Bug ID 727031: TMM restart in B2250 vCMP systems, and ping/monitor failures in non-B2250 vCMP systems.

Last Modified: Jul 12, 2023

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

Known Affected Versions:
15.1.2, 13.1.3.5

Fixed In:
14.1.0, 13.1.3.6, 12.1.5.3

Opened: Jul 06, 2018

Severity: 1-Blocking

Symptoms

-- On B2250 blades the Traffic Management Microkernel (TMM) may experience a series of panics and restarts. The system reports a related message in /var/log/tmm: vdag failed to attach -- On other types of BIG-IP systems (e.g., iSeries) ICMP monitors (or manual pings from the command line) may fail, indicating incorrectly that a node which is known to be up is down, and a packet capture will show ICMP echo reply packets arriving, but being ignored.

Impact

-- For B2250 configurations, tmm restarts. Traffic disrupted while tmm restarts. -- For non-B2250 configurations, ICMP monitors are incorrectly marked down.

Conditions

-- Guest BIG-IP instance in a vCMP configuration. -- Guest is running BIG-IP v13.1.3.5. -- Host is running a software version other than BIG-IP v13.1.3.5. -- For the tmm panic issue, the host is a B2250 blade.

Workaround

None. Note: If installing 13.1.3.5 on a vCMP host or guest, please contact F5 Support and ask for an engineering hotfix with the fix for this issue. Install it on the vCMP host and all guests on that host running 13.1.3.5.

Fix Information

TMM disruption no longer occurs from disaggregation in vCMP systems.

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips