Or, How can I make my users' browsers use my cache without configuring the browsers for proxying?
- Concepts of Interception Caching
- Requirements and methods for Interception Caching
Steps involved in configuring Interception Caching
- Compile a version of Squid which accepts connections for other addresses
- Getting your traffic to the right port on your Squid Cache
Get the packets from the end clients to your cache server
- Interception Caching packet redirection with Cisco routers using policy routing (NON WCCP)
- Interception Caching packet redirection with Foundry L4 switches
- Interception Caching packet redirection with an Alcatel OmnySwitch 7700
- Interception Caching packet redirection with Cabletron/Entrasys products
- Interception Caching packet redirection with ACC Tigris digital access server
WCCP - Web Cache Coordination Protocol
- Does Squid support WCCP?
- Do I need a cisco router to run WCCP?
- Can I run WCCP with the Windows port of Squid?
- Where can I find out more about WCCP?
- Cisco router software required for WCCP
- Cache/Host configuration of WCCP
- TProxy Interception
- Other Configuration Examples
- Troubleshooting and Questions
- Further information about configuring Interception Caching with Squid
- Issues with HotMail
Concepts of Interception Caching
Interception Caching goes under many names - Interception Caching, Transparent Proxying, URL rewriting, SSL-Bump and Cache Redirection. Interception Caching is the process by which HTTP connections coming from remote clients are redirected to a cache server, without their knowledge or explicit configuration.
There are some good reasons why you may want to use this technique:
- There is no client configuration required.
- This is the most popular reason for investigating HTTP intercept - there are many client software who fail to implement HTTP proxy support.
- There is no client SSL proxy connection required.
- This is the most popular reason for investigating HTTPS intercept - there are very few clients which support SSL proxy connections.
- The server may be using URL-based virtual hosting.
This is the only known reason to need URL-rewrite form of intercept - there are two popular CMS systems which depend on this form of URL manipulation to operate. When the public URL is presented by the web server as a re-writer is needed to 'fix' the broken web server system.
Note that all other known uses of re-write can be avoided with better alternatives!
However there are also significant disadvantages for this strategy, as outlined by Mark Elsen:
- Intercepting HTTP breaks TCP/IP standards because user agents think they are talking directly to the origin server.
- Requires IPv4 with NAT on most operating systems, although some now support TPROXY or NAT for IPv6 as well.
- It causes path-MTU (PMTUD) to fail, possibly making some remote sites inaccessible. This is not usually a problem if your client machines are connected via Ethernet or DSL PPPoATM where the MTU of all links between the cache and client is 1500 or more. If your clients are connecting via DSL PPPoE then this is likely to be a problem as PPPoE links often have a reduced MTU (1472 is very common).
- On older IE versions before version 6, the ctrl-reload function did not work as expected.
- Connection multiplexing does not work. Clients aware of the proxy can send requests for multiple domains down one proxy connection and save resources while letting teh proxy do multiple backend connections. When talking to an origin clients are not permitted to do this and will open many TCP connections for resources. This causes intercepting proxy to consume more network sockets than a regular proxy.
- Proxy authentication does not work.
- IP based authentication by the origin fails because the users are all seen to come from the Interception Cache's own IP address.
- You can't use IDENT lookups (which are inherently very insecure anyway)
- ARP relay breaks at the proxy machine.
- Interception Caching only supports the HTTP protocol, not gopher, SSL, or FTP. You cannot setup a redirection-rule to the proxy server for other protocols other than HTTP since the client will not know how to deal with it.
- Intercepting Caches are incompatible with IP filtering designed to prevent address spoofing.
- Clients are still expected to have full Internet DNS resolving capabilities; in certain intranet/firewalling setups, this is not always wanted.
- Related to above: suppose the users browser connects to a site which is down. However, due to the transparent proxying, it gets a connected state to the interceptor. The end user may get wrong error messages or a hung browser, for seemingly unknown reasons to them.
- DNS load is doubled, as clients do one DNS lookup, and the interception proxy repeats it.
- protocol tunnelling over the intercepted port 80 or 443 breaks.
WebSockets connectivity does not work.
- SPDY connectivity does not work (HTTPS interception proxy).
URL-rewriting and SSL-Bump forms of interception are usually not compatible. SSL-Bump generates a fake server certificate to match what the server presents. If URL-rewrite alters what sever is being contacted the client will receive wrong certificates. OR, attempting to re-write a HTTPS URL to http::// - the server will not present any SSL certificate. Both of these will result in user visible errors.
If you feel that the advantages outweigh the disadvantages in your network, you may choose to continue reading and look at implementing Interception Caching.
Requirements and methods for Interception Caching
- You need to have a good understanding of what you are doing before you start. This involves understanding at a TCP layer what is happening to the connections. This will help you both configure the system and additionally assist you if your end clients experience problems after you have deployed your solution.
- A current Squid (2.5+). You should run the latest version of Squid that is available at the time.
- A current OS may make things easier.
- Quite likely you will need a network device which can redirect the traffic to your cache. If your Squid box is also functioning as a router and all traffic from and to your network is in the path, you can skip this step. If your cache is a standalone box on a LAN that does not normally see your clients web browsing traffic, you will need to choose a method of redirecting the HTTP traffic from your client machines to the cache. This is typically done with a network appliance such as a router or Layer 3 switch which either rewrite the destination MAC address or alternatively encapsulate the network traffic via a GRE or WCCP tunnel to your cache.
NOTE: NAT configuration will only work when used on the squid box. This is required to perform intercept accurately and securely. To intercept from a gateway machine and direct traffic at a separate squid box use policy routing.
NB: If you are using Cisco routers and switches in your network you may wish to investigate the use of WCCP. WCCP is an extremely flexible way of redirecting traffic and is intelligent enough to automatically stop redirecting client traffic if your cache goes offline. This may involve you upgrading your router or switch to a release of IOS or an upgraded featureset which supports WCCP. There is a section written specifically on WCCP below.
Steps involved in configuring Interception Caching
- Building a Squid with the correct options to ./configure to support the redirection and handle the clients correctly.
- Routing the traffic from port 80 to the port your Squid is configured to accept the connections on
- Decapsulating the traffic that your network device sends to Squid (only if you are using GRE or WCCP to intercept the traffic)
- Configuring your network device to redirect the port 80 traffic.
The first two steps are required and the last two may or may not be required depending on how you intend to route the HTTP traffic to your cache.
!It is critical to read the full comments in the squid.conf file and in this document in it's entirety before you begin. Getting Interception Caching to work with Squid is non-trivial and requires many subsystems of both Squid and your network to be configured exactly right or else you will find that it will not work and your users will not be able to browse at all. You MUST test your configuration out in a non-live environment before you unleash this feature on your end users.
Compile a version of Squid which accepts connections for other addresses
Firstly you need to build Squid with the correct options to ./configure, and then you need to configure squid.conf to support Intercept Caching.
Choosing the right options to pass to ./configure
All supported versions of Squid currently available support Interception Caching, however for this to work properly, your operating system and network also need to be configured. For some operating systems, you need to have configured and built a version of Squid which can recognize the hijacked connections and discern the destination addresses.
For Linux configure Squid with the --enable-linux-netfilter option.
For *BSD-based systems with IP filter configure Squid with the --enable-ipf-transparent option.
If you're using OpenBSD's PF configure Squid with --enable-pf-transparent.
Do a make clean if you previously configured without that option, or the correct settings may not be present.
Squid-2.6+ and Squid-3.0+ support both WCCPv1 and WCCPv2 by default (unless explicitly disabled).
Configure Squid to accept and process the redirected port 80 connections
You have to change the Squid configuration settings to recognize the hijacked connections and discern the destination addresses.
A number of different interception methods and their specific configuration is detailed at ConfigExamples/Intercept
You can usually manually configure browsers to connect to the IP address and port which you have specified as intercepted. The only drawback is that there will be a very slight (and probably unnoticeable) performance hit as a syscall done to see if the connection is intercepted. If no interception state is found it is processed just like a normal connection.
Getting your traffic to the right port on your Squid Cache
You have to configure your cache host to accept the redirected packets - any IP address, on port 80 - and deliver them to your cache application. This is typically done with IP filtering/forwarding features built into the kernel.
On Linux 2.4 and above this is called iptables
On FreeBSD its called ipfw.
Other BSD systems may use ip filter, ipnat or pf.
On most systems, it may require rebuilding the kernel or adding a new loadable kernel module. If you are running a modern Linux distribution and using the vendor supplied kernel you will likely not need to do any rebuilding as the required modules will have been built by default.
Interception Caching packet redirection for Solaris, SunOS, and BSD systems
You don't need to use IP Filter on FreeBSD. Use the built-in ipfw feature instead. See the FreeBSD subsection below.
Install IP Filter
First, get and install the IP Filter package.
Put these lines in /etc/ipnat.rules:
# Redirect direct web traffic to local web server. rdr de0 188.8.131.52/32 port 80 -> 184.108.40.206 port 80 tcp # Redirect everything else to squid on port 8080 rdr de0 0.0.0.0/0 port 80 -> 220.127.116.11 port 8080 tcp
Modify your startup scripts to enable ipnat. For example, on FreeBSD it looks something like this:
/sbin/modload /lkm/if_ipl.o /sbin/ipnat -f /etc/ipnat.rules chgrp nobody /dev/ipnat chmod 644 /dev/ipnat
Thanks to Quinton Dolan.
Interception Caching packet redirection for OpenBSD PF
<After having compiled Squid with the options to accept and process the redirected port 80 connections enumerated above, either manually or with FLAVOR=transparent for /usr/ports/www/squid, one needs to add a redirection rule to pf (/etc/pf.conf). In the following example, sk0 is the interface on which traffic you want transparently redirected will arrive:
i = "sk0" rdr on $i inet proto tcp from any to any port 80 -> $i port 3128 pass on $i inet proto tcp from $i:network to $i port 3128
Or, depending on how recent your implementation of PF is:
i = "sk0" rdr pass on $i inet proto tcp to any port 80 -> $i port 3128
Also, see Daniel Hartmeier's page on the subject.
Get the packets from the end clients to your cache server
There are several ways to do this. First, if your proxy machine is already in the path of the packets (i.e. it is routing between your proxy users and the Internet) then you don't have to worry about this step as the Interception Caching should now be working. This would be true if you install Squid on a firewall machine, or on a UNIX-based router. If the cache is not in the natural path of the connections, then you have to divert the packets from the normal path to your cache host using a router or switch.
If you are using an external device to route the traffic to your Cache, there are multiple ways of doing this. You may be able to do this with a Cisco router using WCCP, or the "route map" feature. You might also use a so-called layer-4 switch, such as the Alteon ACE-director or the Foundry Networks ServerIron.
Finally, you might be able to use a stand-alone router/load-balancer type product, or routing capabilities of an access server.
Interception Caching packet redirection with Cisco routers using policy routing (NON WCCP)
This works with at least IOS 11.1 and later. If your router is doing anything more complicated that shuffling packets between an ethernet interface and either a serial port or BRI port, then you should work through if this will work for you.
First define a route map with a name of proxy-redirect (name doesn't matter) and specify the next hop to be the machine Squid runs on.
! route-map proxy-redirect permit 10 match ip address 110 set ip next-hop 18.104.22.168 !
Define an access list to trap HTTP requests. The second line allows the Squid host direct access so an routing loop is not formed. By carefully writing your access list as show below, common cases are found quickly and this can greatly reduce the load on your router's processor.
! access-list 110 deny tcp any any neq www access-list 110 deny tcp host 22.214.171.124 any access-list 110 permit tcp any any !
Apply the route map to the ethernet interface.
! interface FastEthernet0/0 ip policy route-map proxy-redirect !
Shortcomings of the cisco ip policy route-map method
Bruce Morgan notes that there is a Cisco bug relating to interception proxying using IP policy route maps, that causes NFS and other applications to break. Apparently there are two bug reports raised in Cisco, but they are not available for public dissemination.
The problem occurs with o/s packets with more than 1472 data bytes. If you try to ping a host with more than 1472 data bytes across a Cisco interface with the access-lists and ip policy route map, the icmp request will fail. The packet will be fragmented, and the first fragment is checked against the access-list and rejected - it goes the "normal path" as it is an icmp packet - however when the second fragment is checked against the access-list it is accepted (it isn't regarded as an icmp packet), and goes to the action determined by the policy route map!
John notes that you may be able to get around this bug by carefully writing your access lists. If the last/default rule is to permit then this bug would be a problem, but if the last/default rule was to deny then it won't be a problem. I guess fragments, other than the first, don't have the information available to properly policy route them. Normally TCP packets should not be fragmented, at least my network runs an MTU of 1500 everywhere to avoid fragmentation. So this would affect UDP and ICMP traffic only.
Basically, you will have to pick between living with the bug or better performance. This set has better performance, but suffers from the bug:
access-list 110 deny tcp any any neq www access-list 110 deny tcp host 10.1.2.3 any access-list 110 permit tcp any any
Conversely, this set has worse performance, but works for all protocols:
access-list 110 deny tcp host 10.1.2.3 any access-list 110 permit tcp any any eq www access-list 110 deny tcp any any
Interception Caching packet redirection with Foundry L4 switches
First, configure Squid for interception caching as detailed at the beginning of this section.
Next, configure the Foundry layer 4 switch to redirect traffic to your Squid box or boxes. By default, the Foundry redirects to port 80 of your squid box. This can be changed to a different port if needed, but won't be covered here.
In addition, the switch does a "health check" of the port to make sure your squid is answering. If you squid does not answer, the switch defaults to sending traffic directly thru instead of redirecting it. When the Squid comes back up, it begins redirecting once again.
This example assumes you have two squid caches:
squid1.foo.com 192.168.1.10 squid2.foo.com 192.168.1.11
We will assume you have various workstations, customers, etc, plugged into the switch for which you want them to be intercepted and sent to Squid. The squid caches themselves should be plugged into the switch as well. Only the interface that the router is connected to is important. Where you put the squid caches or ther connections does not matter.
This example assumes your router is plugged into interface 17 of the switch. If not, adjust the following commands accordingly.
