While it is true that the maximum TLS record size is 16k, this doea not
relate directly to the TCP handling.
For example, the server hello, certificate request, etc are individual
records but they are all bundled into the one TCP record because it is
(apparently) more efficient to perform fewer reads.
Does the TLS spec prohibit writing four 16k TLS records as one 64k TCP
record?
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [SMTP:[EMAIL PROTECTED]]
> Sent: Saturday, May 29, 1999 7:59 AM
> To: [EMAIL PROTECTED]
> Subject: Re: SSL_read problem?
>
> "Gang Lu" <[EMAIL PROTECTED]>:
>
> > Recently, I built a simple client/server program when trying
> > openssl-0.9.2b. However, when I try to send 64K bytes from client to
> > server, I found that server has to get those data by 4 times usingh
> > SSL_read. That means I use SSL_read 4 times to get all 64K data
> because
> > SSL_read only return 16K no matter how big I specify it to read.
>
> > Should I set some cache size somewhere before I call SSL_read or
> there
> > is other problem?
>
> It's not a problem at all. You never can expect that a single read()
> will return all data -- the library (just like the read() system call)
> cannot know where to stop.
>
> The reason for the behaviour you observed is not just some peculiarity
> of SSL_read: That data is split in 16K blocks already when it is
> written. This is the maximum record size allowed by RFC 2246.
> ______________________________________________________________________
> OpenSSL Project http://www.openssl.org
> Development Mailing List [EMAIL PROTECTED]
> Automated List Manager [EMAIL PROTECTED]
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
