nav-img
Advanced

Checking and Handling Intrusion Events

HSS displays alarm and event statistics and their summary all on one page. You can have a quick overview of alarms, including the numbers of servers with alarms, handled alarms, unhandled alarms, blocked IP addresses, and isolated files.

The Events page displays the alarm events generated in the last 30 days.

The status of a handled event changes from Unhandled to Handled.

Constraints and Limitations

  • To skip the checks on high-risk command execution, privilege escalation, 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.
  • Other detection items cannot be manually disabled.

Checking Alarm Events

  1. Log in to the management console.
  2. In the upper left corner of the page, select a region, click , and choose Security > Host Security Service.
  3. In the navigation pane, choose Intrusions > Events, as shown in Figure 1.

    Figure 1 Events page


    Table 1 Alarm events

    Alarm Event

    Description

    Affected Servers

    Number of servers for which alarms are generated.

    Alarms to be Handled

    Number of alarms to be handled.

    By default, all unhandled alarms are displayed on the Events page. For more information, see Handling Alarm Events.

    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.

    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:

    After a blocked IP address is unblocked, HSS will no longer block the operations performed by the IP address.

    Isolated Files

    HSS can isolate detected threat files. Files that have been isolated are displayed on a slide-out panel on the Events 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 event in the list to view the affected servers and occurrence time of the event, as shown in Figure 2. The following information is displayed:

    • Total number of alarms
    • Number of each type of alarms

    Figure 2 Alarm event statistics


  5. Click an alarm name to view its details, as shown in Figure 3.

    Figure 3 Alarm details


Handling Alarm Events

This section describes how you should handle alarm events to ensure server security.

Note

Do not fully rely on alarms 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 > Host Security Service.
  3. In the navigation pane, choose Intrusions > Events.

    Figure 4 Events page


  4. Click an event type, select events, and click Handle, as shown in Figure 5. Table 2 describes the processing methods you can choose from.

    Note

    You can also click Handle in the row where an alarm resides.

    Figure 5 Handling alarm events


    Alarm events are displayed on the Events page. Here you can check up to 30 days of historical events.

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

    Table 2 Event handling methods

    Method

    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.

    The following types of alarm events support online isolation and killing:

    • Malicious program (cloud scan)
    • Abnormal process behavior
    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 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 whitelisted items.

    Add to alarm whitelist

    Add false alarmed items of the following types to the login whitelist.

    HSS will no longer report alarm on the whitelisted items.

    • Reverse shell
    • Web shell
    • Abnormal process behavior
    • Process privilege escalation
    • File privilege escalation
    • High-risk command
    • Malicious program

Handling Suggestion

Alarm Name

Suggestion

Brute-force attack

Pay special attention to such events.

If you receive a brute-force attack alarm, detected events will probably be but are not limited to:

  • The system uses weak passwords and is under brute-force attacks.
  • Attackers correctly guess the password and log in after several failed attempts (before their login IP addresses are blocked).

You are advised to check whether the alarmed login IP address is valid.

  • If the source IP address is valid, ignore the alarms and manually unblock the IP addresses.Alternatively, whitelist the alarmed IP address. This IP address will no longer trigger alarms.
  • If the source login IP address are unknown, your servers may have been intruded by hackers.
    1. You are advised to mark the event as Handled.
    2. Immediately log in to the intruded account and set a strong password.
    3. Check all the accounts and delete suspicious accounts to prevent attackers from creating new accounts or changing account permissions.
    4. Check for malicious programs on servers. Then, log in to the servers where the malicious programs are running and stop them immediately.

Abnormal login

If an abnormal login is detected, you are advised to immediately check whether the source IP address is valid.

  • If it is valid, you can ignore this event.

    If the login location is valid, you can add the location to the list of common login locations.

  • If it is invalid or unknown, your servers have been intruded.

    In this case, you are advised to mark the event as Handled, immediately change the account password, and scan the entire system for risks to prevent further damage.

