|
Distributed computing systems, whether in a local data center or on the cloud, are extremely difficult to troubleshoot. When a problem occurs, gathering basic information involves connecting to each machine individually (this can mean dozens or hundreds of machines), generating a system dump and downloading logs. What's more, it is almost impossible to get this information in real time, immediately after an error occurs - not to mention getting it when it's most needed, before the error occurs.
Instant Snapshot of Entire Distributed Systems
With GigaSpaces XAP 7.1, administrators can do just that. This Early Access release showcases the ability to generate distributed dump and log gathering at the click of a button. Administrators can instantly generate a snapshot of an entire distributed system, including JVM thread and heap dumps, network utilization report, logs, and GigaSpaces XAP middleware status, no matter how many machines there are or where they are physically hosted.

It is also possible to view constantly-updated logs for any machine, within the administration UI - including the ability to filter log entries based on a regular expression (e.g. view only messages that contain the word INFO or WARN).

Event-Driven Troubleshooting
XAP is not limited to on-demand dump and log gathering. Using the administration API, new in XAP 7.0, administrators can request dumps and logs to be generated automatically when events occur anywhere in the system. For example, each time a machine exceeds a certain CPU or memory utilization threshold, dumps and logs can be automatically taken.
This enables true real-time troubleshooting of a distributed system. Administrators can view status of entire distributed systems at the precise moment a problem occurs, making it much easier to track down the cause. They can also become aware of problems across the distributed system before they occur, and take targeted corrective action.
Download the latest milestone of XAP 7.1
Learn more about GigaSpaces XAP
|