Re: Squid connection retry problem -- fd leak?

On Mon, 15 Sep 1997, John Line wrote:
>
> Duane's analysis of the original problem was that a connection problem
> occurred at a stage in connection setup when Squid didn't have a timeout
> set to allow it to clean up after problems, leaving things in a mess. It
> looks as if the retry patch may introduce a similar problem.

Duane, or someone, do you have the details of what was done to correct the
problem originally? I thought I was taking care of timeouts correctly in
the connection-retry patch, but maybe not. The timeouts I'm setting
relate to the connection establishment, and I think that the connection
timeout handler is set when the filehandle is first initialized, right?
Perhaps when the fd is being dup2()'d for the next attempt, something odd
is happening.

One important question is: when you see these stuck write fd's, do you
see anything in the log file pertaining to connection retries on that
address?

I also find it interesting that there's some VM activity in the NOVM
version... I *did* design the patch solely on the VM version of Squid,
and I'm not quite up on all the ways in which NOVM is different. There
might be a nuance that's not permitting the VM patch to be applied
straight to the NOVM code, because of different base assumptions.

I'll do a bit of poking around and see what I can come up with. Thanks
for all the information, everyone. Feel free to send me more details if
you find them.

        -Mike Pelletier.
Received on Mon Sep 15 1997 - 07:25:42 MDT