Conditional Route Advertising on SRX

Junos is a very powerful networking operating system, and by harnessing it we can perform more unusual tasks than we could with other alternatives. Today I will discuss a more unusual scenario to utilize conditional router advertisements and NAT to provide access to services. When the network is unavailable then the SRX will automatically disable its advertised routes.

Continue reading

Quick Tip – DHCP Relay on MC-LAG

In a recent post I described the cutover procedure used to move from a Virtual Chassis (VC) configuration to a Multi-Chassis Link Aggregation Group (MC-LAG) configuration. One of the other requirements on this cutover was to utilize DHCP Relay. However I ran into a problem getting the DHCP Relay to work. When reviewing the statistics on Bootp Helper I noticed that the dropped packet counter “Due to no interface in fud database” was increasing exponentially. Naturally, I enabled traceoptions under forwarding-options helpers with the following command:

forwarding-options {
    helpers {
        traceoptions {
            file fud.log;
            level all;
            flag all;
        }
    }
}

In the debug log we could see the following:

vndcpwar029@USDCA-SWJ-P01> show log fud.log 
May 6 21:51:24 fud_ifdb_get_addr(): getting database entry for irb.1221, port 67
May 6 21:51:28 received packet from 0.0.0.0 port 68 interface irb.1221 l2 interface ae21 rt_inst default
May 6 21:51:28 fud_ifdb_get_addr(): getting database entry for irb.1221, port 67
May 6 21:51:31 received packet from 0.0.0.0 port 68 interface irb.1224 l2 interface ae21 rt_inst default
May 6 21:51:31 fud_ifdb_get_addr(): getting database entry for irb.1224, port 67
May 6 21:51:36 received packet from 0.0.0.0 port 68 interface irb.1221 l2 interface ae21 rt_inst default
May 6 21:51:36 fud_ifdb_get_addr(): getting database entry for irb.1221, port 67

The unusual part about this is that the debug log did not indicate any unknown interface that was dropping the packet, and that the MX was receiving the packets. After a quick call to JTAC they recommended we enable the broadcast option on each bootp interface so that it looked like the following:

forwarding-options {
    helpers {
        traceoptions {
            file fud.log;
            level all;
            flag all;
        }
        bootp {
            interface {
                irb.1 {
                    broadcast;
                    server 1.2.3.4;
                }
            }
        }
    }
}

This broadcast option enables the MX to forward DHCP relay packets if the interface is unknown. Once this was added and committed DHCP Relay started to work as expected.

News, and site update!

So I have moved the blog from Tumblr over to WordPress, and the new DNS propagation should be completed by the end of the day.

On a sad note, I did not pass the JNCIE-ENT Beta exam. Back to the drawing board on that.

In a future post I will talk about migrating a MX Virtual Chassis to MX MC-LAG configuration. This should be exciting.
– Clay

JNCIS-SP – Intro to IS-IS

While waiting to hear back on the results of my JNCIE-ENT exam I have decided to start learning about the service provider realm. In particular there are many techniques in the service provider realm that could be used to enhance the toolset of any engineer, and potentially can be used to re-think the architecture of any enterprise. In this post I will discuss IS-IS, and how to perform a basic configuration of it.

What is IS-IS?

IS-IS is an interior gateway protocol used in many networking environments to dynamically update routes. It utilizes link-state information to make routing decisions, and uses Dijkstra’s algorithm to calculate the shortest path on the network – similar to OSPF.

An IS-IS network is called an Autonomous System (AS) and is divided up into segments called Areas. A single AS can contain multiple areas, and each area contains End Systems (ESs) and Intermediate Systems (ISs). ESs send and receive packets while ISs send, receive, and relay packets to other ISs. For simplicity’s sake we will consider an IS a router from this point forward.

In each area a router can be configured to relay routes into its own area (Level 1), relay routes to another area or AS (Level 2), or act as both a Level 1 and Level 2 router. This is very similar to OSPF as Level 2 router could be considered a backbone area, and a Level 1&2 router behaves like an area border router.

How Do I Configure IS-IS?

