perfSONAR User Q&A and Other Discussion

["Uhl, George D. (GSFC-423.0)[Arctic Slope Technical Services, Inc.]" <>]

Hi Tim,

 

I was using nagios to generate email alerts when my mesh tests were below thresholds.  I say “was” because I’ve moved to a homegrown alerting system that (I hope) works better for my needs.  I leveraged the same checks used by Maddash such as check_loss.pl, check_throughput.pl, check_ping_loss.pl, with thresholds and check intervals specific to each test host pair.   The Nagios checks weren’t in sync with Maddash check intervals but when I received an alert email that was a trigger for me to check the Maddash graphs related to the related source/destination/test.

 

My motivation was to move away from Maddash dashboard alerting system because of the fixed threshold settings within any given grid.  My dashboard contains grids of test nodes with common associations but not necessarily common performance thresholds.  Thus I have grids with cells that display a variety of colors because the performance thresholds vary among the test host pairs contained in the grid.  By establishing Nagios checks with test pair specific thresholds I could customize alerting for any test pair I wanted to monitor.

 

George Uhl

 

 

From: <> on behalf of Tim Chown <>
Reply-To: Tim Chown <>
Date: Wednesday, October 14, 2020 at 12:30 PM
To: "" <>
Subject: [EXTERNAL] [perfsonar-user] Nagios monitoring of performance

 

Hi,

I’m interested in hearing from anyone who is using Nagios to monitor the performance measurements taken by perfSONAR, be that directly to a given measurement point, or via a MaDDash mesh. 

I see there are some Nagios tools at  https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_perfsonar_nagios&d=DwIGaQ&c=ApwzowJNAKKw3xye91w7BE1XMRKi2LN9kiMk5Csz9Zk&r=vgbMJbo5EYOocL9LEZO2YwEKDaSiiYNeAZ6_pFTA-nM&m=FmFlAyT4JoFyVjy6C8IlksGSlQ9CX0ZxjJlBSEZxgGw&s=qFvwUfFbIcu7S_--bhkRx_oG45HHU6Qj6BlPylJWsH8&e= , which seem to do a variety of checks and include performance thresholds, I’ve not dug into what’s there in detail.

We have sites who have deployed perfSONAR, but would like to monitor the results via the Nagios platform they use for other systems, rather than checking MaDDash. One option for example might be to have an alarm should a specific measurement fall below a certain threshold for a certain period of time.

So we’d like to hear from anyone who is using Nagios this way to monitor a community mesh of results, and what their experiences are.  A live example would be great.

Thanks,
Tim

--
To unsubscribe from this list: https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.internet2.edu_sympa_signoff_perfsonar-2Duser&d=DwIDaQ&c=ApwzowJNAKKw3xye91w7BE1XMRKi2LN9kiMk5Csz9Zk&r=vgbMJbo5EYOocL9LEZO2YwEKDaSiiYNeAZ6_pFTA-nM&m=FmFlAyT4JoFyVjy6C8IlksGSlQ9CX0ZxjJlBSEZxgGw&s=uITKuGKz94m1Iof13IICFdde7PWvHBAMkDeaoiJgbiE&e=