Long receive time for chunked ssl response

[Scott Willey <sw...@soasta.com>]

Processing a chunked SSL response of 221 characters, I found I was waiting over three seconds for the entire response to be read. Upon investigation, this appears to be happening because the decrypt method in SSLIOSession does not decrypt everything in the encrypted buffer at one go. Because of this, portions of my response didn't get processed via the processing of the events, but rather via the BaseIOReactor's validate method which means I experienced multiple select interval timeouts prior to the message being fully processed. In reviewing the java nio documentation for the unwrap method of the sslEngine, the comments in the javadoc state: "This method will attempt to consume one complete SSL/TLS network packet". It appears that one time through the decrypt method decrypted small packets with almost no data in them, leaving encrypted data waiting. As an experiment, I changed the code in isAppInputReady in SSLIOSession to call decrypt in a loop until everything in the encrypted buffer was processed and bingo the response processed in under 5ms.

I was surprised not to find any other comments on this topic. Perhaps I simply a related topic somewhere? In any case, I'd be curious to know what the "correct" solution to this problem is.

Thank you very much

Scott

Re: Long receive time for chunked ssl response

[Scott Willey <sw...@soasta.com>]

Done.

Unfortunately, I there is no public access to the service in question. In theory any connection involving SSL and chunked responses would reproduce-but that's in theory. If I can detect this problem in a public site, I'll add it to the ticket (HTTPCORE-196).

Thank you

Scott


On 5/13/09 12:26 PM, "Oleg Kalnichevski" <ol...@apache.org> wrote:

Processing a chunked SSL response of 221 characters, I found I was waiting over three seconds for the entire response to be read. Upon investigation, this appears to be happening because the decrypt method in SSLIOSession does not decrypt everything in the encrypted buffer at one go. Because of this, portions of my response didn't get processed via the processing of the events, but rather via the BaseIOReactor's validate method which means I experienced multiple select interval timeouts prior to the message being fully processed. In reviewing the java nio documentation for the unwrap method of the sslEngine, the comments in the javadoc state: "This method will attempt to consume one complete SSL/TLS network packet". It appears that one time through the decrypt method decrypted small packets with almost no data in them, leaving encrypted data waiting. As an experiment, I changed the code in isAppInputReady in SSLIOSession to call decrypt in a loop until everything in the en
crypted buffer was processed and bingo the response processed in under 5ms.

Re: Long receive time for chunked ssl response

[Oleg Kalnichevski <ol...@apache.org>]

Scott Willey wrote:
> Processing a chunked SSL response of 221 characters, I found I was waiting over three seconds for the entire response to be read. Upon investigation, this appears to be happening because the decrypt method in SSLIOSession does not decrypt everything in the encrypted buffer at one go. Because of this, portions of my response didn't get processed via the processing of the events, but rather via the BaseIOReactor's validate method which means I experienced multiple select interval timeouts prior to the message being fully processed. In reviewing the java nio documentation for the unwrap method of the sslEngine, the comments in the javadoc state: "This method will attempt to consume one complete SSL/TLS network packet". It appears that one time through the decrypt method decrypted small packets with almost no data in them, leaving encrypted data waiting. As an experiment, I changed the code in isAppInputReady in SSLIOSession to call decrypt in a loop until everything in the en
crypted buffer was processed and bingo the response processed in under 5ms.
> 
> I was surprised not to find any other comments on this topic. Perhaps I simply a related topic somewhere? In any case, I'd be curious to know what the "correct" solution to this problem is.
> 
> Thank you very much
> 
> Scott
> 

Hi Scott

This does sound like an issue with SSLIOSession. Could please raise a 
JIRA for it?

https://issues.apache.org/jira/browse/HTTPCORE

You are also very welcome to attach your local changes as a patch to the 
JIRA ticket. If the server is publicly accessible could you please let 
me know the URL I could use to reproduce the problem?

Cheers

Oleg

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org