Re: Disable specifc OIDs in MIB

Robert,

Thanks for your reply. More in-line

On Tue, Jul 3, 2018 at 10:01 PM, Robert Story <[email protected]> wrote:

> On Tue, 3 Jul 2018 19:05:59 +0530 Thommandra wrote:
> TG> - How can we disable an OID from snmpwalk output?
>
> For snmpd, you can configure VACM views in snmpd.conf to exclude
> certain OIDS. It's not very intuitive, but should be possible.
>

Can we remove the OIDs at compile time (using configure option) like we do
for modules? Changing VACM views is tedious and has to be configures for
all communities/users etc.
And if the list of OIDs is huge, it is not feasible.


>
> For the client, there isn't a way to not query for something during
> a snmpwalk.
>
> TG> c) Can we disable an entire MIB during compile time? Ex: TCP or
> TG> UDP-MIB
>
> Yes. You can disable MIBs via configure and at run-time.
>

Was able to figure that out. Thanks.


>
> TG> Also,  I see that after moving to 5.7.3 snmpwalk times-out
> TG> after walking the VRRP-MIB though there is data beyond that
> TG> OID. And this is happening always when there is data in
> TG> VrrpOperTable.
>
> We don't implement the VRRP-MIB, so you'll have to check with the
> device vendor. Increasing the timeout may help.
>

Actually I understand that. I went further and enabled debug on snmpgetnext
to find that the failure is due to the error during snmp_parse()
And it is because the getnext PDU contains an IP_ADDR as a value
(vrrpOperMasterIpAddr OID)

On debugging further, I found that the below change in 5.7.3 is actually
causing the issue

net-snmp-5.7.3/snmplib/snmp_api.c

4594         case ASN_IPADDRESS:
4595             if (vp->val_len != 4)
4596                 return -1;
4597             /* fallthrough */
4598         case ASN_OCTET_STR:
4599         case ASN_OPAQUE:
4600         case ASN_NSAP:

We fail if ASN_IPADDRRESS len is not 4. Earlier in 5.7.2, we did not have
this check.

And looking at the history, it seems like you have modified/added those
lines. May I know if this is a bug?
BTW, removing that val_len check from snmp_api.c makes the snmpwalk work
smooth.

Please note that our implementation of VRRP-MIB was done using mib2c
utility. And so vrrpOperMasterIpAddr is implemented as u_long in the code.

Regards,
Gowtham


>
> Robert
>

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders