Hi Graham,
On Wed, Jan 29, 2014 at 2:51 PM, Graham Leggett wrote:
> On 29 Jan 2014, at 11:17 PM, Erik Pearson wrote:
>
> > Actually, the more I've delved and actually used mod_session and
> friends, the more fundamental the changes have become. For instance, a lot
> of the code that lives in mo
Following Jeff's advice to create a dedicated thread, I'm forwarding
my message sent to the wrong place (clearly)...
On the mod_proxy_http side, when the backend closes the connection (as
EOS), the socket is recycled w/o being closed.
The following patch avoids this, freeing the resource quickly
On Wed, Jan 29, 2014 at 5:31 PM, Yann Ylavic wrote:
> On the mod_proxy_http side, when the backend closes the connection (as
> EOS), the socket is recycled w/o being closed either.
>
> The following patch avoids this, freeing the resource quickly and saving
> the is_socket_connected() call later
On 29 Jan 2014, at 11:17 PM, Erik Pearson wrote:
> Actually, the more I've delved and actually used mod_session and friends, the
> more fundamental the changes have become. For instance, a lot of the code
> that lives in mod_session_cookie and mod_session_dbd seems more appropriate
> for mod_c
On the mod_proxy_http side, when the backend closes the connection (as
EOS), the socket is recycled w/o being closed either.
The following patch avoids this, freeing the resource quickly and saving
the is_socket_connected() call later when re-acquired.
Since it is quite symetric to the current fix
Actually, the more I've delved and actually used mod_session and friends,
the more fundamental the changes have become. For instance, a lot of the
code that lives in mod_session_cookie and mod_session_dbd seems more
appropriate for mod_cookie -- including session name, session caching,
session obje
On 29 Jan 2014, at 16:24, kbr...@apache.org wrote:
> Author: kbrand
> Date: Wed Jan 29 16:24:43 2014
> New Revision: 1562500
>
> URL: http://svn.apache.org/r1562500
> Log:
> propose SSLCertificate[Key]File/SSLCertificateChainFile overhaul for mod_ssl
Would it be possible to do the same for the S
It's also worth noting that we don't actually *do*
an attach()...
On Jan 29, 2014, at 10:02 AM, Jim Jagielski wrote:
> But I would prefer others testing as well...
>
> It's worth noting that it's just slotmem which did
> this and so, if it was "required" any other use of
> shared mem which child
But I would prefer others testing as well...
It's worth noting that it's just slotmem which did
this and so, if it was "required" any other use of
shared mem which child processes needed to have access
to cross forks would be broken. And it was just the SysV
shm which did this.
I *still* think th
Yes, it still works (at least in all my testing) :)
On Jan 28, 2014, at 12:48 PM, Jeff Trawick wrote:
> On Sat, Jan 25, 2014 at 3:11 PM, wrote:
> Author: jim
> Date: Sat Jan 25 20:11:50 2014
> New Revision: 1561385
>
> URL: http://svn.apache.org/r1561385
> Log:
> Remove wonky and unneeded code
10 matches
Mail list logo
