Either nntpcached has a memory corruption somewhere, or my linux sscanf() function is broken. I traced the bug I found earlier with an articles that generated a SEGV, nntpcached is doing nothing strange, all points to a bad sscanf function (libc.so.5.4.7). Here is the gdb session: in article.c 745 if (!*msgid) (gdb) 747 msgid_header = find_header_arg(buf, "Message-ID:", body? body-buf: len); (gdb) 748 if (!msgid_header) /* shouldn't be in the body! */ (gdb) p msgid_header $1 = 0x400cb2b1 "<53t4mn$gnc@tofu.alt.net>\r\nX-Newsreader: News Xpress Version 1.0 Beta #3\r\nXref: artemis.ibernet.es alt.binaries.pictures:13325 alt.binaries.pictures.erotica:473319 alt.binaries.pictures.erotica.blonde"... (gdb) n 750 if (sscanf(msgid_header, "<%127[^>\r\n]", msgid)!=1) (gdb) n Program received signal SIGSEGV, Segmentation fault. -- Jon. <jon@gte.esi.us.es, http://www.esi.us.es/~jon>
