On Wed, Sep 27, 2006, Adrian Chadd wrote:
> On Wed, Sep 27, 2006, Adrian Chadd wrote:
> > On Wed, Sep 27, 2006, Adrian Chadd wrote:
> >
> > > If Temp2 is around the while() loop then it matches the runtime for the
> > > comm close handlers within half a percent - ie, 13.13%.
>
> 4% of that 13% is
On Wed, Sep 27, 2006, Adrian Chadd wrote:
> On Wed, Sep 27, 2006, Adrian Chadd wrote:
>
> > If Temp2 is around the while() loop then it matches the runtime for the
> > comm close handlers within half a percent - ie, 13.13%.
4% of that 13% is being tied up in ClientHttpRequest::logRequest(). No, I
On Wed, Sep 27, 2006, Adrian Chadd wrote:
> If Temp2 is around the while() loop then it matches the runtime for the
> comm close handlers within half a percent - ie, 13.13%.
>
> Where it is there, it only accounts for 2.18% of the CPU.
Its because its refcounted. The call to connIsFinished() isn
Hiya,
I'm profiling squid-3 and I'm trying to quell the heavy CPU users.
Specifically, I'm trying to find why comm_close() is profiling as taking
~19% of the CPU.
Bout 3% is due to close(); which is annoying but understandable.
The other 16%? Good question.
I ended up here.
void
ConnStateData:
Thanks,
I 'll try to help
Regards,
Christos
>
> Please do! :)
>
> Feel free to send squid-dev whats broken with the ICAP code.
> You're also welcome in the #squiddev channel on Freenode IRC where you'll
> find us trying to squish bugs.
>
>
>
> adrian
>
On Wed, Sep 27, 2006, Tsantilas Christos wrote:
> Hi all,
>
> Now the ICAP client in squid3 does not work very well. I am testing it
> and looks that only partially works.
> Are you planning to include ICAP client in squid 3.0 release?
> Is anybody working on that?
> It is not so difficult for
Hi all,
Now the ICAP client in squid3 does not work very well. I am testing it
and looks that only partially works.
Are you planning to include ICAP client in squid 3.0 release?
Is anybody working on that?
It is not so difficult for someone to understand the ICAP related code,
looks good. I do
On Wed, Sep 27, 2006, Henrik Nordstrom wrote:
> >> 1. Add minimal HTTP/1.1 support as a requirement for 3.0.STABLE. It's
> >> seriously starting to be a bigh hinderance for Squid being HTTP/1.0.
> >> At minimum we need chunked encoding and expect 100.
>
> > I dont agree - at the moment noone is de
>> 1. Add minimal HTTP/1.1 support as a requirement for 3.0.STABLE. It's
>> seriously starting to be a bigh hinderance for Squid being HTTP/1.0.
>> At minimum we need chunked encoding and expect 100.
> I dont agree - at the moment noone is developing or focusing this, and
> we dont know how big an