It’s been a while since I wrote up a new post, so I thought I would come back with a nice post about the changes in configuring DHCP Servers on the SRX’s. Since Junos 12.x a new DHCP process came out to help fix some long standing issues with the existing feature. In this post I will discuss the old configuration, some of the problems I would regularly encounter, and the configuration of the new DHCP process.

The SRX platform has had DHCP available since the platform originally was deployed. It is pretty simple to configure under System > Services > DHCP:

[edit system services]
    dhcp {
        pool 192.168.1.0/24 {
            address-range low 192.168.1.5 high 192.168.1.100;
            default-lease-time 3600;
            domain-name alostrealist.com;
            name-server {
                192.168.1.1;
            }
            router {
                192.168.1.1;
            }
        }
    }

[edit security zones security-zone trust]

interfaces { ge-0/0/1.0 { host-inbound-traffic { system-services { dhcp; } } } }

So this is pretty straightforward. In the case the SRX is passing out addresses for the 192.168.1.0/24 network, providing a name-server, domain name, and a default gateway.

This solution had its problems, and the big issue was related to setting up DHCP in routing instances. The SRX would drop the DHCP packets if the interface was bound to a routing an instance. In order to get around the limitation the interface would have to remain in the default routing instance, and a firewall filter would redirect DHCP packets from the routing instance into the default routing instance. This KB Article explains it in detail. In simple deployments this usually was not an issue, but for larger deployments with overlapping subnets this posed a huge problem without routing instance support.

The new DHCP daemon, JDHCP, was introduced in 12.1 for the SRX’s. This new daemon resolved the issue with DHCP in routing instances. Moreover the configuration was moved to the routing instance instead of globally.

[edit]
  routing-instances {
      JDHCP-VR {
          system {
              services {
                  dhcp-local-server {
                      group JDHCP-GROUP {
                          interface ge-0/0/1.0;
                      }
                  }
              }
          }
          access {
              address-assignment {
                  pool JDHCPVR-POOL {
                      family inet {
                          network 192.168.1.0/24;
                          range JDHCPVR-RANGE {
                              low 192.168.1.5;
                              high 192.168.1.100;
                          }
                          dhcp-attributes {
                              name-server {
                                  192.168.1.1;
                              }
                              router {
                                  192.168.1.1;
                              }
                          }
                      }
                  }
              }
          }
      }                                
  }

Now while setting up a DHCP Server in a cluster is still not officially supported I have found that JDHCP seems to be more stable on clustered SRX’s as well.

UPDATE: As of 12.1×47, DHCP on a SRX Cluster is now fully supported!

2 comments

  1. I hope that maybe you can help me out with a question. I had a single SRX210 configured with 2 routing-instances, both with JDHCP working. It was working great. Now I have 2 SRX210s in a cluster and updated to 12.1×47. I cannot get the JDHCP to work anymore. I put the 2 vlans that I had before into reth0 with vlan-tagging. You say that you have had DHCP working in a cluster before. Maybe you can see the issue that I am having? Here are the routing instance and the interfaces:

    ge-0/0/0 {
    gigether-options {
    redundant-parent reth0;
    }
    }
    ge-2/0/0 {
    gigether-options {
    redundant-parent reth0;
    }
    }

    reth0 {
    vlan-tagging;
    redundant-ether-options {
    redundancy-group 1;
    }
    unit 1 {
    vlan-id 1;
    family inet {
    address 10.10.1.1/24;
    }
    }
    unit 2 {
    vlan-id 2;
    family inet {
    address 10.10.2.1/24;
    }
    }
    }

    vlan {
    unit 1 {
    family inet {
    address 10.10.1.1/24;
    }
    }
    unit 2 {
    family inet {
    address 10.10.2.1/24;
    }
    }
    }

    vlan1_vr {
    instance-type virtual-router;
    system {
    services {
    dhcp-local-server {
    group VLAN1-JDHCP {
    interface reth0.1;
    interface vlan.1;
    }
    }
    }
    }
    access {
    address-assignment {
    pool VLAN1VR-POOL {
    family inet {
    network 10.10.1.0/24;
    range VLAN1VR-RANGE {
    low 10.10.1.11;
    high 10.10.1.254;
    }
    dhcp-attributes {
    router {
    10.10.1.1;
    }
    }
    }
    }
    }
    }
    interface reth0.1;
    interface vlan.1;
    routing-options {
    instance-import from_internet;
    }
    }

    Sorry the long post. Maybe I should use the srx as a dhcp server and just off-load that the a vm or even to my EX switches?

    Any advice you can provide is appreciated!

    1. What does your security zone configuration look like? It may be something as simple as host-inbound-traffic isn’t set properly.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.