Last Modified: Oct 09, 2026
Affected Product(s):
BIG_IP_NEXT(CNF) CNF
Known Affected Versions:
1.4.0, 1.4.1
Fixed In:
1.4.2
Opened: Mar 11, 2026 Severity: 3-Major
An ICMP health monitor can transition a pool member to DOWN even when probe failures are not consecutive. In environments with intermittent packet loss, the monitor can mark the endpoint down after a single failed probe, even if successful probes occur between failures. With the default down threshold of 3, the sequence FAIL ? OK ? FAIL ? OK ? FAIL triggers a DOWN state instead of requiring three consecutive failures.
Pool members trigger DOWN status even when the endpoint is reachable. This leads to incorrect pool member health state reporting.
Issue occurs when the default value of monitor DOWN threshold is set to 3 and - An ICMP health monitor is configured for a pool member. - Probe failures occur intermittently (e.g., due to packet filtering, network instability, or packet loss). - The system counts failed probes cumulatively rather than resetting the counter after successful probes.
If the endpoint responds successfully soon after being marked DOWN, that suggests intermittent probe failures rather than a persistent outage. If the DOWN and subsequent UP events align with the monitor probe interval, the endpoint is likely failing individual probes intermittently, not experiencing a continuous outage.
Add a fail_count field to tma_activity. Reset fail_count on each successful probe and increment it by 1 on each failed probe. Transition the pool member to DOWN only after 3 consecutive failed probes.