Intermediate iRules: Nested Conditionals
Conditionals are a pretty standard tool in every programmer's toolbox. They are the functions that allow us to decided when we want certain actions to happen, based on, well, conditions that can be determined within our code. This concept is as old as compilers. Chances are, if you're writing code, you're going to be using a slew of these things, even in an Event based language like iRules. iRules is no different than any other programming/scripting language when it comes to conditionals; we have them. Sure how they're implemented and what they look like change from language to language, but most of the same basic tools are there: if, else, switch, elseif, etc. Just about any example that you might run across on DevCentral is going to contain some example of these being put to use. Learning which conditional to use in each situation is an integral part to learning how to code effectively. Once you have that under control, however, there's still plenty more to learn. Now that you're comfortable using a single conditional, what about starting to combine them? There are many times when it makes more sense to use a pair or more of conditionals in place of a single conditional along with logical operators. For example: if { [HTTP::host] eq "bob.com" and [HTTP::uri] starts_with "/uri1" } { pool pool1 } elseif { [HTTP::host] eq "bob.com" and [HTTP::uri] starts_with "/uri2" } { pool pool2 } elseif { [HTTP::host] eq "bob.com" and [HTTP::uri] starts_with "/uri3" } { pool pool3 } Can be re-written to use a pair of conditionals instead, making it far more efficient. To do this, you take the common case shared among the example strings and only perform that comparison once, and only perform the other comparisons if that result returns as desired. This is more easily described as nested conditionals, and it looks like this: if { [HTTP::host] eq "bob.com" } { if {[HTTP::uri] starts_with "/uri1" } { pool pool1 } elseif {[HTTP::uri] starts_with "/uri2" } { pool pool2 } elseif {[HTTP::uri] starts_with "/uri3" } { pool pool3 } } These two examples are logically equivalent, but the latter example is far more efficient. This is because in all the cases where the host is not equal to "bob.com", no other inspection needs to be done, whereas in the first example, you must perform the host check three times, as well as the uri check every single time, regardless of the fact that you could have stopped the process earlier. While basic, this concept is important in general when coding. It becomes exponentially more important, as do almost all optimizations, when talking about programming in iRules. A script being executed on a server firing perhaps once per minute benefits from small optimizations. An iRule being executed somewhere in the order of 100,000 times per second benefits that much more. A slightly more interesting example, perhaps, is performing the same logical nesting while using different operators. In this example we'll look at a series of if/elseif statements that are already using nesting, and take a look at how we might use the switch command to even further optimize things. I've seen multiple examples of people shying away from switch when nesting their logic because it looks odd to them or they're not quite sure how it should be structured. Hopefully this will help clear things up. First, the example using if statements: when HTTP_REQUEST { if { [HTTP::host] eq "secure.domain.com" } { HTTP::header insert "Client-IP:[IP::client_addr]" pool sslServers } elseif { [HTTP::host] eq "www.domain.com" } { HTTP::header insert "Client-IP:[IP::client_addr]" pool httpServers } elseif { [HTTP::host] ends_with "domain.com" and [HTTP::uri] starts_with "/secure"} { HTTP::header insert "Client-IP:[IP::client_addr]" pool sslServers } elseif {[HTTP::host] ends_with "domain.com" and [HTTP::uri] starts_with "/login"} { HTTP::header insert "Client-IP:[IP::client_addr]" pool httpServers } elseif { [HTTP::host] eq "intranet.myhost.com" } { HTTP::header insert "Client-IP:[IP::client_addr]" pool internal } } As you can see, this is completely functional and would do the job just fine. There are definitely some improvements that can be made, though. Let's try using a switch statement instead of several if comparisons for improved performance. To do that, we're going to have to use an if nested inside a switch comparison. While this might be new to some or look a bit odd if you're not used to it, it's completely valid and often times the most efficient you’re going to get. This is what the above code would look like cleaned up and put into a switch: when HTTP_REQUEST { HTTP::header insert "Client-IP:[IP::client_addr]" switch -glob [HTTP::host] { "secure.domain.com" { pool sslServers } "www.domain.com" { pool httpServers } "*.domain.com" { if { [HTTP::uri] starts_with "/secure" } { pool sslServers } else { pool httpServers } } "intranet.myhost.com" { pool internal } } } As you can see this is not only easier to read and maintain, but it will also prove to be more efficient. We've moved to the more efficient switch structure, we've gotten rid of the repeat host comparisons that were happening above with the /secure vs /login uris, and while I was at it I got rid of all those examples of inserting a header, since that was happening in every case anyway. Hopefully the benefit this technique can offer is clear, and these examples did the topic some justice. With any luck, you'll nest those conditionals with confidence now.iRule Security 101 - #07 - FTP Proxy
We get questions all the time about custom application protocols and how one would go about writing an iRule to "understand" what's going on with that protocol. In this article, I will look at the FTP protocol and show you how one could write the logic to understand that application flow and selectively turn on and off support for various commands within the protocol. Other articles in the series: iRule Security 101 – #1 – HTTP Version iRule Security 101 – #02 – HTTP Methods and Cross Site Tracing iRule Security 101 – #03 – HTML Comments iRule Security 101 – #04 – Masking Application Platform iRule Security 101 – #05 – Avoiding Path Traversal iRule Security 101 – #06 – HTTP Referer iRule Security 101 – #07 – FTP Proxy iRule Security 101 – #08 – Limiting POST Data iRule Security 101 – #09 – Command Execution FTP FTP, for those who don't know, stands for File Transfer Protocol. FTP is designed to allow for the remote uploading and downloading of documents. I'm not going to dig deep into the protocol in this document, but for those who want to explore further, it is defined in RFC959. The basics of FTP are as follows. Requests are made with single line requests formatted as: COMMAND COMMAND_ARGS CRLF Some FTP commands include USER, PASS, & ACCT for authentication, CWD for changing directories, LIST for requesting the contents of a directory, and QUIT for terminating a session. Responses to commands are made in two ways. Over the main "control" connection, the server will process the request and then return a response in this format CODE DESCRIPTION CRLF Where code is the status code defined for the given request command. These have some similarity to HTTP response codes (200 -> OK, 500 -> Error), but don't count on them being exactly the same for each situation. For commands that do not requests content from the server (USER, PASS, CWD, etc), the control connection is all that is uses. But, there are other commands that specifically request data from the server. RETR (downloading a file), STOR (uploading a file), and LIST (for requesting a current directory listing) are examples of these types of commands. For these commands, the status is still returned in the control channel, but the data is passed back in a separate "data" channel that is configured by the client with either the PORT or PASV commands. Writing the Proxy We'll start of the iRule with a set of global variables that are used across all connections. In this iRule will will only inspect on the following FTP commands: USER, PASV, RETR, STOR, RNFR, FNTO, PORT, RMD, MKD, LIST, PWD, CWD, and DELE. This iRule can easily be expanded to include other commands in the FTP command set. In the RULE_INIT event we will set some global variables to determine how we want the proxy to handle the specific commands. A value of 1 for the "block" options will make the iRule deny those commands from reaching the backend FTP server. Setting a value of 0 for the block flag, will allow the command to pass through. when RULE_INIT { set DEBUG 1 #------------------------------------------------------------------------ # FTP Commands #------------------------------------------------------------------------ set sec_block_anonymous_ftp 1 set sec_block_passive_ftp 0 set sec_block_retr_cmd 0 set sec_block_stor_cmd 0 set sec_block_rename_cmd 0 set sec_block_port_cmd 0 set sec_block_rmd_cmd 0 set sec_block_mkd_cmd 0 set sec_block_list_cmd 0 set sec_block_pwd_cmd 0 set sec_block_cwd_cmd 0 set sec_block_dele_cmd 1 } Since we will not be relying on a BIG-IP profile to handle the application parsing, we'll be using the low level TCP events to capture the requests and responses. When a client establishes a connection, the CLIENT_ACCPETED event will occur, from within this event we'll have to trigger a collection of the TCP data so that we can