- observium - lists.observium.org

(no subject)
by Pedro Serotto 04 Jun '13

04 Jun '13

Hi all, I would like to hide some unused ports on some device. I see the filters configuration "$config['bad_if'] = array" but if I correctly understand this filter is apply to all devices. Is there a way to do different filter for different device ? BR Pedro

2 5

ASR9k users?
by Robert Williams 03 Jun '13

03 Jun '13

Hi, Just curious if anyone is currently using Observium to monitor any Cisco ASR9k based devices? I can see that it has correctly identified all the various hardware modules, line cards and temperature sensors etc. but isn’t graphing things like temperatures or light levels like it does for our Cat6K workforce. Since this is the first ASR device I’ve tried polling on Observium I’m not sure if it’s just not fully supported or if we have some SNMP restrictions on the Cisco which are preventing it from getting what it needs. Any feedback or other experiences are welcome! Cheers all, Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 5

Memcached updates
by Britt Treece 03 Jun '13

03 Jun '13

Starting to use Memcached and found some of the current Memcached bits in Observium not working optimally. I wrote a new unix-agent script using python rather than PHP. The previous script relied on PECL memcached but didn't do anything for installations using PECL memcache. After a bunch of trial and error I determined that it was easiest to write the script in Python which requires the python-memcached library and is readily available on different OS types and versions. I also modified the applications/memcached.inc.php and polling/unix-agent.inc.php to accommodate the new script. They will also accurately separate different instances of memcached running on the same server as reported by the script. I'm still trying to work out a couple things. The main thing is when you enable Memcached as an application for a device Observium automatically creates an entry in the database but the instance is NULL. When the unix-agent reports on the Memcached instance(s) running on the server new entries get created in the database for each instance and the app with the NULL app_instance never gets used. I've been manually removing it from the database. If possible I would like to work it out so that it doesn't get created automatically. If that's not possible I was going to look into updating that entry with the first real instance reported by the agent. Also wanting to get the Apps tab for the Device to better identify each instance and maybe work on some additional metrics reporting for Memcached specific stats. Patch attached, let me know if it is acceptable or not... Britt Treece

3 4

Re: [Observium] (no subject)
by Adam Armstrong 02 Jun '13

02 Jun '13

Observium is designed for use in a service provider environment with a main design assumption being that one does not necessarily know what one wants to monitor until the data is needed. We purposefully make it difficult to delete or obscure port information, because that fits our requirements. Adam. Jason Lixfeld <jason(a)lixfeld.ca> wrote: >Why do you say that, Adam? I can sympathize with Pedro because I find myself faced with the same issues. > >In my specific use case, I don't have any interest in seeing interfaces in Device > Ports view that are down/down, dwdm (which I don't use and aren't active anyway, yet they still show up for some reason), loopback or vlan, personally. They tend to make it difficult to efficiently scroll through the ports page to find information that I am interested in (ie: which of my customers just started sending 4Gbps of traffic). > >Those types of ports I never care about seeing, so ignoring them in a global scope would be wonderful :) > >If I'm using Observium wrong too, I'd love to know what the intended right usage is :) > >On 2013-06-02, at 3:06 PM, Adam Armstrong <adama(a)memetic.org> wrote: > >> If you are hiding ports, you are using observium wrong. >> >> Adam. >> >> Pedro Serotto <pedro.serotto(a)yahoo.es> wrote: >> >> Hi all, >> I would like to hide some unused ports on some device. >> >> I see the filters configuration "$config['bad_if'] = array" but if I correctly understand this filter is apply to all devices. >> >> Is there a way to do different filter for different device ? >> >> BR >> >> Pedro >> >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Re: [Observium] (no subject)
by Adam Armstrong 02 Jun '13

02 Jun '13

If you are hiding ports, you are using observium wrong. Adam. Pedro Serotto <pedro.serotto(a)yahoo.es> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 2

Switching back to old location behavior
by Bryan Socha 01 Jun '13

01 Jun '13

I just updated and locations are all over the place. Is there a way to set it back to the old behavior where it groups by the snmp-location setup in each device? Thanks, Bryan Socha

2 1

Cisco 4500X VSS, no cpu values
by Bryan Socha 01 Jun '13

01 Jun '13

Hello, We added some Cisco 4500X switches to our network with a VSS setup and observium shows 0 for the cpu usage. Any ideas on how to get it to populate? Bryan Socha

1 0

Polling
by Wes Kennedy 01 Jun '13

01 Jun '13

I recently added in 25 new routers to be monitored and as of two hours later, they haven't been polled. They have all been Discovered properly, but not polled. All of my existing hosts are polling just fine. Has anyone seen this? -Wes

6 7

add power usuages
by Khapare Joshi 31 May '13

31 May '13

I added one of my HP enclouser, and ran the ./discovery.php -h it does not discover these : Temperatures: Humidity : Voltages: Frequencies: Current: Power: Fanspeeds : how do i get these into my observium ? also, is there a way to integrate total power in observium ? I get the total by snmpwalk : snmpwalk -c public -v2c test01.local enterprises.232.22.2.5.1.1.1.10 |awk '{ sum+=$4} END {print sum}' thank KV

3 2

switch errors meaning
by Mihai - Cristian Satmarean 31 May '13

31 May '13

Hi, I would like to ask what the switch errors mean, I have approx. 300 m Discards In on one several ports constantly and on other ports Discards Out. Please help me understand what is that. Thank you!

5 9