Aww shucks :/ Out of curiosity, how does the _push method on the Cryptostream prototype know where one message ends and the other begins?
It seems to me that the while/do loop continues to call _pusher (which calls SSL_read under the hood) until either the buffer is all used up or the SSL_read function can't read anything (it returns -1). Then it'll emit a data event only if it has actually read a non-negative number of bytes. When I was debugging, it occurred to me that maybe the data was coming in slowly in chunks so that after it had read the first one or two chunks from SSL_read, the next tick of the while loop fired so quickly that it returned -1 before the next chunk of data had reached the C buffer. Then it emitted this first part of the data as if it were its own isolated grouping. Is this possible? On Sunday, October 7, 2012 4:59:57 PM UTC-4, Ben Noordhuis wrote: > > On Sun, Oct 7, 2012 at 2:50 AM, Justin Meltzer > <[email protected]<javascript:>> > wrote: > > Thanks Ben! > > > > Patch[1] gets rid of the "data length too long" error, and is better > than > > the hack I did. However, it does not fix the breaking up of data. > > > > Reverting commit f210530f unfortunately does not prevent the data from > being > > broken up either. > > > > Any other ideas? > > I'm afraid not. > > I can land a patch that turns on SSL_OP_MICROSOFT_BIG_SSLV3_BUFFER but > I can't tell if the socket.io error is a bug in socket.io or node. > > Try contacting the socket.io authors - if it turns out it's a bug in > node, they can forward the bug report to us. > -- Job Board: http://jobs.nodejs.org/ Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines You received this message because you are subscribed to the Google Groups "nodejs" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/nodejs?hl=en?hl=en