inspect it in the CLIENT_DATA event. when CLIENT_ACCEPTED { if { $::DEBUG } { log local0. "client accepted" } TCP::collect TCP::release } In the CLIENT_DATA event, we will look at the request with the TCP::payload command. We will then feed that value into a switch statement with options for each of the commands. For commands that are found that we want to disallow, we will issue an FTP error response code with description string, empty out the payload, and return from the iRule - thus breaking the connection. For all other cases, we allow the TCP engine to continue on with it's processing and then enter into data collect mode again. when CLIENT_DATA { if { $::DEBUG } { log local0. "----------------------------------------------------------" } if { $::DEBUG } { log local0. "payload [TCP::payload]" } set client_data [string trim [TCP::payload]] #--------------------------------------------------- # Block or alert specific commands #--------------------------------------------------- switch -glob $client_data { "USER anonymous*" - "USER ftp*" { if { $::DEBUG } { log local0. "LOG: Anonymous login detected" } if { $::sec_block_anonymous_ftp } { TCP::respond "530 Guest user not allowed\r\n"; reject } } "PASV*" { if { $::DEBUG } { log local0. "LOG: passive request detected" } if { $::sec_block_passive_ftp } { TCP::respond "502 Passive commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "RETR*" { if { $::DEBUG } { log local0. "LOG: RETR request detected" } if { $::sec_block_retr_cmd } { TCP::respond "550 RETR commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "STOR*" { if { $::DEBUG } { log local0. "LOG: STOR request detected" } if { $::sec_block_stor_cmd } { TCP::respond "550 STOR commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "RNFR*" - "RNTO*" { if { $::DEBUG } { log local0. "LOG: RENAME request detected" } if { $::sec_block_rename_cmd } { TCP::respond "550 RENAME commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "PORT*" { if { $::DEBUG } { log local0. "LOG: PORT request detected" } if { $::sec_block_port_cmd } { TCP::respond "550 PORT commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "RMD*" { if { $::DEBUG } { log local0. "LOG: RMD request detected" } if { $::sec_block_rmd_cmd } { TCP::respond "550 RMD commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "MKD*" { if { $::DEBUG } { log local0. "LOG: MKD request detected" } if { $::sec_block_mkd_cmd } { TCP::respond "550 MKD commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "LIST*" { if { $::DEBUG } { log local0. "LOG: LIST request detected" } if { $::sec_block_list_cmd } { TCP::respond "550 LIST commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "PWD*" { if { $::DEBUG } { log local0. "LOG: PWD request detected" } if { $::sec_block_pwd_cmd } { TCP::respond "550 PWD commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "CWD*" { if { $::DEBUG } { log local0. "LOG: CWD request detected" } if { $::sec_block_cwd_cmd } { TCP::respond "550 CWD commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } "DELE*" { if { $::DEBUG } { log local0. "LOG: DELE request detected" } if { $::sec_block_dele_cmd } { TCP::respond "550 DELE commands not allowed\r\n" TCP::payload replace 0 [string length $client_data] "" return } } } TCP::release TCP::collect } Once a connection has been made to the backend server, the SERVER_CONNECTED event will be raised. In this event we will release the context and issue a collect to occur for the server data. The server data will then be returned, and optionally logged, in the SERVER_DATA event. when SERVER_CONNECTED { if { $::DEBUG } { log "server connected" } TCP::release TCP::collect } when SERVER_DATA { if { $::DEBUG } { log local0. "payload <[TCP::payload]>" } TCP::release TCP::collect } And finally when the client closes it's connection,. the CLIENT_CLOSED event will be fired and we will log the fact that the session is over. when CLIENT_CLOSED { if { $::DEBUG } { log local0. "client closed" } } Conclusion This article shows how one can use iRules to inspect, and optionally secure, an application based on command sets within that application. Not all application protocols behave like FTP (TELNET for instance sends one character at a time and it's up to the proxy to consecutively request more data until the request is complete). But this should give you the tools you need to start inspection on your TCP based application. Get the Flash Player to see this player.iRules 101 - #12 - The Session Command
