Re: Adding row indices for a new table row created by a manager

2011-01-13 Thread Dave Shield
On 13 January 2011 23:16, Homan, Dave wrote: >> What does "snmpgetnext" report when given the same OID as "snmpwalk" ? >> i.e. >>    snmpgetnext -v 2c -c localhost .1.3.6.1.4.1.31409.3.19.1 > snmpgetnext -v 2c -c localhost .1.3.6.1.4.1.31409.3.19.1 > SNMPv2-MIB::snmpSetSerialNo.0 = INTEGER: 176

RE: Adding row indices for a new table row created by a manager

2011-01-13 Thread Homan, Dave
snmpgetnext -v 2c -c localhost .1.3.6.1.4.1.31409.3.19.1 SNMPv2-MIB::snmpSetSerialNo.0 = INTEGER: 176107532 -Original Message- From: dave.shi...@googlemail.com [mailto:dave.shi...@googlemail.com] On Behalf Of Dave Shield Sent: Thursday, January 13, 2011 6:12 PM To: Homan, Dave Cc: net-sn

RE: Adding row indices for a new table row created by a manager

2011-01-13 Thread Homan, Dave
On 12 January 2011 16:02, Homan, Dave wrote: > It appears that the NetSNMP table_dataset.c code DOES create a new row, based > on the 'createAndGo' setting of my RowStatus object, but does not change it > to 'active'. OK - so one issue is the handling of RowStatus values. > The problem seems to

Re: Adding row indices for a new table row created by a manager

2011-01-13 Thread Dave Shield
On 13 January 2011 22:44, Homan, Dave wrote: > snmpwalk -v 2c -c localhost .1.3.6.1.4.1.31409.3.19.1 > SNMPv2-SMI::enterprises.31409.3.19.1 = No Such Object available on this agent > at this OID > > Repeated calls to snmpgetnext seem to work: > > snmpgetnext -v 2c -c localhost .1.3.6.1.4.1.3140

Re: Objects for huge configuration on snmpd.

2011-01-13 Thread Wes Hardaker
hs> I would like to perform an huge configuration on NET-SNMP agent hs> (snmpd), using Set PDUs carried by SNMP messages. I would like to use hs> the HOST RESOURCES MIB for that, but it does not contain much hs> ReadWrite objects. hs>Can someone show me what are the MIBs used to configure s

Re: Convert SNMPv1 traps to SNMPv3

2011-01-13 Thread Wes Hardaker
KK> We have a requirement of SNMPv3 communications on our network. Yay! KK> We have recently acquired some software from a vendor that currently KK> only supports v1 traps. Boo! KK> My plan was to have it send to the loopback or some other secondary KK> IP and have a Perl script receive the tr

Re: Response size needed to get too big error response

2011-01-13 Thread Wes Hardaker
> On Wed, 12 Jan 2011 13:42:09 +0100, "JORDI LOPEZ" > said: JL> Actually I'm getting +7000 bytes responses without problems but, I JL> must pass a set of conformance tests and one of them is sending a JL> tooBig error if responses are bigger than 1500 bytes. Well... the Net-SNMP agent

Re: Monitoring "CPU steal"?

2011-01-13 Thread Wes Hardaker
> On Thu, 13 Jan 2011 11:01:19 -0800, Krisztian Mizser > said: KM> Pardon my ignorance, but who's responsible for changing the MIBs to KM> support new features as technology advances? That's ok! MIBs that exist in the enterprise space are "private", or appropriately, "controlled by the

Re: Monitoring "CPU steal"?

2011-01-13 Thread Krisztian Mizser
On 07/15/2010 11:39 AM, Wes Hardaker wrote: > KM> In today's virtualized environments it's getting more important to > KM> monitor the "CPU steal" parameter. Top and vmstat have been supporting > KM> this for a while. I'm wondering if there's any work being done to > KM> implement this in net-s

Re: hrStorageUsed and integer overflow

2011-01-13 Thread Troels Arvin
Hello, Dave Shield wrote: > For large storage devices, the answer is to report the size/usage values > using a larger allocation unit. This should keep things within the > 32-bit limit. Do you know of an authoritative spec stating this that I can refer to in a bug report to IBM regarding AIX'

Re: hrStorageUsed and integer overflow

2011-01-13 Thread Dave Shield
On 13 January 2011 10:47, Troels Arvin wrote: > HOST-RESOURCES-MIB::hrStorageAllocationUnits.12 = INTEGER: 4096 Bytes > HOST-RESOURCES-MIB::hrStorageSize.12 = INTEGER: -1682702336 > HOST-RESOURCES-MIB::hrStorageUsed.12 = INTEGER: -1956338944 > Clearly, this is a case of an integer overflow: > B

hrStorageUsed and integer overflow

2011-01-13 Thread Troels Arvin
Hello, (Posted on both Net-SNMP's and nagios-snmp's mailing lists.) From a Linux server, I monitor disk usage on an AIX server, using a Nagios plugin called check_snmp_storage[1] which links to Net-SNMP libraries. We recently overlooked a filesystem which had run full. When running snmpwalk

Re: snmpwalk stops at "no such instance currently exist at this OID"

2011-01-13 Thread Dave Shield
2011/1/13 Yufeng Zhu : > But I tried “snmpgetnext” command on the failed OID and it can return the > next OID correctly. The real test is what happens if you try "snmpgetnext" on the OID immediately *before* the failed one. What does that report? > Is the “getnext” command different from “snmpw

RE: snmpwalk stops at "no such instance currently exist at this OID"

2011-01-13 Thread Yufeng Zhu
Sir Thank you very much for your reply. I am querying a vendor’s MIB and got this error and I have submit a defect to them. But I tried “snmpgetnext” command on the failed OID and it can return the next OID correctly. Is the “getnext” command different from “snmpwalk” and “snmpgetnext”