HTTP Compliance: Reply with an error if required validation failed
RFC 2616 says that proxy MUST not use stale entries that have s-maxage, 
proxy-revalidate, or must-revalidate cache-directive.
Add new fail_on_validation_err request flag to store result from 
refreshCheck().  It is needed to avoid refreshLimits() recalculation in 
clientReplyContext::handleIMSReply().
Split LOG_TCP_REFRESH_FAIL into LOG_TCP_REFRESH_FAIL_OLD (stale reply 
sent) and LOG_TCP_REFRESH_FAIL_ERR (error forwarded). Though both logged 
as TCP_REFRESH_FAIL for backward-compatibility with external scripts, etc.
Co-Advisor test cases:
     test_case/rfc2616/noSrv-hit-must-reval-s-maxage-resp
     test_case/rfc2616/noSrv-hit-must-reval-proxy-revalidate-resp
     test_case/rfc2616/noSrv-hit-must-reval-must-revalidate-resp
-------------
Is it a good idea to log new LOG_TCP_REFRESH_FAIL_OLD (stale reply sent) 
and LOG_TCP_REFRESH_FAIL_ERR (error forwarded) as 
"LOG_TCP_REFRESH_FAIL"? There are probably a lot of scripts out there 
that cannot easily handle new outcomes, but I do not know whether we 
should be kept hostage to them as the two outcomes are rather different. 
Will we have to use the old tag forever?
Thank you,
Alex.
This archive was generated by hypermail 2.2.0 : Tue Sep 21 2010 - 12:00:06 MDT