healthcheck

Documentation

Index

• Variables
• func GetRebootable(prom promtest.PromClient, minutes int) ([]node.Node, error)

Variables

var (
	// NodeQuery is a Prometheus query to determine what nodes aren't reachable
	// on any of the currently used SSH ports (22 or 806). Additional filtering
	// based on gmx maintenance, lame-duck mode, or the presence of recent NDT
	// tests on the node is applied. This makes sure we never reboot a node
	// unnecessarily or lose data.
	NodeQuery = `` /* 653-byte string literal not displayed */

	// SwitchQuery is a prometheus query to determine what switches are
	// offline.  To determine if a switch is offline, pings are generally
	// more reliable than SNMP scraping.
	SwitchQuery = `sum_over_time(probe_success{instance=~"s1.*", module="icmp"}[15m]) == 0 unless on(site) gmx_site_maintenance == 1`
)

Functions

func GetRebootable

func GetRebootable(prom promtest.PromClient, minutes int) ([]node.Node, error)