- Enter configuration mode:
- Configure each squid on the Foundry:
telnet@ServerIron(config)# server cache-name squid1 192.168.1.10 telnet@ServerIron(config)# server cache-name squid2 192.168.1.11
- Add the squids to a cache-group:
telnet@ServerIron(config)#server cache-group 1 telnet@ServerIron(config-tc-1)#cache-name squid1 telnet@ServerIron(config-tc-1)#cache-name squid2
- Create a policy for caching http on a local port
telnet@ServerIron(config)# ip policy 1 cache tcp http local
- Enable that policy on the port connected to your router
telnet@ServerIron(config)#int e 17 telnet@ServerIron(config-if-17)# ip-policy 1
Since all outbound traffic to the Internet goes out interface 17 (the router), and interface 17 has the caching policy applied to it, HTTP traffic is going to be intercepted and redirected to the caches you have configured.
The default port to redirect to can be changed. The load balancing algorithm used can be changed (Least Used, Round Robin, etc). Ports can be exempted from caching if needed. Access Lists can be applied so that only certain source IP Addresses are redirected, etc. This information was left out of this document since this was just a quick howto that would apply for most people, not meant to be a comprehensive manual of how to configure a Foundry switch. I can however revise this with any information necessary if people feel it should be included.
Interception Caching packet redirection with an Alcatel OmnySwitch 7700
by Pedro A M Vazquez
On the switch define a network group to be intercepted:
policy network group MyGroup 10.1.1.0 mask 255.255.255.0
Define the tcp services to be intercepted:
policy service web80 destination tcp port 80 policy service web8080 destination tcp port 8080
Define a group of services using the services above:
policy service group WebPorts web80 web8080
And use these to create an intercept condition:
policy condition WebFlow source network group MyGroup service group WebPorts
Now, define an action to redirect the traffic to the host running squid:
policy action Redir alternate gateway ip 10.1.2.3
Finally, create a rule using this condition and the corresponding action:
policy rule Intercept condition WebFlow action Redir
Apply the rules to the QoS system to make them effective
Don't forget that you still need to configure Squid and Squid's operating system to handle the intercepted connections. See above for Squid and OS-specific details.
Interception Caching packet redirection with Cabletron/Entrasys products
By Dave Wintrip, dave at purevanity dot net, June 3, 2004.
I have verified this configuration as working on a Cabletron SmartSwitchRouter 2000, and it should work on any layer-4 aware Cabletron or Entrasys product.
You must first configure Squid to enable interception caching, outlined earlier.
Next, make sure that you have connectivity from the layer-4 device to your squid box, and that squid is correctly configured to intercept port 80 requests thrown it's way.
I generally create two sets of redirect ACLs, one for cache, and one for bypassing the cache. This method of interception is very similar to Cisco's route-map.
Log into the device, and enter enable mode, as well as configure mode.
ssr> en Password: ssr# conf ssr(conf)#
I generally create two sets of redirect ACLs, one for specifying who to cache, and one for destination addresses that need to bypass the cache. This method of interception is very similar to Cisco's route-map in this way. The ACL cache-skip is a list of destination addresses that we do not want to transparently redirect to squid.
ssr(conf)# acl cache-skip permit tcp any 192.168.1.100/255.255.255.255 any http
The ACL cache-allow is a list of source addresses that will be redirected to Squid.
ssr(conf)# acl cache-allow permit tcp 10.0.22.0/255.255.255.0 any any http
Save your new ACLs to the running configuration.
ssr(conf)# save a
Next, we need to create the ip-policies that will work to perform the redirection. Please note that 10.0.23.2 is my Squid server, and that 10.0.24.1 is my standard default next hop. By pushing the cache-skip ACL to the default gateway, the web request is sent out as if the squid box was not present. This could just as easily be done using the squid configuration, but I would rather Squid not touch the data if it has no reason to.
ssr(conf)# ip-policy cache-allow permit acl cache-allow next-hop-list 10.0.23.2 action policy-only ssr(conf)# ip-policy cache-skip permit acl cache-skip next-hop-list 10.0.24.1 action policy-only
Apply these new policies into the active configuration.
ssr(conf)# save a
We now need to apply the ip-policies to interfaces we want to cache requests from. Assuming that localnet-gw is the interface name to the network we want to cache requests from, we first apply the cache-skip ACL to intercept requests on our do-not-cache list, and forward them out the default gateway. We then apply the cache-allow ACL to the same interface to redirect all other requests to the cache server.
ssr(conf)# ip-policy cache-skip apply interface localnet-gw ssr(conf)# ip-policy cache-allow apply interface localnet-gw
We now need to apply, and permanently save our changes. Nothing we have done before this point would effect anything without adding the ip-policy applications into the active configuration, so lets try it.
ssr(conf)# save a ssr(conf)# save s
Provided your Squid box is correct configured, you should now be able to surf, and be transparently cached if you are using the localnet-gw address as your gateway.
Some Cabletron/Entrasys products include another method of applying a web cache, but details on configuring that is not covered in this document, however is it fairly straight forward.
