Cisco Headend System Release 2.7 テクニカルリファレンス
Field Notice:
Preventing Failures of the SRM
Processes
Background
Our engineers are aware of a condition that may corrupt the memory of the
System Resource Management (SRM) processes of the Digital Network Control
System (DNCS). This condition can occur under all system releases of the Digital
Broadband Delivery System (DBDS) software. Memory of the SRM processes
may sometimes become corrupted if the processes are not stopped and restarted
in a specific order.
Note: The dsm, drm, qamManager, pkeManager, camEx, and camPsm processes
together make up the SRM processes.
The SRM processes are dependent upon one another. The SRM processes risk
corrupting system memory should any SRM process be stopped and restarted
without a stop and restart of the other SRM processes, as well. System operators
will know that an SRM process has failed when the indicator light on the DNCS
Control window changes from green to red for any of the individual SRM
processes. New sessions will not start during the time the SRM processes are
down.
System Resource Management (SRM) processes of the Digital Network Control
System (DNCS). This condition can occur under all system releases of the Digital
Broadband Delivery System (DBDS) software. Memory of the SRM processes
may sometimes become corrupted if the processes are not stopped and restarted
in a specific order.
Note: The dsm, drm, qamManager, pkeManager, camEx, and camPsm processes
together make up the SRM processes.
The SRM processes are dependent upon one another. The SRM processes risk
corrupting system memory should any SRM process be stopped and restarted
without a stop and restart of the other SRM processes, as well. System operators
will know that an SRM process has failed when the indicator light on the DNCS
Control window changes from green to red for any of the individual SRM
processes. New sessions will not start during the time the SRM processes are
down.
Recommendation
If system operators notice that any of the SRM processes are stopped, they should
ideally stop and restart all DNCS processes.
However, if a stop and a restart of all DNCS processes is not desirable, then a stop
and restart of the aforementioned SRM processes ought to be conducted in the
specific order set forth in the following sections. Furthermore, the stopping and
restarting of the SRM processes should be done while within a maintenance
window. No new sessions can be set up while the SRM processes are down.
ideally stop and restart all DNCS processes.
However, if a stop and a restart of all DNCS processes is not desirable, then a stop
and restart of the aforementioned SRM processes ought to be conducted in the
specific order set forth in the following sections. Furthermore, the stopping and
restarting of the SRM processes should be done while within a maintenance
window. No new sessions can be set up while the SRM processes are down.
Stop Remaining SRM Processes
System operators should stop any remaining SRM process that is still running in
the following order:
1
the following order:
1
dsm
2
drm