On Thu, 4 Sep 2008 10:46:44 +0200, Gabriele Messineo
<[EMAIL PROTECTED]> wrote:
> Hi,
> I think you can fix this bug by applying this patch:
>
> --- python/netsnmp/client_intf.c
> +++ python/netsnmp/client_intf.c
> 320c320
> < snprintf(buf,"%ld", buf_len, *var->val.integer);
> ---
>> snprintf(buf, buf_len, "%ld", *var->val.integer);
>
> The bug has been introduced in the SVN 16962 and it's applied in the
> patch:
> debian/patches/49_cve2008_2292_python.patch
>
> It's a simple type that cause every walk on integer MIBs to cause a
> segfault
> from python.
>
> I've verified it is present in the latest experimental package
> (5.4.1~dfsg-9)
> and in the SVN Trunk of net-snmp (actually 17206).
Hi Gabriele!
Thanks for your quick reply. You are right. Here is a more detailed
backtrace:
#4 0x00007fa69711626a in __snprint_value (buf=0x7fffa02df3a0 "",
buf_len=4096, var=0x222af70, tp=0x21bbb80, type=16, flag=0)
at netsnmp/client_intf.c:341
341 snprintf(buf,"%ld", buf_len, *var->val.integer);
After applying your patch, the segfault is gone.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
