monitor veth count on OpenShift nodes

Description

We've got an issue with veth count rising on certain bare metal nodes which causes dnsmask to fail on them and break the node.

It's quite trivial to monitor this from a master using even a bash one-liner such as:
oc get nodes | cut -d" " -f1 | grep -v NAME | xargs -I % -n 1 ssh % "ip a | grep veth| wc -l"

Let's add this to nagios to have the information before it causes problems.

Assignee

infra

Reporter

Evgheni Dereveanchin

Blocked By

None

Priority

Medium
Configure