[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: sb_sasl_pkt_length



At 11:11 PM 2002-06-11, Howard Chu wrote:
>That's the last time I pay attention to comments in the code.

The comment is actually correct.  See RFC 2222.

The problem, I suspect, is that there are implementations
which overrun the negotiated buffer size.  I suggest we
kick out a warning when the negotiated buffer size is
exceeded, and error out when the hard buffer size is
exceeded.

Kurt


>I think this
>fix
>was a mistake. However, I haven't been able to trigger the errors that are
>being reported on the -software list. Anyone else?
>
>@@ -157,10 +161,7 @@
>                | buf[2] << 8
>                | buf[3];
>
>-       /* we really should check against actual buffer size set
>-        * in the secopts.
>-        */
>-       if ( size > SASL_MAX_BUFF_SIZE ) {
>+       if ( size > max ) {
>                /* somebody is trying to mess me up. */
>                ber_log_printf( LDAP_DEBUG_ANY, debuglevel,
>                        "sb_sasl_pkt_length: received illegal packet length
>"
>
>  -- Howard Chu
>  Chief Architect, Symas Corp.       Director, Highland Sun
>  http://www.symas.com               http://highlandsun.com/hyc
>  Symas: Premier OpenSource Development and Support