expected to run in a Pod and produce “OUTPUT”:

u@pod$ COMMAND
OUTPUT

If the command “COMMAND” is expected to run on a Node and produce “OUTPUT”:

u@node$ COMMAND
OUTPUT

If the command is “kubectl ARGS”:

$ kubectl ARGS
OUTPUT

Running commands in a Pod

For many steps here you will want to see what a Pod running in the cluster sees. You can start a busybox Pod and run commands in it:

$ kubectl run -i --tty busybox --image=busybox --generator="run-pod/v1"
Waiting for pod default/busybox to be running, status is Pending, pod ready: false

Hit enter for command prompt

/ #

If you already have a running Pod, run a command in it using:

$ kubectl exec <POD-NAME> -c <CONTAINER-NAME> -- <COMMAND>

or run an interactive shell with:

$ kubectl exec -ti <POD-NAME> -c <CONTAINER-NAME> sh
/ #

Setup

For the purposes of this walk-through, let’s run some Pods. Since you’re probably debugging your own Service you can substitute your own details, or you can follow along and get a second data point.

$ kubectl run hostnames --image=gcr.io/google_containers/serve_hostname \
                        --labels=app=hostnames \
                        --port=9376 \
                        --replicas=3
deployment "hostnames" created

kubectl commands will print the type and name of the resource created or mutated, which can then be used in subsequent commands. Note that this is the same as if you had started the Deployment with the following YAML:

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: hostnames
spec:
  selector:
    app: hostnames
  replicas: 3
  template:
    metadata:
      labels:
        app: hostnames
    spec:
      containers:
      - name: hostnames
        image: gcr.io/google_containers/serve_hostname
        ports:
        - containerPort: 9376
          protocol: TCP

Confirm your Pods are running:

$ kubectl get pods -l app=hostnames
NAME                        READY     STATUS    RESTARTS   AGE
hostnames-632524106-bbpiw   1/1       Running   0          2m
hostnames-632524106-ly40y   1/1       Running   0          2m
hostnames-632524106-tlaok   1/1       Running   0          2m

Does the Service exist?

The astute reader will have noticed that we did not actually create a Service yet - that is intentional. This is a step that sometimes gets forgotten, and is the first thing to check.

So what would happen if I tried to access a non-existent Service? Assuming you have another Pod that consumes this Service by name you would get something like:

u@pod$ wget -qO- hostnames
wget: bad address 'hostname'

or:

u@pod$ echo $HOSTNAMES_SERVICE_HOST

So the first thing to check is whether that Service actually exists:

$ kubectl get svc hostnames
Error from server (NotFound): services "hostnames" not found

So we have a culprit, let’s create the Service. As before, this is for the walk-through - you can use your own Service’s details here.

$ kubectl expose deployment hostnames --port=80 --target-port=9376
service "hostnames" exposed

And read it back, just to be sure:

$ kubectl get svc hostnames
NAME        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
hostnames   10.0.0.226   <none>        80/TCP    5s

As before, this is the same as if you had started the Service with YAML:

apiVersion: v1
kind: Service
metadata:
  name: hostnames
spec:
  selector:
    app: hostnames
  ports:
  - name: default
    protocol: TCP
    port: 80
    targetPort: 9376

Now you can confirm that the Service exists.

Does the Service work by DNS?

From a Pod in the same Namespace:

u@pod$ nslookup hostnames
Server:         10.0.0.10
Address:        10.0.0.10#53

Name:   hostnames
Address: 10.0.1.175

If this fails, perhaps your Pod and Service are in different Namespaces, try a namespace-qualified name:

u@pod$ nslookup hostnames.default
Server:         10.0.0.10
Address:        10.0.0.10#53

Name:   hostnames.default
Address: 10.0.1.175

If this works, you’ll need to ensure that Pods and Services run in the same Namespace. If this still fails, try a fully-qualified name:

u@pod$ nslookup hostnames.default.svc.cluster.local
Server:         10.0.0.10
Address:        10.0.0.10#53

Name:   hostnames.default.svc.cluster.local
Address: 10.0.1.175

Note the suffix here: “default.svc.cluster.local”. The “default” is the Namespace we’re operating in. The “svc” denotes that this is a Service. The “cluster.local” is your cluster domain.

You can also try this from a Node in the cluster (note: 10.0.0.10 is my DNS Service):

u@node$ nslookup hostnames.default.svc.cluster.local 10.0.0.10
Server:         10.0.0.10
Address:        10.0.0.10#53

Name:   hostnames.default.svc.cluster.local
Address: 10.0.1.175

If you are able to do a fully-qualified name lookup but not a relative one, you need to check that your kubelet is running with the right flags. The --cluster-dns flag needs to point to your DNS Service’s IP and the --cluster-domain flag needs to be your cluster’s domain - we assumed “cluster.local” in this document, but yours might be different, in which case you should change that in all of the commands above.