Also note, that if your Squid box is plugged directly into a port on your layer-4 switch, and that port is part of its own VLAN, and its own subnet, if that port were to change states to down, or the address becomes uncontactable, then the switch will automatically bypass the ip-policies and forward your web request though the normal means. This is handy, might I add.
Interception Caching packet redirection with ACC Tigris digital access server
This is to do with configuring interception proxy for an ACC Tigris digital access server (like a CISCO 5200/5300 or an Ascend MAX 4000). I've found that doing this in the NAS reduces traffic on the LAN and reduces processing load on the CISCO. The Tigris has ample CPU for filtering.
Step 1 is to create filters that allow local traffic to pass. Add as many as needed for all of your address ranges.
ADD PROFILE IP FILTER ENTRY local1 INPUT 10.0.3.0 255.255.255.0 0.0.0.0 0.0.0.0 NORMAL ADD PROFILE IP FILTER ENTRY local2 INPUT 10.0.4.0 255.255.255.0 0.0.0.0 0.0.0.0 NORMAL
Step 2 is to create a filter to trap port 80 traffic.
ADD PROFILE IP FILTER ENTRY http INPUT 0.0.0.0 0.0.0.0 0.0.0.0 0.0.0.0 = 0x6 D= 80 NORMAL
Step 3 is to set the "APPLICATION_ID" on port 80 traffic to 80. This causes all packets matching this filter to have ID 80 instead of the default ID of 0.
SET PROFILE IP FILTER APPLICATION_ID http 80
Step 4 is to create a special route that is used for packets with "APPLICATION_ID" set to 80. The routing engine uses the ID to select which routes to use.
ADD IP ROUTE ENTRY 0.0.0.0 0.0.0.0 PROXY-IP 1 SET IP ROUTE APPLICATION_ID 0.0.0.0 0.0.0.0 PROXY-IP 80
Step 5 is to bind everything to a filter ID called transproxy. List all local filters first and the http one last.
ADD PROFILE ENTRY transproxy local1 local2 http
With this in place use your RADIUS server to send back the "Framed-Filter-Id = transproxy" key/value pair to the NAS.
You can check if the filter is being assigned to logins with the following command:
display profile port table
WCCP - Web Cache Coordination Protocol
WCCP is a very common and indeed a good way of doing Interception Caching as it adds additional features and intelligence to the traffic redirection process. WCCP is a dynamic service in which a cache engine communicates to a router about it's status, and based on that the router decides whether or not to redirect the traffic. This means that if your cache becomes unavailable, the router will automatically stop attempting to forward traffic to it and end users will not be affected (and likely not even notice that your cache is out of service).
For WCCP to work, you firstly need to configure your Squid Cache, and additionally configure the host OS to redirect the HTTP traffic from port 80 to whatever port your Squid box is listening to the traffic on. Once you have done this you can then proceed to configure WCCP on your router.
Does Squid support WCCP?
Cisco's Web Cache Coordination Protocol V1.0 and WCCPv2 are both supported in all current versions of Squid.
Do I need a cisco router to run WCCP?
No. Originally WCCP support could only be found on cisco devices, but some other network vendors now support WCCP as well. If you have any information on how to configure non-cisco devices, please post this here.
Can I run WCCP with the Windows port of Squid?
Technically it may be possible, but we have not heard of anyone doing so. The easiest way would be to use a Layer 3 switch and doing Layer 2 MAC rewriting to send the traffic to your cache. If you are using a router then you will need to find out a way to decapsulate the GRE/WCCP traffic that the router sends to your Windows cache (this is a function of your OS, not Squid).
Where can I find out more about WCCP?
Cisco have some good content on their website about WCCP. One of the better documents which lists the features and describes how to configure WCCP on their routers can be found on there website here.
There is also a more technical document which describes the format of the WCCP packets at Colasoft
Cisco router software required for WCCP
This depends on whether you are running a switch or a router.
IOS support in Cisco Routers
Almost all Cisco routers support WCCP provided you are running IOS release 12.0 or above, however some routers running older software require an upgrade to their software feature sets to a 'PLUS' featureset or better. WCCPv2 is supported on almost all routers in recent IPBASE releases.
Cisco's Feature Navigator at http://www.cisco.com/go/fn runs an up to date list of which platforms support WCCPv2.
Generally you should run the latest release train of IOS for your router that you can. We do not recommend you run T or branch releases unless you have fully tested them out in a test environment before deployment as WCCP requires many parts of IOS to work reliably. The latest mainline 12.1, 12.2, 12.3 and 12.4 releases are generally the best ones to use and should be the most trouble free.
Note that you will need to set up a GRE or WCCP tunnel on your cache to decapsulate the packets your router sends to it.