One of the things that makes iRules so incredibly powerful is the fact that it is a true scripting language, or at least based on one. The fact that they give you the tools that TCL brings to the table - regular expressions, string functions, even things as simple as storing, manipulating and recalling variable data - sets iRules apart from the rest of the crowd. It also makes it possible to do some pretty impressive things with connection data and massaging/directing it the way you want it. Other articles in the series: Getting Started with iRules: Intro to Programming with Tcl | DevCentral Getting Started with iRules: Control Structures & Operators | DevCentral Getting Started with iRules: Variables | DevCentral Getting Started with iRules: Directing Traffic | DevCentral Getting Started with iRules: Events & Priorities | DevCentral Intermediate iRules: catch | DevCentral Intermediate iRules: Data-Groups | DevCentral Getting Started with iRules: Logging & Comments | DevCentral Advanced iRules: Regular Expressions | DevCentral Getting Started with iRules: Events & Priorities | DevCentral iRules 101 - #12 - The Session Command | DevCentral Intermediate iRules: Nested Conditionals | DevCentral Intermediate iRules: Handling Strings | DevCentral Intermediate iRules: Handling Lists | DevCentral Advanced iRules: Scan | DevCentral Advanced iRules: Binary Scan | DevCentral Sometimes, though, a simple variable won't do. You've likely heard of global variables in one of the earlier 101 series and read the warning there, and are looking for another option. So here you are, you have some data you need to store, which needs to persist across multiple connections. You need it to be efficient and fast, and you don't want to have to do a whole lot of complex management of a data structure. One of the many ways that you can store and access information in your iRule fits all of these things perfectly, little known as it may be. For this scenario I'd recommend the usage of the session command. There are three main permutations of the session command that you'll be using when storing and referencing data within the session table. These are: session add: Stores user's data under the specified key for the specified persistence mode session lookup: Returns user data previously stored using session add session delete: Removes user data previously stored using session add A simple example of adding some information to the session table would look like: when CLIENTSSL_CLIENTCERT { set ssl_cert [SSL::cert 0] session add ssl $ssl_cert 90 } By using the session add command, you can manually place a specific piece of data into the LTM's session table. You can then look it up later, by unique key, with the session lookup command and use the data in a different section of your iRule, or in another connection all together. This can be helpful in different situations where data needs to be passed between iRules or events that it might not normally be when using a simple variable. Such as mining SSL data from the connection events, as below: when CLIENTSSL_CLIENTCERT { # Set results in the session so they are available to other events session add ssl [SSL::sessionid] [list [X509::issuer] [X509::subject] [X509::version]] 180 } when HTTP_REQUEST { # Retrieve certificate information from the session set sslList [session lookup ssl [SSL::sessionid]] set issuer [lindex sslList 0] set subject [lindex sslList 1] set version [lindex sslList 2] } Because the session table is optimized and designed to handle every connection that comes into the LTM, it's very efficient and can handle quite a large number of items. Also note that, as above, you can pass structured information such as TCL Lists into the session table and they will remain intact. Keep in mind, though, that there is currently no way to count the number of entries in the table with a certain key, so you'll have to build all of your own processing logic for now, where necessary. It's also important to note that there is more than one session table. If you look at the above example, you'll see that before we listed any key or data to be stored, we used the command session add ssl. Note the "ssl" portion of this command. This is a reference to which session table the data will be stored in. For our purposes here there are effectively two session tables: ssl, and uie. Be sure you're accessing the same one in your session lookup section as you are in your session add section, or you'll never find the data you're after. This is pretty easy to keep straight, once you see it. It looks like: session add uie ... session lookup uie Or: session add ssl ... session lookup ssl You can find complete documentation on the session command here, in the iRules, as well as some great examplesthat depict some more advanced iRules making use of the session command to great success. Check out Codeshare for more examples.Getting Started with Bigsuds–a New Python Library for iControl