IS-IS is relatively straightforward to configure:

  1. Enable family iso on each logical interface that will participate in IS-IS
  2. Configure a loopback interface with an ISO Network Address
  3. Enable the interface under protocols isis

Configuring the interface:

ge-0/0/0 {
    unit 0 {
        family inet {
            address 10.0.0.2/24;
        }
        family iso;
    }
}

Configuring the loopback:

lo0 {
    unit 0 {
        family inet {
            address 192.168.254.3/32;
        }
        family iso {
            address 49.0001.1921.6800.2543.00;
        }
    }
}

Enabling the interfaces:

protocols {
    isis {
        interface ge-0/0/0.0;
        interface lo0.0
    }
}

Now let us move on to a little more advanced design. The diagram below shows two Juniper SRX’s in packet mode that will be participating in IS-IS. In particular we will be configuring two virtual routers, and two IS-IS areas that will share routing information:

image

Let’s first start with the interface configuration.

SRX1:

interfaces {
    ge-0/0/0 {
        unit 0 {
            family inet {
                address 10.0.0.1/24;
            }
            family iso;
        }
    }
    lt-0/0/0 {
        unit 1 {
            encapsulation ethernet;
            peer-unit 2;
            family inet {
                address 172.16.1.0/31;
            }
            family iso;
        }
        unit 2 {
            encapsulation ethernet;
            peer-unit 1;
            family inet {
                address 172.16.1.1/31;
            }
            family iso;
        }
    }
    ge-0/0/1 {
        unit 0 {
            family inet {
                address 10.1.0.1/24;
            }
            family iso;
        }                               
    }
    lo0 {
        unit 1 {
            family inet {
                address 192.168.254.1/32;
            }
            family iso {
                address 49.0001.1921.6800.2541.00;
            }
        }
        unit 2 {
            family inet {
                address 192.168.254.2/32;
            }
            family iso {
                address 49.0002.1921.6800.2542.00;
            }
        }
    }
}

SRX2:

interfaces {
    ge-0/0/0 {
        unit 0 {
            family inet {
                address 10.0.0.2/24;
            }
            family iso;
        }
    }
    lt-0/0/0 {
        unit 1 {
            encapsulation ethernet;
            peer-unit 2;
            family inet {
                address 172.16.1.2/31;
            }
            family iso;
        }
        unit 2 {
            encapsulation ethernet;
            peer-unit 1;
            family inet {
                address 172.16.1.3/31;
            }
            family iso;
        }
    }
    ge-0/0/1 {
        unit 0 {
            family inet {
                address 10.1.0.2/24;
            }
            family iso;
        }                               
    }
    lo0 {
        unit 1 {
            family inet {
                address 192.168.254.3/32;
            }
            family iso {
                address 49.0001.1921.6800.2543.00;
            }
        }
        unit 2 {
            family inet {
                address 192.168.254.4/32;
            }
            family iso {
                address 49.0001.1921.6800.2544.00;
            }
        }
    }
}

In this example the SRX’s have multiple loopback addresses configured. These will be used to identify the connections inside each AS belonging to each router. The address can be variable in length ranging from eight octets to twenty octets, and consists of 5 major parts:

  • Authority and Format Identifier (AFI)
  • Domain ID
  • Area ID
  • System ID
  • Selector

In the case of the configured ISO addresses they are omitting the Domain ID. In the example of SRX1 lo0.1 and its address 49.0001.1921.6800.2541.00:

  • AFI: 49
  • Area ID: 0001
  • System ID: 1921.6800.2541
  • Selector: 00

The AFI and Area ID on each loopback address should match up to the area listed in each diagram.

Now onto the virtual router configuration, and enabling IS-IS in each virtual router.

SRX1:

routing-instances {
    r1 {
        instance-type virtual-router;
        interface ge-0/0/0.0;
        interface lt-0/0/0.1;
        interface lo0.1;
        protocols {
            isis {
                interface ge-0/0/0.0 {
                    level 2 disable;
                }
                interface lt-0/0/0.1;
                interface lo0.1 {
                    level 2 disable;
                }
            }
        }
    }
    r2 {
        instance-type virtual-router;
        interface lt-0/0/0.2;
        interface ge-0/0/1.0;
        interface lo0.2;
        protocols {                     
            isis {
                interface lt-0/0/0.2;
                interface ge-0/0/1.0 {
                    level 2 disable;
                }
                interface lo0.2 {
                    level 2 disable;
                }
            }
        }
    }
}

SRX2:

routing-instances {
    r1 {
        instance-type virtual-router;
        interface ge-0/0/0.0;
        interface lt-0/0/0.1;
        interface lo0.1;
        protocols {
            isis {
                interface ge-0/0/0.0 {
                    level 2 disable;
                }
                interface lt-0/0/0.1;
                interface lo0.1 {
                    level 2 disable;
                }
            }
        }
    }
    r2 {
        instance-type virtual-router;
        interface lt-0/0/0.2;
        interface ge-0/0/1.0;
        interface lo0.2;
        protocols {                     
            isis {
                interface lt-0/0/0.2;
                interface ge-0/0/1.0 {
                    level 2 disable;
                }
                interface lo0.2 {
                    level 2 disable;
                }
            }
        }
    }
}

