====== Server stability issues ======
===== Monitoring stops =====
==== Symptoms ====
  * Monitoring stops
  * No metrics generated
  * Application is unreachable

==== Possible causes ====
  * Not enough heap memory vs application needs
  * Memory outage caused by memory leak or hanging network connections

==== Resolution ====
  * Check configured startup memory matches recommended amount (see [[..:..:installguide:prerequisites]]) and update [[..:..:installguide:installredpeaks#application_memory|setenv.sh file]]
  * If the problem keeps happening even after increasing the memory, we might be looking at a memory leak issue.
  * This can be investigated by doing a [[..:..:troubleshooting:procedures#heap_dump|memory dump]] and a [[..:..::troubleshooting:procedures#stack_trace|stack trace]] as soon as the problem happens
  * Open a ticket describing your problem with full details and attach heap dump and stack trace files.

===== The server seems to restart on its own regularly =====
==== Possible causes ====
  * On linux, a possible cause is simply that the server is running out of memory and the OS decided to kill the service.
  * You can check this by running:
  * <code>dmesg -T| grep -E -i -B100 'killed process'</code>
  * You will see if the promonitor process has been killed because of out of memory problems
  * You can also check about available memory by running:
  * <code>free -m</code>
  * The memory truly available to application is stated below **available** in MB. Make sure there is at least enough memory for the monitoring process to start, according to its startup max memory (1 / 2 / 3 GB or more)

==== Resolution ====
  * The only way to solve this is to increase the memory of the host.