Openshift & Kubernetes monitoring

Openmaru Cloud APM optimized for Openshift & Kubernetes

OPENMARU APM is the first product in Korea that can monitor Docker containers that cannot be monitored with existing APM and WAS running on Docker containers.

Monitoring issues in PaaS-type cloud

Difficult to connect directly to the virtual OS environment
- Because it is a virtual OS environment, CPU/Memory/Disk/Network monitoring is not possible.
- Difficulty accessing or modifying configuration files or log files
POD has no state information
- Disabled POD status information cannot be checked / root cause cannot be found in case of failure
- Project and APM monitoring standards, which are open shift management units, are different
- IP changes every time POD is restarted
- Possible correlation between POD and node
- Impossible to monitor as it is operated based on a POD that does not have a state
- Inability to access historical information about POD
- Impossible to transmit monitoring information via UDP

Issues when auto scaling occurs
- During auto-scaling, it is not possible to know the load distribution and processing status of each pod
- How to view the CPU usage of Pods based on autoscaling
WAS Operational Issues
- In case of failure, the thread dump file required to identify the main cause is not created.
- Unable to analyze out-of-memory situations, which are particularly problematic when operating Java-based systems
- Unable to get information about Java Heap usage

OPENMARU APM optimized for PaaS-type cloud

Provide information necessary for operation in virtual OS environment

Real-time provision of CPU/Memory/Disk/Network information in virtual OS environment

Auto scaling visual tool provided

Load distribution during auto scaling and identification of processing status by pod

POD provides status information

Provide search and status information for retired POD instances

Provide WAS essential monitoring tools

Provide essential analysis tools in WAS failure situations

Container monitoring function

In a container environment, overall monitoring of distributed pod locations, container resources, etc. is required. With Openmaru Cloud APM, operators can conveniently monitor containers (pods).

Provides the number of container images and container status
Provide the container’s CPU utilization
Providing container and image information
Provides the container’s memory utilization

Performance monitoring function of WAS operated in container

When monitoring WAS resources and WAS applications in a container environment, you must use a specific tool or access the Docker container to check logs. Openmaru Cloud APM monitors WAS resources and applications connected by specific groups. Provides WAS resource and WAS application monitoring function in container.

Container WAS resource (HEAP memory) monitoring (check container HEAP memory usage)
Real-time monitoring of container WAS applications (check real-time TPS, active user, and response time for each container)
Container WAS error rate monitoring (check the error rate (4xx, 5xx) in container WAS)
Container WAS APDEX monitoring (monitoring the APDEX Score of container WAS)

Monitoring function for stateless WAS instance in PaaS environment

Openmaru Cloud APM provides a way to obtain information to identify the cause of failure after a non-existent container stops in the PaaS environment. Since information about containers that do not exist is stored, it is possible to identify the cause of failure and take corrective action.

Support for monitoring non-existent containers
Real-time data collection (record non-state information)
- Stateless instances can be retrieved
- Possible to search data at the point of failure that disappeared
- Can identify the cause of instance failure

Volatile containers can be tracked and monitored

When stopped in a container environment, it is impossible to analyze logs and track the cause of container failure, but Openmaru Cloud APM keeps all records of volatile instances, so it is possible to accurately identify the cause in case of failure by tracking container data in the past.

OPENMARU Cloud APM specialized in troubleshooting

Openmaru Cloud APM is a monitoring solution that can operate the service in an optimal state through application performance monitoring and diagnosis, and failure cause analysis.

Thread dump analyzer (Lock traceable, URL information displayed)
JVM memory object analyzer (analysis/comparison of objects occupying Java memory)
Network Status Analyzer (Analyzes Java processes, networks the system is using)
Open file analyzer (Analyzes files opened by Java process)
System Process Analyzer (Analyze, compare process CPU, memory usage on your system)
Data trend analysis (a function that analyzes the trend of increase or decrease in past data)

Application error monitoring function

Openmaru Cloud APM can track exceptions in applications.
You can check which type of exception occurred by tracing multiple exceptions. Openmaru Cloud APM can also track WARN and Error Level values output from the logging library.

Warning messages can be checked according to the logging level
When SQL message occurs, trace and check exception and error message
Error Message can be checked when an Out of Memory Error occurs
Runtime exception can be confirmed as Error Message

Even container metering with Openmaru Cloud APM!

Every time a container is deployed, the existing container is discarded and a new container is created, and it is very difficult to track because it is auto-scaling according to the load.

In some cases, a container floats for only 10 or 20 minutes a day and then disappears, so applying a simple daily billing policy may result in an unreasonable amount.

However, OpenMaru APM calculates the billing based on the average value of CPU and memory usage per hour, so accurate billing can be done.

with cloud native
Optimized for hybrid environments
OPENMARU Cloud APM

15-day free start