Create Performance SLAs¶
A Performance SLA measures the health of each SD-WAN member via active probes (ping, HTTP, DNS, TCP echo). Members that fail the SLA are temporarily removed from routing. SLAs are what make SD-WAN react to brownouts, not just hard failures.
Steps¶
- Network → SD-WAN → Performance SLAs → + Create New.
- Fill in:
- Name — e.g.
Google-DNS-Health. - Protocol —
Ping(most common),HTTP,DNS,TCP echo,UDP echo,TWAMP. - Server — the target. For ping:
8.8.8.8. For HTTP: a URL. For DNS: a server IP. - Participants — which SD-WAN members to test.
- Check Interval (s) —
500ms (default). - Failures before Inactive —
5consecutive failures = mark member inactive. - Restore Link After —
5consecutive successes = mark active again. - SLA Targets — define thresholds:
- Latency Threshold — e.g.
100ms. - Jitter Threshold — e.g.
30ms. - Packet Loss Threshold — e.g.
1%.
- Latency Threshold — e.g.
- Member SLAs out of: members that meet ALL thresholds are "in SLA."
- Name — e.g.
- OK.
📸 Screenshot needed
Performance SLA create form with thresholds + Participants picker.
CLI Equivalent¶
config system sdwan
config health-check
edit "Google-DNS-Health"
set server "8.8.8.8"
set protocol ping
set interval 500
set failtime 5
set recoverytime 5
set members 1 2
config sla
edit 1
set latency-threshold 100
set jitter-threshold 30
set packetloss-threshold 1
next
end
next
end
end
Verify¶
diagnose sys sdwan health-check
Shows live latency/jitter/loss per member with SLA pass/fail.
In GUI: Network → SD-WAN → Performance SLAs — graph of metrics over time.
Common Issues¶
- All members fail SLA. Threshold too tight, or target unreachable. Loosen, or change target.
- Flapping in/out. Threshold too close to actual values; small variance flips state. Increase the failure-count.
- SLA passes but actual app feels bad. Threshold too generous. Drop latency threshold.