VM Container Monitor

Use

You can use this monitor to check the utilization of the VM Container, to identify bottlenecks, and to investigate errors. For an introduction to the SAP spool system see the documentation, SAP Virtual Machine Container.

Features

In the monitoring infrastructure, there are various subtrees for monitoring the VM Container. These are integrated in this monitor:

For each instance of the system there are the VMC subtree - VMC General, Java Virtual Machine, Client Requests, Application Cache, and Shared Garbage Collector.

Note

You can get further details on each node by choosing Properties.

Choose F1 help on each node to get detailed information.

Subtree VMC General

Here you can find general information on the VM Container and on the resources it uses (memory areas, VM pool, and so on).

The following table provides information about the monitoring tree elements (MTEs) of this subtree:

MTE Name

(MTE class)

Meaning

(Details in long text for F1 help)

Status

(vmcStatus)

VMC status - active or inactive

Shared Pool Size

(vmcMemory)

Size of the shared pool (shared memory for all VMs)

Shared Memory Usage

(vmcSharedMemUsage)

Size of the memory used from shared pool

Shared Memory Fragmentation

(vmcMemFragment)

Total fragmentation of the shared memory

Shared Memory Classes

(vmcMemForSharedClasses)

Memory consumed by shared classes as a percentage of the entire shared pool

Shared Classes Fragmentation

(vmcMemFragmentClasses)

Fragmentation of the memory area for shared classes

Shared Memory Closures

(vmcMemForSharedClosures)

Memory consumed by shared closures as a percentage of the entire shared pool

Shared Closures Fragmentation

(vmcMemFragmentClosures)

Fragmentation of the memory area for shared closures

Shared Code Cache

(vmcSharedCodeCache)

Memory size of the shared code cache

Shared Code Cache Usage

(vmcMemForSharedCodeCache)

Shared Code Cache Belegung

No. Free VMs

(vmcAvailableVms)

No. Free VMs

No. Pooled VMs

(vmcNoOfPooledVms)

Number of VMs in the VM pool

Sticky VMs

(vmcPartStickyVms)

Number of VMs in "sticky" mode (fixed assignment to a user context based on high memory requriement)

No. VM Errors

(vmcErrors)

Cumulated number of VM errors

More Information

Shared pool, shared classes, shared closures, and so on: Memory Management in the VM Container
VM Properties: VM Overview

Subtree Java Virtual Machines

This monitor provides information about the operation of all active VMs with respect to efficiency and stability.

Some of the values are mean values, others are maximum, minimum, or total values. You can find more information in the F1 help for each node.

The subtree VM Operation provides information about the monitoring tree elements (MTEs):

MTE Name `(MTE class)`	Meaning (Details in long text for F1 help)
Java Heap Size `(vmcJavaHeap)`	Size of the Java Heap for each VM
CPU Usage `(vmcCPUUsageVM)`	CPU used by all active VMs as a percentage
CPU Usage GC `(vmcCPUUsageGCVM)`	Average effective CPU usage for the garbage collection of the VMs as a percentage
Memory Usage `(vmcMemoryUsageVM)`	Average memory consumption of each VM
Max. Memory Usage `(vmcMaxMemoryUsageVM)`	Maximum memory consumption of each VM
Avg. Full GC Rate `(vmcAvgFullGCRateVM)`	Average frequency of full GCs of all VMs
Max. Full GC Rate `(vmcMaxFullGCRateVM)`	Maximum frequency of full GCs of individual VMs
Avg. GC Rate `(vmcAvgGCRateVM)`	Average frequency of GC runs for all VMs
Max. GC Rate `(vmcMaxGCRateVM)`	Maximum rate of GC runs for individual VMs
No. GC Page Faults `(vmcPageFaultRateGC)`	Rate of failed attempts to read the GC memory
Mem. Availability (single VM) `(vmcMaxGCWaterLevelVM)`	Memory availability for individual VMs
Average Memory Availability `(vmcAvgGCWaterLevel)`	Average memory availability for all VMs

More Information

Subtree Application Cache

The VM Container cache is a cache for any Java objects.

MTE Name

(MTE class)

Meaning

(Details in long text for F1 help)

Max. Size

(vmcCacheSize)

Current size of the application cache

Access Rate

(vmcCacheAccessRate)

Frequency of accesses to cache

Success Rate

(vmcCacheSuccessRate)

Frequency of Successfuly Access to Cache

No. Sync. Evictions

(vmcSyncEvictionNo)

Number of deleted old cache entries when adding new entries

Size Sync. Evictions

(vmcSyncEvictionSize)

Size of deleted old cache entries when adding new entries

No. Async. Evictions

(vmcAsyncEvictionNo)

Number of deleted old cache entries in the background

Size Async. Evictions

(vmcAsncEvictionSize)

Size of deleted old cache entries in the background

More Information

Caching in the VM Container

Subtree Shared Garbage Collection

The shared garbage collector (SGC) removes objects no longer needed from the shared pool (shared closures).

MTE Name

(MTE class)

Meaning

(Details in long text for F1 help)

Max. Water Level

(vmcMaxSharedGCWaterLevel)

Maximum relative occupancy of shared pool by shared GC.

Shared Memory Availability

(vmcSharedMemoryAvailable)

Memory availability in the shared pool

Shared GC Rate

(vmcSharedGCRate)

Frequency of the shared garbage collector runs

Shared GC Duration

(vmcSharedGCDuration)

Total duration of specified shared GCs

CPU Usage Shared GC

(vmcCPUUsage)

CPU usage for all specified shared GCs

Avg. Reclaimed Memory

(vmcShGCMemoryReclaimed)

Average size of released memory for each shared GC (in kB)

More Information

Subtree Client Requests

The node RFC Container contains the following subtrees.

MTE Name

(MTE class)

Meaning

(Details in long text for F1 help)

State

(vmcRFCContainerState

Container status

RFC Request Rate

(vmcRFCRequestRate)

Frequency of client requests

Response Time

(vmcRFCProcessingTime)

Average processing time of each client request

Activities

To start the monitor, follow the procedure below:

Start the Alert Monitor using transaction RZ20 or choose CCMS Control/Monitoring Alert Monitor .
On the CCMS Monitor Sets screen, expand the SAP CCMS Monitor Templates set.
Start the VM Container monitor from the list by double-clicking it.