Download PDF
Download page Container Metrics.
Container Metrics
To access the Containers Dashboard, select: AppDynamics Home > Servers > Containers.
The Containers Dashboard shows resource utilization metrics of your containerized applications and the underlying host machine. You can access the Containers Dashboard from the Servers tab in the Controller UI. The Containers page lists all monitored containers used by an application that is registered to the Controller.
Container Metrics
CPU
| Name | Description | Default Monitoring Mode |
|---|---|---|
| %Busy | Percentage of time the CPU was busy processing system or user requests from this container; this metric includes CPU Stolen time. | KPI |
| %Busy Scaled | If the -cpus=<value> option is set in the container, this metric measures the scaled value of the CPUs property. If this value is not set, this metric measures the scaled value of the Logical CPUs property. To enable CPU Scaling:
| KPI |
| Logical Cores | Number of logical cores on the host machine that are available to the container. | KPI |
| Shares | The -c or cpu-shares setting used to start the container. | KPI |
| Throttle Count | Number of times the container CPU is throttled. Use Advanced mode to view this metric as Throttled Time. If Throttled Time is less and the container is limited on computation, then this metric is available for you to review. | KPI |
| Throttled Time (ms) | Total amount of time the container CPU was throttled but did not receive enough host CPU. High throttle times indicate that the -c or cpu-shares setting (used to start the container) was too low. | KPI |
| Total % | Average utilization, of all available CPUs, by the container. | KPI |
Disks
| Name | Description | Default Monitoring Mode |
|---|---|---|
| Average IO Utilization % | Average disk utilization for read/write operations. This metric may be unavailable on some platforms. | KPI |
<disk> Reads/sec | Number of read operations by the container on <disk>. | Diagnostic |
<disk> Writes/sec | Number of write operations by the container on <disk>. | Diagnostic |
| KB Read/sec | The rate of data read by the container from all disks and partitions. | KPI |
| KB Written/sec | The rate of data written by the container to all disks and partitions. | KPI |
| MB Reads/sec (MB) | Number of read operations by the container on all disks and partitions. | KPI |
| MB Writes/sec | Number of write operations by the container on all disks and partitions. | KPI |
| Queue Size | Average number of container requests queued during the time bucket. Consistently high queue sizes indicate that the container is not receiving enough IO resources. This metric may be unavailable on some platforms. | KPI |
Memory
| Name | Description | Default Monitoring Mode |
|---|---|---|
| Cache Size (MB) | Total amount of data cached in memory. | Diagnostic |
| Failed Count | Number of times a memory allocation failure occurred because it exceeded the pre-set memory limit for that container; or, number of times the hard limit for the container was reached. | KPI |
| Page Major Fault Count | Number of major page faults caused by the container. Docker metric is pgmajfault. A high rate of page faults could indicate that the specified memory constraints for the container are too low. | KPI |
| Reservation | Soft limit based on the --memory-reservation used to start the container. | KPI |
| Resident Set Size (MB) | Total amount of memory that belongs to stacks, heaps, and anonymous memory maps. | KPI |
| Usage | Memory usage by the container, in bytes. | KPI |
| Used % | Memory utilization by the container as a percentage of the memory hard limit, based on the --memory option used to start the container. | KPI |
| Used MB | Memory utilization by the container, in MB. | KPI |
Network
| Name | Description | Default Monitoring Mode1 |
|---|---|---|
| Count (KB) | Network Volume, count (KB), incoming from an external network and outgoing to an external network. | KPI |
| Rate (KB/s) | Network Rate, rate (KB/s), incoming from an external network and outgoing to an external network. | Diagnostic |
| Incoming (KB) | Network traffic received by the container. | KPI |
| Outgoing (KB) | Network traffic sent by the container. | KPI |
| Incoming (KB/sec) | Rate of network traffic received by the container. | Diagnostic |
| Outgoing (KB/sec) | Rate of network traffic sent by the container. | Diagnostic |
| Incoming Packets | Number of packets received by the container. | Diagnostic |
| Outgoing Packets | Number of packets sent by the container. | Diagnostic |
| Incoming Packets Dropped | Number of incoming packets dropped before reaching the container. A high count of dropped packets may indicate high congestion or an issue on the network path. | KPI |
| Outgoing Packets Dropped | Number of outgoing packets dropped after leaving the container. A high count of dropped packets may indicate high congestion or an issue on the network path. | KPI |
| Incoming Errors | Number of packets received with errors. | KPI |
| Outgoing Errors | Number of packets sent with errors. | KPI |
<eth0> Incoming Errors | Number of collision errors when <eth0> could not receive a frame due to a problem on the interface. | Diagnostic |
<eth0> Outgoing Errors | Number of collision errors when <eth0> could not send a frame due to a problem on the interface. | Diagnostic |
<eth0> Incoming KB | Average KB of traffic received on <eth0>. | Diagnostic |
<eth0> Outgoing KB | Average KB of traffic sent on <eth0>. | Diagnostic |
<eth0> Incoming KB/sec | Average rate of traffic received on <eth0>. | Diagnostic |
<eth0> Outgoing KB/sec | Average rate of traffic sent on <eth0>. | Diagnostic |
<eth0>Incoming Errors | Number of packets received on <eth0> with errors. | Diagnostic |
<eth0> Outgoing Errors | Number of packets sent on <eth0> with errors. | Diagnostic |
<eth0> Incoming Packets Dropped | Number of incoming packets dropped before reaching <eth0>. A high count of dropped packets may indicate high congestion or an issue on the network path. | Diagnostic |
<eth0> Outgoing Packets Dropped | Number of outgoing packets dropped after leaving <eth0>. A high count of dropped packets may indicate high congestion or an issue on the network path. | Diagnostic |