Bug ID 1472745: TMM container is restarted by Kubernetes

Last Modified: Nov 29, 2025

Affected Product(s):
BIG_IP_NEXT(CNF) CNF(all modules)

Fixed In:
1.2.1

Opened: Jan 02, 2024

Severity: 2-Critical

Symptoms

The sock driver experiences tx_errors sending out packets on the veth interface (tmctl -d blade -w 250 tmm/xnet/sock/stats), as shown below: ifname q_id mmap_ring_drops rx_buf_err rx_io_err tx_err ------ ---- --------------- ---------- --------- -------- xeth0 0 0 0 0 70828320

Impact

Traffic disrupted while TMM restarts.

Conditions

The TMM container's veth mtu uses a value less than 1500.

Workaround

None

Fix Information

To ensure TMM container doesn't see any disruption in control plane and k3s traffic that receives on veth interface, TMM matches the mtu value between TMM container's veth interface (eth0) and TMM's internally created VLAN on top of the veth interface.

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips