Fwd: kio: slave process dies when using https

[Michael Cardenas]

no need to apologize, every bit of effort helps!

I thought of doing this, but earlier, it seemed lke a bad idea because
of the performance hit. Now that it's so late at night, it seems far
more attractive. We're testing this patch now. 

I also tried just sending a MSG_NOSIGNAL to the socket in
TCPSlaveBase::isConnectionValid, which would prevent the slave from
dying in such an ugly way. This stopped the error from occuring, but it
seemed to slow down the web page loading drastically, and is not as
safe. 

I'll let you know what we decide to go with. Thanks again!

michael

On Fri, May 31, 2002 at 12:35:49AM -0400, Dawit A. wrote:
> Hi Michael,
> 
> Truely sorry about this. Please apply this after applying the patch
> I sent in the second email.  The first part of the patch creeped was
> not supposed to be there!  Basically the fix is to add && !m_bIsSSL in 
> HTTPProtocol::httpClose where we check to see if the connection 
> should be persistent.
> 
> Again my appologies....
> 
> Regards,
> Dawit A.



-- 
michael cardenas
lead windows compatibility engineer
lindows.com
-
hyperpoem.net
-
"Be the change you want to see in the world"
-Mahatma Gandhi