On Wed, 6 May 1998, Ian Simpson wrote:
> if cache_stoplist forces certain sites never to be cached, how come I get
> an error with a cgi script from the apache log files telling me that access
> to the file for the squid cache failed. eg:
>
> [Wed May 6 15:25:32 1998] access to /usr/httpd/html/cgi-bin/uncgi failed
> for squid.cache.net. reason: Premature end of script headers
Hang on... think you're confusing some things here.
cache_stoplist tells squid to not *store* urls matching these, ie cgi-bin.
hierarchy_stoplist tells squid to not ask any of its known
siblings/parents for the object in question.
Neither tell squid to *avoid* getting the object, merely how to get it and
what to do with it afterwards (throw it away).
The question you've asked is more of 'why is my cgi script failing - look,
I have an example of it being requested from my cache'.
The fact that apache is showing that the cgi-bin was requested from the
proxy shows that the proxy is working. As Duane has pointed out, your cgi
looks to be failing, perhaps all the time, perhaps only when squid asks
for it. either way, not related to cache_stoplist.
> In cache_stoplist I have "cgi-bin", "uncgi" etc.
>
> Can I force the bloody cgi script to completely bypass the cache?
the cgi script? then how does it know how to pass the data back to the
client? the client itself? sure, tell it not to use the cache for
anything on that site.
hope dis helps ;)
--==--
Bruce.
Received on Thu May 07 1998 - 03:28:08 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:40:06 MST