I imagine the progression for you, the reader, will be something like this in the first six- or seven-hundred milliseconds after reading the title: Oh cool! Wait, what? Don’t we already have like two libraries for python? Really, a third library for python? Yes. An emphatic yes. The first iteration of pycontrol (pc1) was based on the zsi library, which hasn’t been updated in years and was abandoned with the development of the second iteration, pycontrol v2 (pc2), which switched to the active and well-maintained suds library. Bigsuds, like pycontrol v2, is also based on the suds library. So why bigsuds? There are several advantages to using the bigsuds library. No need to specify which WSDLs to download In pycontrol v2, any iControl interface you wish to work with must be specified when you instantiate the BIG-IP, as well as specifying the local directory or loading from URL for the WSDLs. In bigsuds, just specify the host, username, and password (username and password optional if using test box defaults of admin/admin) and you’re good to go. Currently in pycontrol v2: >>> import pycontrol.pycontrol as pc >>> b = pc.BIGIP( ... hostname = '192.168.6.11', ... username = 'admin', ... password = 'admin', ... fromurl = True, ... wsdls = ['LocalLB.Pool']) >>> b.LocalLB.Pool.get_list() [/Common/p1, /Common/p2, /Common/p3, /Common/p5] And here in bigsuds: >>> import bigsuds >>> b = bigsuds.BIGIP(hostname = '192.168.6.11') >>> b.LocalLB.Pool.get_list() ['/Common/p1', '/Common/p2', '/Common/p3', '/Common/p5'] >>> b.GlobalLB.Pool.get_list() ['/Common/p2', '/Common/p1'] No need to define the typefactory for write operations. This was the most challenging aspect of pycontrol v2 for me personally. I would get them correct sometimes. Often I’d bang my head against the wall wondering what little thing I missed to prevent success. The cool thing with bigsuds is you are just passing lists for sequences and lists of dictionaries for structures. No object creation necessary before making the iControl calls. It’s a thing of beauty. Creating a two member pool in pycontrol v2: lbmeth = b.LocalLB.Pool.typefactory.create('LocalLB.LBMethod') # This is basically a stub holder of member items that we need to wrap up. mem_sequence = b.LocalLB.Pool.typefactory.create('Common.IPPortDefinitionSequence') # Now we'll create some pool members. mem1 = b.LocalLB.Pool.typefactory.create('Common.IPPortDefinition') mem2 = b.LocalLB.Pool.typefactory.create('Common.IPPortDefinition') # Note how this is 'pythonic' now. We set attributes agains the objects, then # pass them in. mem1.address = '1.2.3.4' mem1.port = 80 mem2.address = '1.2.3.4' mem2.port = 81 # Create a 'sequence' of pool members. mem_sequence.item = [mem1, mem2] # Let's create our pool. name = 'PC2' + str(int(time.time())) b.LocalLB.Pool.create(pool_names = [name], lb_methods = \ [lbmeth.LB_METHOD_ROUND_ROBIN], members = [mem_sequence]) In contrast, here is a two member pool in bigsuds. >>> b.LocalLB.Pool.create_v2(['/Common/Pool1'],['LB_METHOD_ROUND_ROBIN'],[[{'port':80, 'address':'1.2.3.4'},{'port':81, 'address':'1.2.3.4'}]]) Notice above that I did not use the method parameters. They are not required in bigsuds, though you can certainly include them. This could be written in the long form as: >>> b.LocalLB.Pool.create_v2(pool_names = ['/Common/Pool1'],lb_methods = ['LB_METHOD_ROUND_ROBIN'], members = [[{'port':80, 'address':'1.2.3.4'},{'port':81, 'address':'1.2.3.4'}]]) Standard python data types are returned There’s no more dealing with data returned like this: >>> p2.LocalLB.Pool.get_statistics(pool_names=['/Common/p2']) (LocalLB.Pool.PoolStatistics){ statistics[] = (LocalLB.Pool.PoolStatisticEntry){ pool_name = "/Common/p2" statistics[] = (Common.Statistic){ type = "STATISTIC_SERVER_SIDE_BYTES_IN" value = (Common.ULong64){ high = 0 low = 0 } time_stamp = 0 }, (Common.Statistic){ type = "STATISTIC_SERVER_SIDE_BYTES_OUT" value = (Common.ULong64){ high = 0 low = 0 } time_stamp = 0 }, Data is standard python types: strings, lists, dictionaries. That same data returned by bigsuds: >>> b.LocalLB.Pool.get_statistics(['/Common/p1']) {'statistics': [{'pool_name': '/Common/p1', 'statistics': [{'time_stamp': 0, 'type': 'STATISTIC_SERVER_SIDE_BYTES_IN', 'value': {'high': 0, 'low': 0}}, {'time_stamp': 0, 'type': 'STATISTIC_SERVER_SIDE_BYTES_OUT', 'value': {'high': 0, 'low': 0}} Perhaps not as readable in this form as with pycontrol v2, but far easier to work programmatically. Better session and transaction support George covered the benefits of sessions in his v11 iControl: Sessions article in fine detail, so I’ll leave that to the reader. Regarding implementations, bigsuds handles sessions with a built-in utility called with_session_id. Example code: >>> bigip2 = b.with_session_id() >>> bigip2.System.Session.set_transaction_timeout(99) >>> print b.System.Session.get_transaction_timeout() 5 >>> print bigip2.System.Session.get_transaction_timeout() 99 Also, with transactions, bigsuds has built-in transaction utilities as well. In the below sample code, creating a new pool that is dependent on a non-existent pool being deleted results in an error as expected, but also prevents the pool from the previous step from being created as show in the get_list method call. >>> try: ... with bigsuds.Transaction(bigip2): ... bigip2.LocalLB.Pool.create_v2(['mypool'],['LB_METHOD_ROUND_ROBIN'],[[]]) ... bigip2.LocalLB.Pool.delete_pool(['nonexistent']) ... except bigsuds.OperationFailed, e: ... print e ... Server raised fault: 'Exception caught in System::urn:iControl:System/Session::submit_transaction() Exception: Common::OperationFailed primary_error_code : 16908342 (0x01020036) secondary_error_code : 0 error_string : 01020036:3: The requested pool (/Common/nonexistent) was not found.' >>> bigip2.LocalLB.Pool.get_list() ['/Common/Pool1', '/Common/p1', '/Common/p2', '/Common/p3', '/Common/p5', '/Common/Pool3', '/Common/Pool2'] F5 maintained Community member L4L7, the author of the pycontrol v2 library, is no longer with F5 and just doesn’t have the cycles to maintain the library going forward. Bigsuds author Garron Moore, however, works in house and will fix bugs and enhance as time allows. Note that all iControl libraries are considered experimental and are not officially supported by F5 Networks. Library maintainers for all the languages will do their best to fix bugs and introduce features as time allows. Source is provided though, and bugs can and are encouraged to be fixed by the community! Installing bigsuds Make sure you have suds installed and grab a copy of bigsuds (you’ll need to log in) and extract the contents. You can use the easy setup tools to install it to python’s site-packages library like this: jrahm@jrahm-dev:/var/tmp$ tar xvfz bigsuds-1.0.tar.gz bigsuds-1.0/ bigsuds-1.0/setup.py bigsuds-1.0/bigsuds.egg-info/ bigsuds-1.0/bigsuds.egg-info/top_level.txt bigsuds-1.0/bigsuds.egg-info/requires.txt bigsuds-1.0/bigsuds.egg-info/SOURCES.txt bigsuds-1.0/bigsuds.egg-info/dependency_links.txt bigsuds-1.0/bigsuds.egg-info/PKG-INFO bigsuds-1.0/setup.cfg bigsuds-1.0/bigsuds.py bigsuds-1.0/MANIFEST.in bigsuds-1.0/PKG-INFO jrahm@jrahm-dev:/var/tmp$ cd bigsuds-1.0/ jrahm@jrahm-dev:/var/tmp/bigsuds-1.0$ python setup.py install Doing it that way, you can just enter the python shell (or run your script) with a simple ‘import bigsuds’ command. If you don’t want to install it that way, you can just extract the bigsuds.py from the download and drop it in a directory of your choice and make a path reference in the shell or script: >>> import bigsuds Traceback (most recent call last): File "<stdin>", line 1, in <module> ImportError: No module named bigsuds >>> import sys >>> sys.path.append(r'/home/jrahm/dev/bigsuds-1.0') >>> import bigsuds >>> Conclusion Garron Moore's bigsuds contribution is a great new library for python users. There is work to be done to convert your pycontrol v2 samples, but the flexibility and clarity in the new library makes it worth it in this guy’s humble opinion. A new page in the iControl wiki has been created for bigsuds developments. Please check it out, community! For now, I’ve converted a few scripts to bigsuds, linked in the aforementioned page as well as directly below: Get GTM Pool Status Get LTM Pool Status Get or Set GTM Pool TTL Create or Modify an LTM PooliRule Security 101 - #06 - HTTP Referer
In this article, I'm going to talk about the HTTP "Referer" header, how it's used, and how you can use iRules to ensure that an access request to a website is coming from where you want it to come from. Other articles in the series: iRule Security 101 – #1 – HTTP Version iRule Security 101 – #02 – HTTP Methods and Cross Site Tracing iRule Security 101 – #03 – HTML Comments iRule Security 101 – #04 – Masking Application Platform iRule Security 101 – #05 – Avoiding Path Traversal iRule Security 101 – #06 – HTTP Referer iRule Security 101 – #07 – FTP Proxy iRule Security 101 – #08 – Limiting POST Data iRule Security 101 – #09 – Command Execution First, let me say that I know that "Referer" is misspelled. For some reason, the authors of the HTTP specification (RFC 2616, section 14.36) didn't run a spell checker on the specification and now that every browser and web server has implemented this with the wrong spelling it's too late to