That is kind of weird. I had a 7 3650 running the recommended IOS XE release (3.3.5SE) and no issue with the default polling options.

Were these particularly busy switches?

On 3/15/2015 1:32 AM, Joseph L. Brunner wrote:

Thanks – clearing out some modules now..

With a stock Cisco 3850 switch – the CPU goes insane… that’s really a buggy box though – the IOS XE has been a bust… the old 3750X’s were so much more stable and reliable…

We are getting high cpu polling them

*From:*observium [mailto:observium-bounces@observium.org] *On Behalf Of *Tristan Rhodes *Sent:* Sunday, March 15, 2015 01:07 AM *To:* Observium Network Observation System *Subject:* Re: [Observium] High CPU on Cisco Switches

Joseph,

Thanks for the feedback! You can also accomplish something similar by disabling the poller modules that cause high CPU on a device.

1. Using the web-interface, click on the device with high CPU.

2. Click on the "Gear" icon on the right side to edit device settings.

3. Click on the "Modules" tab

4. Now you can disable the modules that you don't want Observium to

attempt on this device

Note that you can also disable poller modules globally by editing the "config.php" file. Here are some of the options you can set:

http://observium.org/wiki/Configuration_Options#Poller_and_Discovery_Modules

Cheers!

Tristan

*Tristan Rhodes* Network Engineer

Weber State University

801.626.8549

On Sat, Mar 14, 2015 at 9:47 PM, Joseph L. Brunner <joe@affirmedsystems.com mailto:joe@affirmedsystems.com> wrote:

Team,

We had some cisco switches dropping packets with 99% cpu being polled by Observium.

We had to create a view which hides some arp and routing table objects to bring the CPU back down to normal levels

snmp-server view cutdown at excluded

snmp-server view cutdown snmpUsmMIB excluded

snmp-server view cutdown snmpVacmMIB excluded

snmp-server view cutdown snmpCommunityMIB excluded

snmp-server view cutdown ip.21 excluded

snmp-server view cutdown ip.22 excluded

snmp-server community ourcomm1 view cutdown RO

---

observium mailing list observium@observium.org mailto:observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

---

observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

Yup – you’re both right – the view actually BROKE Observium – the view disabled too much for Observium to poll at all –

We ended up just going this we found in the forums

(in /opt/observium/config.php)

// Disable EtherLike-MIB $config['enable_ports_etherlike'] = 0; # Enable Polling EtherLike-MIB (doubles interface processing time)

// Disable FDB table polling $config['poller_modules']['fdb-table'] = 0;

// Disable ARP table polling $config['discovery_modules']['arp-table'] = 0;

Quite a big issue – without this – the 3850 switch CPU was going BONKERS and dropping packets… insane.

From: observium [mailto:observium-bounces@observium.org] On Behalf Of Derek Sent: Sunday, March 15, 2015 01:09 AM To: Observium Network Observation System Subject: Re: [Observium] High CPU on Cisco Switches

You have a lot of VLAN's? You can disable certain types of polling through Observium too.

On Sat, Mar 14, 2015 at 10:47 PM, Joseph L. Brunner <joe@affirmedsystems.commailto:joe@affirmedsystems.com> wrote: Team,

We had some cisco switches dropping packets with 99% cpu being polled by Observium.

We had to create a view which hides some arp and routing table objects to bring the CPU back down to normal levels

snmp-server view cutdown at excluded snmp-server view cutdown snmpUsmMIB excluded snmp-server view cutdown snmpVacmMIB excluded snmp-server view cutdown snmpCommunityMIB excluded snmp-server view cutdown ip.21 excluded snmp-server view cutdown ip.22 excluded snmp-server community ourcomm1 view cutdown RO

_______________________________________________ observium mailing list observium@observium.orgmailto:observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium