2.7 KiB
| layout | page_title | sidebar_current |
|---|---|---|
| docs | Multiple Datacenters | docs-guides-datacenters |
Multi-Datacenter Deploys
One of the key features of Consul is it's support for multiple datacenters. The architecture of Consul is designed to promote a low-coupling of datacenters, so that connectivity issues or failure of any datacenter does not impact the availability of Consul in other regions. This means each datacenter runs independently, with a dedicated group of servers and a private LAN gossip pool.
To get started, we initially just bootstrap each
datacenter individually. We should have at least two datacenters now, which
we can refer to as dc1 and dc2, although the names are opaque to Consul.
The next step is to ensure that all the server nodes join the WAN gossip pool.
To query the known WAN nodes, we use the members command:
$ consul members -wan
...
This will provide a list of all known members in the WAN gossip pool. This should only contain server nodes. Client nodes do not participate, since they send requests to a datacenter-local server, which then forwards the request to a server in the target datacenter.
The next step is to simply join all the servers in the WAN pool:
$ consul join -wan <server 1> <server 2> ...
...
The join command is used with the -wan flag to indicate we are attempting
to join a server in the WAN gossip pool. As with the LAN gossip, you only need
to join a single existing member, and the gossip protocol will be used to exchange
information about all known members. For the initial setup however, each server
will only know about itself and must be added to the cluster.
Once this is done the members command can be used to verify that
all server nodes are known about. We can also verify that both datacenters
are known using the HTTP API:
$ curl http://localhost:8500/v1/catalog/datacenters
["dc1", "dc2"]
As a simple test, you can try to query the nodes in each datacenter:
$ curl http://localhost:8500/v1/catalog/nodes?dc=dc1
...
$ curl http://localhost:8500/v1/catalog/nodes?dc=dc2
...
There are a few networking requirements that must be satisfied for this to work. Clearly, all server nodes must be able to talk to each other. Otherwise, the gossip protocol as well as RPC forwarding will not work. If service discovery is to be used across datacenters, then the network must be able to route traffic between IP addresses across regions as well. Usually, this means that all datacenters must be connected using a VPN or other tunneling mechanism. Consul does not handle VPN, address rewriting, or NAT traversal for you.