Foundation Topics
Mitigation Technology for E-mail-Based Threats
Users are no longer accessing e-mail from the corporate network or from a single device. Cisco provides cloud-based, hybrid, and on-premises ESA-based solutions that can help protect any dynamic environment. This section introduces these solutions and technologies explaining how users can use threat intelligence to detect, analyze, and protect against both known and emerging threats.
E-mail-Based Threats
There are several types of e-mail-based threats. The following are the most common:
- Spam: Unsolicited e-mail messages that can be advertising a service or (typically) a scam or a message with malicious intent. E-mail spam continuous to be a major threat because it can be used to spread malware.
- Malware attachments: E-mail messages containing malicious software (malware).
- Phishing: An attacker’s attempt to fool a user that such e-mail communication comes from a legitimate entity or site, such as banks, social media websites, online payment processors, or even corporate IT communications. The goal of the phishing e-mail is to steal user’s sensitive information such as user credentials, bank accounts, and so on.
- Spear phishing: Phishing attempts that are more targeted. These phishing e-mails are directed to specific individuals or organizations. For instance, an attacker may perform a passive reconnaissance on the individual or organization by gathering information from social media sites (for example, Twitter, LinkedIn, Facebook) and other online resources. Then the attacker may tailor a more directed and relevant message to the victim increasing the probability of such user being fooled to follow a malicious link, click an attachment containing malware, or simply reply to the e-mail providing sensitive information. There is another phishing-based attack called whaling. These attacks specifically target executives and high-profile users within a given organization.
Cisco Cloud E-mail Security
Cisco cloud e-mail security provides a cloud-based solution that allows companies to outsource the management of their e-mail security management. The service provides e-mail security instances in multiple Cisco data centers to enable high availability. Figure 18-1 illustrates the Cisco cloud e-mail security solution.
)
Figure 18-1 Cisco Cloud E-mail Security Architecture
In Figure 18-1, three organizations (a large enterprise, a university, and a small- to medium-size business) leverage the Cisco hosted (cloud) environment. The solution also supports mobile workers.
Cisco Hybrid E-mail Security
The Cisco hybrid e-mail security solution combines both cloud-based and on-premises ESAs. This hybrid solution helps Cisco customers reduce their on-site e-mail security footprint, outsourcing a portion of their e-mail security to Cisco, while still allowing them to maintain control of confidential information within their physical boundaries. Many organizations need to stay compliant to many regulations that may require them to keep sensitive data physically on their premises. The Cisco hybrid e-mail security solution allows network security administrators to remain compliant and to maintain advanced control with encryption, data loss prevention (DLP), and on-site identity-based integration.
Cisco E-mail Security Appliance
The following are the different ESA models:
Cisco X-Series E-mail Security Appliances
- Cisco X1070: High-performance ESA for service providers and large enterprises
Cisco C-Series E-mail Security Appliances
- Cisco C680: The high-performance ESA for service providers and large enterprises
- Cisco C670: Designed for medium-size enterprises
- Cisco C380: Designed for medium-size enterprises
- Cisco C370: Designed for small- to medium-size enterprises
- Cisco C170: Designed for small businesses and branch offices
The Cisco ESA runs the Cisco AsyncOS operating system. The Cisco AsyncOS supports numerous features that will help mitigate e-mail-based threats. The following are examples of the features supported by the Cisco ESA:
- Access control: Controlling access for inbound senders according to the sender’s IP address, IP address range, or domain name.
- Antispam: Multilayer filters based on Cisco SenderBase reputation and Cisco antispam integration. The antispam reputation and zero-day threat intelligence are fueled by Cisco’s security intelligence and research group named Talos.
- Network Antivirus: Network antivirus capabilities at the gateway. Cisco partnered with Sophos and McAfee, supporting their antivirus scanning engines.
- Advanced malware protection (AMP): Allows security administrators to detect and block malware and perform continuous analysis and retrospective alerting.
- DLP: The ability to detect any sensitive e-mails and documents leaving the corporation. The Cisco ESA integrates RSA e-mail DLP for outbound traffic.
- E-mail encryption: The ability to encrypt outgoing mail to address regulatory requirements. The administrator can configure an encryption policy on the Cisco ESA and use a local key server or hosted key service to encrypt the message.
- E-mail authentication: A few e-mail authentication mechanisms are supported, including Sender Policy Framework (SPF), Sender ID Framework (SIDF), and DomainKeys Identified Mail (DKIM) verification of incoming mail, as well as DomainKeys and DKIM signing of outgoing mail.
- Outbreak filters: Preventive protection against new security outbreaks and e-mail-based scams using Cisco’s Security Intelligence Operations (SIO) threat intelligence information.
The Cisco ESA acts as the e-mail gateway to the organization, handling all e-mail connections, accepting messages, and relaying them to the appropriate systems. The Cisco ESA can service e-mail connections from the Internet to users inside your network, and from systems inside your network to the Internet. E-mail connections use Simple Mail Transfer Protocol (SMTP). The ESA services all SMTP connections by default acting as the SMTP gateway.
The Cisco ESA uses listeners to handle incoming SMTP connection requests. A listener defines an e-mail processing service that is configured on an interface in the Cisco ESA. Listeners apply to e-mail entering the appliance from either the Internet or from internal systems.
The following listeners can be configured:
- Public listeners for e-mail coming in from the Internet.
- Private listeners for e-mail coming from hosts in the corporate (inside) network. These e-mails are typically from an internal groupware, Exchange, POP, or IMAP e-mail servers.
Figure 18-2 illustrates the concept of Cisco ESA listeners.
)
Figure 18-2 Cisco ESA Listeners
Cisco ESA listeners are often referred to as SMTP daemons running on a specific Cisco ESA interface. When a listener is configured, the following information must be provided:
- Listener properties such as a specific interface in the Cisco ESA and the TCP port that will be used. The listener properties must also indicate whether it is a public or a private listener.
- The hosts that are allowed to connect to the listener using a combination of access control rules. An administrator can specify which remote hosts can connect to the listener.
- The local domains for which public listeners accept messages.
Cisco ESA Initial Configuration
To perform the initial Cisco ESA configuration, complete the following steps:
- Step 1. Log in to the Cisco ESA. The default username is admin, and the default password is ironport.
- Step 2. Use the systemsetup command in the command-line interface (CLI) of the Cisco ESA to initiate the System Setup Wizard, as shown in Example 18-1.
Example 18-1 Initial Setup with the systemsetup Command
IronPort> systemsetup WARNING: The system setup wizard will completely delete any existing ‘listeners’ and all associated settings including the ‘Host Access Table’ - mail operations may be interrupted. Are you sure you wish to continue? [Y]> Y You are now going to configure how the IronPort C60 accepts mail by creating a “Listener”. Please create a name for this listener (Ex: “InboundMail”): []> InboundMail Please choose an IP interface for this Listener. 1. Management (192.168.42.42/24: mail3.example.com) 2. PrivateNet (192.168.1.1/24: mail3.example.com) 3. PublicNet (192.168.2.1/24: mail3.example.com) [1]>3 Enter the domains or specific addresses you want to accept mail for. Hostnames such as “example.com” are allowed. Partial hostnames such as “.example.com” are allowed. Usernames such as “postmaster@” are allowed. Full email addresses such as “joe@example.com” or “joe@[1.2.3.4]” are allowed. Separate multiple addresses with commas []> securemeinc.org Would you like to configure SMTP routes for example.com? [Y]> y Enter the destination mail server which you want mail for example.com to be delivered. Separate multiple entries with commas. []> exchange.securemeinc.org Do you want to enable rate limiting for this listener? (Rate limiting defines the maximum number of recipients per hour you are willing to receive from a remote domain.) [Y]> y Enter the maximum number of recipients per hour to accept from a remote domain. []> 4500 Default Policy Parameters ========================== Maximum Message Size: 100M Maximum Number Of Connections From A Single IP: 1,000 Maximum Number Of Messages Per Connection: 1,000 Maximum Number Of Recipients Per Message: 1,000 Maximum Number Of Recipients Per Hour: 4,500 Maximum Recipients Per Hour SMTP Response: 452 Too many recipients received this hour Use SenderBase for Flow Control: Yes Virus Detection Enabled: Yes Allow TLS Connections: No Would you like to change the default host access policy? [N]> n Listener InboundMail created. Defaults have been set for a Public listener. Use the listenerconfig->EDIT command to customize the listener. ***** Do you want to configure the C60 to relay mail for internal hosts? [Y]> y Please create a name for this listener (Ex: “OutboundMail”): []> OutboundMail Please choose an IP interface for this Listener. 1. Management (192.168.42.42/24: mail3.example.com) 2. PrivateNet (192.168.1.1/24: mail3.example.com) 3. PublicNet (192.168.2.1/24: mail3.example.com) [1]> 2 Please specify the systems allowed to relay email through the IronPort C60. Hostnames such as “example.com” are allowed. Partial hostnames such as “.example.com” are allowed. IP addresses, IP address ranges, and partial IP addressed are allowed. Separate multiple entries with commas. []> .securemeinc.org Do you want to enable rate limiting for this listener? (Rate limiting defines the maximum number of recipients per hour you are willing to receive from a remote domain.) [N]> n Default Policy Parameters ========================== Maximum Message Size: 100M Maximum Number Of Connections From A Single IP: 600 Maximum Number Of Messages Per Connection: 10,000 Maximum Number Of Recipients Per Message: 100,000 Maximum Number Of Recipients Per Hour: Disabled Use SenderBase for Flow Control: No Virus Detection Enabled: Yes Allow TLS Connections: No Would you like to change the default host access policy? [N]> n Listener OutboundMAil created. Defaults have been set for a Private listener. Use the listenerconfig->EDIT command to customize the listener. ***** Congratulations! System setup is complete. For advanced configuration, please refer to the User Guide. mail3.securemeinc.org >
In Example 18-1, the inside (private) and outside (public) listeners are configured. The domain name of securemeinc.org is used in this example.
To verify the configuration, you can use the mailconfig command to send a test e-mail containing the system configuration data that was entered in the System Setup Wizard, as shown in Example 18-2.
Example 18-2 Verifying the Configuration with the mailconfig Command
mail3.securemeinc.org> mailconfig Please enter the email address to which you want to send the configuration file. Separate multiple addresses with commas. []> admin@securemeinc.org The configuration file has been sent to admin@securemeinc.org. mail3.securemeinc.org>
In Example 18-2, the e-mail is sent to the administrator (admin@securemeinc.org).
Mitigation Technology for Web-Based Threats
For any organization to be able to protect its environment against web-based security threats, the security administrators need to deploy tools and mitigation technologies that go far beyond traditional blocking of known bad websites. Nowadays, you can download malware through compromised legitimate websites, including social media sites, advertisements in news and corporate sites, gaming sites, and many more. Cisco has developed several tools and mechanisms to help their customers combat these threats. The core solutions for mitigating web-based threats are the Cisco Cloud Web Security (CWS) offering and the integration of advanced malware protection (AMP) to the Cisco Web Security Appliance (WSA). Both solutions enable malware detection and blocking, continuous monitoring, and retrospective alerting. The following sections cover the Cisco CWS and Cisco WSA in detail.
