Skip to content

Commit

Permalink
Update Multi-cluster architecture doc
Browse files Browse the repository at this point in the history 
Update the Multi-cluster architecture doc with descriptions about
Multi-cluster Gateway, and change the diagrams' format from png to svg.

Signed-off-by: Lan Luo <[email protected]>
Co-authored-by: Jianjun Shen <[email protected]>
  • Loading branch information
@luolanzone @jianjuns
luolanzone and jianjuns committed Jun 14, 2022
1 parent ea99ee5 commit 9e34e8a
Show file tree
Hide file tree
Showing 6 changed files with 2,414 additions and 80 deletions.
255 changes: 175 additions & 80 deletions docs/multicluster/architecture.md
View file
Original file line number Diff line number Diff line change
@@ -1,18 +1,21 @@
# Antrea Multi-cluster Architecture

Antrea Multi-cluster implements [Multi-cluster Service API](https://github.com/kubernetes/enhancements/tree/master/keps/sig-multicluster/1645-multi-cluster-services-api),
which allows users to create multi-cluster Services that can be accessed cross clusters in a
ClusterSet. Antrea Multi-cluster also supports Antrea ClusterNetworkPolicy replication.
Multi-cluster admins can define ClusterNetworkPolicies to be replicated across the entire
ClusterSet, and enforced in all member clusters.
which allows users to create multi-cluster Services that can be accessed cross
clusters in a ClusterSet. Antrea Multi-cluster also supports Antrea
ClusterNetworkPolicy replication. Multi-cluster admins can define
ClusterNetworkPolicies to be replicated across the entire ClusterSet, and
enforced in all member clusters.

The diagram below depicts a basic multi-cluster topology in Antrea.
An Antrea Multi-cluster ClusterSet includes a leader cluster and multiple member
clusters. Antrea Multi-cluster Controller needs to be deployed in the leader and
all member clusters. A cluster can serve as the leader, and meanwhile also be a
member cluster of the ClusterSet.

<img src="assets/basic-topology.png" width="500" alt="Antrea Multi-cluster Topology">
The diagram below depicts a basic Antrea Multi-cluster topology with one leader
cluster and two member clusters.

Given a set of Kubernetes clusters, there will be a leader cluster and several member clusters.
By default, a leader cluster itself is also a member cluster of a ClusterSet. A cluster
can also be configured as a dedicated leader cluster of multiple ClusterSets.
<img src="assets/basic-topology.svg" width="650" alt="Antrea Multi-cluster Topology">

## Terminology

Expand All @@ -22,8 +25,8 @@ Namespace sameness applies, which means all Namespaces with a given name are con
be the same Namespace. The ClusterSet Custom Resource Definition(CRD) defines a ClusterSet
including the leader and member clusters information.

The ClusterClaim CRD is used to claim a cluster itself as a member of a ClusterSet with
unique cluster ID, to claim a ClusterSet with unique ClusterSet ID.
The ClusterClaim CRD is used to claim a ClusterSet with a unique ClusterSet ID, and to
claim the cluster itself as a member of a ClusterSet with a unique cluster ID.

The MemberClusterAnnounce CRD declares a member cluster configuration to the leader cluster.

Expand All @@ -34,75 +37,167 @@ given ClusterSet.

## Antrea Multi-cluster Controller

In a member cluster, Antrea Multi-cluster creates a Deployment that runs Antrea Multi-cluster
Controller which is responsible for exporting resource to and importing resource from a leader
cluster in a ClusterSet.

In a leader cluster, Antrea Multi-cluster creates a Deployment that runs Antrea Multi-cluster
Controller which is responsible for converting resources from different member clusters into one
encapsulated resource as long as these resources have the same kind and match Namespace sameness.

In ClusterSet initialization, Antrea Multi-cluster Controller in a member cluster watches
ClusterSet, ClusterClaim, and creates a MemberClusterAnnounce in the leader cluster.

For exporting resources from the member cluster, it watches ServiceExport, Service and Endpoints
resources, encapsulates Services and Endpoints into ResourceExports according to ServiceExports,
and writes the ResourceExports to leader cluster. For resource importing, it watches ResourceImports
from leader cluster, and creates multi-cluster Services and Endpoints with a prefix `antrea-mc-`
plus exported Service name, and also ServiceImports which have the same name as the Service name.

