[twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Fastream Technologies
Hello, Has anybody found the chance to test ICS-SSL on one of those? What's needed to make sure they (sw+hw) are cooperating? I think this issue is getting more and more important every day, agree? Currently I get 20 SSL TPS local-to-local. When keep-alive is enabled, this goes up to 2000. Best

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Paul
What exactly do you mean ? Our hosting company installed a hardware SSL offloader. The server is now http server instead of https. Works exactly as before. No problems with the clients Paul - Original Message - From: Fastream Technologies ga...@fastream.com To: ICS support mailing

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Fastream Technologies
Hello, Sorry for being ambigous. I want the __OpenSSL accelerator_ on the _host_ to speed up ICS FTPS/HTTPS server. HTH. Regards, SZ On Sun, Feb 15, 2009 at 6:55 PM, Paul paul.blommae...@telenet.be wrote: What exactly do you mean ? Our hosting company installed a hardware SSL offloader.

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Paul
Then this is a good but very expensive choise. The speed increase was visible. But the accelerator they have costs about 20.000 ? Paul -- To unsubscribe or change your settings for TWSocket mailing list please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket Visit our website at

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Arno Garrels
Fastream Technologies wrote: Currently I get 20 SSL TPS local-to-local. When keep-alive is enabled, this goes up to 2000. Do you use a SSL session cache, either the built-in OpenSSl session cache or (IMHO faster) the ICS AVL session cache? -- Arno Garrels -- To unsubscribe or change your

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Paul
Do you use a SSL session cache, either the built-in OpenSSl session cache or (IMHO faster) the ICS AVL session cache? Would that help for the keep-alives ? Paul btw: I can see the keep-alive (httpcli) working after 5 seconds and afterwards 2 times per second. -- To unsubscribe or change

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Arno Garrels
Paul wrote: Do you use a SSL session cache, either the built-in OpenSSl session cache or (IMHO faster) the ICS AVL session cache? Would that help for the keep-alives ? No, only with non-persistent connections since just the SSL handshake becomes faster. Paul btw: I can see the

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Paul
I don't understand? Last week, I got complaints about a user that there was too much traffic in wait state. The HttpCli checks for comands on a server, but if there is nothing available, the server keeps the connection open for 20s before answering. I checked the traffic between

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Arno Garrels
Paul wrote: I don't understand? Last week, I got complaints about a user that there was too much traffic in wait state. The HttpCli checks for comands on a server, OK, got it ;-) but if there is nothing available, the server keeps the connection open for 20s before answering. If

Re: [twsocket] OpenSSL hardware accelerators

2009-02-15 Thread Paul
Both client and server are ICS (application server) Client asks server for the next command that the client should execute locally. If the server has no command available for the client, he postpones the answer for 20 seconds. During that 20 seconds, there should no communication between

Re: [twsocket] Bad http redirection

2009-02-15 Thread Maurizio Lotauro
Scrive Maurizio Lotauro lotauro.mauri...@dnet.it: On the same pc using a virtual machine (XPpro SP3) this (the redirection to my provider) doesn't happen. It is strange, isn't it? Ignore this sentence, it works in the VM exactly the same way. Bye, Maurizio.

Re: [twsocket] Bad http redirection

2009-02-15 Thread Maurizio Lotauro
Scrive Francois PIETTE francois.pie...@skynet.be: Hello, I think this is the expected behaviour. You can disable automatic relocation (FollowRelocation property) or use the location change event to know your request has been redirected. I don't think so. I get a 302 status code but in