Cisco CWS
Cisco CWS is a cloud-based security service from Cisco that provides worldwide threat intelligence, advanced threat defense capabilities, and roaming user protection. The Cisco CWS service uses web proxies in Cisco’s cloud environment that scan traffic for malware and policy enforcement. Cisco customers can connect to the Cisco CWS service directly by using a proxy autoconfiguration (PAC) file in the user endpoint or through connectors integrated into the following Cisco products:
- Cisco ISR G2 routers
- Cisco ASA
- Cisco WSA
- Cisco AnyConnect Secure Mobility Client
Organizations using the transparent proxy functionality through a connector can get the most out of their existing infrastructure. In addition, the scanning is offloaded from the hardware appliances to the cloud, reducing the impact to hardware utilization and reducing network latency. Figure 18-3 illustrates how the transparent proxy functionality through a connector works.
)
Figure 18-3 Cisco ASA with Cisco CWS Connector Example
In Figure 18-3, the Cisco ASA is enabled with the Cisco CWS connector at a branch office. The following steps explain how Cisco CWS protects the corporate users at the branch office:
- An internal user makes an HTTP request to an external website (securemeinc.org).
- The Cisco ASA forwards the request to Cisco CWS global cloud infrastructure.
- It notices that securemeinc.org had some web content (ads) that were redirecting the user to a known malicious site.
- Cisco CWS blocks the request to the malicious site.
Cisco WSA
The Cisco WSA uses cloud-based intelligence from Cisco to help protect the organization before, during, and after an attack. This “lifecycle” is what is referred to as the attack continuum. The cloud-based intelligence includes web (URL) reputation and zero-day threat intelligence from Cisco’s security intelligence and research group named Talos. This threat intelligence helps security professionals to stop threats before they enter the corporate network, while also enabling file reputation and file sandboxing to identify threats during an attack. Retrospective attack analysis allows security administrators to investigate and provide protection after an attack when advanced malware might have evaded other layers of defense.
The Cisco WSA can be deployed in explicit proxy mode or as a transparent proxy using the Web Cache Communication Protocol (WCCP). WCCP is a protocol originally developed by Cisco, but several other vendors have integrated it in their products to allow clustering and transparent proxy deployments on networks using Cisco infrastructure devices (routers, switches, firewalls, and so on).
Figure 18-4 illustrates a Cisco WSA deployed as an explicit proxy.
)
Figure 18-4 Explicit Proxy Configuration
The following are the steps illustrated in Figure 18-4:
- An internal user makes an HTTP request to an external website. The client browser is configured to send the request to the Cisco WSA.
- The Cisco WSA connects to the website on behalf of the internal user.
- The firewall (Cisco ASA) is configured to only allow outbound web traffic from the Cisco WSA, and it forwards the traffic to the web server.
Figure 18-5 shows a Cisco WSA deployed as a transparent proxy.
)
Figure 18-5 Transparent Proxy Configuration
The following are the steps illustrated in Figure 18-5:
- An internal user makes an HTTP request to an external website.
- The internal router (R1) redirects the web request to the Cisco WSA using WCCP.
- The Cisco WSA connects to the website on behalf of the internal user.
- Also in this example, the firewall (Cisco ASA) is configured to only allow outbound web traffic from the WSA. The web traffic is sent to the Internet web server.
Figure 18-6 demonstrates how the WCCS registration works. The Cisco WSA is the WCCP client, and the Cisco router is the WCCP server.
)
Figure 18-6 WCCP Registration
During the WCCP registration process, the WCCP client sends a registration announcement (“Here I am”) every 10 seconds. The WCCP server (the Cisco router in this example) accepts the registration request and acknowledges it with an “I See You” WCCP message. The WCCP server waits 30 seconds before it declares the client as “inactive” (engine failed). WCCP can be used in large-scale environments. Figure 18-7 shows a cluster of Cisco WSAs, where internal Layer 3 switches redirect web traffic to the cluster.
)
Figure 18-7 Cisco WSA Cluster Example
The Cisco WSA comes in different models. The following are the different Cisco WSA models:
Cisco WSA S680
- It is a high-performance WSA designed for large organizations with 6000 to 12,000 users.
