You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by David Reid <dr...@jetnet.co.uk> on 2000/09/11 23:54:42 UTC
FreeBSD 4.0 strangeness
Is anyone else seeing a lot of sockets hanging on FreeBSD 4 with the current
CVS code? If I run ab against the loopback I get the exact number of
requests +1 socket in TIMED_WAIT and they never exit. Could this be due to
the chunking again? BTW, I've seen this on 2 different installations with
mpmt_pthread.
david
"I try never to get involved in my own life,
it's way too complicated."
http://www.david-reid.com
Re: FreeBSD 4.0 strangeness
Posted by David Reid <dr...@jetnet.co.uk>.
OK. Maybe someone who knows can look at some point? It'll be the weekend
before I can look again as I'll be away in Genoa for the next 2 nights then
Edinburgh on Friday...
david
----- Original Message -----
From: <rb...@covalent.net>
To: <ne...@apache.org>
Sent: Monday, September 11, 2000 11:42 PM
Subject: Re: FreeBSD 4.0 strangeness
>
> Maybe. The chunking filter doesn't do much though. Currently, all it
> does is add a bucket at the head of the bucket_brigade with the length of
> the current brigade. There is a bunch of other magic in the function, but
> all that does is read from the buckets correctly.
>
> I don't think the lingering_close function was ever actually fixed.
>
> Ryan
>
> On Mon, 11 Sep 2000, David Reid wrote:
>
> > Yeah, that was my initial reaction but as I thought we'd fixed that one
ages
> > ago I discounted it. :(
> >
> > david
> >
> > ----- Original Message -----
> > From: <rb...@covalent.net>
> > To: "Apache (new-httpd)" <ne...@apache.org>
> > Sent: Monday, September 11, 2000 11:21 PM
> > Subject: Re: FreeBSD 4.0 strangeness
> >
> >
> > >
> > > It's MUCH more likely this is due to lingering close.
> > >
> > > Ryan
> > >
> > > On Mon, 11 Sep 2000, David Reid wrote:
> > >
> > > > Is anyone else seeing a lot of sockets hanging on FreeBSD 4 with the
> > current
> > > > CVS code? If I run ab against the loopback I get the exact number
of
> > > > requests +1 socket in TIMED_WAIT and they never exit. Could this be
due
> > to
> > > > the chunking again? BTW, I've seen this on 2 different
installations
> > with
> > > > mpmt_pthread.
> > > >
> > > > david
> > > >
> > > > "I try never to get involved in my own life,
> > > > it's way too complicated."
> > > >
> > > > http://www.david-reid.com
> > > >
> > >
> > >
> > >
> >
____________________________________________________________________________
> > ___
> > > Ryan Bloom rbb@apache.org
> > > 406 29th St.
> > > San Francisco, CA 94131
> >
> --------------------------------------------------------------------------
> > -----
> > >
> > >
> > >
> >
>
>
>
____________________________________________________________________________
___
> Ryan Bloom rbb@apache.org
> 406 29th St.
> San Francisco, CA 94131
> --------------------------------------------------------------------------
-----
>
>
Re: FreeBSD 4.0 strangeness
Posted by rb...@covalent.net.
Maybe. The chunking filter doesn't do much though. Currently, all it
does is add a bucket at the head of the bucket_brigade with the length of
the current brigade. There is a bunch of other magic in the function, but
all that does is read from the buckets correctly.
I don't think the lingering_close function was ever actually fixed.
Ryan
On Mon, 11 Sep 2000, David Reid wrote:
> Yeah, that was my initial reaction but as I thought we'd fixed that one ages
> ago I discounted it. :(
>
> david
>
> ----- Original Message -----
> From: <rb...@covalent.net>
> To: "Apache (new-httpd)" <ne...@apache.org>
> Sent: Monday, September 11, 2000 11:21 PM
> Subject: Re: FreeBSD 4.0 strangeness
>
>
> >
> > It's MUCH more likely this is due to lingering close.
> >
> > Ryan
> >
> > On Mon, 11 Sep 2000, David Reid wrote:
> >
> > > Is anyone else seeing a lot of sockets hanging on FreeBSD 4 with the
> current
> > > CVS code? If I run ab against the loopback I get the exact number of
> > > requests +1 socket in TIMED_WAIT and they never exit. Could this be due
> to
> > > the chunking again? BTW, I've seen this on 2 different installations
> with
> > > mpmt_pthread.
> > >
> > > david
> > >
> > > "I try never to get involved in my own life,
> > > it's way too complicated."
> > >
> > > http://www.david-reid.com
> > >
> >
> >
> >
> ____________________________________________________________________________
> ___
> > Ryan Bloom rbb@apache.org
> > 406 29th St.
> > San Francisco, CA 94131
> > --------------------------------------------------------------------------
> -----
> >
> >
> >
>
_______________________________________________________________________________
Ryan Bloom rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------
Re: FreeBSD 4.0 strangeness
Posted by David Reid <dr...@jetnet.co.uk>.
Yeah, that was my initial reaction but as I thought we'd fixed that one ages
ago I discounted it. :(
david
----- Original Message -----
From: <rb...@covalent.net>
To: "Apache (new-httpd)" <ne...@apache.org>
Sent: Monday, September 11, 2000 11:21 PM
Subject: Re: FreeBSD 4.0 strangeness
>
> It's MUCH more likely this is due to lingering close.
>
> Ryan
>
> On Mon, 11 Sep 2000, David Reid wrote:
>
> > Is anyone else seeing a lot of sockets hanging on FreeBSD 4 with the
current
> > CVS code? If I run ab against the loopback I get the exact number of
> > requests +1 socket in TIMED_WAIT and they never exit. Could this be due
to
> > the chunking again? BTW, I've seen this on 2 different installations
with
> > mpmt_pthread.
> >
> > david
> >
> > "I try never to get involved in my own life,
> > it's way too complicated."
> >
> > http://www.david-reid.com
> >
>
>
>
____________________________________________________________________________
___
> Ryan Bloom rbb@apache.org
> 406 29th St.
> San Francisco, CA 94131
> --------------------------------------------------------------------------
-----
>
>
>
Re: FreeBSD 4.0 strangeness
Posted by rb...@covalent.net.
It's MUCH more likely this is due to lingering close.
Ryan
On Mon, 11 Sep 2000, David Reid wrote:
> Is anyone else seeing a lot of sockets hanging on FreeBSD 4 with the current
> CVS code? If I run ab against the loopback I get the exact number of
> requests +1 socket in TIMED_WAIT and they never exit. Could this be due to
> the chunking again? BTW, I've seen this on 2 different installations with
> mpmt_pthread.
>
> david
>
> "I try never to get involved in my own life,
> it's way too complicated."
>
> http://www.david-reid.com
>
_______________________________________________________________________________
Ryan Bloom rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------