IOS support in Cisco Switches
High end Cisco switches support Layer 2 WCCPv2, which means that instead of a GRE tunnel transport, the ethernet frames have their next hop/destination MAC address rewritten to that of your cache engine. This is far faster to process by the hardware than the router/GRE method of redirection, and in fact on some platforms such as the 6500s may be the only way WCCP can be configured. L2 redirection is supposedly capable of redirecting in excess of 30 million PPS on the high end 6500 Sup cards.
Cisco switches known to be able to do WCCPv2 include the Catalyst 3550 (very basic WCCP only), Catalyst 4500-SUP2 and above, and all models of the 6000/6500.
Note that the Catalyst 2900, 3560, 3750 and 4000 early Supervisors do NOT support WCCP (depending from iOS version).
Layer 2 WCCP is a WCCPv2 feature and does not exist in cisco's WCCPv1 implementation.
WCCPv2 Layer 2 redirection was added in 12.1E and 12.2S.
It is always advisable to read the release notes for the version of software you are running on your switch before you deploy WCCP.
Software support in Cisco Firewalls (PIX OS)
Version 7.2(1) of the cisco PIX software now also supports WCCP, allowing you to do WCCP redirection with this appliance rather than having to have a router do the redirection.
7.2(1) has been tested and verified to work with Squid-2.6.
What about WCCPv2?
WCCPv2 is a new feature to Squid-2.6 and Squid-3.0. WCCPv2 configuration is similar to the WCCPv1 configuration. The directives in squid.conf are slightly different but are well documented within that file. Router configuration for WCCPv2 is identical except that you must not force the router to use WCCPv1 (it defaults to WCCPv2 unless you tell it otherwise).
Configuring your router
There are two different methods of configuring WCCP on Cisco routers. The first method is for routers that only support V1.0 of the protocol. The second is for routers that support both.
Cache/Host configuration of WCCP
There are two parts to this. Firstly you need to configure Squid to talk WCCP, and additionally you need to configure your operating system to decapsulate the WCCP traffic as it comes from the router.
Configuring Squid to talk WCCP
The configuration directives for this are well documented in squid.conf.
For WCCPv1, you need these directives:
wccp_router a.b.c.d wccp_version 4 wccp_incoming_address e.f.g.h wccp_outgoing_address e.f.g.h
- a.b.c.d is the address of your WCCP router
- e.f.g.h is the address that you want your WCCP requests to come and go from. If you are not sure or have only a single IP address on your cache, do not specify these.
Note: do NOT configure both the WCCPv1 directives (wccp_*) and WCCPv2 (wccp2_*) options at the same time in your squid.conf. Squid only supports configuration of one version at a time, either WCCPv1 or WCCPv2. With no configuration, the unconfigured version(s) are not enabled. Unpredictable things might happen if you configure both sets of options.
For WCCPv2, then you will want something like this:
wccp2_router a.b.c.d wccp2_version 4 wccp2_forwarding_method 1 wccp2_return_method 1 wccp2_service standard 0 wccp2_outgoing_address e.f.g.h
Use a wccp_forwarding_method and wccp2_return_method of 1 if you are using a router and GRE/WCCP tunnel, or 2 if you are using a Layer 3 switch to do the forwarding.
Your wccp2_service should be set to standard 0 which is the standard HTTP redirection.
- a.b.c.d is the address of your WCCP router
- e.f.g.h is the address that you want your WCCP requests to come and go from. If you are not sure or have only a single IP address on your cache, do not specify these parameters as they are usually not needed.
Now you need to read on for the details of configuring your operating system to support WCCP.
FreeBSD first needs to be configured to receive and strip the GRE encapsulation from the packets from the router. The steps depend on your kernel version.
FreeBSD 4.8 and later
The operating system now comes standard with some GRE support. You need to make a kernel with the GRE code enabled:
And then configure the tunnel so that the router's GRE packets are accepted:
# ifconfig gre0 create # ifconfig gre0 $squid_ip $router_ip netmask 255.255.255.255 up # ifconfig gre0 tunnel $squid_ip $router_ip # route delete $router_ip
Alternatively, you can try it like this:
ifconfig gre0 create ifconfig gre0 $squid_ip 10.20.30.40 netmask 255.255.255.255 link1 tunnel $squid_ip $router_ip up
Since the WCCP/GRE tunnel is one-way, Squid never sends any packets to 10.20.30.40 and that particular address doesn't matter.
FreeBSD 6.x and later
FreeBSD 6.x has GRE support in kernel by default. It also supports both WCCPv1 and WCCPv2. From gre(4) manpage: "Since there is no reliable way to distinguish between WCCP versions, it should be configured manually using the link2 flag. If the link2 flag is not set (default), then WCCP version 1 is selected." The rest of configuration is just as it was in 4.8+
Standard Linux GRE Tunnel
Linux versions earlier than 2.6.9 may need to be patched to support WCCP. That is why we strongly recommend you run a recent version of the Linux kernel, as if you are you simply need to modprobe the module to gain it's functionality.
