System Alerts¶

On this page

Overview
System Alerts
Procedures

Overview¶

System alerts are internal health checks that monitor the health of Ops Manager itself, including the health of backing databases, Backup Daemons, and backed-up deployments. Ops Manager runs health checks every 5 minutes.

To view system alerts, click the Admin link at the top of Ops Manager, select the Alerts tab, and then select the Open Alerts link under System Alerts.

Users with the Global Owner or Global Monitoring Admin roles can modify notification settings or disable an alert.

System Alerts¶

Ops Manager provides the following system alerts:

System detects backup oplog TTL was resized¶

Sends an alert if the Backup Daemon has fallen so far behind in applying oplog entries that Ops Manager has extended the period of time it will store the entries. By default, Ops Manager stores oplog entries in the Oplog Store for 24 hours. If the Daemon has not yet applied an entry an hour before its expiration, Ops Manager extends the storage period by three hours. If the entry again reaches an hour from expiration, Ops Manager continues to extend the storage period, up to seven days. The system sets the storage period in the Daemon’s mms.backup.restore.snapshotPITExpirationHours setting.

If you receive this alert, check that your Backup Daemon is running and that it has sufficiently performant hardware to apply oplog entries in a timely manner.

System detects backing database startup warnings¶: Sends an alert if the MongoDB process hosting a backing database contains startupWarnings in its log files. Check the logs on the server running the MongoDB process.

System detects an unhealthy database backing the system¶: Sends an alert if Ops Manager cannot connect to a backing database and run the ping command.

System detects backup daemon is down¶: Sends an alert the Backup Daemon has not pinged Ops Manager for more than 15 minutes.

System detects backup daemon has low free head space¶

Sends an alert if the disk partition on which the local copy of a backed-up replica set is stored has less than 10 GB of free space remaining.

The Ops Manager Daemons Page displays head space used for each daemon. The mms.alerts.LowHeadFreeSpace.minimumHeadFreeSpaceGB setting controls the alert threshold, which has a default value of 10 GB.

System detects backup was not moved successfully¶: Sends an alert if you attempt to move a job to a new Backup Daemon but the move failed. The job will continue to run in its original location. For more information on moving jobs, see Jobs Page.

Procedures¶

Modify Notification Settings for a System Alert¶

1

Click the Admin link in the top right corner of Ops Manager.¶

Ops Manager displays the Admin link only if you have administrative privileges.

2

Select the Alerts tab and then the System Alert Settings page.¶

3

On the line for the alert, click the ellipsis icon and select Edit.¶

4

Select the alert recipients and delivery methods.¶

In the Send to section, select the notification methods, recipients, and alert frequencies. To add methods or recipients, click Add and select from the following:

Notification Method	Description
User	Sends the alert by email or SMS to a specified Ops Manager user. If you select SMS, Ops Manager sends the text message to the number configured on the user’s Account page.
Email	Sends the alert to a specified email address.
SMS	Sends the alert to a specified mobile number. Ops Manager removes all punctuation and letters and uses only the digits. If you are outside of the United States or Canada, include ‘011’ and the country code. For example, for New Zealand enter ‘01164’ before your phone number. As an alternative, use a Google Voice number. Ops Manager uses the U.S.-based Twilio to send SMS messages.
HipChat	Sends the alert to a HipChat room message stream. Enter the HipChat room name and API token.
PagerDuty	Enter only the service key. Define escalation rules and alert assignments in PagerDuty.
Administrators	Sends the alert to the email address specified in the `mms.adminEmailAddr` setting in the conf-mms.properties configuration file.

5

Click Save.¶

Disable a System Alert¶

1

Click the Admin link in the top right corner of Ops Manager.¶

Ops Manager displays the Admin link only if you have administrative privileges.

2

Select the Alerts tab and then the System Alert Settings page.¶

3

Disable the alert.¶

On the line for the system alert that you want to disable, click the ellipsis icon and select Disable.

← Global Alerts Back Up and Restore Deployments →