Navigation
This version of the documentation is archived and no longer supported. To learn how to upgrade your version of MongoDB Ops Manager, refer to the upgrade documentation.
You were redirected from a different version of the documentation. Click here to go back.

Edit a Group’s Configuration

Overview

You can configure your Ops Manager groups from the Settings tab. The tab gives access to group’s settings, users and agents. To access group settings and agents, see the procedures on this page. To access the group’s users, see Manage Ops Manager Users and Roles.

Edit Group Settings

To modify group settings, click the Settings tab, then click Group Settings. For descriptions of the settings, see Group Settings.

If you have Global Owner access, Ops Manager displays a second Group Settings link under the Admin Only section. For information on these settings, see Admin-Only Group Settings.

Group Settings

The following settings in the Settings tab’s Group Settings page apply to all users in the group:

Setting Description
Group Time Zone Sets your group’s time zone.
Collect Logs For All Hosts Activates or deactivates the collection of log data for all hosts. This overwrites the statuses set on the individual hosts.
Collect Profiling Information for All Hosts

Activates or deactivates Ops Manager collection of data from the MongoDB database profilers running on your mongod instances. A mongod instance must have its profiler enabled in order for Ops Manager to collect data from it.

When you change this setting, Ops Manager applies the change globally to all mongod processes in the group. For example, if you disable this setting, Ops Manager disables the collection of profiling data for all the group’s processes. This setting does not affect whether the profiler is enabled on a given mongod process, only whether Ops Manager collects profiling data.

To enable the collection of profiling data on a process-by-process basis, see Profile Databases.

When profiling is enabled, Ops Manager collects data from MongoDB’s profiler to provide statistics about performance and database operations. Ensure exposing profile data to Ops Manager is consistent with your information security practices. Also be aware the profiler can consume resources which may adversely affect MongoDB performance.

For more information, see Profile Databases.

Collect Database Specific Statistics Allows you to enable or disable the collection of database statistics. For more information, see How does Ops Manager gather database statistics?.
Reset Duplicates Allows you to reset and remove all detected duplicate hosts. This is useful if your server environment has drastically changed and you believe a host is incorrectly marked as a duplicate.
Preferred Hostnames

Allows you to specify resolvable hostnames or IP address for your deployment’s servers. Ops Manager keeps a list of the multiple ways to which each server is referred (hostname, FQDN, IP address) and uses heuristics to determine the best choice. Use this setting to guarantee Ops Manager uses a resolvable method. The method you choose will also be the method used to display the servers in Ops Manager.

To specify a preferred hostname, click Add and do one of the following:

  • To specify hostnames that end with a particular string, click the Ends With button and enter the string.

  • To specify hostnames that match a pattern, click the Regex button and enter a regular expression. An expression that uses “starts with” behavior must have .* at the end in order to correctly match. For example, to specify hostnames that start with acme-, enter:

    ^acme-.*
    
Suppress Mongos Automatic Discovery Suppresses automatic discovery of all mongos processes in your deployment’s sharded clusters.
Public Key for SCP Restores If you use Ops Manager Backup, this setting allows you to generate a public key for SCP backup restoration. If you restore a snapshot through SCP, Ops Manager uses the key to transmit the snapshot. For more information on restores, see how to validate an SCP restore and other SCP FAQs.
PagerDuty Service Key

Sets a default service key for alert notifications sent to a PagerDuty account. Ops Manager enters the key by default when you add a PagerDuty notification to an alert configuration.

You can add PagerDuty notifications only for alerts that require user acknowledgement. Informational alerts, such as the alert that a replica set has elected a new primary, cannot use PagerDuty notification.

Users can acknowledge PagerDuty alert notifications only from the PagerDuty dashboard.

Flowdock Settings

Sets default values for alert notifications sent to Flowdock. Ops Manager enters the values by default when you add a Flowdock notification to an alert configuration. Set the following:

  • Org Name: The Flowdock organization name in lower-case letters. This is the name that appears after www.flowdock.com/app/ in the URL string.

  • Flow Name: The flow name in lower-case letters. The flow name appears after the org name in the URL string: www.flowdock.com/app/<org-name>/<flow-name>.

    The flow name also appears in the “flow email address” setting in Flowdock. For example: <flow-name>@example.flowdock.com.

  • API Token: Your Flowdock “personal API token” found on the https://www.flowdock.com/account/tokens page of your Flowdock account.

HipChat Settings Sets a default room and API token for alert notifications sent to a HipChat account. Ops Manager enters the values by default when you add a HipChat notification to an alert configuration.
Slack Settings

Sets a default channel and token for alert notifications sent to a Slack account. You can use either an API token or a Bot token. To create an API token, see the https://api.slack.com/web page in your Slack account. For information on Bot users in Slack, see https://api.slack.com/bot-users.

Ops Manager will enter these values by default when you add a Slack notification to an alert configuration.

Webhook Settings

Adds a Webhook URL endpoint to which Ops Manager can send alert notifications for programmatic processing. Ops Manager sends an alert notification as an HTTP POST request in which the request body contains a JSON document that uses the same format as the Public API’s Alerts resource.

Ops Manager adds a request header called X-MMS-Event to distinguish between various alert states. The possible values for this header are:

  • alert.open: The alert was just opened.
  • alert.close: The alert was resolved.
  • alert.update: A previously opened alert is still open.
  • alert.cancel: The alert became invalid and was canceled.
  • alert.inform: Represents an informational alert, which is a point-in-time event, such as “Primary Elected.”

If you specify a key in the Webhook Secret field, Ops Manager adds the X-MMS-Signature request header, which contains the hex-encoded HMAC signature of the request body. The signature is created using the provided secret.

To send alert notifications to a Webhook, select the Webhook notification option when creating or editing an alert configuration.

Admin-Only Group Settings

The following group settings in the Admin Only section of the Settings tab could, in certain situations, affect more than the group. For example, setting logging to a high verbosity would cause system logs to roll over faster. Only users with the Global Owner role can edit these settings:

Setting Description
Mongos Deactivation Threshold Change the amount of time before Ops Manager stops monitoring an unreachable mongos. By default, the Monitoring Agent stops monitoring an unreachable mongos after 24 hours. Set this to the amount of time in hours to wait before deactivation.
Monitoring Agent Log Level Change the verbosity of the Monitoring Agent log.
Automation Agent Log Level Change the verbosity of the Automation Agent log.

View Agent Status

Click Settings, then Agents to display the following information about your agents:

Field Description
Status The time of the last ping from the agent.
Type The type of agent.
Hostname The hostname for the agent and any warnings, such as that the agent is down or out-of-date.
State Indicates whether the agent is active.
Ping Count The number of pings (i.e. data payloads) sent by the agent since midnight GMT. Typically agents send pings every minute.
Version The version of the agent software running on this agent instance.
Log Click view logs to view the agent’s log.

If you have more than one Monitoring Agent, only one agent actively monitors MongoDB instances at a time. See Monitoring Architecture for more information.