With the exception of the lt interfaces, all interfaces have level 2 explicitly disabled. This is to ensure that those interfaces cannot establish adjacencies to routers outside of their respective areas. Let’s take a look at the interfaces and adjacencies below:

SRX1:

root@SRX-1> show isis interface instance r1    
IS-IS interface database:
Interface             L CirID Level 1 DR        Level 2 DR        L1/L2 Metric
ge-0/0/0.0            1   0x2 SRX-1.02          Disabled               10/10
lo0.1                 0   0x1 Passive           Disabled                0/0
lt-0/0/0.1            3   0x1 SRX-1.00          1921.6800.2542.03      10/10

root@SRX-1> show isis interface instance r2    
IS-IS interface database:
Interface             L CirID Level 1 DR        Level 2 DR        L1/L2 Metric
ge-0/0/1.0            1   0x2 1921.6800.2542.00 Disabled               10/10
lo0.2                 0   0x1 Passive           Disabled                0/0
lt-0/0/0.2            3   0x3 1921.6800.2542.00 1921.6800.2542.03      10/10

SRX2:

root@SRX-2> show isis interface instance r1    
IS-IS interface database:
Interface             L CirID Level 1 DR        Level 2 DR        L1/L2 Metric
ge-0/0/0.0            1   0x1 SRX-1.02          Disabled               10/10
lo0.1                 0   0x1 Passive           Disabled                0/0
lt-0/0/0.1            3   0x1 SRX-2.02          SRX-2.02               10/10

root@SRX-2> show isis interface instance r2    
IS-IS interface database:
Interface             L CirID Level 1 DR        Level 2 DR        L1/L2 Metric
ge-0/0/1.0            1   0x1 SRX-2.00          Disabled               10/10
lo0.2                 0   0x1 Passive           Disabled                0/0
lt-0/0/0.2            3   0x2 SRX-2.02          SRX-2.02               10/10

It appears that all the interfaces are configured correctly. Let’s check the adjacencies:

SRX1:

root@SRX-1> show isis adjacency instance r1 
Interface             System         L State        Hold (secs) SNPA
ge-0/0/0.0            SRX-2          1  Up                   24  0:c:29:48:6d:a4
lt-0/0/0.1            SRX-1          2  Up                    8  2:96:14:10:1:34

root@SRX-1> show isis adjacency instance r2    
Interface             System         L State        Hold (secs) SNPA
lt-0/0/0.2            1921.6800.2541 2  Up                   22  2:96:14:10:1:33

SRX2:

root@SRX-2> show isis adjacency instance r1 
Interface             System         L State        Hold (secs) SNPA
ge-0/0/0.0            SRX-1          1  Up                    7  0:c:29:a5:26:b3
lt-0/0/0.1            SRX-2          1  Up                    8  2:96:14:10:1:34
lt-0/0/0.1            SRX-2          2  Up                    8  2:96:14:10:1:34

root@SRX-2> show isis adjacency instance r2    
Interface             System         L State        Hold (secs) SNPA
lt-0/0/0.2            1921.6800.2543 1  Up                   24  2:96:14:10:1:33
lt-0/0/0.2            1921.6800.2543 2  Up                   24  2:96:14:10:1:33

That is odd; it appears that IS-IS established adjacencies across ge-0/0/0 and the lt interfaces, but not ge-0/0/1. Let’s monitor traffic on ge-0/0/1 and see what is going on:

SRX1:

root@SRX-1> monitor traffic interface ge-0/0/1 detail no-resolve 
Address resolution is OFF.
Listening on ge-0/0/1, capture size 1514 bytes

--snip--

01:06:46.358359  In IS-IS, length 48
        L1 Lan IIH, hlen: 27, v: 1, pdu-v: 1, sys-id-len: 6 (0), max-area: 3 (0)
          source-id: 1921.6800.2544,  holding time: 27s, Flags: [Level 1 only]
          lan-id:    1921.6800.2544.00, Priority: 64, PDU length: 48
            Protocols supported TLV #129, length: 2
              NLPID(s): IPv4 (0xcc), IPv6 (0x8e)
            IPv4 Interface address(es) TLV #132, length: 4
              IPv4 interface address: 10.1.0.2
            Area address(es) TLV #1, length: 4
              Area address (length: 3): 49.0001
            Restart Signaling TLV #211, length: 3
              Flags [none], Remaining holding time 0s

--snip--

I think we found our problem; as per the diagram the address should be 49.0002 when coming from SRX2’s lo0.2 interface. Let’s correct that:

SRX2:

root@SRX-2> edit 
Entering configuration mode

[edit]
root@SRX-2# delete interfaces lo0.2 family iso address 49.0001.1921.6800.2544.00 

[edit]
root@SRX-2# set interfaces lo0.2 family iso address 49.0002.1921.6800.2544.00  

[edit]
root@SRX-2# commit and-quit

Now let’s check the adjacencies one more time:

SRX1:

root@SRX-1> show isis adjacency instance r1                         
Interface             System         L State        Hold (secs) SNPA
ge-0/0/0.0            SRX-2          1  Up                   20  0:c:29:48:6d:a4
lt-0/0/0.1            1921.6800.2542 2  Up                    8  2:96:14:10:1:34

root@SRX-1> show isis adjacency instance r2    
Interface             System         L State        Hold (secs) SNPA
ge-0/0/1.0            1921.6800.2544 1  Up                   25  0:c:29:48:6d:ae
lt-0/0/0.2            SRX-1          2  Up                   20  2:96:14:10:1:33

SRX2:

root@SRX-2> show isis adjacency instance r1    
Interface             System         L State        Hold (secs) SNPA
ge-0/0/0.0            1921.6800.2541 1  Up                    7  0:c:29:a5:26:b3
lt-0/0/0.1            1921.6800.2544 2  Up                    7  2:96:14:10:1:34

root@SRX-2> show isis adjacency instance r2    
Interface             System         L State        Hold (secs) SNPA
ge-0/0/1.0            SRX-1          1  Up                    6  0:c:29:a5:26:bd
lt-0/0/0.2            SRX-2          2  Up                   25  2:96:14:10:1:33

Perfect! And now look at the routing:

SRX1:

r1.inet.0: 10 destinations, 10 routes (10 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both

10.1.0.0/24        *[IS-IS/18] 00:04:13, metric 20
                    > to 172.16.1.1 via lt-0/0/0.1
172.16.1.2/31      *[IS-IS/15] 01:55:23, metric 20
                    > to 10.0.0.2 via ge-0/0/0.0
192.168.254.2/32   *[IS-IS/18] 01:54:31, metric 10
                    > to 172.16.1.1 via lt-0/0/0.1
192.168.254.3/32   *[IS-IS/15] 01:55:23, metric 10
                    > to 10.0.0.2 via ge-0/0/0.0
192.168.254.4/32   *[IS-IS/18] 00:04:07, metric 20
                    > to 172.16.1.1 via lt-0/0/0.1

r2.inet.0: 10 destinations, 10 routes (10 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both

10.0.0.0/24        *[IS-IS/18] 01:54:31, metric 20
                    > to 172.16.1.0 via lt-0/0/0.2
172.16.1.2/31      *[IS-IS/15] 00:04:08, metric 20
                    > to 10.1.0.2 via ge-0/0/1.0
192.168.254.1/32   *[IS-IS/18] 01:54:31, metric 10
                    > to 172.16.1.0 via lt-0/0/0.2
192.168.254.3/32   *[IS-IS/18] 01:54:31, metric 20
                    > to 172.16.1.0 via lt-0/0/0.2
192.168.254.4/32   *[IS-IS/15] 00:04:08, metric 10
                    > to 10.1.0.2 via ge-0/0/1.0

We have now successfully configured IS-IS inside Junos!

More to come soon.

JNCIE-ENT Beta Exam

I recently took the JNCIE-ENT Beta exam in Sunnyvale, CA, and have been meaning to post my thoughts on the experience.

Back in September I took the JNCIE-ENT Exam in Herndon, VA. After being involved in several switching projects, and putting in over 200 hours of labs on my own I thought I would be ready to pass this exam. The study materials I had available are listed below:

I began to study and understand the wide world of L2/L3. It was a major change from the Security track that I was so familiar with, but all in all it turned out to be a very exciting and enlightening experience. It also was my first real exposure of IPv6 outside of a HE Tunnel on my SSG at home.

Unfortunately the exam itself was an eye opening experience for me in what I did know, and more importantly what I did not know. I fell short of the requirements needed to pass the exam, and got the unpleasant email that I failed the exam. It was a humbling experience, and after a few months I began to hit the books again for attempt #2.

In the meantime Juniper had put out an announcement for candidates for the JNCIE-ENT Beta exam in February, and they accepted my request to participate in the beta. This was an excellent opportunity to attempt the exam again as well as provide feedback on the exam team. A huge THANK YOU goes out to Liz Burns and her team on providing such an opportunity!

After finding out some of my friends and Internet colleagues were also accepted into the beta we began to collaborate to truly understand some of the techniques needed to understand the objectives of the JNCIE-ENT exams. We were able to provide scenarios/tools used to test many of the requirements needed to pass the JNCIE-ENT. Individuals such as Tim Hoffman were valuable as well as instrumental in leading the collaboration efforts.

February finally rolls around to sit the Beta Exam. It is important to note that these exams are under a very strict NDA so I cannot detail the exam itself. That being said I can say that the topics extremely thorough, and the tasks to configure/troubleshoot each topic would likely be seen in a real-world scenario. That being said even after taking the exam once already I almost got tripped up on a few requirements to get the network up and running. Fortunately Juniper does provide the entire configuration/examples guide, which was instrumental in helping figure the steps necessary to accomplish such tasks. The trick is combining PDF searches with CLI commands such as help apropos or help reference – quick tools to help find that correct command.

All in all I had plenty of time to review my work, and continually test portions of the configuration to ensure that new changes did not break existing functionality. I completed the exam with about an hour to spare, and had plenty of opportunities to review each task to ensure that I understood what was being asked as part of the requirements. It was another great learning experience, and I am hopeful that this time I will earn a passing grade. Since this exam was a beta, I will have to wait for 2 months before I learn the results of the exam.

Those two months give me plenty of time to start working towards JNCIE-SP…

Packet Mode and Selective Packet Services

Packet mode enables a SRX firewall to act strictly as a router, forwarding packets from a source to a destination without tracking sessions. This is useful for an engineer in certain situations such as high throughput applications that do not need full firewall functionality, or asymmetric traffic flows. We can also enable this mode on interesting traffic which is called Selective Packet Services. More details on Selective Packet Services is available on the following PDF as well as the rest of this post.

Enabling packet mode does change some of the behaviors of the SRX. For starters, it will bypass practically everything under the security stanza of the SRX. This means that the following features are disabled:

  • NAT
  • VPN Termination
  • Firewall Policies under the security stanza
  • IPS
  • UTM
  • Screens

Furthermore, when enabling packet mode for all traffic you will need to use stateless firewall filters to manage traffic which is located under the firewall stanza in the config.

On to business, shall we?

How to turn on Packet Mode for all traffic:

set security forwarding-options family mpls mode packet-based
commit and-quit
request system reboot

How to turn on Packet Mode for selective traffic:

  1. Set up a firewall filter to accept interesting traffic with an action of packet-mode
  2. Apply filter to interface that traffic is ingresses the firewall
  3. Set up 2nd firewall filter to accept return traffic with an action of packet-mode
  4. Apply 2nd filter to interface that return traffic ingresses the firewall

Below is a slightly fancier version of Selective Packet Services as it includes forwarding the traffic out of a forwarding instance:

interfaces {
    gr-0/0/0 {
        unit 0 {
            tunnel {
                source 1.1.1.1;  
                destination 2.2.2.2;
            }
            family inet {
                filter {
                    input redirect-return;
                }
                address 172.25.0.1/30;
            }
        }
    ge-0/0/1 {
        description "Trust Zone";
        unit 0 {
            family inet {
                filter {
                    input redirect-packet;
                }
                address 192.168.0.1/24;
            }
        }
    }
}

routing-options {
    interface-routes {
        rib-group inet global-rib;
    }
    rib-groups {
        global-rib {
            import-rib [ r1.inet.0 inet.0 ];
        }
    }
}

firewall {
    filter redirect-packet {
        term packet {
            from {
                destination-port [ http https ];
            }
            then {
                packet-mode;
                routing-instance r1;
            }
        }
        term allow-everything-else {
            then accept;
        }
    }
    filter redirect-return {
        term packet {
            from {
                source-port [ http https ];
            }
            then {
                packet-mode;
            }
        }
        term allow-everything-else {
            then accept;
        }
    }
}

routing-instances {
    r1 {
        instance-type forwarding;
        routing-options {
            static {
                route 0.0.0.0/0 next-hop gr-0/0/0.0;
            }
        }
    }
}