Re: Problem with squid interpretting gzipped content from Apache

Hi All,

I kind of have a temporary solution.
I maintain the httprequest object for that url requested.
I use this to request object to fetch me the actual entry from the cache
whenever requested.
So temporarily, my problem of wrong cache fetch is solved for objects that
are stored in a compressed format by squid.

I have a new question now:
1. I am expecting a race condition in my fix. I am currently forcing all
requests to go through the full prefetch cycle - i.e. it will not stop if
the object is already present in the cache.
Of course in the end of it, only the first request will record a miss while
the rest of it will record a HIT as the object is already present in the
cache for them. This I predict, will happen if the requests are processed
sequentially.

But what if there are simultaneous requests from 2 different clients?

For this, I want to know how the original squid code (3.1.16) handles this
case? Will it also face a race condition to fetch the object and update the
cache?
I may want to emulate this behaviour for my prefetch logic.

Thanks!

--
View this message in context: http://squid-web-proxy-cache.1019090.n4.nabble.com/Problem-with-squid-interpretting-gzipped-content-from-Apache-tp4659542p4659766.html
Sent from the Squid - Development mailing list archive at Nabble.com.