Does any Service exist in DNS?

If the above still fails - DNS lookups are not working for your Service - we can take a step back and see what else is not working. The Kubernetes master Service should always work:

u@pod$ nslookup kubernetes.default
Server:    10.0.0.10
Address 1: 10.0.0.10

Name:      kubernetes
Address 1: 10.0.0.1

If this fails, you might need to go to the kube-proxy section of this doc, or even go back to the top of this document and start over, but instead of debugging your own Service, debug DNS.

Does the Service work by IP?

The next thing to test is whether your Service works at all. From a Node in your cluster, access the Service’s IP (from kubectl get above).

u@node$ curl 10.0.1.175:80
hostnames-0uton

u@node$ curl 10.0.1.175:80
hostnames-yp2kp

u@node$ curl 10.0.1.175:80
hostnames-bvc05

If your Service is working, you should get correct responses. If not, there are a number of things that could be going wrong. Read on.

Is the Service correct?

It might sound silly, but you should really double and triple check that your Service is correct and matches your Pods. Read back your Service and verify it:

$ kubectl get service hostnames -o json
{
    "kind": "Service",
    "apiVersion": "v1",
    "metadata": {
        "name": "hostnames",
        "namespace": "default",
        "selfLink": "/api/v1/namespaces/default/services/hostnames",
        "uid": "428c8b6c-24bc-11e5-936d-42010af0a9bc",
        "resourceVersion": "347189",
        "creationTimestamp": "2015-07-07T15:24:29Z",
        "labels": {
            "app": "hostnames"
        }
    },
    "spec": {
        "ports": [
            {
                "name": "default",
                "protocol": "TCP",
                "port": 80,
                "targetPort": 9376,
                "nodePort": 0
            }
        ],
        "selector": {
            "app": "hostnames"
        },
        "clusterIP": "10.0.1.175",
        "type": "ClusterIP",
        "sessionAffinity": "None"
    },
    "status": {
        "loadBalancer": {}
    }
}

Is the port you are trying to access in spec.ports[]? Is the targetPort correct for your Pods? If you meant it to be a numeric port, is it a number (9376) or a string “9376”? If you meant it to be a named port, do your Pods expose a port with the same name? Is the port’s protocol the same as the Pod’s?

Does the Service have any Endpoints?

If you got this far, we assume that you have confirmed that your Service exists and is resolved by DNS. Now let’s check that the Pods you ran are actually being selected by the Service.

Earlier we saw that the Pods were running. We can re-check that:

$ kubectl get pods -l app=hostnames
NAME              READY     STATUS    RESTARTS   AGE
hostnames-0uton   1/1       Running   0          1h
hostnames-bvc05   1/1       Running   0          1h
hostnames-yp2kp   1/1       Running   0          1h

The “AGE” column says that these Pods are about an hour old, which implies that they are running fine and not crashing.

The -l app=hostnames argument is a label selector - just like our Service has. Inside the Kubernetes system is a control loop which evaluates the selector of every Service and saves the results into an Endpoints object.

$ kubectl get endpoints hostnames
NAME        ENDPOINTS
hostnames   10.244.0.5:9376,10.244.0.6:9376,10.244.0.7:9376

This confirms that the control loop has found the correct Pods for your Service. If the hostnames row is blank, you should check that the spec.selector field of your Service actually selects for metadata.labels values on your Pods.

Are the Pods working?

At this point, we know that your Service exists and has selected your Pods. Let’s check that the Pods are actually working - we can bypass the Service mechanism and go straight to the Pods.

u@pod$ wget -qO- 10.244.0.5:9376
hostnames-0uton

pod $ wget -qO- 10.244.0.6:9376
hostnames-bvc05

u@pod$ wget -qO- 10.244.0.7:9376
hostnames-yp2kp

We expect each Pod in the Endpoints list to return its own hostname. If this is not what happens (or whatever the correct behavior is for your own Pods), you should investigate what’s happening there. You might find kubectl logs to be useful or kubectl exec directly to your Pods and check service from there.

Is the kube-proxy working?

If you get here, your Service is running, has Endpoints, and your Pods are actually serving. At this point, the whole Service proxy mechanism is suspect. Let’s confirm it, piece by piece.

Is kube-proxy running?

Confirm that kube-proxy is running on your Nodes. You should get something like the below:

u@node$ ps auxw | grep kube-proxy
root  4194  0.4  0.1 101864 17696 ?    Sl Jul04  25:43 /usr/local/bin/kube-proxy --master=https://kubernetes-master --kubeconfig=/var/lib/kube-proxy/kubeconfig --v=2

Next, confirm that it is not failing something obvious, like contacting the master. To do this, you’ll have to look at the logs. Accessing the logs depends on your Node OS. On some OSes it is a file, such as /var/log/kube-proxy.log, while other OSes use journalctl to access logs. You should see something like:

I1027 22:14:53.995134    5063 server.go:200] Running in resource-only container "/kube-proxy"
I1027 22:14:53.998163    5063 server.go:247] Using iptables Proxier.
I1027 22:14:53.999055    5063 server.go:255] Tearing down userspace rules. Errors here are acceptable.
I1027 22:14:54.038140    5063 proxier.go:352] Setting endpoints for "kube-system/kube-dns:dns-tcp" to [10.244.1.3:53]
I1027 22:14:54.038164    5063 proxier.go:352] Setting endpoints for "kube-system/kube-dns:dns" to [10.244.1.3:53]
I1027 22:14:54.038209    5063 proxier.go:352] Setting endpoints for "default/kubernetes:https" to [10.240.0.2:443]
I1027 22:14:54.038238    5063 proxier.go:429] Not syncing iptables until Services and Endpoints have been received from master
I1027 22:14:54.040048    5063 proxier.go:294] Adding new service "default/kubernetes:https" at 10.0.0.1:443/TCP
I1027 22:14:54.040154    5063 proxier.go:294] Adding new service "kube-system/kube-dns:dns" at 10.0.0.10:53/UDP
I1027 22:14:54.040223    5063 proxier.go:294] Adding new service "kube-system/kube-dns:dns-tcp" at 10.0.0.10:53/TCP

If you see error messages about not being able to contact the master, you should double-check your Node configuration and installation steps.

Is kube-proxy writing iptables rules?

One of the main responsibilities of kube-proxy is to write the iptables rules which implement Services. Let’s check that those rules are getting written.

The kube-proxy can run in either “userspace” mode or “iptables” mode. Hopefully you are using the newer, faster, more stable “iptables” mode. You should see one of the following cases.

Userspace

u@node$ iptables-save | grep hostnames
-A KUBE-PORTALS-CONTAINER -d 10.0.1.175/32 -p tcp -m comment --comment "default/hostnames:default" -m tcp --dport 80 -j REDIRECT --to-ports 48577
-A KUBE-PORTALS-HOST -d 10.0.1.175/32 -p tcp -m comment --comment "default/hostnames:default" -m tcp --dport 80 -j DNAT --to-destination 10.240.115.247:48577

There should be 2 rules for each port on your Service (just one in this example) - a “KUBE-PORTALS-CONTAINER” and a “KUBE-PORTALS-HOST”. If you do not see these, try restarting kube-proxy with the -V flag set to 4, and then look at the logs again.

Iptables

u@node$ iptables-save | grep hostnames
-A KUBE-SEP-57KPRZ3JQVENLNBR -s 10.244.3.6/32 -m comment --comment "default/hostnames:" -j MARK --set-xmark 0x00004000/0x00004000
-A KUBE-SEP-57KPRZ3JQVENLNBR -p tcp -m comment --comment "default/hostnames:" -m tcp -j DNAT --to-destination 10.244.3.6:9376
-A KUBE-SEP-WNBA2IHDGP2BOBGZ -s 10.244.1.7/32 -m comment --comment "default/hostnames:" -j MARK --set-xmark 0x00004000/0x00004000
-A KUBE-SEP-WNBA2IHDGP2BOBGZ -p tcp -m comment --comment "default/hostnames:" -m tcp -j DNAT --to-destination 10.244.1.7:9376
-A KUBE-SEP-X3P2623AGDH6CDF3 -s 10.244.2.3/32 -m comment --comment "default/hostnames:" -j MARK --set-xmark 0x00004000/0x00004000
-A KUBE-SEP-X3P2623AGDH6CDF3 -p tcp -m comment --comment "default/hostnames:" -m tcp -j DNAT --to-destination 10.244.2.3:9376
-A KUBE-SERVICES -d 10.0.1.175/32 -p tcp -m comment --comment "default/hostnames: cluster IP" -m tcp --dport 80 -j KUBE-SVC-NWV5X2332I4OT4T3
-A KUBE-SVC-NWV5X2332I4OT4T3 -m comment --comment "default/hostnames:" -m statistic --mode random --probability 0.33332999982 -j KUBE-SEP-WNBA2IHDGP2BOBGZ
-A KUBE-SVC-NWV5X2332I4OT4T3 -m comment --comment "default/hostnames:" -m statistic --mode random --probability 0.50000000000 -j KUBE-SEP-X3P2623AGDH6CDF3
-A KUBE-SVC-NWV5X2332I4OT4T3 -m comment --comment "default/hostnames:" -j KUBE-SEP-57KPRZ3JQVENLNBR

There should be 1 rule in KUBE-SERVICES, 1 or 2 rules per endpoint in KUBE-SVC-(hash) (depending on SessionAffinity), one KUBE-SEP-(hash) chain per endpoint, and a few rules in each KUBE-SEP-(hash) chain. The exact rules will vary based on your exact config (including node-ports and load-balancers).

Is kube-proxy proxying?

Assuming you do see the above rules, try again to access your Service by IP:

u@node$ curl 10.0.1.175:80
hostnames-0uton

If this fails and you are using the userspace proxy, you can try accessing the proxy directly. If you are using the iptables proxy, skip this section.

Look back at the iptables-save output above, and extract the port number that kube-proxy is using for your Service. In the above examples it is “48577”. Now connect to that:

u@node$ curl localhost:48577
hostnames-yp2kp

If this still fails, look at the kube-proxy logs for specific lines like:

Setting endpoints for default/hostnames:default to [10.244.0.5:9376 10.244.0.6:9376 10.244.0.7:9376]

If you don’t see those, try restarting kube-proxy with the -V flag set to 4, and then look at the logs again.

Services provide load balancing across a set of pods. There are several common problems that can make services not work properly. The following instructions should help debug service problems.

First, verify that there are endpoints for the service. For every service object, the apiserver makes an endpoints resource available.

You can view this resource with:

$ kubectl get endpoints ${SERVICE_NAME}

Make sure that the endpoints match up with the number of containers that you expect to be a member of your service. For example, if your service is for an nginx container with 3 replicas, you would expect to see three different IP addresses in the service’s endpoints.

My service is missing endpoints

If you are missing endpoints, try listing pods using the labels that service uses. Imagine that you have a service where the labels are:

...
spec:
  - selector:
     name: nginx
     type: frontend

You can use:

$ kubectl get pods --selector=name=nginx,type=frontend

to list pods that match this selector. Verify that the list matches the pods that you expect to provide your service.

If the list of pods matches expectations, but your endpoints are still empty, it’s possible that you don’t have the right ports exposed. If your service has a containerPort specified, but the pods that are selected don’t have that port listed, then they won’t be added to the endpoints list.

Verify that the pod’s containerPort matches up with the service’s containerPort.

Network traffic is not forwarded

If you can connect to the service, but the connection is immediately dropped, and there are endpoints in the endpoints list, it’s likely that the proxy can’t contact your pods.

There are three things to check:

A Pod cannot reach itself via Service IP

This mostly happens when kube-proxy is running in iptables mode and Pods are connected with bridge network. The Kubelet exposes a hairpin-mode flag that allows endpoints of a Service to loadbalance back to themselves if they try to access their own Service VIP. The hairpin-mode flag must either be set to hairpin-veth or promiscuous-bridge.

The common steps to trouble shoot this are as follows:

u@node$ ps auxw|grep kubelet
root      3392  1.1  0.8 186804 65208 ?        Sl   00:51  11:11 /usr/local/bin/kubelet --enable-debugging-handlers=true --config=/etc/kubernetes/manifests --allow-privileged=True --v=4 --cluster-dns=10.0.0.10 --cluster-domain=cluster.local --configure-cbr0=true --cgroup-root=/ --system-cgroups=/system --hairpin-mode=promiscuous-bridge --runtime-cgroups=/docker-daemon --kubelet-cgroups=/kubelet --babysit-daemons=true --max-pods=110 --serialize-image-pulls=false --outofdisk-transition-frequency=0

I0629 00:51:43.648698    3252 kubelet.go:380] Hairpin mode set to "promiscuous-bridge"
u@node$ for intf in /sys/devices/virtual/net/cbr0/brif/*; do cat $intf/hairpin_mode; done
1
1
1
1
u@node$ ifconfig cbr0 |grep PROMISC
UP BROADCAST RUNNING PROMISC MULTICAST  MTU:1460  Metric:1

Seek help

If you get this far, something very strange is happening. Your Service is running, has Endpoints, and your Pods are actually serving. You have DNS working, iptables rules installed, and kube-proxy does not seem to be misbehaving. And yet your Service is not working. You should probably let us know, so we can help investigate!

Contact us on Slack or email or GitHub.

More information

Visit troubleshooting document for more information.

Analytics

Create an Issue Edit this Page