On 26 January 2011 10:57, Timo Schoeler wrote:
>> One other option might be to disable the whole ARP table module at
>> runtime, by adding "-I-at" to the agent startup flags.
>
> Hm, that doesn't work for me
Do you mean that it doesn't disable reporting of the ARP table?
Or that it doesn't solve
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
thus Dave Shield spake:
> On 26 January 2011 10:35, Timo Schoeler wrote:
>> After Wes mentioned VACM, I tried to exclude the OIDs that I thought to
>> be source of the problem from being polled via SNMP.
>>
>> However, it didn't solve the problem. Not
On 26 January 2011 10:35, Timo Schoeler wrote:
> After Wes mentioned VACM, I tried to exclude the OIDs that I thought to
> be source of the problem from being polled via SNMP.
>
> However, it didn't solve the problem. Not active polling by another host
> via SNMP creates the load spikes, but net-s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
thus Timo Schoeler spake:
> Hi,
Hi,
answering myself for the archives:
> I have a medium catastrophe happening:
>
> A server on which I run net-snmp (it's a CentOS machine, so it runs a
> certainly not-too-recent net-snmp-5.3.2.2-9.el5_5.1 from the
> On Tue, 21 Dec 2010 18:12:38 +0100, Timo Schoeler
> said:
TS> A server on which I run net-snmp (it's a CentOS machine, so it runs a
TS> certainly not-too-recent net-snmp-5.3.2.2-9.el5_5.1 from the packages)
TS> has a rather big ARP table (>5000 entries). I suspect this the reason to
Hi,
I have a medium catastrophe happening:
A server on which I run net-snmp (it's a CentOS machine, so it runs a
certainly not-too-recent net-snmp-5.3.2.2-9.el5_5.1 from the packages)
has a rather big ARP table (>5000 entries). I suspect this the reason to
have snmpd eat almost 100% CPU when p
