Setting Rules for Event Forwarding

In systems management, many servers may need access to forward logs, traps and Netflows from network devices and servers, but it is often resource intensive for network devices and servers to forward logs, traps and netflows to multiple destinations. For example, most Cisco routers can forward Netflow to two locations at most. However, AccelOps can forward/relay specific logs, traps and Netflows to one or more destinations. If you want to send a log to multiple destinations, you can send it to AccelOps, which will use an event forwarding rule to send it to the desired locations.

1. Log in to your Supervisor node.
2. Go to Admin > General Settings > Event Handling.
3. Under Event Forwarding Rule, for multi-tenant deployments, select the organization for which the rule will apply.
4. Click Add.
5. For Sender IP, enter the IP address of the device that will be sending the logs.
6. For Severity, select an operator and enter a severity level that must match for the log to be forwarded.
7. Select the Traffic Type to which the rule should apply.

The Forward To > Port field will be populated based on your selection here.

8. For Forward to > IP, enter the IP address to which the event should be forwarded.
9. Click OK.

Setting Rules for Event Dropping

Some devices and applications generate a significant number of logs, which may be very verbose, contain little valuable information, and consume storage resources. You can configure Event Dropping rules that will drop events just after they have been received by AccelOps, preventing these event logs from being collected and processed. Implementing these rules may require some thought to accurately set the event type, reporting device type, and event regular expression match, for example. However, dropped events do not count towards licensed Events per Second (EPS), and are not stored in the Event database. Dropped event also do not appear in reports, and do not trigger rules. You can also specify that events should be dropped but stored, so event information will be available for searches and reports, but will not trigger rules. And example of an event type that you might want to store but not have trigger any rules would be an IPS event that is a false positive.

Procedure

1. Log in to your Supervisor node.

For multi-tenant deployments you should log in to the Super/Global account if you want to set a system-wide event dropping rule. If you want to set an event-dropping rule for a specific organization, either log in as an administrator for that organization, or or log in using the Super/Global Account and then select the organization to which the rule should apply when you are creating it.

2. Go to Admin > General Settings > Event Handling.
3. Under Event Dropping Rule, click Add.
4. Next to Reporting Device, click Edit, and use the CMDB Browser to find device group or individual device that you want to create the rule for.
5. Next to Event Type, click Edit, and use the Event Type Browser to find the group of event types, or a specific event type, that you want to create the rule for.
6. If the event type you select has an Source IP or Destination IP attribute, you can enter specific IP addresses to which the rule should apply.
7. For Regex Filter, enter any regular expressions you want to use to filter the log files.

If any matches are made against your regular expression, then the event will be dropped.

8. For multi-tenant deployments, select the Organization to which the rule should apply.
9. Select the Action that should be taken when the event dropping rule is triggered.
10. Enter any Description for the rule.
11. Click Save.

Notes

1. All matching rules are implemented by AccelOps, and inter-rule order is not important. If you create a duplicate of an event dropping rule, the first rule is in effect.
2. If you leave a rule definition field blank, then that field is not evaluated. For example, leaving Event Type left blank is the same as selecting All Event Types.
3. AccelOps drops the event at the first entry point. If your deployment uses Collectors, events are dropped by the Collectors. If your deployment doesn’t use Collectors, then the event will be droppedby the Worker or Supervisor where the event is received.
4. You can use the report System Event Processing Statistics to view the statistics for dropped events. When you run the report, select AVG(Policy Dropped Event Rate(/sec) as one of the dimensions for Chart For to see events that have been dropped to this policy.

Running Historical Searches to Test Rule Sub Patterns

If you are trying to analyze why a rule is triggering an excessive number of incidents, or why it isn’t triggering any, you can run an historical search with the rule sub patterns to see how the sub pattern behaves in relation to past events. If the search has interesting results, you can then generate a report for further investigation. This is a way that you can test rules without having to deactivate them.

1. Go to Analytics > Rules.
2. Select a rule and then click Edit.
3. Click Edit next to the sub pattern you want to use in the search.
4. Click Run as Query.
5. Enter information for the time period you want to search.
6. Click OK.

An historical search will run based on the sub pattern filters, aggregate conditions, and group by conditions.

Using a Sub Pattern in a Report

If the search includes results that you want to share or investigate further, you can save the rule as a report.

1. In the sub pattern you want to save, click Save as Report.

The report will be saved in Analytics > Reports, and will have the phrase From Rule in the report name.

2. Select the report and click Run Now to generate a report from the sub pattern.

Cloning a Rule

You can clone a rule to use it as the basis for creating another rule, or to use in testing.