- A 2 rack-unit (RU) appliance with 16 (2 octa core) CPUs, 32 GB of memory, and 4.8 TB of disk space.
Cisco WSA S670
- A high-performance WSA designed for large organizations with 6000 to 12,000 users
- A 2 RU appliance with 8 (2 octa core) CPUs, 8 GB of memory, and 2.7 TB of disk space.
Cisco WSA S380
- Designed for medium-size organizations with 1500 to 6000 users.
- A 2 RU appliance with 6 (1 hexa core) CPUs, 16 GB of memory, and 2.4 TB of disk space.
Cisco WSA S370
- Designed for medium-size organizations with 1500 to 6000 users.
- A 2 RU appliance with 4 (1 quad core) CPUs, 4 GB of memory, and 1.8 TB of disk space.
Cisco WSA S170
- Designed for small- to medium-size organizations with up to 1500 users.
- A 1 RU appliance with 2 (1 dual core) CPUs, 4 GB of memory, and 500 GB of disk space.
The Cisco WSA runs Cisco AsyncOS operating system. The Cisco AsyncOS supports numerous features that will help mitigate web-based threats. The following are examples of these features:
- Real-time antimalware adaptive scanning: The Cisco WSA can be configured to dynamically select an antimalware scanning engine based on URL reputation, content type, and scanner effectiveness. Adaptive scanning is a feature designed to increase the “catch rate” of malware that is embedded in images, JavaScript, text, and Adobe Flash files. Adaptive scanning is an additional layer of security on top of Cisco WSA Web Reputation Filters that include support for Sophos, Webroot, and McAfee.
- Layer 4 traffic monitor: Used to detect and block spyware. It dynamically adds IP addresses of known malware domains to a database of sites to block.
- Third-party DLP integration: Redirects all outbound traffic to a third-party DLP appliance, allowing deep content inspection for regulatory compliance and data exfiltration protection. It enables an administrator to inspect web content by title, metadata, and size and to even prevent users from storing files to cloud services, such as Dropbox, Google Drive, and others.
- File reputation: Using threat information from Cisco Talos. This file reputation threat intelligence is updated every 3 to 5 minutes.
- File sandboxing: If malware is detected, the Cisco AMP capabilities can put files in a sandbox to inspect its behavior, combining the inspection with machine-learning analysis to determine the threat level. Cisco Cognitive Threat Analytics (CTA) uses machine-learning algorithms to adapt over time.
- File retrospection: After a malicious attempt or malware is detected, the Cisco WSA continues to cross-examine files over an extended period of time.
- Application visibility and control: Allows the Cisco ASA to inspect and even block applications that are not allowed by the corporate security polity. For example, an administrator can allow users to use social media sites like Facebook but block micro-applications such as Facebook games.
Cisco Content Security Management Appliance
Cisco Security Management Appliance (SMA) is a Cisco product that centralizes the management and reporting for one or more Cisco ESAs and Cisco WSAs. Cisco SMA has consistent enforcement of policy, and enhances threat protection. Figure 18-8 shows a Cisco SMA that is controlling Cisco ESA and Cisco WSAs in different geographic ocations (New York, Raleigh, Chicago, and Boston).
)
Figure 18-8 Cisco SMA Centralized Deployment
The Cisco SMA comes in different models. These models are physical appliances or the Cisco Content Security Management Virtual Appliance (SMAV). The following are the different Cisco SMA models:
- Cisco SMA M680: Designed for large organizations with over 10,000 users
- Cisco SMAV M600v: Designed for large enterprises or service providers
- Cisco SMA M380: Designed for organizations with 1000 to 10,000 users
- Cisco SMAV M300v: Designed for organizations with 1000 to 5000 users
- Cisco SMA M170: Designed for small business or branch offices with up to 1000 users
- Cisco SMAV M100v: Designed for small business or branch offices with up to 1000 users