Networking Reference

• 1: Protocols for Services
• 2: Ports and Protocols
• 3: Virtual IPs and Service Proxies

1 - Protocols for Services

If you configure a Service, you can select from any network protocol that Kubernetes supports.

Kubernetes supports the following protocols with Services:

• SCTP
• TCP (the default)
• UDP

When you define a Service, you can also specify the application protocol that it uses.

This document details some special cases, all of them typically using TCP as a transport protocol:

• HTTP and HTTPS
• PROXY protocol
• TLS termination at the load balancer

Supported protocols

There are 3 valid values for the protocol of a port for a Service:

SCTP

When using a network plugin that supports SCTP traffic, you can use SCTP for most Services. For type: LoadBalancer Services, SCTP support depends on the cloud provider offering this facility. (Most do not).

SCTP is not supported on nodes that run Windows.

Support for multihomed SCTP associations

The support of multihomed SCTP associations requires that the CNI plugin can support the assignment of multiple interfaces and IP addresses to a Pod.

NAT for multihomed SCTP associations requires special logic in the corresponding kernel modules.

TCP

You can use TCP for any kind of Service, and it's the default network protocol.

UDP

You can use UDP for most Services. For type: LoadBalancer Services, UDP support depends on the cloud provider offering this facility.

Special cases

HTTP

If your cloud provider supports it, you can use a Service in LoadBalancer mode to configure a load balancer outside of your Kubernetes cluster, in a special mode where your cloud provider's load balancer implements HTTP / HTTPS reverse proxying, with traffic forwarded to the backend endpoints for that Service.

Typically, you set the protocol for the Service to TCP and add an annotation (usually specific to your cloud provider) that configures the load balancer to handle traffic at the HTTP level. This configuration might also include serving HTTPS (HTTP over TLS) and reverse-proxying plain HTTP to your workload.

You might additionally want to specify that the application protocol of the connection is http or https. Use http if the session from the load balancer to your workload is HTTP without TLS, and use https if the session from the load balancer to your workload uses TLS encryption.

PROXY protocol

If your cloud provider supports it, you can use a Service set to type: LoadBalancer to configure a load balancer outside of Kubernetes itself, that will forward connections wrapped with the PROXY protocol.

The load balancer then sends an initial series of octets describing the incoming connection, similar to this example (PROXY protocol v1):

PROXY TCP4 192.0.2.202 10.0.42.7 12345 7\r\n

The data after the proxy protocol preamble are the original data from the client. When either side closes the connection, the load balancer also triggers a connection close and sends any remaining data where feasible.

Typically, you define a Service with the protocol to TCP. You also set an annotation, specific to your cloud provider, that configures the load balancer to wrap each incoming connection in the PROXY protocol.

TLS

If your cloud provider supports it, you can use a Service set to type: LoadBalancer as a way to set up external reverse proxying, where the connection from client to load balancer is TLS encrypted and the load balancer is the TLS server peer. The connection from the load balancer to your workload can also be TLS, or might be plain text. The exact options available to you depend on your cloud provider or custom Service implementation.

Typically, you set the protocol to TCP and set an annotation (usually specific to your cloud provider) that configures the load balancer to act as a TLS server. You would configure the TLS identity (as server, and possibly also as a client that connects to your workload) using mechanisms that are specific to your cloud provider.

2 - Ports and Protocols

When running Kubernetes in an environment with strict network boundaries, such as on-premises datacenter with physical network firewalls or Virtual Networks in Public Cloud, it is useful to be aware of the ports and protocols used by Kubernetes components.

Control plane

Although etcd ports are included in control plane section, you can also host your own etcd cluster externally or on custom ports.

Worker node(s)

† Default port range for NodePort Services.

All default port numbers can be overridden. When custom ports are used those ports need to be open instead of defaults mentioned here.

One common example is API server port that is sometimes switched to 443. Alternatively, the default port is kept as is and API server is put behind a load balancer that listens on 443 and routes the requests to API server on the default port.

3 - Virtual IPs and Service Proxies

Every node in a Kubernetes cluster runs a kube-proxy (unless you have deployed your own alternative component in place of kube-proxy).

The kube-proxy component is responsible for implementing a virtual IP mechanism for Services of type other than ExternalName. Each instance of kube-proxy watches the Kubernetes control plane for the addition and removal of Service and EndpointSlice objects. For each Service, kube-proxy calls appropriate APIs (depending on the kube-proxy mode) to configure the node to capture traffic to the Service's clusterIP and port, and redirect that traffic to one of the Service's endpoints (usually a Pod, but possibly an arbitrary user-provided IP address). A control loop ensures that the rules on each node are reliably synchronized with the Service and EndpointSlice state as indicated by the API server.

A question that pops up every now and then is why Kubernetes relies on proxying to forward inbound traffic to backends. What about other approaches? For example, would it be possible to configure DNS records that have multiple A values (or AAAA for IPv6), and rely on round-robin name resolution?

There are a few reasons for using proxying for Services:

• There is a long history of DNS implementations not respecting record TTLs, and caching the results of name lookups after they should have expired.
• Some apps do DNS lookups only once and cache the results indefinitely.
• Even if apps and libraries did proper re-resolution, the low or zero TTLs on the DNS records could impose a high load on DNS that then becomes difficult to manage.

Later in this page you can read about how various kube-proxy implementations work. Overall, you should note that, when running kube-proxy, kernel level rules may be modified (for example, iptables rules might get created), which won't get cleaned up, in some cases until you reboot. Thus, running kube-proxy is something that should only be done by an administrator which understands the consequences of having a low level, privileged network proxying service on a computer. Although the kube-proxy executable supports a cleanup function, this function is not an official feature and thus is only available to use as-is.

Some of the details in this reference refer to an example: the backend Pods for a stateless image-processing workloads, running with three replicas. Those replicas are fungible—frontends do not care which backend they use. While the actual Pods that compose the backend set may change, the frontend clients should not need to be aware of that, nor should they need to keep track of the set of backends themselves.

Proxy modes

The kube-proxy starts up in different modes, which are determined by its configuration.

On Linux nodes, the available modes for kube-proxy are:

iptables

A mode where the kube-proxy configures packet forwarding rules using iptables.

ipvs

a mode where the kube-proxy configures packet forwarding rules using ipvs.

nftables

a mode where the kube-proxy configures packet forwarding rules using nftables.

There is only one mode available for kube-proxy on Windows:

kernelspace

a mode where the kube-proxy configures packet forwarding rules in the Windows kernel

iptables proxy mode

This proxy mode is only available on Linux nodes.

In this mode, kube-proxy configures packet forwarding rules using the iptables API of the kernel netfilter subsystem. For each endpoint, it installs iptables rules which, by default, select a backend Pod at random.

Example

As an example, consider the image processing application described earlier in the page. When the backend Service is created, the Kubernetes control plane assigns a virtual IP address, for example 10.0.0.1. For this example, assume that the Service port is 1234. All of the kube-proxy instances in the cluster observe the creation of the new Service.

When kube-proxy on a node sees a new Service, it installs a series of iptables rules which redirect from the virtual IP address to more iptables rules, defined per Service. The per-Service rules link to further rules for each backend endpoint, and the per- endpoint rules redirect traffic (using destination NAT) to the backends.

When a client connects to the Service's virtual IP address the iptables rule kicks in. A backend is chosen (either based on session affinity or randomly) and packets are redirected to the backend without rewriting the client IP address.

This same basic flow executes when traffic comes in through a node-port or through a load-balancer, though in those cases the client IP address does get altered.

Optimizing iptables mode performance

In iptables mode, kube-proxy creates a few iptables rules for every Service, and a few iptables rules for each endpoint IP address. In clusters with tens of thousands of Pods and Services, this means tens of thousands of iptables rules, and kube-proxy may take a long time to update the rules in the kernel when Services (or their EndpointSlices) change. You can adjust the syncing behavior of kube-proxy via options in the iptables section of the kube-proxy configuration file (which you specify via kube-proxy --config <path>):

...
iptables:
  minSyncPeriod: 1s
  syncPeriod: 30s
...

minSyncPeriod

The minSyncPeriod parameter sets the minimum duration between attempts to resynchronize iptables rules with the kernel. If it is 0s, then kube-proxy will always immediately synchronize the rules every time any Service or Endpoint changes. This works fine in very small clusters, but it results in a lot of redundant work when lots of things change in a small time period. For example, if you have a Service backed by a Deployment with 100 pods, and you delete the Deployment, then with minSyncPeriod: 0s, kube-proxy would end up removing the Service's endpoints from the iptables rules one by one, for a total of 100 updates. With a larger minSyncPeriod, multiple Pod deletion events would get aggregated together, so kube-proxy might instead end up making, say, 5 updates, each removing 20 endpoints, which will be much more efficient in terms of CPU, and result in the full set of changes being synchronized faster.

The larger the value of minSyncPeriod, the more work that can be aggregated, but the downside is that each individual change may end up waiting up to the full minSyncPeriod before being processed, meaning that the iptables rules spend more time being out-of-sync with the current API server state.

The default value of 1s should work well in most clusters, but in very large clusters it may be necessary to set it to a larger value. Especially, if kube-proxy's sync_proxy_rules_duration_seconds metric indicates an average time much larger than 1 second, then bumping up minSyncPeriod may make updates more efficient.

Updating legacy minSyncPeriod configuration

Older versions of kube-proxy updated all the rules for all Services on every sync; this led to performance issues (update lag) in large clusters, and the recommended solution was to set a larger minSyncPeriod. Since Kubernetes v1.28, the iptables mode of kube-proxy uses a more minimal approach, only making updates where Services or EndpointSlices have actually changed.

If you were previously overriding minSyncPeriod, you should try removing that override and letting kube-proxy use the default value (1s) or at least a smaller value than you were using before upgrading.

If you are not running kube-proxy from Kubernetes 1.32, check the behavior and associated advice for the version that you are actually running.

syncPeriod

The syncPeriod parameter controls a handful of synchronization operations that are not directly related to changes in individual Services and EndpointSlices. In particular, it controls how quickly kube-proxy notices if an external component has interfered with kube-proxy's iptables rules. In large clusters, kube-proxy also only performs certain cleanup operations once every syncPeriod to avoid unnecessary work.

For the most part, increasing syncPeriod is not expected to have much impact on performance, but in the past, it was sometimes useful to set it to a very large value (eg, 1h). This is no longer recommended, and is likely to hurt functionality more than it improves performance.

IPVS proxy mode

This proxy mode is only available on Linux nodes.

In ipvs mode, kube-proxy uses the kernel IPVS and iptables APIs to create rules to redirect traffic from Service IPs to endpoint IPs.

The IPVS proxy mode is based on netfilter hook function that is similar to iptables mode, but uses a hash table as the underlying data structure and works in the kernel space. That means kube-proxy in IPVS mode redirects traffic with lower latency than kube-proxy in iptables mode, with much better performance when synchronizing proxy rules. Compared to the iptables proxy mode, IPVS mode also supports a higher throughput of network traffic.

IPVS provides more options for balancing traffic to backend Pods; these are:

• rr (Round Robin): Traffic is equally distributed amongst the backing servers.

• wrr (Weighted Round Robin): Traffic is routed to the backing servers based on the weights of the servers. Servers with higher weights receive new connections and get more requests than servers with lower weights.

• lc (Least Connection): More traffic is assigned to servers with fewer active connections.

• wlc (Weighted Least Connection): More traffic is routed to servers with fewer connections relative to their weights, that is, connections divided by weight.

• lblc (Locality based Least Connection): Traffic for the same IP address is sent to the same backing server if the server is not overloaded and available; otherwise the traffic is sent to servers with fewer connections, and keep it for future assignment.

• lblcr (Locality Based Least Connection with Replication): Traffic for the same IP address is sent to the server with least connections. If all the backing servers are overloaded, it picks up one with fewer connections and add it to the target set. If the target set has not changed for the specified time, the most loaded server is removed from the set, in order to avoid high degree of replication.

• sh (Source Hashing): Traffic is sent to a backing server by looking up a statically assigned hash table based on the source IP addresses.

• dh (Destination Hashing): Traffic is sent to a backing server by looking up a statically assigned hash table based on their destination addresses.

• sed (Shortest Expected Delay): Traffic forwarded to a backing server with the shortest expected delay. The expected delay is (C + 1) / U if sent to a server, where C is the number of connections on the server and U is the fixed service rate (weight) of the server.

• nq (Never Queue): Traffic is sent to an idle server if there is one, instead of waiting for a fast one; if all servers are busy, the algorithm falls back to the sed behavior.

• mh (Maglev Hashing): Assigns incoming jobs based on Google's Maglev hashing algorithm, This scheduler has two flags: mh-fallback, which enables fallback to a different server if the selected server is unavailable, and mh-port, which adds the source port number to the hash computation. When using mh, kube-proxy always sets the mh-port flag and does not enable the mh-fallback flag. In proxy-mode=ipvs mh will work as source-hashing (sh), but with ports.

These scheduling algorithms are configured through the ipvs.scheduler field in the kube-proxy configuration.

nftables proxy mode

This proxy mode is only available on Linux nodes, and requires kernel 5.13 or later.

In this mode, kube-proxy configures packet forwarding rules using the nftables API of the kernel netfilter subsystem. For each endpoint, it installs nftables rules which, by default, select a backend Pod at random.

The nftables API is the successor to the iptables API and is designed to provide better performance and scalability than iptables. The nftables proxy mode is able to process changes to service endpoints faster and more efficiently than the iptables mode, and is also able to more efficiently process packets in the kernel (though this only becomes noticeable in clusters with tens of thousands of services).

As of Kubernetes 1.32, the nftables mode is still relatively new, and may not be compatible with all network plugins; consult the documentation for your network plugin.

Migrating from iptables mode to nftables

Users who want to switch from the default iptables mode to the nftables mode should be aware that some features work slightly differently the nftables mode:

• NodePort interfaces: In iptables mode, by default, NodePort services are reachable on all local IP addresses. This is usually not what users want, so the nftables mode defaults to --nodeport-addresses primary, meaning NodePort services are only reachable on the node's primary IPv4 and/or IPv6 addresses. You can override this by specifying an explicit value for that option: e.g., --nodeport-addresses 0.0.0.0/0 to listen on all (local) IPv4 IPs.

• NodePort services on 127.0.0.1: In iptables mode, if the --nodeport-addresses range includes 127.0.0.1 (and the option --iptables-localhost-nodeports false option is not passed), then NodePort services are reachable even on "localhost" (127.0.0.1). In nftables mode (and ipvs mode), this will not work. If you are not sure if you are depending on this functionality, you can check kube-proxy's iptables_localhost_nodeports_accepted_packets_total metric; if it is non-0, that means that some client has connected to a NodePort service via 127.0.0.1.

• NodePort interaction with firewalls: The iptables mode of kube-proxy tries to be compatible with overly-agressive firewalls; for each NodePort service, it will add rules to accept inbound traffic on that port, in case that traffic would otherwise be blocked by a firewall. This approach will not work with firewalls based on nftables, so kube-proxy's nftables mode does not do anything here; if you have a local firewall, you must ensure that it is properly configured to allow Kubernetes traffic through (e.g., by allowing inbound traffic on the entire NodePort range).

• Conntrack bug workarounds: Linux kernels prior to 6.1 have a bug that can result in long-lived TCP connections to service IPs being closed with the error "Connection reset by peer". The iptables mode of kube-proxy installs a workaround for this bug, but this workaround was later found to cause other problems in some clusters. The nftables mode does not install any workaround by default, but you can check kube-proxy's iptables_ct_state_invalid_dropped_packets_total metric to see if your cluster is depending on the workaround, and if so, you can run kube-proxy with the option --conntrack-tcp-be-liberal to work around the problem in nftables mode.

kernelspace proxy mode

This proxy mode is only available on Windows nodes.

The kube-proxy configures packet filtering rules in the Windows Virtual Filtering Platform (VFP), an extension to Windows vSwitch. These rules process encapsulated packets within the node-level virtual networks, and rewrite packets so that the destination IP address (and layer 2 information) is correct for getting the packet routed to the correct destination. The Windows VFP is analogous to tools such as Linux nftables or iptables. The Windows VFP extends the Hyper-V Switch, which was initially implemented to support virtual machine networking.

When a Pod on a node sends traffic to a virtual IP address, and the kube-proxy selects a Pod on a different node as the load balancing target, the kernelspace proxy mode rewrites that packet to be destined to the target backend Pod. The Windows Host Networking Service (HNS) ensures that packet rewriting rules are configured so that the return traffic appears to come from the virtual IP address and not the specific backend Pod.

Direct server return for kernelspace mode

As an alternative to the basic operation, a node that hosts the backend Pod for a Service can apply the packet rewriting directly, rather than placing this burden on the node where the client Pod is running. This is called direct server return.

To use this, you must run kube-proxy with the --enable-dsr command line argument and enable the WinDSR feature gate.

Direct server return also optimizes the case for Pod return traffic even when both Pods are running on the same node.

Session affinity

In these proxy models, the traffic bound for the Service's IP:Port is proxied to an appropriate backend without the clients knowing anything about Kubernetes or Services or Pods.

If you want to make sure that connections from a particular client are passed to the same Pod each time, you can select the session affinity based on the client's IP addresses by setting .spec.sessionAffinity to ClientIP for a Service (the default is None).

Session stickiness timeout

You can also set the maximum session sticky time by setting .spec.sessionAffinityConfig.clientIP.timeoutSeconds appropriately for a Service. (the default value is 10800, which works out to be 3 hours).

IP address assignment to Services

Unlike Pod IP addresses, which actually route to a fixed destination, Service IPs are not actually answered by a single host. Instead, kube-proxy uses packet processing logic (such as Linux iptables) to define virtual IP addresses which are transparently redirected as needed.

When clients connect to the VIP, their traffic is automatically transported to an appropriate endpoint. The environment variables and DNS for Services are actually populated in terms of the Service's virtual IP address (and port).

Avoiding collisions

One of the primary philosophies of Kubernetes is that you should not be exposed to situations that could cause your actions to fail through no fault of your own. For the design of the Service resource, this means not making you choose your own IP address if that choice might collide with someone else's choice. That is an isolation failure.

In order to allow you to choose an IP address for your Services, we must ensure that no two Services can collide. Kubernetes does that by allocating each Service its own IP address from within the service-cluster-ip-range CIDR range that is configured for the API Server.

IP address allocation tracking

To ensure each Service receives a unique IP address, an internal allocator atomically updates a global allocation map in etcd prior to creating each Service. The map object must exist in the registry for Services to get IP address assignments, otherwise creations will fail with a message indicating an IP address could not be allocated.

In the control plane, a background controller is responsible for creating that map (needed to support migrating from older versions of Kubernetes that used in-memory locking). Kubernetes also uses controllers to check for invalid assignments (for example: due to administrator intervention) and for cleaning up allocated IP addresses that are no longer used by any Services.

IP address allocation tracking using the Kubernetes API

If you enable the MultiCIDRServiceAllocator feature gate and the networking.k8s.io/v1alpha1 API group, the control plane replaces the existing etcd allocator with a revised implementation that uses IPAddress and ServiceCIDR objects instead of an internal global allocation map. Each cluster IP address associated to a Service then references an IPAddress object.

Enabling the feature gate also replaces a background controller with an alternative that handles the IPAddress objects and supports migration from the old allocator model. Kubernetes 1.32 does not support migrating from IPAddress objects to the internal allocation map.

One of the main benefits of the revised allocator is that it removes the size limitations for the IP address range that can be used for the cluster IP address of Services. With MultiCIDRServiceAllocator enabled, there are no limitations for IPv4, and for IPv6 you can use IP address netmasks that are a /64 or smaller (as opposed to /108 with the legacy implementation).

Making IP address allocations available via the API means that you as a cluster administrator can allow users to inspect the IP addresses assigned to their Services. Kubernetes extensions, such as the Gateway API, can use the IPAddress API to extend Kubernetes' inherent networking capabilities.

Here is a brief example of a user querying for IP addresses:

kubectl get services

NAME         TYPE        CLUSTER-IP        EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   2001:db8:1:2::1   <none>        443/TCP   3d1h

kubectl get ipaddresses

NAME              PARENTREF
2001:db8:1:2::1   services/default/kubernetes
2001:db8:1:2::a   services/kube-system/kube-dns

Kubernetes also allow users to dynamically define the available IP ranges for Services using ServiceCIDR objects. During bootstrap, a default ServiceCIDR object named kubernetes is created from the value of the --service-cluster-ip-range command line argument to kube-apiserver:

kubectl get servicecidrs

NAME         CIDRS         AGE
kubernetes   10.96.0.0/28  17m

Users can create or delete new ServiceCIDR objects to manage the available IP ranges for Services:

cat <<'EOF' | kubectl apply -f -
apiVersion: networking.k8s.io/v1beta1
kind: ServiceCIDR
metadata:
  name: newservicecidr
spec:
  cidrs:
  - 10.96.0.0/24
EOF

servicecidr.networking.k8s.io/newcidr1 created

kubectl get servicecidrs

NAME             CIDRS         AGE
kubernetes       10.96.0.0/28  17m
newservicecidr   10.96.0.0/24  7m

IP address ranges for Service virtual IP addresses

Kubernetes divides the ClusterIP range into two bands, based on the size of the configured service-cluster-ip-range by using the following formula min(max(16, cidrSize / 16), 256). That formula paraphrases as never less than 16 or more than 256, with a graduated step function between them.

Kubernetes prefers to allocate dynamic IP addresses to Services by choosing from the upper band, which means that if you want to assign a specific IP address to a type: ClusterIP Service, you should manually assign an IP address from the lower band. That approach reduces the risk of a conflict over allocation.

Traffic policies

You can set the .spec.internalTrafficPolicy and .spec.externalTrafficPolicy fields to control how Kubernetes routes traffic to healthy (“ready”) backends.

Internal traffic policy

You can set the .spec.internalTrafficPolicy field to control how traffic from internal sources is routed. Valid values are Cluster and Local. Set the field to Cluster to route internal traffic to all ready endpoints and Local to only route to ready node-local endpoints. If the traffic policy is Local and there are no node-local endpoints, traffic is dropped by kube-proxy.

External traffic policy

You can set the .spec.externalTrafficPolicy field to control how traffic from external sources is routed. Valid values are Cluster and Local. Set the field to Cluster to route external traffic to all ready endpoints and Local to only route to ready node-local endpoints. If the traffic policy is Local and there are are no node-local endpoints, the kube-proxy does not forward any traffic for the relevant Service.

If Cluster is specified all nodes are eligible load balancing targets as long as the node is not being deleted and kube-proxy is healthy. In this mode: load balancer health checks are configured to target the service proxy's readiness port and path. In the case of kube-proxy this evaluates to: ${NODE_IP}:10256/healthz. kube-proxy will return either an HTTP code 200 or 503. kube-proxy's load balancer health check endpoint returns 200 if:

1. kube-proxy is healthy, meaning:
   • it's able to progress programming the network and isn't timing out while doing so (the timeout is defined to be: 2 × iptables.syncPeriod); and
2. the node is not being deleted (there is no deletion timestamp set for the Node).

The reason why kube-proxy returns 503 and marks the node as not eligible when it's being deleted, is because kube-proxy supports connection draining for terminating nodes. A couple of important things occur from the point of view of a Kubernetes-managed load balancer when a node is being / is deleted.

While deleting:

• kube-proxy will start failing its readiness probe and essentially mark the node as not eligible for load balancer traffic. The load balancer health check failing causes load balancers which support connection draining to allow existing connections to terminate, and block new connections from establishing.

When deleted:

• The service controller in the Kubernetes cloud controller manager removes the node from the referenced set of eligible targets. Removing any instance from the load balancer's set of backend targets immediately terminates all connections. This is also the reason kube-proxy first fails the health check while the node is deleting.

It's important to note for Kubernetes vendors that if any vendor configures the kube-proxy readiness probe as a liveness probe: that kube-proxy will start restarting continuously when a node is deleting until it has been fully deleted. kube-proxy exposes a /livez path which, as opposed to the /healthz one, does not consider the Node's deleting state and only its progress programming the network. /livez is therefore the recommended path for anyone looking to define a livenessProbe for kube-proxy.

Users deploying kube-proxy can inspect both the readiness / liveness state by evaluating the metrics: proxy_livez_total / proxy_healthz_total. Both metrics publish two series, one with the 200 label and one with the 503 one.

For Local Services: kube-proxy will return 200 if

1. kube-proxy is healthy/ready, and
2. has a local endpoint on the node in question.

Node deletion does not have an impact on kube-proxy's return code for what concerns load balancer health checks. The reason for this is: deleting nodes could end up causing an ingress outage should all endpoints simultaneously be running on said nodes.

The Kubernetes project recommends that cloud provider integration code configures load balancer health checks that target the service proxy's healthz port. If you are using or implementing your own virtual IP implementation, that people can use instead of kube-proxy, you should set up a similar health checking port with logic that matches the kube-proxy implementation.

Traffic to terminating endpoints

If the ProxyTerminatingEndpoints feature gate is enabled in kube-proxy and the traffic policy is Local, that node's kube-proxy uses a more complicated algorithm to select endpoints for a Service. With the feature enabled, kube-proxy checks if the node has local endpoints and whether or not all the local endpoints are marked as terminating. If there are local endpoints and all of them are terminating, then kube-proxy will forward traffic to those terminating endpoints. Otherwise, kube-proxy will always prefer forwarding traffic to endpoints that are not terminating.

This forwarding behavior for terminating endpoints exist to allow NodePort and LoadBalancer Services to gracefully drain connections when using externalTrafficPolicy: Local.

As a deployment goes through a rolling update, nodes backing a load balancer may transition from N to 0 replicas of that deployment. In some cases, external load balancers can send traffic to a node with 0 replicas in between health check probes. Routing traffic to terminating endpoints ensures that Node's that are scaling down Pods can gracefully receive and drain traffic to those terminating Pods. By the time the Pod completes termination, the external load balancer should have seen the node's health check failing and fully removed the node from the backend pool.

Traffic Distribution

The spec.trafficDistribution field within a Kubernetes Service allows you to express preferences for how traffic should be routed to Service endpoints. Implementations like kube-proxy use the spec.trafficDistribution field as a guideline. The behavior associated with a given preference may subtly differ between implementations.

PreferClose with kube-proxy

For kube-proxy, this means prioritizing sending traffic to endpoints within the same zone as the client. The EndpointSlice controller updates EndpointSlices with hints to communicate this preference, which kube-proxy then uses for routing decisions. If a client's zone does not have any available endpoints, traffic will be routed cluster-wide for that client.

In the absence of any value for trafficDistribution, the default routing strategy for kube-proxy is to distribute traffic to any endpoint in the cluster.

Comparison with service.kubernetes.io/topology-mode: Auto

The trafficDistribution field with PreferClose and the service.kubernetes.io/topology-mode: Auto annotation both aim to prioritize same-zone traffic. However, there are key differences in their approaches:

• service.kubernetes.io/topology-mode: Auto: Attempts to distribute traffic proportionally across zones based on allocatable CPU resources. This heuristic includes safeguards (such as the fallback behavior for small numbers of endpoints) and could lead to the feature being disabled in certain scenarios for load-balancing reasons. This approach sacrifices some predictability in favor of potential load balancing.

• trafficDistribution: PreferClose: This approach aims to be slightly simpler and more predictable: "If there are endpoints in the zone, they will receive all traffic for that zone, if there are no endpoints in a zone, the traffic will be distributed to other zones". While the approach may offer more predictability, it does mean that you are in control of managing a potential overload.

If the service.kubernetes.io/topology-mode annotation is set to Auto, it will take precedence over trafficDistribution. (The annotation may be deprecated in the future in favour of the trafficDistribution field).

Interaction with Traffic Policies

When compared to the trafficDistribution field, the traffic policy fields (externalTrafficPolicy and internalTrafficPolicy) are meant to offer a stricter traffic locality requirements. Here's how trafficDistribution interacts with them:

• Precedence of Traffic Policies: For a given Service, if a traffic policy (externalTrafficPolicy or internalTrafficPolicy) is set to Local, it takes precedence over trafficDistribution: PreferClose for the corresponding traffic type (external or internal, respectively).

• trafficDistribution Influence: For a given Service, if a traffic policy (externalTrafficPolicy or internalTrafficPolicy) is set to Cluster (the default), or if the fields are not set, then trafficDistribution: PreferClose guides the routing behavior for the corresponding traffic type (external or internal, respectively). This means that an attempt will be made to route traffic to an endpoint that is in the same zone as the client.

Considerations for using traffic distribution control

• Increased Probability of Overloaded Endpoints: The PreferClose heuristic will attempt to route traffic to the closest healthy endpoints instead of spreading that traffic evenly across all endpoints. If you do not have a sufficient number of endpoints within a zone, they may become overloaded. This is especially likely if incoming traffic is not proportionally distributed across zones. To mitigate this, consider the following strategies:

  • Pod Topology Spread Constraints: Use Pod Topology Spread Constraints to distribute your pods more evenly across zones.

  • Zone-specific Deployments: If you expect to see skewed traffic patterns, create a separate Deployment for each zone. This approach allows the separate workloads to scale independently. There are also workload management addons available from the ecosystem, outside the Kubernetes project itself, that can help here.

• Implementation-specific behavior: Each dataplane implementation may handle this field slightly differently. If you're using an implementation other than kube-proxy, refer the documentation specific to that implementation to understand how this field is being handled.

What's next

To learn more about Services, read Connecting Applications with Services.

You can also:

• Read about Services as a concept
• Read about Ingresses as a concept
• Read the API reference for the Service API