1. Log in to your Supervisor node.
2. Go to Analytics > Rules.
3. Search or browse to select the rule you want to clone.
4. Click Clone.
5. Enter a new name for the cloned rule and click OK.

The cloned rule will be added to the same group as the original rule but will be inactive.

Adding a Watch List to a Rule

1. Go to Analytics > Rules.
2. Select the rule you want to add the watch list to, and then click Edit.
3. Next to Watch Lists, click Edit.
4. Select the watch list you want to add, and use the Add >> button to add it to the rule.
5. For Incident Attribute, select the incident information you want to add to the watch list.

Watch List Attribute Type Must Match Incident Attribute

The Type that you set for the watch list must match the Incident Attribute Types for the rule. For example, if your watch list Type is IP, and the Incident Attribute Type for the rule is string, you will not be able to associate the watch list to the rule.

6. Click OK.

Next to Watch Lists, you will see Watch List has been defined.

Activating and Deactivating Rules

When you create a new rule, you must activate it before it will start to monitor events. You may also want to deactivate a rule, for example to test it, instead of deleting it from the system.

1. Log in to your Supervisor node.
2. Go to Analytics > Rules.
3. Browse or search to find the rule that you want to activate or deactivate.
4. Select Active for the rule you want to activate, or clear the Active option if you want to deactivate a rule.

Testing a Rule

After you’ve created or a edited a rule, you should test it to see if behave as expected before you activate it. This topic describes how to test a rule using synthetic events.

Procedure

Test Results

Test Example

Troubleshooting for Rule Testing

Rule Syntax Error

Rule Semantics Error

Event Parsing Error

Procedure

1. Go to Analytics > Rules, and deactivate the rule you want to test.
2. Select the rule, and then click Test Rule.

This will open the Rule Debugger.

3. Enter a Reporting IP where the synthetic event should originate from.
4. Under Raw Event, enter the raw event log text that contains the triggering conditions for the rule.
5. Under Pause, enter the number of seconds before the next test event will be sent, and then click + under Action to enter additional test events.

You will need to create as many events as are necessary to trigger the rule conditions.

6. Click Run Test.

If the test succeeds you are now ready to activate the rule.

Test Results

The test will run through a four stage process, which you can observe in the Test Results tab of the rule. A yellow icon will also appear in the Stat us column for the rule to indicate that the test is running.

1. Rules are checked for syntax errors.
2. Events are parsed and sent to Rule Workers.

If there are errors in the rule syntax or event parsing errors, see the examples under Troubleshooting for Rule Testing for suggestions on how to correct them. As events are being parsed, you can view their Event Details by clicking on the Raw Event Log icon next to the event.

3. Rule Worker nodes evaluate the events against the rule conditions, and if they match, they are sent to the Rule Master.
4. The Rule Master creates incidents, which then appear in the Incidents dashboards.

When the test successfully completes, a green icon will appear in the Status column next to the rule name.

Test Example

This screenshot shows the example of a test for the rule Multiple Admin Login Failures: Net Devices. The conditions for this rule are that the the Reporting IP must belong to a network device, and there must be 3 login failure events from the same IP and user.

Troubleshooting for Rule Testing

If the test fails, a red icon will appear under the Status column next to the rule name, and you will see the error message in the Test Results tab for the rule.

Rule Syntax Error

Rule Semantics Error

This means that the conditions of the rule were not met by the event. For example, if five events were required to meet the condition, but only one was sent.

Event Parsing Error

This means that some text in the raw event log did not pass the event parser. For example, if “denied” is the term expected by the parser in the test example, but the raw event log contains the term “deny,” then the event will not pass the parser.

Defining Clear Conditions

Clear conditions specify conditions in which incidents will have their status changed from Active to Cleared. You can set the time period that must elapse for the clear condition to occur, and then set the conditions based on the triggering of the original rule, or on a sub pattern based on t he Incident Attributes.

1. In Analytics > Rules, select the rule you want to add the clear condition to, and click Edit.
2. Next to Clear Condition, click Edit.
3. Set the Time Period that should elapse for the clear condition to go into effect.
4. If you want the clear condition to go into effect based on the firing of the original rule, select the Original Rule Does Not Trigger. For example, if you wanted the clear condition to change the status of Active incidents to Cleared after the original rule had not been triggered for ten minutes, you would set Cleared Within to 10 Minutes and select this option.
5. If you want to base the clear condition on a sub-pattern of the incident attributes, select the following conditions are met.

The incident attributes from your rule will load and the clear condition attributes will be set to match.

6. Define the pattern to use by clicking the Edit icon next to the clear sub pattern.
7. Click Save.