Client task rebooting, out of kernel queues

We have a Citect SCADA 2016 (Aug13,2019) update client/server HMI1 that periodically reboots the client task, syslog shows "Out of kernel queues Mem.Grp 5839"

Parameter [KERNEL] Queue = 6000

How can we fix this ?

  • The are several causes which could explain the out of kernel queues hardware alarm restarting:
    - Make sure you close handles where necessary, and avoid creating new handles if you cannot close them (e.g. devopen and devclose).
    - Make sure you no not have increasing numbers of endless cicode loops, or cicode which fails to complete.
    This would need some further investigation. Check using the kernel:
    - CPU Usage of the client process on the Page General
    - The Task Late time (on the Page General) to see if the hardware is fast enough
    - Page Table Cicode to see if there are a lot of Cicode tasks running in the background.
    - Page Table Stats, to get an overview of the statistics
    - Page Table Profile, to see if functions are called with extreme high rates
    There are a lot of things you can monitor inside the kernel, check the Citect Help.