Cisco ACE 101: Tony’s 5 Steps to a Happy VIP

I’ve been teaching Cisco ACE for over four years now, and I developed a quick trick/check list to teach students the minimum configuration to get a virtual service (VIP) up and running. And since the CCIE Data Center lab will soon be upon us, I’m sharing this little trick with you. I call it “Tony’s 5 Steps to a Happy VIP”. And here it is:

Step #1: ACL
Step #2: class-map: Defines the VIP address and port
Step #3: policy-map: Which server farm(s) do we send traffic to
Step #4: policy-map: Multi-match, will pair every class-map to its policy-map
Step #5: service-policy: Apply step #4 to the VLAN interface

Using that checklist, you can quickly troubleshoot/understand most ACE configurations. So what does that list mean?

First off, let’s define what a VIP even is: In load balancing terms, it refers to an IP and TCP or UDP port combination. In that regard, it’s a bit of a misnomer, since VIP is an acronym for “Virtual IP”, and only implies an IP address. Depending on the vendor, a VIP can be called a “Virtual Server”, “Virtual Service”, although it’s commonly referred to simply as “VIP”. It’s whatever you point the firehouse of network traffic to.

I’m not anti-GUI (in fact, I think the GUI is increasingly necessary in the network world), but in the case of the ACE (and CCIE DC) you’re going to want to use the CLI. It’s just faster, and you’re going to feel the need for speed in that 8 hour window. Also, when things go wrong, the CLI (and config file) is going to allow you to troubleshoot much more quickly than the GUI in the case of the ACE.

The CLI for Cisco ACE can be a little overwhelming. For some reason, Cisco decided to use the Modular QoS CLI (MQC) configuration framework. To me, it seems overly complicated.  Other vendors have CLIs that tend to make a lot more sense, or at least is a lot easier to parse with your eyes. If you’re familiar with class-maps, policy-maps, and service-policies, the transition to the ACE CLI won’t be all that difficult. It works very similar to setting up QoS. However, if you’re new to MQC, it’s going to be a bit of a bumpy ride.

How I felt learning MQC for the first time

The Configuration

Here is a very basic configuration for an ACE:

access-list ANYANY line 10 extended permit ip any any 

rserver host SERVER1 ip address 192.168.10.100
  inservice 
rserver host SERVER2 ip address 192.168.10.101 
  inservice 
rserver host SERVER3 ip address 192.168.10.101 
  inservice

serverfarm host SERVERFARM1
  rserver SERVER1
    inservice
  rserver SERVER2
    inservice
  rserver SERVER3
    inservice 

class-map match-all VIP1-80 
  2 match virtual-address 192.168.1.200 tcp eq http

class-map match-all VIP1-443
  2 match virtual-address 192.168.1.200 tcp eq https

policy-map type loadbalance first-match VIP1-POLICY
  class class-default 
    serverfarm SERVERFARM1 

policy-map multi-match CLIENT-VIPS 
  class VIP1-80
    loadbalance vip inservice 
    loadbalance policy VIP1-POLICY
  class VIP1-443
    loadbalance vip inservice
    loadbalance policy VIP1-POLICY

interface vlan 200 
  description Client-facing interface 
  ip address 192.168.1.10 255.255.255.0 
  access-group input ANYANY
  service-policy input CLIENT-VIPS 
  no shutdown
interface vlan 100
  description Server VLAN
  ip address 192.168.10.1 255.255.255.0
  no shutdown

Step #1: ACL

It’s not necessarily part of the VIP setup, but you do need to have an ACL rule in before a VIP will work. The reason is that the ACE, unlike most load balancers, is deny all by default. Without an ACL you can’t pass any traffic through the ACE. (However, ACLs have no effect on traffic to the ACE for management.)

Many an ACE configuration problem has been caused by forgetting to put an ACL rule in. My recommendation? Even if you plan on using specific ACLs, start out with an “any/any” rule.

access-list ANYANY line 10 extended permit ip any any

And don’t forget to put them on the interface facing the client (outside VLAN).

interface vlan 200 
  description Client-facing interface 
  ip address 192.168.1.10 255.255.255.0 
  access-group ANYANY input 
  service-policy input CLIENT-VIPS 
  no shutdown

Once you get everything working, then you can make a more nailed-down ACL if required, although most don’t since there is likely a firewall in place anyway (even the Cisco example configurations typically only have an any-any rule in place).

If you do use a more specific ACL, it’s often a good idea to switch back to any-any for troubleshooting. Put the more specific rule in place only when you’re sure your config works.

Step #2: class-map (VIP declaration)

The next step is to create a class-map that will catch traffic destined for the VIP. You should always include an IP address as well as a single TCP or UDP port. I’ve seen configurations that match any TCP/UDP port on a specific IP address, and this is usually a really, really bad idea.

class-map match-all VIP1-80
  2 match virtual-address 192.168.1.200 tcp eq http

This defines a VIP with an address of 192.168.1.200 on port http (port 80). Even if you set up multiple ports on the same IP address, such as port 80 and 443, use different class-maps and configure them separately.

Step #3: policy-map (what do we do with traffic hitting the VIP)

Here is where the VIP is defined as either a Layer 4 VIP or a Layer 7 VIP. The example below is a simple Layer 4 VIP (the ACE is not aware of anything that happens above Layer 4). You can get a lot fancier in this section, such as sending certain matched traffic to one server farm, and other traffic to others, and/or setting up persistence. Again, this is the most basic configuration.

policy-map type loadbalance first-match VIP1-POLICY
  class class-default <-- This matches everything
    serverfarm SERVERFARM1 <-- And sends it all right here

Step #4: policy-map (round-up policy-map, pairs a VIP with a decision process, and all the pairs are joined into a single statement)

You will typically have multiple Step 2’s and Step 3’s, but they exist as independent declarations so you’ll need something to round them all up into a single place and join them. In most configurations, you will typically only have one multi-match policy-map. This multi-match is where you marry a Step 2 class-map to a Step 3 policy-map. In this example, two separate class-maps use the same policy-map (which is fine).

policy-map multi-match CLIENT-VIPS 
  class VIP1-80 <-- This VIP...
    loadbalance vip inservice 
    loadbalance policy VIP1-POLICY <-- ...sends traffic to this policy
  class VIP1-443 <-- This VIP...
    loadbalance vip inservice
    loadbalance policy VIP1-POLICY <-- ...sends traffic to this policy

Step #5: service-policy (apply the round-up to the client-facing interface)

interface vlan 200 

 description Client-facing interface 
 ip address 192.168.1.10 255.255.255.0 
 access-group input ANYANY <-- Step 1's ACL is applied
 service-policy input CLIENT-VIPS <-- Step 5's multi-match policy map is applied
 no shutdown <-- Don't forget the no shut!

Hope this helps with demystifying the ACE configuration. A short little check list can really help save time, especially in a time-constrained environment like a CCIE lab.

Cisco ACE: Insert Client IP Address

Source-NAT (also referred to as one-armed mode) is a common way of implementing load balancers into a network. It has several advantages over routed-mode (where the load balancer is the default gateway of the servers), most importantly that the load balancer doesn’t need to be Layer 2 adjacent/on the same subnet as the servers.  As long as the SNAT IP address of the load balancer has bi-directional communication with the IP address of the servers, the load balancer can be anywhere. A different subnet, a different data center, even a different continent.

However, one drawback is that with Source NAT the client’s IP address is obscured. The server’s logs will show only the IP address of the SNAT address(s).

There is a way to remedy that if the traffic is HTTP/HTTPS, and that’s by having the load balancer insert the true source IP address into the HTTP request header from the client. You can do it with the ACE by putting it into the load balance policy-map.

policy-map type loadbalance http first-match VIP1_L7_POLICY
  class class-default
     serverfarm FARM1
     insert-http x-forwarded-for header-value "%is"

But alone is not enough. There are two extra steps you need to take.

The first step is you need to tell the web server to log the x-forwarded-for. For Apache, it’s a configuration file change. For IIS, you need to run an ISAPI filter in IIS.

The other thing you need to do is fix the ACE’s attention span. You see, by default the ACE has a short attention span. The HTTP protocol allows you to make multiple HTTP requests on a single TCP connection. By default, the ACE will only evaluate/manipulate the first HTTP request in a TCP connection.

So your log files will look like this:

1.1.1.1 "GET /lb/archive/10-2002/index.htm"
- "GET /lb/archive/10-2003/index.html"
- "GET /lb/archive/05-2004/0100.html HTTP/1.1"
2.2.2.2 "GET /lb/archive/10-2007/0010.html"
- "GET /lb/archive/index.php"
- "GET /lb/archive/09-2002/0001.html"

The “-” indicates Apache couldn’t find the header, because the ACE didn’t insert it. The ACE did add the first source IP address, but every request after it in the same TCP connection was ignored.

Why does the ACE do this? It’s less work for one, only evaluating/manipulating the first request in a connection. Since browsers will make dozens or even hundreds of requests over a single connection, this would be  a significant saving of resources. After all, most of the time when L7 configurations are used, it’s for cookie-based persistence. If that’s the case, all the requests in the same TCP connection are going to contain the same cookies anyway.

How do you fix it? By using a very ill-named feature called persistence-rebalance. This gives the ACE a longer attention span, telling the ACE to look at every HTTP request in the TCP connection.

First, create an HTTP parameter-map.

parameter-map type http HTTP_LONG_ATTENTION_SPAN
  persistence-rebalance

Then apply the parameter-map to the VIP in the multi-match policy map.

policy-map multi-match VIPsOnInterface
  class VIP1
    loadbalance vip inservice
    loadbalance policy VIP1_L7_POLICY
    appl-parameter http advanced-options HTTP_LONG_ATTENTION_SPAN

When that happens, the IP address will show up in all of the log entries.

1.1.1.1 "GET /lb/archive/10-2002/index.htm"
2.2.2.2 "GET /lb/archive/10-2003/index.html"
1.1.1.1 "GET /lb/archive/05-2004/0100.html HTTP/1.1"
2.2.2.2 "GET /lb/archive/10-2007/0010.html"
1.1.1.1 "GET /lb/archive/index.php"
2.2.2.2 "GET /lb/archive/09-2002/0001.html"

But remember, configuring the ACE (or load balancer in general) isn’t the only step you need to perform. You also need to tell the web service (Apache, Nginx, IIS) to use the header as well. None of them automatically use the X-Forwarded-for header.

I don’t know if they’ll try to trick you with this in the CCIE Lab, but it’s something to keep in mind for the CCIE and for implementations.