Anyone notice the strange spikes/valleys on the Cacti graphs lately? Looks like a heartbeat every hour on the hour. Thanks, Dean Dean Bahls COO | Fiber Minnesota 14 Main St SW | PO Box J | Menahga, MN 56464 218.564-1321 (Direct) | 218.564.3000 (Main Office) dean.bahls@fiberminnesota.com<mailto:dean.bahls@fiberminnesota.com> [cid:image001.png@01DA3298.4466DA60] [cid:image002.png@01DA3298.4466DA60]
I see what you mean on the Aggregate graphs on the MICE Looking Glass. http://micelg.usinternet.com/cacti/graph.php?action=view&local_graph_id=2625&rra_id=all It could be a real member port doing something--but those are ~100 Gbps spikes. What I also notice is that each spike is followed by a dip of roughly the same magnitude. This makes me think of sampling jitter (or artifacts, or noise) It's hard to describe but port counters don't update every second, sometimes it's every 30 seconds. Then Cacti comes along every approx 5 minutes (300 seconds) to poll the counters. The slight mismatch in timing means that Cacti usually gets 10 counter updates per 300 seconds--but every hour it gets 11 on one interval, and 9 on the next. Overall, the graph is accurate, but it shows this 5% noise every hour. This is guesswork in the case of MICE, but I've seen this in other networks, Cheers, Jonathan Jonathan Stewart Network Engineer LES.NET - AS18451 Desk: 1-204-666-6191 Mobile: 1-204-990-2120 130 Portage Avenue E Winnipeg, MB R3C 0A1 CANADA On Tue, 2023-12-19 4:27 p.m., Dean Bahls wrote:
Anyone notice the strange spikes/valleys on the Cacti graphs lately?
Looks like a heartbeat every hour on the hour.
Thanks,
Dean
*Dean Bahls*
**COO | Fiber Minnesota
14 Main St SW | PO Box J | Menahga, MN 56464
*218.564-1321 (Direct)* | 218.564.3000 (Main Office)
dean.bahls@fiberminnesota.com <mailto:dean.bahls@fiberminnesota.com>
------------------------------------------------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 <http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1>
participants (2)
-
Dean Bahls
-
Jonathan Stewart