these scripts are what i use to permanently cache downloads from microsoft (including windows update). note that this does not "clone" windowsupdate, rather it only stops you re-downloading files that have already been fetched (ie. it's a cache not a _clone_).
if you're a windows house, you may want to play with sus server instead.
another alternative is to put the following lines in your squid.conf:
refresh_pattern windowsupdate.com/.*\.(cab|exe) 4320 100% 43200 reload-into-ims refresh_pattern download.microsoft.com/.*\.(cab|exe) 4320 100% 43200 reload-into-ims refresh_pattern au.download.windowsupdate.com/.*\.(cab|exe) 4320 100% 43200 reload-into-ims
note that i haven't tested this.
last updated 6 nov 2006
you'll need squid, apache and perl. and you'll need to know how to configure squid and apache.
you'll also need wget and the perl Date::Parse module (part of the TimeDate package).
there's three scripts
redir.pl- redirector script for squid
download- called by redir.pl
check_store- nightly cron job to stop stale and duplicate entries
the flow of redir.pl is ..
if the request is for a microsoft cab/exe if the file doesn't exist in the store spawn the download script to download into the store else rewrite the url to redirect to the store endif endif
create a directory under your apache's htdocs area, and then a directory called
storeunder that. i use:
modify paths in
$cache_path= path to the store
$cache_url= internal url to the store
$downloader= full path to the download script
configure squid to use
set up a cron job to run
the download script will need permissions to create a file called
(it will run as the same user that runs squid). it also needs permissions
to write to the
there's a bug in the version of autoupdate that ships with a default winxp build – it will go into a loop fetching the same file over and over. make sure you manually upgrade to at least SP1a before running windowsupdate on xp boxes. (large thanks to chris hoypoy for figuring that one out).