Re: [squid-users] unexplainable MISSes (squid 2.7stable9)

On Mon, 08 Nov 2010 14:33:31 +0200, Adrian Dascalu
<adrian.dascalu_at_eea.europa.eu> wrote:
> Hi,
>
> I'm out of ideeas trying to debug cache misses that I cannot explain. As
a
> last resort I'm sending this problem to the list with the hope that you
> could come up with some explanation and/or cure for this.
>
> the setup is: squid 2.7stable9 on RHEL 5, configured as accel, 12
parents
> 1 sibling (another squid). Apache in front zope as parents.
>
> For the root page I send requests from the same browser. The page is
> supposed to stay in cache for 1h. I've seen it behaving correctly one
time
> (at squid startup) afterwards if i keep requesting the page a few times
I
> will get a MISS long before the 3600s have passed.
>
> I have checked and there is no PURGE for this URL in the mean time.
There
> are some for other URL's deeper in the structure.

Does not need to be an explicit PURGE. Merely a required alternative ETag,
or a force-reload request.

Can you please provide a new set of headers, for a given object initially
when its a HIT and afterwards when its a MISS.
NP: a full new HIT set is required to correlate exact times and tags, the
ones below are too old now to be reliably compared to any MISS.

>
> here's a request:
>
> Host www.somewebsite.com
>
> User-Agent Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.9)
> Gecko/20100330 Fedora/3.5.9-1.fc11
>
Firefox/3.5.9Accepttext/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
>
> Accept-Language en-us,en;q=0.5
>
> Accept-Encoding gzip,deflate
>
> Accept-Charset ISO-8859-1,utf-8;q=0.7,*;q=0.7
>
> Keep-Alive 300
>
> Connection keep-alive
>
> Referer http://www.somewebsite.com/
>
> Cookie
>
__utma=173508663.4134765344646281700.1250060356.1271487209.1289208944.50;
> __utmb=173508663.59.10.1289208944; __utmc=173508663;
>
__utmz=173508663.1289208944.50.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none)
>
> here's a HIT reply:
>
> Date Mon, 08 Nov 2010 12:06:29 GMT
>
> Server Zope/(Zope 2.9.10-final, python 2.4.3, linux2) ZServer/1.1
> Plone/2.5.5
>
> Content-Length 10131
>
> Content-Language en
>
> Content-Encoding gzip
>
> Expires Fri, 10 Nov 2000 12:05:48 GMT

Hmm, object is cacheable for *2 days*. Not the hour you said.

>
> Vary Accept-Encoding,Accept,If-None-Match,X-Username
>
> X-Caching-Rule-Id plone-containers
>
> Cache-Control max-age=0, s-maxage=3600

Irrelevant, Expires: header overrides these.

>
> Content-Type text/html;charset=utf-8
>
> X-Header-Set-Id cache-in-proxy-1-hour
>
> Age 40
>
> X-Cache HIT from squid1.somewebsite.com
>
> X-Cache-Lookup HIT from squid1.somewebsite.com:3128
>
> Via 1.0 squid1.somewebsite.com:3128 (squid/2.7.STABLE9)
>
> Keep-Alive timeout=8, max=100
>
> Connection Keep-Alive
>
> Long before the 3600s have passed ,from the same browser, I would get a
> MISS. The request headers are IDENTICAL and there is no PURGE. What else
> might invalidate the cached object?
>
>
> Thank you,
> Adrian
Received on Tue Nov 09 2010 - 02:25:03 MST