Ensure that the GRE code is either built as static or as a module by chosing the appropriate option in your kernel config. Then rebuild your kernel. If it is a module you will need to:
The next step is to tell Linux to establish an IP tunnel between the router and your host.
ip tunnel add wccp0 mode gre remote <Router-External-IP> local <Host-IP> dev <interface> ip addr add <Host-IP>/32 dev wccp0 ip link set wccp0 up
or if using the older network tools
iptunnel add wccp0 mode gre remote <Router-External-IP> local <Host-IP> dev <interface> ifconfig wccp0 <Host-IP> netmask 255.255.255.255 up
<Router-External-IP> is the extrnal IP address of your router that is intercepting the HTTP packets. <Host-IP> is the IP address of your cache, and <interface> is the network interface that receives those packets (probably eth0).
Note that WCCP is incompatible with the rp_filter function in Linux and you must disable this if enabled. If enabled any packets redirected by WCCP and intercepted by Netfilter/iptables will be silendly discarded by the TCP/IP stack due to their "unexpected" origin from the gre interface.
echo 0 >/proc/sys/net/ipv4/conf/wccp0/rp_filter
And then you need to tell the Linux NAT kernel to redirect incoming traffic on the wccp0 interface to Squid
iptables -t nat -A PREROUTING -i wccp0 -j REDIRECT --redirect-to 3128
TProxy is a new feature in Squid-2.6 which enhances standard Interception Caching so that it further hides the presence of your cache. Normally with Interception Caching the remote server sees your cache engine as the source of the HTTP request. TProxy takes this a step further by hiding your cache engine so that the end client is seen as the source of the request (even though really they aren't).
Here are some notes by StevenWilton on how to get TProxy working properly:
I've got TProxy + WCCPv2 working with squid 2.6. There are a few things that need to be done:
- The kernel and iptables need to be patched with the tproxy patches (and the tproxy include file needs to be placed in /usr/include/linux/netfilter_ipv4/ip_tproxy.h or include/netfilter_ipv4/ip_tproxy.h in the squid src tree).
- The iptables rule needs to use the TPROXY target (instead of the REDIRECT target) to redirect the port 80 traffic to the proxy. ie:
iptables -t tproxy -A PREROUTING -i eth0 -p tcp -m tcp --dport 80 -j TPROXY --on-port 80
- The kernel must strip the GRE header from the incoming packets (either using the ip_wccp module, or by having a GRE tunnel set up in Linux pointing at the router (no GRE setup is required on the router)).
- Two WCCP services must be used, one for outgoing traffic and an inverse for return traffic from the Internet. We use the following WCCP definitions in squid.conf:
wccp2_service dynamic 80 wccp2_service_info 80 protocol=tcp flags=src_ip_hash priority=240 ports=80 wccp2_service dynamic 90 wccp2_service_info 90 protocol=tcp flags=dst_ip_hash,ports_source priority=240 ports=80
It is highly recommended that the above definitions be used for the two WCCP services, otherwise things will break if you have more than one cache (specifically, you will have problems when the name of a web server resolves to multiple ip addresses).
- The http port that you are redirecting to must have the transparent and tproxy options enabled as follows (modify the port as appropriate): http_port 80 transparent tproxy
There must be a tcp_outgoing address defined. This will need to be valid to satisfy any non-tproxied connections.
On the router, you need to make sure that all traffic going to/from the customer will be processed by both WCCP rules. The way we have
implemented this is to apply WCCP service 80 to all traffic coming in from a customer-facing interface, and WCCP service 90 applied to all traffic going out a customer-facing interface. We have also applied the WCCP exclude-in rule to all traffic coming in from the proxy-facing interface although this will probably not normally be necessary if all your caches have registered to the WCCP router. ie:
interface GigabitEthernet0/3.100 description ADSL customers encapsulation dot1Q 502 ip address x.x.x.x y.y.y.y ip wccp 80 redirect in ip wccp 90 redirect out interface GigabitEthernet0/3.101 description Dialup customers encapsulation dot1Q 502 ip address x.x.x.x y.y.y.y ip wccp 80 redirect in ip wccp 90 redirect out interface GigabitEthernet0/3.102 description proxy servers encapsulation dot1Q 506 ip address x.x.x.x y.y.y.y ip wccp redirect exclude in
- It's highly recommended to turn httpd_accel_no_pmtu_disc on in the squid.conf.
The homepage for the TProxy software is at balabit.com.
Starting with Squid 3.1 support for TProxy is closely tied into the netfilter component of Linux kernels. see TProxy v4.1 Feature for current details.
Other Configuration Examples
Contributed by users who have working installations can be found in the ConfigExamples/Intercept section for most current details.
If you have managed to configure your operating system to support WCCP with Squid please contact us or add the details to this wiki so that others may benefit.
By now if you have followed the documentation you should have a working Interception Caching system. Verify this by unconfiguring any proxy settings in your browser and surfing out through your system. You should see entries appearing in your access.log for the sites you are visiting in your browser. If your system does not work as you would expect, you will want to read on to our troubleshooting section below.
Troubleshooting and Questions
It doesn't work. How do I debug it?
- Start by testing your cache. Check to make sure you have configured Squid with the right configure options - squid -v will tell you what options Squid was configured with.
- Can you manually configure your browser to talk to the proxy port? If not, you most likely have a proxy configuration problem.
- Have you tried unloading ALL firewall rules on your cache and/or the inside address of your network device to see if that helps? If your router or cache are inadvertently blocking or dropping either the WCCP control traffic or the GRE, things won't work.
- If you are using WCCP on a cisco router or switch, is the router seeing your cache? Use the command show ip wccp web-cache detail
- Look in your logs both in Squid (cache.log), and on your router/switch where a show log will likely tell you if it has detected your cache engine registering.
On your Squid cache, set debug_options ALL,1 80,3 or for even more detail debug_options ALL,1 80,5 . The output of this will be in your cache.log.
- On your cisco router, turn on WCCP debugging:
router#term mon router#debug ip wccp events WCCP events debugging is on router#debug ip wccp packets WCCP packet info debugging is on router#
!Do not forget to turn this off after you have finished your debugging session as this imposes a performance hit on your router.
- Run tcpdump or ethereal on your cache interface and look at the traffic, try and figure out what is going on. You should be seeing UDP packets to and from port 2048 and GRE encapsulated traffic with TCP inside it. If you are seeing messages about "protocol not supported" or "invalid protocol", then your GRE or WCCP module is not loaded, and your cache is rejecting the traffic because it does not know what to do with it.
- Have you configured both wccp_ and wccp2_ options? You should only configure one or the other and NOT BOTH.
- The most common problem people have is that the router and cache are talking to each other and traffic is being redirected from the router but the traffic decapsulation process is either broken or (as is almost always the case) misconfigured. This is often a case of your traffic rewriting rules on your cache not being applied correctly (see section 2 above - Getting your traffic to the right port on your Squid Cache).
- Run the most recent General Deployment (GD) release of the software train you have on your router or switch. Broken IOS's can also result in broken redirection. A known good version of IOS for routers with no apparent WCCP breakage is 12.3(7)T12. There was extensive damage to WCCP in 12.3(8)T up to and including early 12.4(x) releases. 12.4(8) is known to work fine as long as you are not doing ip firewall inspection on the interface where your cache is located.
If none of these steps yield any useful clues, post the vital information including the versions of your router, proxy, operating system, your traffic redirection rules, debugging output and any other things you have tried to the squid-users mailing list.
Why can't I use authentication together with interception proxying?
Interception Proxying works by having an active agent (the proxy) where there should be none. The browser is not expecting it to be there, and it's for all effects and purposes being cheated or, at best, confused. As an user of that browser, I would require it not to give away any credentials to an unexpected party, wouldn't you agree? Especially so when the user-agent can do so without notifying the user, like Microsoft browsers can do when the proxy offers any of the Microsoft-designed authentication schemes such as NTLM (see ../ProxyAuthentication and Features/NegotiateAuthentication).
In other words, it's not a squid bug, but a browser security feature.
Can I use ''proxy_auth'' with interception?
No, you cannot. See the answer to the previous question. With interception proxying, the client thinks it is talking to an origin server and would never send the Proxy-authorization request header.
"Connection reset by peer" and Cisco policy routing
Fyodor has tracked down the cause of unusual "connection reset by peer" messages when using Cisco policy routing to hijack HTTP requests.
When the network link between router and the cache goes down for just a moment, the packets that are supposed to be redirected are instead sent out the default route. If this happens, a TCP ACK from the client host may be sent to the origin server, instead of being diverted to the cache. The origin server, upon receiving an unexpected ACK packet, sends a TCP RESET back to the client, which aborts the client's request.
To work around this problem, you can install a static route to the null0 interface for the cache address with a higher metric (lower precedence), such as 250.
Then, when the link goes down, packets from the client just get dropped instead of sent out the default route. For example, if 126.96.36.199 is the IP address of your Squid cache, you may add:
ip route 188.8.131.52 255.255.255.255 Null0 250
This appears to cause the correct behaviour.
Further information about configuring Interception Caching with Squid
DuaneWessels has written an O'Reilly book about Web Caching which is an invaluable reference guide for Squid (and in fact non-Squid) cache administrators. A sample chapter on "Interception Proxying and Caching" from his book is up online, at http://www.oreilly.com/catalog/webcaching/chapter/ch05.html.
Issues with HotMail
Hotmail has been known to suffer from the HTTP/1.1 Transfer Encoding problem. see http://squidproxy.wordpress.com/2008/04/29/chunked-decoding/ for more details on that and some solutions.
Back to the SquidFaq