Networking prerequisitesedit

The first host you install ECE on initially requires the ports for all roles to be open, which includes the ports for the coordinator, allocator, director, and proxy roles. After you have brought up your initial ECE installation, only the ports for the roles that the initial host continues to hold need to remain open. Before installing a host, make sure that ports 20000, 21000, and 22000 are open for the installation script checks.

For versions 2.4.0 and 2.4.1, IPv6 should remain enabled on any host with the Proxy role. In 2.4.2 and later, IPv6 can be disabled.

Inbound trafficedit

When there are multiple hosts for each role, the inbound networking and ports can be represented by the following diagram:

ECE networking and ports

Inbound traffic from any source
Number Host role Inbound ports *Purpose*

All

22

Installation and troubleshooting SSH access only (TCP)

2

Coordinator

12300/12343, 12400/12443

Admin API access (HTTP/HTTPS)

3

Proxy

9200, 9243

Elasticsearch REST API. 9200 is plain text and 9243 is with TLS, also required by load balancers

3

Proxy

9300, 9343

Elasticsearch transport client. 9300 is plain text and 9343 is with TLS, also required by load balancers

3

Proxy

9400

Elasticsearch Cross Cluster Replication / Search with TLS authentication, also required by load balancers. Can be blocked if CCR/CCS is not used.

7

Coordinator

12400/12443

Cloud UI console to API (HTTP/HTTPS)

Inbound traffic from internal components of ECE

In addition to the following list, you should open 12898-12908 and 13898-13908 on the director host for Zookeeper leader and election activity.

Number Host role Inbound ports *Purpose*

1

Director

2112

ZooKeeper ensemble discovery/joining (TCP)

4

Director

12191-12201

Client forwarder to ZooKeeper, one port per director (TLS tunnels)

5

Allocator

19000-19999

Elasticsearch node to node (Node Transport 6.x+/TLS 6.x+)

6

Allocator

18000-18999/20000-20999

Proxy to Elasticsearch/Kibana/APM Server instance (HTTPS/Transport Client 6.x+/TLS 6.x+)

7

Coordinator

22191-22195

Connections to initial coordinator from allocators and proxies, one port per coordinator, up to five (TCP)

8

Allocator

9244

Kibana to the services forwarder (HTTP)

9

Proxy

9200/9243, 9300/9343

Kibana and Elasticsearch (HTTP via TLS tunnel)

10

Allocator

18000-18999

Constructor to Elasticsearch cluster (HTTPS)

11

Allocator

18000-18999/20000-20999

Elasticsearch (HTTPS/Transport Client TLS)

Allocator

21000-21999

APM Server (Instance Monitoring)

Proxy

9000/9043

Proxy metrics endpoint (HTTP/HTTPS)

12

Allocator

14000

Proxy to Allocator service endpoint (HTTPS)

13

Proxy

14043

API to Proxy for Allocator service traffic (HTTPS)

Outbound trafficedit

Open these ports for outbound traffic:

Host role Outbound ports Purpose

All

80

Installation script and docker.elastic.co Docker registry access (HTTP)

All

443

Installation script and docker.elastic.co Docker registry access (HTTPS)

Hosts in multiple data centersedit

A typical ECE installation should be contained within a single data center. We recommend that ECE installations not span different data centers, due to variations in networking latency and bandwidth that cannot be controlled.

Installation of ECE across multiple data centers might be feasible with sufficiently low latency and high bandwidth, with some restrictions around what we can support. Based on our experience with our hosted Elastic Cloud service, the following is required:

  • A typical network latency between the data centers of less than 10ms round-trip time during pings
  • A network bandwidth of at least 10 Gigabit

If you choose to deploy a single ECE installation across multiple data centers, you might need to contend with additional disruptions due to bandwidth or latency issues. Both ECE and Elasticsearch are designed to be resilient to networking issues, but this resiliency is intended to handle exceptions and should not be depended on as part of normal operations. If Elastic determines during a support case that an issue is related to an installation across multiple data centers, the recommended resolution will be to consolidate your installation into a single data center, with further support limited until consolidation is complete.