Re: Strange error in multiprocess communication
"Taylor R. Campbell" <[email protected]> Sun, 9 Apr 2006 14:32:56 +0000
| Newsgroups | gmane.lisp.scheme.scsh |
|---|---|
| Message-ID | <[email protected]> |
Date: Sun, 09 Apr 2006 13:37:33 +0530 From: Gautham Anil <[email protected]> Newsgroups: gmane.lisp.scheme.scsh Error: exception os-error (channel-maybe-read 104 '#{Byte-vector (length 4096) 40 115 101 110 100 ---} 0 4096 #t ---) Judging by this error message, if my errno-fu is correct, this means that the connection was reset by peer. You could probably find out for certain if you typed ,CONDITION at the debugger and looked at the last element of the list it prints; CHANNEL-MAYBE-READ puts the OS's error message (in the sense of strerror(3)) at the end of the condition it signals. But I notice that you seem to be using the sockets interface from Scheme48, which scsh happens to come with, and which is very minimal and not very useful. Why aren't you using scsh's much more extensive sockets interface? As it happens, the OS error message will be put at the very front of conditions signalled by the socket syscalls if you use scsh's sockets interface.