httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Ames <grega...@apache.org>
Subject Re: don't try this at home
Date Tue, 28 May 2002 17:22:34 GMT
Justin Erenkrantz wrote:
> 
> On Tue, May 28, 2002 at 10:18:52AM -0400, Jeff Trawick wrote:
> > okay, do try it, but (unlike somebody last night) don't try it on daedalus
> >
> > GET / HTTP/1.1
> > Accept: */*
> > Host: test
> > Content-Type: application/x-www-form-urlencoded
> > Transfer-Encoding: chunked
> >
> > AAAAAAAAAAAAAAAAAAA
> 
> Hmm.  Isn't that legal?  A is a hex digit.
> 
> What should we be doing?  -- justin

well, not this for sure:

(from /usr/local/apache2.0.37dev/corefiles/httpd.core.1)

#0  0x2815e7b4 in kill () from /usr/lib/libc.so.4
#1  0x2819eb26 in abort () from /usr/lib/libc.so.4
#2  0x80699d2 in ap_log_assert (szExp=0x80854db "readbytes > 0",
    szFile=0x8085464 "http_protocol.c", nLine=911) at log.c:691
#3  0x805ffe8 in ap_http_filter (f=0x814bef8, b=0x814b798,
    mode=AP_MODE_READBYTES, block=APR_BLOCK_READ, readbytes=-1431655766)
    at http_protocol.c:911
#4  0x80717fc in ap_get_brigade (next=0x814bef8, bb=0x814b798,
    mode=AP_MODE_READBYTES, block=APR_BLOCK_READ, readbytes=8192)
    at util_filter.c:507
#5  0x8078f4d in net_time_filter (f=0x814b7b8, b=0x814b798,
    mode=AP_MODE_READBYTES, block=APR_BLOCK_READ, readbytes=8192)
    at core.c:3324
#6  0x80717fc in ap_get_brigade (next=0x814b7b8, bb=0x814b798,
    mode=AP_MODE_READBYTES, block=APR_BLOCK_READ, readbytes=8192)
    at util_filter.c:507
#7  0x8061225 in ap_get_client_block (r=0x814b048,
    buffer=0xbfbfd790 ".0-doc/config/ajp.html\n", bufsiz=8192)
    at http_protocol.c:1705
#8  0x8061407 in ap_discard_request_body (r=0x814b048) at http_protocol.c:1801
#9  0x8063153 in ap_die (type=302, r=0x814b048) at http_request.c:150
#10 0x8063376 in ap_process_request (r=0x814b048) at http_request.c:274
#11 0x805e8f3 in ap_process_http_connection (c=0x8134120) at http_core.c:291

Jeff said something about the chunk size being negative.  Clearly that's wrong,
but while we're at it, we should review the RFC and see what the rules are.

Greg

Mime
View raw message