This is on a personal backport to sarge, but it might affect more. The problem seems to be that http_read_headers is very brittle, I started to patch it, but I don't have time to fix it right now.
On the first iteration it does a memset of NULL address for zero bytes (don't know if that is a problem or not, but seems dodgy), then if the socket read fails, most of the following code will segfault since line is NULL. I minimal patch which at a quick glance looks like it will fix it is to put this at the top of the http_read_headers routine (before the while loop): len=2048; line = malloc(len); Adrian -- Adrian Bridgett - [EMAIL PROTECTED] GPG key available on public key servers -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]