Handling Server Alarms

Limitations and Constraints

• To skip the checks on high-risk command execution, privilege escalations, reverse shells, abnormal shells, or web shells, manually disable the corresponding policies in the policy groups on the Policies page. HSS will not check the servers associated with disabled policies. For details, see Viewing a Policy Group.
• Other detection items cannot be manually disabled.
• Servers that are not protected by HSS do not support operations related to alarms and events.

Handling Server Alarm Events

This section describes how you should handle alarms to enhance server security.

Do not fully rely on alarm handling to defend against attacks, because not every issue can be detected in a timely manner. You are advised to take more measures to prevent threats, such as checking for and fixing vulnerabilities and unsafe settings.

1. Log in to the management console.
2. In the upper left corner of the page, select a region, click , and choose Security & Compliance > HSS.
3. In the navigation pane on the left, choose Detection > Alarms and click Server Alarms.
   

   If your servers are managed by enterprise projects, you can select an enterprise project to view or operate the asset and scan information.

   Table 1 Alarm statistics

   Parameter	Description
   Enterprise Project	Select an enterprise project and view alarm details by enterprise project.
   Time range	You can select a fixed period or customize a time range to search for alarms. Only alarms generated within 30 days can be queried. The options are as follows: Last 24 hours Last 3 days Last 7 days Last 30 days
   Urgent Alarms	Number of urgent alarms that need to be handled.
   Total Alarms	Total number of alarms on your assets.
   Affected Servers	Number of servers for which alarms are generated. When checking alarms generated in the last 24 hours, you can click the number of servers to go to the Servers & Quota page and check the corresponding servers.
   Handled Alarms	Number of handled alarms.
   Blocked IP Addresses	Number of blocked IP addresses. You can click the number to check blocked IP address list. The blocked IP address list displays the server name, attack source IP address, login type, blocking status, number of blocks, blocking start time, and the latest blocking time. If a valid IP address is blocked by mistake (for example, after O&M personnel enter incorrect passwords for multiple times), you can manually unblock it. If a server is frequently attacked, you are advised to fix its vulnerabilities in a timely manner and eliminate risks. NOTICE: The agent of Linux 3.2.10 or later supports IPv6 interception. The agent of a version earlier than Linux 3.2.10 supports TCP Wrapper interception, but does not support IPv6 interception using IPTables. After a blocked IP address is unblocked, HSS will no longer block the operations performed by the IP address. A maximum of 10,000 IP addresses can be blocked for each type of software. If your Linux server does not support ipset, a maximum of 50 IP addresses can be clocked for MySQL and vsftp. If your Linux server does not support ipset or hosts.deny, a maximum of 50 IP addresses can be blocked for SSH.
   Isolated Files	HSS can isolate detected threat files. Files that have been isolated are displayed on a slide-out panel on the Server Alarms page. You can click Isolated Files on the upper right corner to check them. You can recover isolated files. For details, see Managing Isolated Files.

4. Click an alarm name to view the alarm details and suggestions.
5. Handle alarms.
   

   Alarms are displayed on the Server Alarms page. Here you can check up to 30 days of historical alarms.

   Check and handle alarms as needed. The status of a handled alarm changes from Unhandled to Handled. HSS will no longer collect its statistics or display them on the Dashboard page.

   • Handling a single alarm

     In the Operation column of an alarm, click Handle.

   • Handling alarms in batches

     Select all alarms and click Batch Handle above the alarm list.

   • Handling all alarms

     In the Alarms to be Handled area on the left pane of the alarm list, select an alarm type and click Handle All above the alarm list.

     Figure 1 Handling all alarms
     

6. In the Handle Event dialog box, select an action. For details about the alarm handling actions, see Table 2.
   When handling a single alarm event or handling alarms in batches, you can select Handle duplicate alarms in batches in the Handle Event dialog box.

   Table 2 Alarm handling methods

   Action	Description
   Ignore	Ignore the current alarm. Any new alarms of the same type will still be reported by HSS.
   Isolate and kill	If a program is isolated and killed, it will be terminated immediately and no longer able to perform read or write operations. Isolated source files of programs or processes are displayed on the Isolated Files slide-out panel and cannot harm your servers. You can click Isolated Files on the upper right corner to check the files. For details, see Managing Isolated Files. For details about events that can be isolated and killed, see Server Alarms. NOTE: When a program is isolated and killed, the process of the program is terminated immediately. To avoid impact on services, check the detection result, and cancel the isolation of or unignore misreported malicious programs (if any).
   Mark as handled	Mark the event as handled. You can add remarks for the event to record more details.
   Add to process whitelist	If you can confirm that a process triggering an alarm can be trusted, you can add it to the process whitelist. HSS will no longer report alarms on whitelisted processes.
   Add to Login Whitelist	Add false alarmed items of the Brute-force attack and Abnormal login types to the Login Whitelist. HSS will no longer report alarm on the Login Whitelist. A whitelisted login event will not trigger alarms. The following alarm events can be added: Brute-force attacks Abnormal logins
   Add to alarm whitelist	Add false alarmed items to the login whitelist. HSS will no longer report alarm on the whitelisted items. A whitelisted alarm will not trigger alarms. After adding an alarm to the alarm whitelist, you can customize a whitelist rule. The custom rule types vary depending on the alarm types, including the file path, process path, process command line, remote IP address, and user name. If a detected alarm event hit the rule you specified, HSS does not generate an alarm. For details about events that can be isolated and killed, see Server Alarms.

7. Click OK.

   You check handled alarms. For details, see Viewing the Handling History.

Canceling Handled Server Alarms

You can cancel the processing of a handled alarm event.

1. In the alarm event list, filter handled alarms.
2. In the Operation column of an alarm, click Handle.
3. In the Handle Alarm Event dialog box, click OK to cancel the last handling.