Usage

Configuration

To configure MetalLB, write a config map to metallb-system/config

There is an example configmap in manifests/example-config.yaml, annotated with explanatory comments.

For a basic configuration featuring one BGP router and one IP address range, you need 4 pieces of information:

  • The router IP address that MetalLB should connect to,
  • The router’s AS number,
  • The AS number MetalLB should use,
  • The IP address range expressed as a CIDR prefix.

As an example, if you want to give MetalLB the range 192.168.10.0/24 and AS number 42, and connect it to a router at 10.0.0.1 with AS number 100, your configuration will look like:

apiVersion: v1
kind: ConfigMap
metadata:
  namespace: metallb-system
  name: config
data:
  config: |
    peers:
    - peer-address: 10.0.0.1
      peer-asn: 100
      my-asn: 42
    address-pools:
    - name: default
      cidr:
      - 192.168.10.0/24
      advertisements:
      - aggregation-length: 32

Simple balancers

Once MetalLB is installed and configured, to expose a service externally, simply create it with spec.type set to LoadBalancer, and MetalLB will do the rest.

MetalLB attaches informational events to the services that it’s controlling. If your LoadBalancer is misbehaving, run kubectl describe service <service name> and check the event log.

Requesting specific IPs

MetalLB respects the spec.loadBalancerIP parameter, so if you want your service to be set up with a specific address, you can request it by setting that parameter. If MetalLB does not own the requested address, or if the address is already in use by another service, assignment will fail and MetalLB will log a warning event visible in kubectl describe service <service name>.

MetalLB also supports requesting a specific address pool, if you want a certain kind of address but don’t care which one exactly. To request assignment from a specific pool, add the metallb.universe.tf/address-pool annotation to your service, with the name of the address pool as the annotation value. For example:

apiVersion: v1
kind: Service
metadata:
  name: nginx
annotations:
  metallb.universe.tf/address-pool: production-public-ips
spec:
  ports:
  - port: 80
    targetPort: 80
  selector:
    app: nginx
  type: LoadBalancer

Traffic policies

MetalLB respects the service’s externalTrafficPolicy option, and implements two different announcement modes depending on what policy you select. If you’re familiar with Google Cloud’s Kubernetes load balancers, you can probably skip this section: MetalLB’s behaviors and tradeoffs are identical.

“Cluster” traffic policy

With the default Cluster traffic policy, every node in your cluster will attract traffic for the service IP. On each node, the traffic is subjected to a second layer of load-balancing (provided by kube-proxy), which directs the traffic to individual pods.

This policy results in uniform traffic distribution across all nodes in your cluster, and across all pods in your service. However, it results in two layers of load-balancing (one at the BGP router, one at kube-proxy on the nodes), which can cause inefficient traffic flows. For example, a particular user’s connection might be sent to node A by the BGP router, but then node A decides to send that connection to a pod running on node B.

The other downside of the “Cluster” policy is that kube-proxy will obscure the source IP address of the connection when it does its load-balancing, so your pod logs will show that external traffic appears to be coming from your cluster’s nodes.

“Local” traffic policy

With the Local traffic policy, nodes will only attract traffic if they are running one or more of the service’s pods locally. The BGP routers will load-balance incoming traffic only across those nodes that are currently hosting the service. On each node, the traffic is forwarded only to local pods by kube-proxy, there is no “horizontal” traffic flow between nodes.

This policy provides the most efficient flow of traffic to your service. Furthermore, because kube-proxy doesn’t need to send traffic between cluster nodes, your pods can see the real source IP address of incoming connections.

The downside of this policy is that it treats each cluster node as one “unit” of load-balancing, regardless of how many of the service’s pods are running on that node. This may result in traffic imbalances to your pods.

For example, if your service has 2 pods running on node A and one pod running on node B, the Local traffic policy will send 50% of the service’s traffic to each node. Node A will split the traffic it receives evenly between its two pods, so the final per-pod load distribution is 25% for each of node A’s pods, and 50% for node B’s pod. In contrast, if you used the Cluster traffic policy, each pod would receive 33% of the overall traffic.

In general, when using the Local traffic policy, it’s recommended to finely control the mapping of your pods to nodes, for example using node anti-affinity, so that an even traffic split across nodes translates to an even traffic split across pods.

In future, MetalLB might be able to overcome the downsides of the Local traffic policy, in which case it would be unconditionally the best mode to use in MetalLB-powered clusters. See #1 for more information.

Example

As an example of how to use all of MetalLB’s options, consider an ecommerce site that runs a production environment and multiple developer sandboxes side by side. The production environment needs public IP addresses, but the sandboxes can use private IP space, routed to the developer offices through a VPN.

Additionally, because the production IPs end up hardcoded in various places (DNS, security scans for regulatory compliance…), we want specific services to have specific addresses in production. On the other hand, sandboxes come and go as developers bring up and tear down environments, so we don’t want to manage assignments by hand.

We can translate these requirements into MetalLB. First, we define two address pools, and set BGP attributes to control the visibility of each set of addresses:

# Rest of config omitted for brevity
communities:
  # Our datacenter routers understand a "VPN only" BGP community.
  # Announcements tagged with this community will only be propagated
  # through the corporate VPN tunnel back to developer offices.
  vpn-only: 1234:1
address-pools:
- # Production services will go here. Public IPs are expensive, so we leased
  # just 4 of them.
  name: production
  cidr:
  - 42.176.25.64/30
  advertisements:
  - aggregation-length: 32

- # On the other hand, the sandbox environment uses private IP space,
  # which is free and plentiful. We give this address pool a ton of IPs,
  # so that developers can spin up as many sandboxes as they need.
  name: sandbox
  cidr:
  - 192.168.144.0/20
  advertisements:
  - communities:
    - vpn-only

In our Helm charts for sandboxes, we tag all services with the annotation metallb.universe.tf/address-pool: sandbox. Now, whenever developers spin up a sandbox, it’ll come up on some IP address within 192.168.144.0/20.

For production, we set spec.loadBalancerIP to the exact IP address that we want for each service. MetalLB will check that it makes sense given its configuration, but otherwise will do exactly as it’s told.

Limitations

MetalLB uses the BGP routing protocol to implement load-balancing. This has the advantage of simplicity, in that you don’t need specialized equipment, but it comes with some downsides as well.

The biggest is that BGP-based load balancing does not react gracefully to changes in the backend set for an address. When a cluster node goes down, you should expect all active connections to your service to be broken (users will see “Connection reset by peer”).

BGP-based routers implement stateless load-balancing. They assign a given packet to a specific next hop by hashing some fields in the packet header, and using that hash as an index into the array of available backends.

The problem is that the hashes used in routers are not stable, so whenever the size of the backend set changes (for example when a node’s BGP session goes down), existing connections will be rehashed effectively randomly, which means that the majority of existing connections will end up suddenly being forwarded to a different backend, one that has no knowledge of the connection in question.

The consequence of this is that any time the IP-to-Node mapping changes for your service, you should expect to see a one-time hit where most active connections to the service break. There’s no ongoing packet loss or blackholing, just a one-time clean break.

Depending on what your services do, there are a couple of mitigation strategies you can employ:

  • Pin your service deployments to specific nodes, to minimize the pool of nodes that you have to be “careful” about.
  • Schedule changes to your service deployments during “trough”, when most of your users are asleep and your traffic is low.
  • Split each logical service into two Kubernetes services with different IPs, and use DNS to gracefully migrate user traffic from one to the other prior to disrupting the “drained” service.
  • Add transparent retry logic on the client side, to gracefully recover from sudden disconnections. This works especially well if your clients are things like mobile apps or rich single page web apps.
  • Put your services behind an ingress controller. The ingress controller itself can use MetalLB to receive traffic, but having a stateful layer between BGP and your services means you can change your services without concern. You only have to be careful when changing the deployment of the ingress controller itself (e.g. when adding more nginx pods to scale up).
  • Accept that there will be occasional bursts of reset connections. For low-availability internal services, this may be acceptable as-is.