Skip to main content
Version: v11.6.2

Alerts


WaveMaker uses the Prometheus AlertManager to listen to the data accumulated at Prometheus and fire an alert based on the query included in the alert.

Accessing Alerts

The list of Alerts can be accessed from Prometheus UI by clicking on the "Alerts" button as shown in the below image

Prometheus UI Alert Button

Available Alerts

Following are the list of alerts included in WaveMaker

Platform Health

Alert NameDescription
InstanceDownInstance is down.
RootDiskSpaceFullRoot Disk space on an instance is full
PlatformDataDiskSpaceFullPlatform Disk space on an instance is full
ExternalInstanceDataDiskSpaceFullExternal Instance (Backup Server) Disk space is full
RuntimeDiskSpaceFullVolume mounted to a container is full
MemoryFullLess Instance Memory is available
HighCPUThe CPU is used extensively
ContainerDownThe container is not sending health signals
UserContainerMemoryUsageThe User workspace container has Less Memory
AppContainerMemoryUsageThe User Application container has Less Memory
TomcatDeploymentFailedA Platform microservice has a deployment failure
JvmThreadsDeadlockedWhen a JVM thread encounters a deadlock
PlatformInstanceDownAn Instance in the Platform is not sending health signals

Slow APIs

Alert NameDescription
LongRunningUserAPICallDetectedAn API call submitted by the user is taking a long time to process
LongRunningPlatformAPICallDetectedAn intercommunication API call is taking a long time to process
PlatformBottleNeckDetectedA microservice has a lot of pending API calls. This will slow down the whole platform

API Failures

Alert NameDescription
TooManyExceptionsA huge number of exceptions are detected in a short period
TooManyUnseenExceptionsA huge number of new exceptions are detected in a short period

Container Orchestration

Alert NameDescription
ContainerActivationFailedThe Container Activation failed
ContainerPassivationFailedThe Container Passivation Failed
ContainerHibernationFailedThe Container Hibernation Failed
ContainerAllocationsFullThere is not enough space in the platform for a new user Container Allocation

VCS operations

Alert NameDescription
VcsPullFailedA VCS pull operation Failed
VcsPushFailedA VCS push operation Failed

Background Jobs

Alert NameDescription
TaskIdleTimeoutThe task of a job submitted to jobs-service has not been served by the worker. More workers are probably needed
TaskExecutionTimeoutThe worker was unable to complete the task of a job submitted to jobs-service

Project failures

Alert NameDescription
ProjectCreationFailedA User could not create a project in Studio
ProjectImportFailedA User could not import a project in Studio
ProjectOpeningFailedA User project opening Failed
ProjectMigrationFailedA User could not migrate his project successfully
note

The Alerts listed above keep changing as new things are discovered and might vary over time

Configure Notifications for Alerts

Prometheus uses an alert manager to send notifications. In WaveMaker Enterprise Setup alert notifications are configured through the Email channel by default. The below link explains the configuration of alert notifications using various other channels like Slack ,Pageduty etc.https://prometheus.io/docs/alerting/latest/configuration/

Configure Alert Notifications using Email Channel

1) SSH to Platform Instance and edit the file /wm-runtime/conf/alertmanager/config/config.yml by replacing the below place holders

  • @SMTP_HOST@ and @SMTP_PORT@ with your Corporate SMTP Server Details.
  • @SMPT_FROM_EMAIL@ and @SMTP_EMAIL@ with Corportate EmailAddress ex: notifications@xyz.com
  • @SMTP_PASSWORD@ with valid password.
  • Change the Receivers EmailAddress replace 'devops-alerts@wavemaker.com' with your Corporate EmailAddress ex:devops@xyz.com 2) Restart AlertManager Container using the below commands.
  • docker exec -it alertmanager supervisorctl stop alertmanager
  • docker exec -it alertmanager supervisorctl start alertmanager