Bug ID 1467273: LCD restarting internal services periodically due to memory allocation error.

Last Modified: Jul 03, 2025

Affected Product(s):
F5OS F5OS-A(all modules)

Known Affected Versions:
F5OS-A 1.5.1, F5OS-A 1.5.2, F5OS-A 1.5.3, F5OS-A 1.7.0

Fixed In:
F5OS-A 1.8.0

Opened: Dec 20, 2023

Severity: 3-Major

Symptoms

On a r2000/r4000 system, the LCD may experience an issue causing it to repeatedly restart the ‘platform_monitor’ or ‘bmcservice’ service due to a memory allocation problem. The lcd.log file displays recurring patterns of log messages such as: lcd platform_monitor[1581] Traceback (most recent call last):#012 File "/usr/sbin/platform_monitor.py", line 540, in <module>#012 raise e#012OSError: [Errno 12] Cannot allocate memory lcd /etc/watchdog.d/platform_monitor_watchdog.sh: found platform_monitor is not running <snip> lcd watchdog[1424]: test binary /etc/watchdog.d/platform_monitor_watchdog.sh returned 255 lcd /etc/watchdog.d/platform_monitor_watchdog.sh: Trying to repair platform_monitor ... <snip> lcd /etc/watchdog.d/platform_monitor_watchdog.sh: platform_monitor successfully repaired, pid = 27676 A similar condition may occur for the 'bmcservice' service. lcd /etc/watchdog.d/bmcservice_watchdog.sh: found bmcservice is not running lcd watchdog[1436]: test binary /etc/watchdog.d/bmcservice_watchdog.sh returned 255 lcd /etc/watchdog.d/bmcservice_watchdog.sh: Trying to repair bmcservice ... lcd /etc/watchdog.d/bmcservice_watchdog.sh: bmcservice successfully repaired, pid = 26226

Impact

While the LCD is operating in this state, it may not be able to obtain the PSU status from the BMC and update the PSU status LEDs accordingly. Furthermore, the LCD may not be able to ascertain the system’s current power state from the BMC.

Conditions

Excessive LCD memory consumption may occur if the appliance is turned off while external power is still connected. This excess memory usage will not be resolved when the appliance is turned on again.

Workaround

The excessive LCD memory consumption can be recovered by rebooting the LCD. 1) Log into the appliance as a root user. 2) Issue `docker exec -it platform-hal psf call POST:lcd/reboot waitForBootup=true` at the Linux prompt. 3) Wait 60–90 seconds for the LCD to complete a reboot. Example: [root@appliance-1:Active] ~ # docker exec -it platform-hal psf call POST:lcd/reboot waitForBootup=true field | value ----------+-------- success | true

Fix Information

The excessive LCD memory consumption is fixed in LCD firmware v1.01.069.00.1 and later.

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips