Apache mod_cache and mod_deflate

apache-2.2http-headersmod-cachemod-deflate

We have a Apache 2.2 with mod_cache and mod_deflate among other modules. The problem is that if we append Vary header as in Apache documentation…

# Make sure proxies don't deliver the wrong content
Header append Vary User-Agent env=!dont-vary

… we end up having copy of the served resource for each user agent variation in our cache. This wastes disk space and lowers the hit rate.

So, what is the preferred solution for this problem? Ditch the Vary header and compress only "safe" resources like plain html?

Best Answer

The reason for setting the header to vary by user agent is that Apache's recommended configuration for mod_deflate serves uncompressed versions of some content to Netscape 4 users. There are probably few enough Netscape 4 users now that you can just serve compressed content to all browsers that claim to support it, and not vary by user agent.

So instead of the recommended configuration:

<Location />
# Insert filter
SetOutputFilter DEFLATE

# Netscape 4.x has some problems...
BrowserMatch ^Mozilla/4 gzip-only-text/html

# Netscape 4.06-4.08 have some more problems
BrowserMatch ^Mozilla/4\.0[678] no-gzip

# MSIE masquerades as Netscape, but it is fine
BrowserMatch \bMSIE !no-gzip !gzip-only-text/html
# Don't compress images
SetEnvIfNoCase Request_URI \
\.(?:gif|jpe?g|png)$ no-gzip dont-vary

# Make sure proxies don't deliver the wrong content
Header append Vary User-Agent env=!dont-vary
</Location>

You can just have:

<Location />
# Insert filter
SetOutputFilter DEFLATE

# Don't compress images
SetEnvIfNoCase Request_URI \
\.(?:gif|jpe?g|png)$ no-gzip dont-vary
</Location>

Related Solutions

Windows – Content Encoding Error when trying to use GZip on Apache

The first issue is that you have the element with a directory from the file system...

This is from the Apache manual:

When to use <Location>. Use <Location> to apply directives to content that lives outside the filesystem. For content that lives in the filesystem, use <Directory> and <Files>. An exception is <Location />, which is an easy way to apply a configuration to the entire server.

When I setup mod_deflate on Apache I use this setup:

# --------------------------------------------------------
# Deflate Module Configuration
# --------------------------------------------------------
<IfModule deflate_module>
  # Set the location to webservers root
  <Location />
    # Insert the filters
    SetOutputFilter DEFLATE

    # Netscape 4.x has some problems
    BrowserMatch ^Mozilla/4 gzip-only-text/html
    # Netscape 4.06-4.08 have some more problems
    BrowserMatch ^Mozilla/4.0[678] no-gzip
    # MSIE masquerades as Netscape, but it is fine
    BrowserMatch \bMSIE !no-gzip !gzip-only-text/html

    # Make sure proxies don't deliver the wrong content
    Header append Vary User-Agent env=!dont-vary

    # Make sure to never zip binaries (ie6 and chrome has issues with this.)
    SetEnvIfNoCase Request_URI \.(?:gif|jpe?g|png)$ no-gzip dont-vary
    SetEnvIfNoCase Request_URI \.(?:exe|t?gz|zip|bz2|sit|rar)$ no-gzip dont-vary
    SetEnvIfNoCase Request_URI \.pdf$ no-gzip dont-vary
  </Location>

    # Enable mod_deflate logging by uncommenting these lines
    #DeflateFilterNote Input instream
    #DeflateFilterNote Output outstream
    #DeflateFilterNote Ratio ratio

    #LogFormat '"%r" %{outstream}n/%{instream}n (%{ratio}n%%)' deflate
    #CustomLog logs/deflate_log deflate 
</IfModule>

Is the Last-Modified HTTP header required for browser caching

Be wary of reading random articles on the net (although Mark Nottingham's stuff is usually sensible). The definitive source should always by the RFCs. And according to RFC 2616 a browser should cache a document with an Expires: header where the timestamp is in the future, or where other valid caching instructions are provided provided that the document is not returned in response to a POST request.

It's perfectly valid to set a max-age without a last-modified - and the spec explicitly addresses that.

Certainly what you describe seems very unusual and implies that FF4.01 will never cache content - I would be amazed that it passed the QC checks with such a glaring omission. Can you provide details of the requests and responses proving this (e.g. with liveheaders)?

Best Answer

Related Solutions

Windows – Content Encoding Error when trying to use GZip on Apache

Is the Last-Modified HTTP header required for browser caching

Related Topic