yCrash Agent
--------------

+ In the previous release we built a yc agent with dynamic link. Due to that, the yc agent was not working on Linux Alpine distribution. Inorder to fix this issue we had to install “libc6-compat” package in Linux Alpine. Now in this release we are building a yc agent with static-link. The yc agent will work without the “libc6-compat” package in Linux Alpine and in other Linux distributions.

+ In Kubernetes' world, pods are considered to be relatively ephemeral (rather than durable) entities. Means, we cannot expect a pod to be a long running resource. When there are any availability issues or SLA breaches, pods will get recycled. Before they get recycled, we should capture troubleshooting artifacts (such as GC logs, thread dump, heap dump, netstat, iostat, top, …) to troubleshoot the problem. We have added support for it. To learn more please visit: https://docs.ycrash.io/ycrash-agent/kubernetes/prestop-hook.html

+ In the previous releases you can only process Id to the '-p' argument. If you don't know process Id, you can also pass a token that will uniquely identify the process amongst all other processes on the device. For example if the application is launched with the argument "webapp-runner-8.0.33.4.jar", you can pass 'webapp-runner-8.0.33.4.jar' as input to this argument. yCrash agent will issue 'ps -ef | grep webapp-runner-8.0.33.4.jar' command, whatever is the resulting process Id, it will use it for capturing the artifacts.

+ Reason why m3 monitoring detected the outage is reported as a 'tag' in the incident report.

yCrash Server
--------------

+ 'pdfReportURL' element is passed in the HeapHero (heap dump analysis) API Response. This element will contain the URL to download the PDF report of the heap dump analysis.

+ License Expiration Date is passed in the '/debug' API response.

+ In m3 monitoring mode, we were taking average of CPU and memory data points. If average crosses the configured threshold value, then alerts were generated. Now we have changed the logic. Only if all data points exceeds the configured threshold limits then alerts are generated.

+ Made enhancements and stabilization to m3 monitoring. 

+ Bug fixes.