OpenShift Ports: Difference between revisions
Jump to navigation
Jump to search
(Created page with "=Internal= * OpenShift Configuration =Overview= * UDP 4789, node to node, node to master and master to node: must be accessible for any...") |
|||
(3 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
=External= | |||
* https://docs.openshift.com/container-platform/3.5/install_config/install/prerequisites.html#required-ports | |||
=Internal= | =Internal= | ||
* [[OpenShift Configuration#Subjects|OpenShift Configuration]] | * [[OpenShift Configuration#Subjects|OpenShift Configuration]] | ||
= | =Internal Acces= | ||
* UDP 4789, node to node, node to master and master to node: must be accessible for any host in the cluster, it is required for [[OpenShift_Concepts#Overlay_Network|SDN]] communication between pods on separated hosts. | * UDP 4789, node to node, node to master and master to node: must be accessible for any host in the cluster, it is required for [[OpenShift_Concepts#Overlay_Network|SDN]] communication between pods on separated hosts. | ||
Line 9: | Line 13: | ||
* TCP 8443 node to master: access to API. | * TCP 8443 node to master: access to API. | ||
* TCP 10250 master to node: a node endpoint to be used by the master for communication. | * TCP 10250 master to node: a node endpoint to be used by the master for communication. | ||
* TCP | * TCP 4001 master to master: 4001 used for non-clustered etcd to accept change in state. | ||
{| | |||
| 2380 || TCP || etcd || clustered etcd uses 2380 to listen for peers, for leader election and peering connection when using standalone etcd | |||
|- | |||
| 2379 || TCP || etcd || etcd uses the port for client requests, used for state change traffic | |||
|} | |||
=External Access= | |||
* TCP 8443 - API access, web console, CLI. | |||
* TCP 80, 443 - ports opened and bound to the [[OpenShift_Concepts#Router|default router]]. |
Latest revision as of 01:16, 25 May 2017
External
Internal
Internal Acces
- UDP 4789, node to node, node to master and master to node: must be accessible for any host in the cluster, it is required for SDN communication between pods on separated hosts.
- TCP 53 or 8053, node to master, master to master: used by internal DNS traffic.
- TCP 8443 node to master: access to API.
- TCP 10250 master to node: a node endpoint to be used by the master for communication.
- TCP 4001 master to master: 4001 used for non-clustered etcd to accept change in state.
2380 | TCP | etcd | clustered etcd uses 2380 to listen for peers, for leader election and peering connection when using standalone etcd |
2379 | TCP | etcd | etcd uses the port for client requests, used for state change traffic |
External Access
- TCP 8443 - API access, web console, CLI.
- TCP 80, 443 - ports opened and bound to the default router.