Merge pull request #151 from claranet/genericjmx

middleware/genericjmx/README.md

@@ -0,0 +1,45 @@
+# MIDDLEWARE JMX SignalFx detectors
+
+## How to use this module
+
+```hcl
+module "signalfx-detectors-middleware-jmx" {
+  source      = "github.com/claranet/terraform-signalfx-detectors.git//middleware/jmx?ref={revision}"
+
+  environment = var.environment
+  notifications = var.notifications
+}
+```
+
+## Purpose
+
+Creates SignalFx detectors with the following checks:
+- JMX GC old generation usage	
+- JMX memory heap usage	
+
+## Notes
+
+This module uses the [GenericJMX](https://docs.signalfx.com/en/latest/integrations/agent/monitors/collectd-genericjmx.html)
+monitor to fetch common Java runtime metrics for every JVM based applications.
+
+You must [enable JMX Remote](https://docs.oracle.com/javadb/10.10.1.2/adminguide/radminjmxenabledisable.html) on your JAVA
+application. Depending on your application you should add following paramters as example:
+
+```
+-Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.port=5000 -Dcom.sun.management.jmxremote.local.only=false -Dcom.sun.management.jmxremote.authenticate=false -Dcom.sun.management.jmxremote.ssl=false -Djava.rmi.server.hostname=127.0.0.1
+```
+
+If there is a native GenericJMX based monitor for your specific application like for
+[Cassandra](https://docs.signalfx.com/en/latest/integrations/agent/monitors/collectd-cassandra.html)
+so you should configure its dedicated monitor and you will automatically retrieve required metrics for this module.
+
+Else if there is no monitor available for your specific application or you simply do not want to collect specific
+application metrics, so you have to configure the GenericJMX directly:
+
+```
+-   type: collectd/genericjmx
+    host: 127.0.0.1
+    port: 5000
+```
+
+Keep in mind you can easily add specific application metrics defining `mBeanDefinitions` parameter.

middleware/genericjmx/common-locals.tf

@@ -0,0 +1 @@
+../../common/locals.tf

middleware/genericjmx/common-variables.tf

@@ -0,0 +1 @@
+../../common/variables.tf

middleware/genericjmx/detectors-genericjmx.tf

@@ -0,0 +1,62 @@
+resource "signalfx_detector" "jmx_memory_heap" {
+  name = format("%s %s", local.detector_name_prefix, "JMX memory heap usage")
+
+  program_text = <<-EOF
+    A = data('jmx_memory.used', filter=filter('plugin_instance', 'memory-heap') and ${module.filter-tags.filter_custom})${var.memory_heap_aggregation_function}${var.memory_heap_transformation_function}
+    B = data('jmx_memory.max', filter=filter('plugin_instance', 'memory-heap') and ${module.filter-tags.filter_custom})${var.memory_heap_aggregation_function}${var.memory_heap_transformation_function}
+    signal = (A/B).scale(100).publish('signal')
+    detect(when(signal > ${var.memory_heap_threshold_critical})).publish('CRIT')
+    detect(when(signal > ${var.memory_heap_threshold_major}) and (signal < ${var.memory_heap_threshold_critical})).publish('MAJOR')
+EOF
+
+  rule {
+    description           = "is too high > ${var.memory_heap_threshold_major}"
+    severity              = "Major"
+    detect_label          = "MAJOR"
+    disabled              = coalesce(var.memory_heap_disabled_major, var.memory_heap_disabled, var.detectors_disabled)
+    notifications         = coalescelist(lookup(var.memory_heap_notifications, "major", []), var.notifications.major)
+    parameterized_subject = local.rule_subject
+    parameterized_body    = local.rule_body
+  }
+  rule {
+    description           = "is too high > ${var.memory_heap_threshold_critical}"
+    severity              = "Critical"
+    detect_label          = "CRIT"
+    disabled              = coalesce(var.memory_heap_disabled_critical, var.memory_heap_disabled, var.detectors_disabled)
+    notifications         = coalescelist(lookup(var.memory_heap_notifications, "critical", []), var.notifications.critical)
+    parameterized_subject = local.rule_subject
+    parameterized_body    = local.rule_body
+  }
+}
+
+resource "signalfx_detector" "jmx_old_gen" {
+  name = format("%s %s", local.detector_name_prefix, "JMX GC old generation usage")
+
+  program_text = <<-EOF
+    A = data('jmx_memory.used', filter=filter('plugin_instance', 'memory_pool-G1 Old Gen') and ${module.filter-tags.filter_custom})${var.gc_old_gen_aggregation_function}${var.gc_old_gen_transformation_function}
+    B = data('jmx_memory.max', filter=filter('plugin_instance', 'memory_pool-G1 Old Gen') and ${module.filter-tags.filter_custom})${var.gc_old_gen_aggregation_function}${var.gc_old_gen_transformation_function}
+    signal = (A/B).scale(100).publish('signal')
+    detect(when(signal > ${var.gc_old_gen_threshold_critical})).publish('CRIT')
+    detect(when(signal > ${var.gc_old_gen_threshold_major}) and (signal < ${var.gc_old_gen_threshold_critical})).publish('MAJOR')
+EOF
+
+  rule {
+    description           = "is too high > ${var.gc_old_gen_threshold_major}"
+    severity              = "Major"
+    detect_label          = "MAJOR"
+    disabled              = coalesce(var.gc_old_gen_disabled_major, var.gc_old_gen_disabled, var.detectors_disabled)
+    notifications         = coalescelist(lookup(var.gc_old_gen_notifications, "major", []), var.notifications.major)
+    parameterized_subject = local.rule_subject
+    parameterized_body    = local.rule_body
+  }
+  rule {
+    description           = "is too high > ${var.gc_old_gen_threshold_critical}"
+    severity              = "Critical"
+    detect_label          = "CRIT"
+    disabled              = coalesce(var.gc_old_gen_disabled_critical, var.gc_old_gen_disabled, var.detectors_disabled)
+    notifications         = coalescelist(lookup(var.gc_old_gen_notifications, "critical", []), var.notifications.critical)
+    parameterized_subject = local.rule_subject
+    parameterized_body    = local.rule_body
+  }
+}
+

