Bug ID 2690389: A DNS cache containing many resource records may cause a tmm restart when the last Virtual Server referencing it is deleted

Last Modified: Oct 22, 2026

Affected Product(s):
BIG-IP DNS(all modules)

Known Affected Versions:
17.1.3.6, 17.5.1.10, 21.1.0.3

Opened: Oct 16, 2026

Severity: 4-Minor

Symptoms

When a DNS cache resolver has a very large number of cached resource records (in the order of millions) and the last Virtual Server referencing the cache is deleted, the tmm process might take a long time to remove the cache and free its memory, might miss its heartbeat signal and might be restarted. See https://my.f5.com/manage/s/article/K9231 for more details on tmm and its heartbeat timeout.

Impact

The TMM process is prevented from servicing its heartbeat and can be restarted

Conditions

- DNS module provisioned - DNS cache resolver configured on a DNS profile and in use - The DNS cache resolver contains a large number of cached resource records (in the order of millions) - The last Virtual Server referencing the DNS profile and the DNS cache resolver is deleted

Workaround

Before deleting the last Virtual Server referencing a DNS cache, wait until most of the Resource Records present in the cache have expired. The number of resource records in the cache can be monitored with this command: "tmsh show ltm dns cache records rrset cache <CACHE_NAME> count-only" Check K42437613 for more details

Fix Information

None

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips