% of Baseline Alternative? Cisco CMTS (Cable Router)

[willieb]

Hello,

I can't get baseline to work correctly. The high is tripping without even being set. And I keep getting un-expected results.

Is it a bug or do I just not understand it? I've researched it but can't find that much info on it.

I'm monitoring cable modems online on a CMTS (Cisco Cable Router) and I want to set a threshhold of 30%. For example, modems on one port may be 200. So I want it to alert me if it drops by 30%. So when 60 modems drop offline, I want to be notified. Or if it was 100 modems on a port, I want to be notified at 30 modems offline. I have 38 ports I need to monitor.

Any ideas? Thanks.

[willieb]

Just for the heck of it, here's my cacti:

http://netmon.wayxcable.com/cacti

User: test
Pass: test

Click "Modems per Upstream"

If I can't monitor % threshholds accurately with thold, and other possibilities with cacti? Thanks.

[aris77]

Hi Willieb

Your CPU usage graphs are very nice. Could you please share the graph template?


Thank you

[Hitch]

Hi Willieb.
I've been using baseline monitoring on my upstreams for over a year and it working very well for me.
I don't remember the default values anymore, but I'm using these #'s in the Threshold templates:
Reference in the past: 86400
Time Range: 10800
Deviation up: 100
Deviation down: 20 (for 20% drop in modems)
Trigger Count: 10

These numbers work well for me.

Roger

[cigamit]

Which reminds me, I really need to change those to drop downs with times and values so they are easier to use.

[willieb]

Hi Willieb

Your CPU usage graphs are very nice. Could you please share the graph template?


Thank you

Sorry it took me so long to reply, but the CPU graph is the built-in fire graph.

[willieb]

Hi Willieb.
I've been using baseline monitoring on my upstreams for over a year and it working very well for me.
I don't remember the default values anymore, but I'm using these #'s in the Threshold templates:
Reference in the past: 86400
Time Range: 10800
Deviation up: 100
Deviation down: 20 (for 20% drop in modems)
Trigger Count: 10

These numbers work well for me.

Roger

Thanks Roger, I am trying your settings now.

[willieb]

Hi Willieb.
I've been using baseline monitoring on my upstreams for over a year and it working very well for me.
I don't remember the default values anymore, but I'm using these #'s in the Threshold templates:
Reference in the past: 86400
Time Range: 10800
Deviation up: 100
Deviation down: 20 (for 20% drop in modems)
Trigger Count: 10

These numbers work well for me.

Roger

Thank you Roger! From your settings, I now understand what they mean. I think the key piece here is setting the dev up to 100.

It's working just as we need. Thanks again!

[Hitch]

Hi WilleB,
I'm glad they worked for you. I've had some personal issues to deal with and haven't been online much lately.

My company, in their ultimate wizdom purchased 2 Arris C4 cmts units instead of Cisco 10k's. Does anyone have any template similar to the Cisco templates, like modems per upstream, snr, bandwidth per upstream, etc? I really don't have time to try to figure out the strangely different mib settings in the C4 right now.