Running and monitoring workload or tasks

FAQs about running and monitoring IBM® Spectrum Symphony workload or MapReduce tasks.

Why won't my service start?
Why are there blocked hosts for my application?
Why can't I view completed session or task information in the CLI or the cluster management console?
I have several clients connected with the same application. Why do I get two different exceptions when I disable the application?
Why is my task still running even after I suspended its owning session?
Will a session be closed or aborted if it becomes inactive for a certain time?
Why are so many of my tasks pending?
After aborting or suspending a session, the SSM dies. Why is the session still open after the SSM recovers?
How do I find the URL for the cluster management console?
Why is my session director not recovering like I want it to?
How are port numbers configured and how do I change them?
Why are tasks running on my primary host?
When a cluster is not configured for failover, by default tasks can execute on any resource group (assuming that the default configuration for resReq and ResourceGroupName is used in Consumer section of the application profile).
Why can't I run symping on the command console when I can run it on the cluster management console?
Why does IBM Spectrum Symphony report task input, task output, and common data sizes that are larger than my actual data sizes?
Why does session recovery fail after the SSM dies?
Why does the SSM remain at a critical memory level when there is enough available memory and not much unfinished workload?
Why is my session demanding fewer or more resources than required for all of its pending tasks?
What happens if one service wants to block the host while there are still other services running tasks on the same slot?
My cluster is large and some SIMs are failing to connect to the SSM with error -64. How do I configure the system to avoid this problem?
Why doesn't my client reconnect to the SSM after SSM failover?
Why do some service instances not start in the 40K SI environment?
What does Rack-local Match Ratio or Off-rack Match Ratio mean?
CPU usage on my Linux on POWER system is always 100%. Why?
How do I fix Address already in use messages in the SIM log?
Why are the static metrics for my GPU application not current?
Why are reduce tasks always pending?

Parent topic: FAQs

Parent topic: Developer Edition FAQs