In a leader cluster, for ClusterSet initialization, Antrea Multi-cluster controller watches and
validates the ClusterSet and Clusterclaim. For resource export/import, it watches ResourceExports
and encapsulates them into ResourceImports.

## Service Export and Import

<img src="assets/resource-export-import-pipeline.png" width="800" alt="Antrea Multi-cluster Resource Export/Import Pipeline">

The current multi-cluster implementation supports Service discovery and Service export/import among
member clusters. The above diagram depicts Antrea Multi-cluster resource export/import pipeline.

Given two Services in the member clusters - `foo.ns.cluster1.local` and `foo.ns.cluster2.local`,
multi-cluster Services may be generated by the following the resource export/import pipeline.

* Administrators create resource ServiceExports `foo` in Namespace `ns` in each
of the clusters.
* The Exporters in member clusters `cluster1` and `cluster2` see ServiceExport `foo`, collect
the associated Services and Endpoints, and create ResourceExports for them in the Common Area of the
leader cluster.
* The controller in the leader cluster sees ResourcesExports in the Common Area, computes multi-cluster
Service `cluster1-ns-foo-service`, `cluster2-ns-foo-service` and associated Endpoints
`cluster1-ns-foo-endpoints`, `cluster2-ns-foo-endpoints`.
* The controller creates ResourcesImport enclosing multi-cluster Service `ns-foo-service` and Endpoints
`ns-foo-endpoints` in the Common Area.
* The Importer in each member cluster watches ResourceImports; decapsulates them and gets Service
`ns/antrea-mc-foo`, Endpoints `ns/antrea-mc-foo`, and creates the resources and as well as a
ServiceImport `ns/foo` locally if they don't exist or updates them if the resources have already
been created by the Importer earlier.

## Antrea Multi-cluster Service

Antrea Multi-cluster Controller only supports Service of type ClusterIP at this moment. In
order to support multi-cluster Service access between member clusters, Antrea requires member
clusters' Pod IPs are reachable and no overlapping between all member clusters.

When Antrea Multi-cluster Controller in member cluster watches ResourceImport's creation event
in leader cluster, it will create multi-cluster Service, Endpoints and ServiceImport locally.
The Service Ports definition will be the same as exported Services, the Endpoints will be Pod
IPs from all member clusters. The new created Antrea Multi-cluster Service is just like a regular
Kubernetes Service, so Pods in a member cluster can access the multi-cluster Service as usual without
any extra setting.
Antrea Multi-cluster Controller implements ClusterSet management and resource
export/import in the ClusterSet. In either a leader or a member cluster, Antrea
Multi-cluster Controller is deployed with a Deployment of a single replica, but
it takes different responsibilities in leader and member clusters.

### ClusterSet Establishment

In a member cluster, Multi-cluster Controller watches and validates the ClusterSet
and ClusterClaim CRs, and creates a MemberClusterAnnounce in the Common Area of the
leader cluster to join the ClusterSet.

In the leader cluster, Multi-cluster controller watches and validates the
ClusterSet and Clusterclaim CRs, and initializes the ClusterSet. It also validates
the MemberClusterAnnounce CR created by a member cluster and updates the member
cluster status to the ClusterSet.

### Resource Export and Import

In a member cluster, Multi-cluster controller watches exported resources (e.g.
ServiceExports, Services, Multi-cluster Gateways), encapsulates an exported
resource into a ResourceExport and creates the ResourceExport CR in the Common
Area of the leader cluster.

In the leader cluster, Multi-cluster Controller watches ResourceExports created
by member clusters (in the case of Service and ClusterInfo export), or by the
ClusterSet admin (in the case of Multi-cluster NetworkPolicy), converts
ResourceExports to ResourceImports, and creates the ResourceImports CRs in the
Common Area for member clusters to import them. Multi-cluster Controller also
merges ResourceExports from different member clusters to a single
ResourceImport, when these exported resources share the same kind, name, and
original Namespace (matching Namespace sameness).

Multi-cluster Controller in a member cluster also watches ResourceImports in the
Common Area of the leader cluster, decapsulates the resources from them, and
creates the resources (e.g. Services, Endpoints, Antrea ClusterNetworkPolicies,
ClusterInfoImports) in the member cluster.

For more information about Multi-cluster Service export/import, also check the
[Service Export and Import](#service-export-and-import) section.

## Multi-cluster Service

### Service Export and Import

<img src="assets/resource-export-import-pipeline.svg" width="1500" alt="Antrea Multi-cluster Service Export/Import Pipeline">

Antrea Multi-cluster Controller implements Service export/import among member
clusters. The above diagram depicts Antrea Multi-cluster resource export/import
pipeline, using Service export/import as an example.

Given two Services with the same name and Namespace in two member clusters -
`foo.ns.cluster-a.local` and `foo.ns.cluster-b.local`, a multi-cluster Service can
be created by the following resource export/import pipeline.

* User creates a ServiceExport `foo` in Namespace `ns` in each of the two
clusters.
* Multi-cluster Controllers in `cluster-a` and `cluster-b` see ServiceExport
`foo`, and both create two ResourceExports for the Service and Endpoints
respectively in the Common Area of the leader cluster.
* Multi-cluster Controller in the leader cluster sees the ResourcesExports in
the Common Area, including the two for Service `foo`: `cluster-a-ns-foo-service`,
`cluster-b-ns-foo-service`; and the two for the Endpoints:
`cluster-a-ns-foo-endpoints`, `cluster-b-ns-foo-endpoints`. It then creates a
ResourceImport `ns-foo-service` for the multi-cluster Service; and a
ResourceImport `ns-foo-endpoints` for the Endpoints, which includes the
exported endpoints of both `cluster-a-ns-foo-endpoints` and
`cluster-b-ns-foo-endpoints`.
* Multi-cluster Controller in each member cluster watches the ResourceImports
from the Common Area, decapsulates them and gets Service `ns/antrea-mc-foo` and
Endpoints `ns/antrea-mc-foo`, and creates the Service and Endpoints, as well as
a ServiceImport `foo` in local Namespace `ns`.

### Service Access Across Clusters

Since Antrea v1.7.0, the Service's ClusterIP is exported as the multi-cluster
Service's Endpoints. Multi-cluster Gateways must be configured to support
multi-cluster Service access across member clusters, and Service CIDRs cannot
overlap between clusters. Please refer to [Antrea Multi-cluster Gateway](#multi-cluster-gateway)
for more information. Before Antrea v1.7.0, Pod IPs are exported as the
multi-cluster Service's Endpoints. Pod IPs must be directly reachable across
clusters for multi-cluster Service access, and Pod CIDRs cannot overlap between
clusters. Antrea Multi-cluster only supports creating multi-cluster Services
for Services of type ClusterIP.

## Multi-cluster Gateway

Antrea started to support Multi-cluster Gateway since v1.7.0. User can choose
one K8s Node as the Gateway in a member cluster. The Gateway Node will be
responsible for routing all cross-clusters traffic from the local cluster to
other member clusters through tunnels. The diagram below depicts Antrea
Mulit-cluster connectivity with Multi-cluster Gateways.

<img src="assets/mc-gateway.svg" width="800" alt="Antrea Multi-cluster Gateway">

Antrea Agent is responsible for setting up tunnels between Gateways of meamber
clusters. At the moment, Multi-cluster Gateway only works with Antrea `encap`
mode. The tunnels between Gateways use Antrea Agent's configured tunnel type.
All member clusters in a ClusterSet need to deploy Antrea with the same tunnel
type.

The Multi-cluster Gateway implementation introduces two new CRDs `Gateway` and
`ClusterInfoImport`. `Gateway` includes the local Multi-cluster Gateway
information including: `internalIP` for tunnels to local Nodes, and `gatewayIP`
for tunnels to remote cluster Gateways. `ClusterInfoImport` includes Gateway
and network information of member clusters, including Gateway IPs and Service
CIDRs. The existing esource export/import pipeline is leveraged to exchange
the cluster network information among member clusters, generating
ClusterInfoImports in each member cluster.