Malicious program (cloud scan)

Common methods to handle the event are as follows:

  • If the programs are normal, ignore the event or whitelist the program. The programs will no longer trigger such events.
  • If the programs are unknown or malicious, you are advised to immediately kill them and isolate their source files.
    • You can isolate and kill detected or suspicious programs in one click. Alternatively, you can mark the event as Handled, immediately log in and stop the program, and scan the entire system for risks to prevent further damage.
    • HSS can isolate and kill malicious programs, including common ransomware, DDoS viruses, and Trojans.

      You are advised to enable this function to harden server security.

  • If the programs are harmless or mandatory for service operation, you can cancel isolation and restore the program source files.

Abnormal process behavior

If abnormal process behaviors are detected, you are advised to check processes immediately.

  • If the processes are normal, ignore the event or whitelist the process. The processes will no longer trigger such events.
  • If the processes are unknown or malicious, you are advised to immediately kill them and isolate their source files.
    • You can isolate and kill detected or suspicious programs in one click. Alternatively, you can mark the event as Handled, immediately log in and stop the program, and scan the entire system for risks to prevent further damage.
    • HSS can isolate and kill malicious programs, including common ransomware, DDoS viruses, and Trojans.

      You are advised to enable this function to harden server security.

  • If the programs are harmless or mandatory for service operation, you can cancel isolation and restore the program source files.

Critical file change

If a key file change is detected, you are advised to check the change immediately.

  • If the change is valid, you can ignore the event.
  • If the change is invalid, critical files have been read, written, or deleted without authorization.

    You are advised to mark the event has Handled and immediately replace the file with the standard version of the OS. Log in to intruded accounts and change their passwords, and scan the entire system for risks to prevent further damage.

Web shell

If a web shell is detected, you are advised to immediately check whether the file is valid.

  • If the file is valid, ignore the event or whitelist the file. The file will no longer trigger such events.
  • If the file is invalid, you are advised to mark the event as Handled and immediately isolate the file.

Reverse/Abnormal shell

If a reverse or abnormal shell is detected, you are advised to check whether executed commands are valid.

  • If they are valid, you can ignore this event.
  • If they are invalid, mark the event as Handled and immediately log in to the system to block invalid connections or stop command execution, and scan the entire system for risks to prevent further damage.

High-risk command execution

If a high-risk command is detected, you are advised to immediately check whether the command is valid.

  • If it is valid, ignore the event or whitelist the command. The command will no longer trigger such events.
  • If it is invalid, mark the event as Handled and immediately log in to the system and check operations performed using the command, and scan the entire system for risks to prevent further damage.

Auto-startup check

If a new auto-started item is detected, you need to check whether the auto-startup item is valid.

  • If it is valid, ignore the event or whitelist the command. The command will no longer trigger such events.
  • If it is invalid, mark the event as Handled and immediately log in to the system to delete the item, and scan the entire system for risks to prevent further damage.

Unsafe account

If an unsafe account is detected, you are advised to immediately check whether the account is valid.

  • If it is valid, you can ignore this event.
  • If it is invalid, mark the event as a Handled and perform the following operations:
    • Deleting suspicious accounts

      Delete unnecessary system login accounts, such as SSH accounts, from the servers.

      Delete unnecessary accounts used by the MySQL and FTP services from the servers.

    • Limiting account permissions

      Specify key configuration items to limit the file access and modification permissions of non-administrators, preventing unauthorized access and operations.

Privilege escalation

If a privilege escalation operation is detected, you are advised to immediately check whether the operation is valid.

  • If it is valid, you can ignore this event.
  • If it is invalid, mark the event as Handled and immediately log in to the system to block invalid connections or stop command execution, and scan the entire system for risks to prevent further damage.

Rootkit

If Rootkit installation is detected, you are advised to immediately check whether the installation is valid.

  • If it is valid, you can ignore this event.
  • If it is invalid, mark the event as Handled and immediately log in to the system to stop Rootkit installation, and scan the entire system for risks to prevent further damage.