5.3.dev snmptrapd after Wes' latest changes doesn't seem to properly
handle SNMPv3/USM authentication failures:
# cat snmptrapd.conf
createUser newtestuser SHA validpass
authuser log newtestuser authNoPriv
# snmptrap -Ci -v3 -u newtestuser -l anp -a SHA -A wrongpass localhost 0
.1.3.6.1.6.3.1.1
Wes Hardaker wrote:
createUser wesx MD5 abcdefgh DES
authuser log,execute,net wesx noAuthNoPriv
Is there a way to allow *all* SNMPv3/USM users to e.g.
"log,execute,net"? If there's not, then we'll effectively ruin the
advantages of snmptrapd usmUserTable management, won't we? One can still
a
Dave Shield wrote:
> On Thu, 2005-10-20 at 08:22 -0700, Wes Hardaker wrote:
>
>
>>Dave>I'd actually be rather
>>Dave> surprised (and somewhat disappointed) if SNMP tools were still
>>Dave> being made that *only* understood SMIv1.
>>
>>Unfortunately this is
I have been following the endianness discussion with some interest, having
been bitten myself by the problems when I had to implement a MIB table
with IpAddress in the index. I have studied the current implementation
and must agree that any solution is complex and not something you wish to
do imme
Dave,
While investigating the nightly build failure for the main branch on
x86-freebsd1, it turns out that sometimes the agent isn't shutting down
properly after a HUP.
The problem is that mteObjects registers a callback for POST_READ_CONFIG,
which inserts defaults into the container without che
> Thanks to both of you -- Wes for the code, Dave for the
> criticism. (oh, and Robert for slipping 5.3.pre1)
That's one of the reasons this project has been so successful.
We all work together to come up with something better.
I'll criticize Robert's code. Robert will criticize my code.
We bot