### Multi-cluster Service Traffic Walk

Let's use the ClusterSet in the above diagram as an example. As shown in the
diagram:

1. Cluster A has a client Pod named `pod-a` running on a regular Node, and a
multi-cluster Service named `antrea-mc-nginx` with ClusterIP `10.112.10.11`
in the `default` Namespace.
2. Cluster B exported a Service named `nginx` with ClusterIP `10.96.2.22` in
the `default` Namespace. The Service has one Endpoint `172.170.11.22` which is
`pod-b`'s IP.
3. Cluster C exported a Service named `nginx` with ClusterIP `10.11.12.33` also
in the `default` Namespace. The Service has one Endpoint `172.10.11.33` which
is `pod-c`'s IP.

The multi-cluster Service `antrea-mc-nginx` in cluster A will have two
Endpoints:

* `nginx` Service's ClusterIP `10.96.2.22` from cluster B.
* `nginx` Service's ClusterIP `10.11.12.33` from cluster C.

When the client Pod `pod-a` on cluster A tries to access the multi-cluster
Service `antrea-mc-nginx`, the request packet will first goes through the
Service load balancing pipeline on the source Node `node-a2`, with one endpoint
of the Multi-cluster Service being chosen as the destination. Let's say endpoint
`10.11.12.33` from cluster C is chosen, then the request packet will be DNAT'd
with IP `10.11.12.33` and tunnelled to the local Gateway Node `node-a1`.
`node-a1` knows the packet is multi-cluster Service traffic destined for Cluster
C from the destination IP `10.11.12.33`, and it will tunnel the packet to
Cluster C's Gateway Node `node-c1`, after performing SNAT and setting the
packet's source IP to its own Gateway IP. On `node-c1`, the packet will go
through the Service load balancing pipeline again with a local endpoint of
Service `nginx` being chosen as the destination. As the Service has only one
local endpoint - `172.10.11.33` of `pod-c`, the request packet will be DNAT'd to
`172.10.11.33` and tunnelled to `node-c2` where `pod-c` is running. Finally,
on `node-c2` the packet will go through the normal Antrea pipeline and be
forwarded to `pod-c`.

## Antrea Multi-cluster NetworkPolicy

At this moment, Antrea does not support Pod-level policy enforcement for cross-cluster traffic. Access
towards Multi-cluster Services can be regulated with Antrea ClusterNetworkPolicy `toService` rules. In
each member cluster, users can create an Antrea ClusterNetworkPolicy selecting Pods in that cluster, with
the imported Mutli-cluster Service name and Namespace in an egress `toService` rule, and the Action to
take for traffic matching this rule. For more information regarding Antrea ClusterNetworkPolicy (ACNP),
refer to [this document](../antrea-network-policy.md).

Multi-cluster admins can also specify certain ClusterNetworkPolicies to be replicated across the entire
ClusterSet. The ACNP to be replicated should be created as a ResourceExport in the leader cluster, and
the resource export/import pipeline will ensure member clusters receive this ACNP spec to be replicated.
Each member cluster's Multi-cluster Controller will then create an ACNP in their respective clusters.
At this moment, Antrea does not support Pod-level policy enforcement for
cross-cluster traffic. Access towards Multi-cluster Services can be regulated
with Antrea ClusterNetworkPolicy `toService` rules. In each member cluster,
users can create an Antrea ClusterNetworkPolicy selecting Pods in that cluster,
with the imported Mutli-cluster Service name and Namespace in an egress
`toService` rule, and the Action to take for traffic matching this rule.
For more information regarding Antrea ClusterNetworkPolicy (ACNP), refer
to [this document](../antrea-network-policy.md).

Multi-cluster admins can also specify certain ClusterNetworkPolicies to be
replicated across the entire ClusterSet. The ACNP to be replicated should
be created as a ResourceExport in the leader cluster, and the resource
export/import pipeline will ensure member clusters receive this ACNP spec
to be replicated. Each member cluster's Multi-cluster Controller will then
create an ACNP in their respective clusters.
Binary file removed docs/multicluster/assets/basic-topology.png
View file
Binary file not shown.
Loading

0 comments on commit 9e34e8a

Please sign in to comment.