On Thu, 14 Oct 2010 20:04:34 +0200, "Lukas Gebauer" <[email protected]> wrote: >> I'm doing this test from HTTPMethod and propably the problem is in >> THTTPsend class. >> >> This problem is only when I'm doing subsequent requests. >> When the KeepAlive := false the speed requests are OK. > > Probably it is "200ms delay" syndrome. Most often TCP performance > problem. I found it nearly everytime when I am hired for repair of > some > slow TCP/IP program. ;-) > > For more information see: http://support.microsoft.com/kb/214397 > > Are you using actual Synapse version? I made optimalizations agains > this > syndrome in THttpSend few versions ago. However server side must be > optimised too! > > In your case it can be server what sending a chunked reply with a few > small chunks. It can cause this syndrome. Disable a chunked reply on > the > server will be best solution. Worst solution is to disable NAGLE or > disable send buffer on the server socket. It is worst, because it can > decrease performance in some different cases. Made a good > communication > strategy in a program is allways best solution. >
Thank you for the answer! Yes, i'm using actual Synapse version (rev.39). this is usefull thx ;) ->http://support.microsoft.com/kb/214397 I'm not sure but It's look like this is going that way: HTTPMethod -> request (keep-alive) LIGHTtpd -> respond with header information (chunked) but in fact there is no chunked packets. HTTPMethod -> receiving first packet and waiting for next part and there is no next part. In the result connection speed is slowing. -- best regards Piotr Polok ------------------------------------------------------------------------------ Download new Adobe(R) Flash(R) Builder(TM) 4 The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly Flex(R) Builder(TM)) enable the development of rich applications that run across multiple browsers and platforms. Download your free trials today! http://p.sf.net/sfu/adobe-dev2dev _______________________________________________ synalist-public mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/synalist-public