middleware/genericjmx/modules.tf

@@ -0,0 +1,8 @@
+module "filter-tags" {
+  source = "github.com/claranet/terraform-signalfx-detectors.git//common/filter-tags"
+
+  filter_defaults        = "filter('env', '${var.environment}') and filter('sfx_monitored', 'true')"
+  filter_custom_includes = var.filter_custom_includes
+  filter_custom_excludes = var.filter_custom_excludes
+}
+

middleware/genericjmx/outputs.tf

@@ -0,0 +1,10 @@
+output "jmx_memory_heap" {
+  description = "Detector resource for jmx_memory_heap"
+  value       = signalfx_detector.jmx_memory_heap
+}
+
+output "jmx_old_gen" {
+  description = "Detector resource for jmx_old_gen"
+  value       = signalfx_detector.jmx_old_gen
+}
+

middleware/genericjmx/variables.tf

@@ -0,0 +1,102 @@
+# Module specific
+
+# memory_heap detector
+
+variable "memory_heap_disabled" {
+  description = "Disable all alerting rules for memory_heap detector"
+  type        = bool
+  default     = null
+}
+
+variable "memory_heap_disabled_critical" {
+  description = "Disable critical alerting rule for memory_heap detector"
+  type        = bool
+  default     = null
+}
+
+variable "memory_heap_disabled_major" {
+  description = "Disable major alerting rule for memory_heap detector"
+  type        = bool
+  default     = null
+}
+
+variable "memory_heap_notifications" {
+  description = "Notification recipients list per severity overridden for memory_heap detector"
+  type        = map(list(string))
+  default     = {}
+}
+
+variable "memory_heap_aggregation_function" {
+  description = "Aggregation function and group by for memory_heap detector (i.e. \".mean(by=['host']).\")"
+  type        = string
+  default     = ""
+}
+
+variable "memory_heap_transformation_function" {
+  description = "Transformation function for memory_heap detector (i.e. \".mean(over='5m')\")"
+  type        = string
+  default     = ".min(over='5m')"
+}
+
+variable "memory_heap_threshold_major" {
+  description = "Warning threshold for memory_heap detector"
+  type        = number
+  default     = 80
+}
+
+variable "memory_heap_threshold_critical" {
+  description = "Critical threshold for memory_heap detector"
+  type        = number
+  default     = 90
+}
+
+# gc_old_gen detector
+
+variable "gc_old_gen_disabled" {
+  description = "Disable all alerting rules for gc_old_gen detector"
+  type        = bool
+  default     = null
+}
+
+variable "gc_old_gen_disabled_critical" {
+  description = "Disable critical alerting rule for gc_old_gen detector"
+  type        = bool
+  default     = null
+}
+
+variable "gc_old_gen_disabled_major" {
+  description = "Disable major alerting rule for gc_old_gen detector"
+  type        = bool
+  default     = null
+}
+
+variable "gc_old_gen_notifications" {
+  description = "Notification recipients list per severity overridden for gc_old_gen detector"
+  type        = map(list(string))
+  default     = {}
+}
+
+variable "gc_old_gen_aggregation_function" {
+  description = "Aggregation function and group by for gc_old_gen detector (i.e. \".mean(by=['host']).\")"
+  type        = string
+  default     = ""
+}
+
+variable "gc_old_gen_transformation_function" {
+  description = "Transformation function for gc_old_gen detector (i.e. \".mean(over='5m')\")"
+  type        = string
+  default     = ".min(over='5m')"
+}
+
+variable "gc_old_gen_threshold_major" {
+  description = "Warning threshold for gc_old_gen detector"
+  type        = number
+  default     = 80
+}
+
+variable "gc_old_gen_threshold_critical" {
+  description = "Critical threshold for gc_old_gen detector"
+  type        = number
+  default     = 90
+}
+

middleware/genericjmx/versions.tf

@@ -0,0 +1,9 @@
+terraform {
+  required_providers {
+    signalfx = {
+      source  = "splunk-terraform/signalfx"
+      version = ">= 4.26.4"
+    }
+  }
+  required_version = ">= 0.12.26"
+}