change it. Take a look at the definition for it on dictionary.com and you'll see for yourself. Nothing like a dictionary 'dissing an Internet spec... Once you can get past the misspelling, the HTTP "Referer" header is defined as the following (RFC 2616, section 14.36) 14.36 Referer The Referer[sic] request-header field allows the client to specify, for the server's benefit, the address (URI) of the resource from which the Request-URI was obtained (the "referrer", although the header field is misspelled.) The Referer request-header allows a server to generate lists of back-links to resources for interest, logging, optimized caching, etc. It also allows obsolete or mistyped links to be traced for maintenance. The Referer field MUST NOT be sent if the Request-URI was obtained from a source that does not have its own URI, such as input from the user keyboard. So basically, when ever you click on a link from a website causing new HTTP request to be made, the URI of the website you are on will be passed in the HTTP request in the form of a HTTP header with the name of "Referer" and a value containing the source URI. Why is this important from a security perspective? I'll give just one example attack and a way to use Referer headers to help block against it. With the massive uptake of blogging by users on the Internet, comments are a useful way to get feedback on your ideas. Unfortunately blog spam, as it's called, has been on the rise. Several ways have been developed to protect against blog spam including comment moderation, CAPTCHA (you know, when you type out the text that is displayed in randomly generated images), as well as online dynamic services such as Akismet that process the content of comments in a very similar way to common email SPAM services. CAPTCHA is the most common form of defense but it is not fool proof and spammers have found ways to build programs to defeat this system. So how does this fit with Referers? I'll get to that in just a minute... If you set a policy on your blog that only the comment form can be accessed by clicking on a feedback link on your blog, then you can make use of this fact by denying all requests that do not contain the URI of your blog post in the Referer header. Sure, there are ways to bypass this since HTTP headers are easily programmed into any HTTP client program. But, there are ways to trick the client into thinking that the post succeeded when it really didn't. Let's take a look at an example. http://www.mycoolblog.com/ - blog site in question http://www.mycoolblog.com/first_post - blog post page that is to be commented on. http://www.mycoolblog.com/PostComment.aspx - Comment post form. Legitimate commenter's will first vist the blog post page and then fill in the comment information and submit it to the PostComment.aspx form. Spammers will try to bypass this step by pulling in these images into a client program, try to determine the CAPTCHA image's text, and then formulate a HTTP POST command directly to the PostComment.aspx page. By enforcing that a Referer header from the same blog site comes in the PostComment.aspx request, we can block out those spammers. when HTTP_REQUEST { switch -glob [HTTP::header "Referer"] { "http://www.mycoolblog.com/*" { # Allow Request to go through... } "" { HTTP::respond 200 content "" } default { HTTP::redirect [HTTP::header "Referer"] } } } Basically any request coming from http://www.mycoolblog.com will be allowed through. Any request with a empty Referer header will be immediately returned with a HTTP 200 response to trick the client that a successful attempt was made, and any other Referer's will be redirected back to the referral site. Caveats: This is by far not a universal blog spam solution as each blogging engine handles comments differently. Some have a different URI for comment posting (as illustrated above) and others use POST data values on the same application page as the blog posting to indicate comment submissions. Also, it is easy for clients to spoof referer values by manually adding the header in the requests. But, it is a good start for those automated bots out there that are just searching for blogs to send their unwanted content to. Also, this solution does not support Trackback/Pingback spam as those solutions typically are programmatic submissions from references in other blogs. Conclusions: Blog Spam was just an example of the type of application security issue that could be addressed by making use of the HTTP Referer header. Hopefully this article has provided some food for thought into how you can use the Referer header to your advantage in protecting your applications. Get the Flash Player to see this player.
