deleteSiteDeviceHaCluster

## Create HA Cluster Both nodes has to be in the same site. We expect the user to configure ha_sync / ha_data port in port_configs already ### SRX cabling see [Chassis Cluster User Guide for SRX Series Devices](https://www.juniper.net/documentation/us/en/software/junos/chassis-cluster-security-devices/topics/concept/chassis-cluster-srx-series-node-interface-understanding.html) Here’s the recommended cabling. #### SRX300 From ZTP / default state, ge-0/0/0 and ge-0/0/7 (SFP) are default WAN ports and will get DHCP IP. However, ge-0/0/0 becomes OOB/fxp0 after cluster is enabled (i.e. using it for reach Mist is not recommended) 1. form cluster in UI 2. configure ge-0/0/7,ge-1/0/7 for WAN (reth0) 3. configure ge-0/0/2,ge-1/0/2 for ha_data 4. configure ge-0/0/3- for LAN or additional WAN e.g. ``` json { "port_config": { "ge-0/0/2,ge-1/0/2": { "usage": "ha_data" }, "ge-0/0/7,ge-1/0/7": { "usage": "wan", "redundant": true, "reth_idx": 0, "ip_config": {"type": "dhcp"} }, } } ``` 1. connect ge-0/0/1 back to back for ha_control 2. connect ge-0/0/2 back to back for ha_data 3. connect both ge-0/0/7 to uplink switch to WAN and to reach Mist 4. power up both devices 5. it takes about 30 minutes for the cluster to form #### SRX320 From ZTP / default state, ge-0/0/0, ge-0/0/7 (SFP) and cl-1/0/0 (LTE) are default WAN ports and will get DHCP IP. However, ge-0/0/0 becomes OOB/fxp0 after cluster is enabled (i.e. using it for reach Mist is not recommended) ##### ZTP via ge-0/0/7 Similar to SRX300 ##### ZTP via cl-1/0/0 (LTE) 1. form cluster in UI 2. configure cl-1/0/0, cl-3/0/0 as WAN (reth0) 3. configure ge-0/0/2,ge-3/0/2 for ha_data 4. same as above #### SRX340 / SRX345 / SRX380 SRX340/SRX345 has dedicated OOB/fxp0 ports 1. form cluster in UI 2. configure ge-0/0/0,ge-5/0/0 for WAN (reth0) 3. configure ge-0/0/2,ge-5/0/2 for ha_data 4. configure ge-0/0/3- for LAN or additional WAN 5. connect ge-0/0/0 to uplink switch to WAN and to reach Mist 6. connect ge-0/0/1 back-to-back for ha_control 7. connect ge-0/0/2 back-to-back for ha_data (fabric); or for SRX380, xe-0/0/16 if 10G SFP+ is used 8. connect ge-0/0/3- to LAN or additional WANs #### SRX550 ge-0/0/0 becomes OOB/fxp0 after cluster is enabled, make enable oob_ip_config as dhcp to maintain cloud connectivity 1. connect ge-0/0/0 to reach Mist (after cluster is fully up, this port becomes OOB/fxp0) 2. connect ge-0/0/1 back-to-back for ha_control 3. connect ge-0/0/2 back-to-back for ha_data (fabric) 4. connect ge-0/0/3 to WAN (after cluster is up, intended to be used for reth0) 5. connect ge-0/0/4- to LAN or additional WANs #### SRX1500 SRX1500 has, additionally, dedicated HA Control port 1. form cluster in UI 2. configure ge-0/0/0,ge-5/0/0 for WAN (reth0) 3. configure ge-0/0/1,ge-5/0/1 for ha_data 4. configure ge-0/0/2- for LAN or additional WAN 5. connect dedicated ha_control back-to-back 6. connect ge-0/0/0 to uplink switch to WAN and to reach Mist 7. connect ge-0/0/1 back-to-back for ha_data 8. connect ge-0/0/2- to LAN or additional WANs #### SRX4100 SRX4100 has dedicated ha_control and ha_data (fabric) ports 1. connect dedicated ha_control back-to-back 2. connect dedicated ha_data back-to-back 3. connect xe-0/0/0 to WAN to reach Mist 4. connect xe-0/0/1- to LAN or additional WANs #### VSRX When standalone, VSRX has fxp0 as first Network Adapter, then ge-0/0/0-N When clustered, VSRX has fxp0, em0, then ge-0/0/0-N 1. connect net0 (fxp0) to WAN to reach Mist 2. connect net1 back-to-back for ha_control 3. connect net2 (ge-0/0/0) back-to-back for ha_data (fab0/fab1) 4. connect net3 (ge-0/0/1) to WAN, intended to be used for reth0 5. connect net4 (ge-0/0/2) to LAN SRX340/SRX345 has dedicated OOB/fxp0 ports VSRX has fxp0 as first Network Adapter, then ge-0/0/0-N 1. connect ge-0/0/0 to WAN to reach Mist 2. connect ge-0/0/1 back-to-back for ha_control 3. connect ge-0/0/2 back-to-back for ha_data (fabric); or for SRX380, xe-0/0/16 if 10G SFP+ is used 4. connect ge-0/0/3- to LAN or additional WANs #### SRX550 ge-0/0/0 becomes OOB/fxp0 after cluster is enabled, make enable oob_ip_config as dhcp to maintain cloud connectivity 1. connect ge-0/0/0 to reach Mist (after cluster is fully up, this port becomes OOB/fxp0) 2. connect ge-0/0/1 back-to-back for ha_control 3. connect ge-0/0/2 back-to-back for ha_data (fabric) 4. connect ge-0/0/3 to WAN (after cluster is up, intended to be used for reth0) 5. connect ge-0/0/4- to LAN or additional WANs #### SRX1500 SRX1500 has, additionally, dedicated HA Control port 1. form cluster in UI 2. configure ge-0/0/0,ge-7/0/0 for WAN (reth0) 3. configure ge-0/0/1,ge-7/0/1 for ha_data 4. configure ge-0/0/2- for LAN or additional WAN 5. connect dedicated ha_control back-to-back 6. connect ge-0/0/0 to uplink switch to WAN and to reach Mist 7. connect ge-0/0/1 back-to-back for ha_data 8. connect ge-0/0/2- to LAN or additional WANs #### SRX1600 SRX1600 has, additionally, two dedicated HA Control port 1. form cluster in UI 2. configure ge-0/0/0,ge-7/0/0 for WAN (reth0) 3. configure ge-0/0/1,ge-7/0/1 for ha_data 4. configure ge-0/0/2- for LAN or additional WAN 5. connect dedicated both ha_control back-to-back 6. connect ge-0/0/0 to uplink switch to WAN and to reach Mist 7. connect ge-0/0/1 back-to-back for ha_data 8. connect ge-0/0/2- to LAN or additional WANs #### SRX4100 SRX4100 has dedicated ha_control and ha_data (fabric) ports 1. connect dedicated ha_control back-to-back 2. connect dedicated ha_data back-to-back 3. connect xe-0/0/0 to WAN to reach Mist 4. connect xe-0/0/1- to LAN or additional WANs ## Replace a Node in a HA Cluster Usually Device Replacement is done by Device Replacement API. For a HA cluster, you can also replace a node by another device in the same site.