I'm attempting to monitor just under 300 nodes. When I specified only
one parent per node, nagios worked fine.
I'm now trying to better represent reality by listing all adjacent
routers potentially on a path to the Nagios
server. I'm not monitoring switches, yet. Nagios 1.2 is core dumping
in the check_for_circular_path routine,
having run out of 8192k of stack space. Could someone please explain
the algorithm being used in that routine?
Should you be interested, our partial mesh adjacency graph looks like
this (including only Nagios and multiply
connected routers, then applying closure):
Node: Parents
Nagios: (no parents)
A: Nagios, B, C, D (e.g., if the interface on Nagios's subnet died,
packets could still get there via B, C or D)
B: Nagios, A, C, D
C: Nagios, A, B, D
D: Nagios, A, B, C
E: A, F, G, H
F: A, E
G: A, E
H: A, E
- Stephen P. Schaefer
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: SSchaefer@rfmd.com