Bug ID 2288173: Some TMMs not ready and failed to bring up full cluster due to failed cmp dag transition

Last Modified: Apr 24, 2026

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

Known Affected Versions:
17.5.1, 17.5.1.2, 17.5.1.3, 17.5.1.4, 17.5.1.5

Opened: Apr 22, 2026

Severity: 3-Major

Symptoms

On VELOS chassis platforms, when you reboot or restart the tenant, the cluster fails to come up fully with some TMMs indicating tmm-not-ready state, and performance is degraded as it fails to bring up the full cluster.

Impact

Performance degraded due to reduced cluster size.

Conditions

VELOS chassis platforms, with scenarios leading to reboot or restart of the tenant, possibly triggered by - some software upgrade - some power reset or - configuration change causes occasional problems in tmm cluster bring-up and reduces the capacity handled by the tenant. When the problem happens, it is observed that - tmctl tmm/cmp shows queue_drops - tmctl tmm/mpi_mem shows tx-full Due to a lot of internal background traffic in the cluster and tmctl tmm/ready_for_world_stat indicates "not read" state for "dag_transition".

Workaround

No Workaround As it is an intermittent problem, reboot/restart the problematic blade may help to recover.

Fix Information

None

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips