F5 F5CAB5 - BIG-IP Administration Support and Troubleshooting (F5CAB5) exam

This scenario describes a classic network performance issue known as the "Mega-Proxy" problem. When an organization routes all employee traffic through a single proxy server, the BIG-IP sees thousands of unique users as having the exact same source IP address. If the administrator has configured "Source Address Affinity" persistence, the BIG-IP will correctly follow the rule but incorrectly route all users to the same single backend pool member. This creates a severe load imbalance where one server is overwhelmed while others remain idle, leading to poor application response times. To resolve this, the administrator must change the persistence profile to "HTTP Cookie". Cookie-based persistence allows the BIG-IP to place a unique identifier in each user's browser, allowing the system to distinguish between individual sessions even if they share the same source IP. This fix ensures that traffic is distributed evenly across the pool members, restoring4 the expect5ed load balancing functionality and resolving the slow performance reported by users behind the corporate proxy.

The exhibit shows the virtual serverapplication_vswith a status indicating it isoffline but enabled. In BIG-IP terminology, this status means the virtual server itself is administratively enabled, but it is unable to pass traffic becauseno usable pool members are available.

Two common and documented causes for this condition are:

Pool member(s) administratively disabled (Option A):When all pool members are administratively disabled, BIG-IP removes them from load-balancing decisions. Even though the virtual server remains enabled, it has no available pool members to send traffic to, resulting in an offline status.

Node(s) administratively disabled (Option C):Pool members inherit the status of their parent nodes. If a node is administratively disabled, all associated pool members are also marked unavailable. This condition causes the virtual server to show as offline, even though the virtual server configuration itself is correct.

The other options are incorrect:

Forced offline pool members (Option B)result in a different operational intent and are explicitly set for maintenance scenarios.

Virtual server administratively disabled (Option D)would show the virtual server as disabled, not enabled/offline.

This behavior is consistent with BIG-IP traffic management logic and is commonly verified by reviewingpool and node availability stateswhen diagnosing virtual server availability issues.

Comprehensive and Detailed 150 to 250 Words Explanation From BIG-IP Administration, Support, and Troubleshooting Documents:

These log messages indicate that the BIG-IP system’sTraffic Management Microkernel (TMM)has enteredaggressive eviction modedue tohigh global memory utilization. When TMM memory consumption reaches critical thresholds, BIG-IP activates thedefault eviction policyto protect system stability and prevent a full traffic processing failure. This condition directly corresponds toOption A, where the global eviction policy is triggered because TMM memory resources are nearing exhaustion.

Once aggressive mode is activated, BIG-IP begins using theconnection sweeper mechanism, which selectively terminates existing connections to free memory. The repeated log entries stating“1 connections killed”confirm that the system isreaping some connections, not all connections. This behavior matchesOption C. The eviction process is incremental and controlled, targeting idle, low-priority, or least-recently-used connections first to minimize impact on active traffic.

Option B is incorrect because BIG-IP doesnotdrop all connections during aggressive mode; it only removes enough connections to relieve memory pressure. Option D is also incorrect because TMM eviction is based onTMM global memory usage, not swap memory utilization. TMM does not rely on swap space in the same way the host Linux system does.

These messages are a critical warning sign that the system is under memory stress and may require traffic optimization, connection limits, or hardware scaling.

Identifying network-level performance issues often starts with investigating hardware-level alerts78. In F5 terminology, a "Link" like "1.0" or "1.1" refers to a physical interface on the appliance79. When an SNMP trap reports that a link is "DOWN," it indicates a loss of signal or an administrative shutdown of the physical port80. To verify this, the administrator must navigate to Network > Interfaces > Interface List81. This screen provides real-time status, showing whether the interface is "up," "down," or "uninitialized," as well as any media speed or duplex mismatches that could be causing performance degradation82. Troubleshooting this is the first step in resolving "pool member down" or "VLAN failsafe" issues, as a down interface will take down any VLANs associated with it, immediately halting all traffic flow for the services relying on that physical path.

When adding new members to an active, high-traffic pool, the Slow Ramp Time setting is critical for maintaining application stability.

Mechanism: The Slow Ramp Time feature (located in the Pool properties) allows the BIG-IP system to gradually increase the number of connection requests sent to a newly added or recently enabled pool member.

The Issue: In a highly utilized pool, if Slow Ramp Time is set to 0 (the default), the BIG-IP immediately begins sending a proportional share of traffic to the new members. If the application requires a "warm-up" period (e.g., to build local caches or establish database connection pools), the sudden influx of traffic can overwhelm the new server, causing it to drop requests or fail to load content for users.

Recommendation: F5 recommends setting a non-zero Slow Ramp Time (measured in seconds) to allow the new member to scale up its processing capacity incrementally.

These log messages indicate that the BIG-IP system is under significant resource pressure and has activated its Adaptive Connection Management.

Global Eviction Policy: The message Aggressive mode ... activated (global memory) confirms that the system has reached a memory utilization threshold that triggers the eviction policy. This happens because TMM (the data plane) is running low on available memory pages.

Sweeper/Reaping: Once in "Aggressive mode," the BIG-IP "sweeper" starts reaping (terminating) connections to free up memory. The log Connections killed confirms this is occurring.

Impact: The system does not drop all connections; it targets connections based on the eviction policy (e.g., oldest connections or those exceeding limits) to bring memory usage back to safe levels. Thus, some connections will be dropped (Option A), and the cause is TMM memory exhaustion (Option C).

Comprehensive and Detailed 150 to 250 Words Explanation From BIG-IP Administration, Support, and Troubleshooting Documents:

AQKViewfile is the primary diagnostic support bundle used by F5 Support to troubleshoot BIG-IP system issues. It contains comprehensive system information, including running configuration, licensing details, module provisioning, hardware status, software versions, log files, statistics, and the output of numerous diagnostic commands. Generating a QKView is a standard and recommended first step when investigating performance problems, configuration issues, or when opening a support case with F5.

In the BIG-IP Configuration Utility (GUI), the correct and supported location to generate a QKView isSystem > Support. This menu is specifically designed for support and troubleshooting operations. From this section, administrators can generate a QKView file, monitor its creation progress, download it locally, or upload it directly to F5 iHealth for automated analysis. This workflow is clearly documented in BIG-IP Administration and Support guides and aligns with F5 best practices.

The other menu options are not appropriate:

System > Configurationis used for system-wide settings such as DNS, NTP, and device identity.

System > Archiveis used to create UCS backup files, which are configuration backups, not diagnostic bundles.

System > Logsis used only for viewing system logs, not generating support files.

Therefore,System > Supportis the correct and only valid answer.

To manually verify a health monitor's "Send String" from the BIG-IP command line, the curl utility is the preferred tool because it allows for custom header insertion.

Matching the Monitor String: The monitor string requires an HTTP/1.1 request which must include a "Host" header. Option A correctly uses the --header (or -H) flag to pass Host: www.example.com to the specific IP and port of the pool member.

Troubleshooting Logic: If curl --header 'Host: www.example.com ' 'http://10.10.3.75:8080/ ' returns a "200 OK" but the BIG-IP monitor still shows "Down," the administrator should check if the Receive String in the monitor configuration matches the output provided by curl.

Invalid Syntax: Option D is incorrect because it tries to append the hostname to the URI path, which the web server will likely reject with a "404 Not Found". tracepath (Options B and C) is a path discovery tool similar to traceroute and cannot validate HTTP response content.

Comprehensive and Detailed Explanation From BIG-IP A41dministration Support and Troubleshooting documents:To confirm functionality across a complex environment, the "Network Map" is the most efficient troubleshooting tool in the Configuration Utility43. It provides a hierarchical, visual representation of the traffic management objects44. A single glance allows the administrator to see the status of a Virtual Server (Green/Red/Yellow), the status of its associated pool, the health of individual pool members, and which iRules are currently attached45. This view is superior to the standard "Virtual Server List" for troubleshooting because it maps the dependencies between objects46. For example, if a Virtual Server is "Red," the Network Map will show if that status is inherited from a failed pool or a specific monitor failing on a pool member. Reviewing these basic stats in the Network Map helps the administrator quickly isolate whether a failure is at the service level (Virtual Server), the logic level (iRule), or the hardware level (Pool Member).