53-1002153-01 29 April 2011 Fabric Watch Administrator’s Guide Supporting Fabric OS v7.0.
Copyright © 2000, 2002-2011 Brocade Communications Systems, Inc. All Rights Reserved. Brocade, the B-wing symbol, BigIron, DCFM, DCX, Fabric OS, FastIron, IronView, NetIron, SAN Health, ServerIron, TurboIron, and Wingspan are registered trademarks, and Brocade Assurance, Brocade NET Health, Brocade One, Extraordinary Networks, MyBrocade, and VCS are trademarks of Brocade Communications Systems, Inc., in the United States and/or in other countries.
Document History Title Publication Number Summary of Changes Date Fabric Watch User’s Guide 53-0001559-02 New document May 2000 Fabric Watch User’s Guide 53-0000186-02 n/a March 2002 Fabric Watch User’s Guide 53-0000504-02 n/a April 2003 Fabric Watch User’s Guide 53-0000524-02 n/a April 2003 Fabric Watch User’s Guide 53-0000524-03 Updated default values and restructured the document.
Title Publication Number Summary of Changes Date Fabric Watch Administrator’s Guide 53-1000601-03 Reorganized many sections to improve clarity. Updates to support Fabric OS v6.2.0: Virtual Fabric, port movement, fan monitoring behavior, link reset, and DCX4S. November 2008 Fabric Watch Administrator’s Guide 53-1001342-01 Updates to support Fabric OS v6.3.0: portThConfig, portFencing, and sysMonitor commands and Brocade 8000 support.
Contents About This Document In this chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv How this document is organized . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Supported hardware and software . . . . . . . . . . . . . . . . . . . . . . . . . . xvi Deprecated hardware platform support. . . . . . . . . . . . . . . . . . xvii What’s new in this document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii New information. . . . . . .
Switch monitoring components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Fabric events monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Performance monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Security monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 SFP monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Port monitoring . . . . . . . . . . . . . . . . . .
Fabric Watch classes, areas, and elements . . . . . . . . . . . . . . . . . . 21 Classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Class areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Chapter 4 Fabric Watch Activation In this chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SFP monitoring guidelines and default settings . . . . . . . . . . . . . . . . 45 SFP class areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 SFP monitoring default settings . . . . . . . . . . . . . . . . . . . . . . . . . 46 Performance monitoring guidelines and default settings . . . . . . . . 47 Performance Monitor class areas. . . . . . . . . . . . . . . . . . . . . . . . 47 Performance monitoring setting guidelines . . . . . . . . . . . . . . . .
Environment monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Environment class area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Environment monitoring setting guidelines . . . . . . . . . . . . . . . 78 Environment class default settings . . . . . . . . . . . . . . . . . . . . . . 78 Resource class settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Resource class area . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Watch reports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Switch Availability Monitor report. . . . . . . . . . . . . . . . . . . . . . . . . . .100 Generating a Switch Availability Monitor report. . . . . . . . . . . .100 Switch Health report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101 Generating a Switch Health report . . . . . . . . . . . . . . . . . . . . . .101 Switch Status Policy report. . . . . . . . . . . . . . . . . . . . .
Tables Table 1 Fabric Watch classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Table 2 Fabric Watch configuration tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Table 3 Fabric class areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Table 4 Fabric class default settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xii Fabric Watch Administrator’s Guide 53-1002153-01
Figures Figure 1 In-between buffer values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Figure 2 Time base set to none . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Figure 3 Event trigger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 4 Above event trigger with buffer zone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xiv Fabric Watch Administrator’s Guide 53-1002153-01
About This Document In this chapter • How this document is organized . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv • Supported hardware and software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi • What’s new in this document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii • Document conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviii • Additional information . . . . . . . . . . . . .
• Chapter 8, “System Monitoring,” describes how to configure system memory and CPU values using the sysMonitor command. This chapter also lists the switch status policy factors that affect the health of the switch, describes how to set and view switch status policies, and details how to configure FRUs. • Chapter 9, “Fabric Watch Configuration Using Web Tools,”provides information about how to use Web Tools to configure Fabric Watch settings, an alternative to using the command line interface.
What’s new in this document Hardware Platform Access Gateway? Brocade DCX 8510-8 Backbone (8-slot) Brocade DCX 8510-4 Backbone (4-slot) • FC8-64 port blade • FC16-32 port blade • FC16-48 port blade • FS8-18 encryption blade • FX8-24 DCX extension blade Deprecated hardware platform support The following hardware platforms are not supported in the Fabric OS 7.0.0 release. These platforms can interoperate with switches running Fabric OS 7.0.0, but cannot load Fabric OS 7.0.0.
What’s new in this document • Packet loss monitoring enhancements on the VE_Port. Changed information • Updated port fencing section and included moderate, aggressive, and conservative threshold settings. • Recommended settings for Port areas using the portThConfig command. • Buffer settings which now enable the user to configure “in-between” buffer values for thresholds that fall between the below high threshold and the above low threshold boundaries.
What’s new in this document Notes, cautions, and warnings The following notices and statements are used in this manual. They are listed below in order of increasing severity of potential hazards. NOTE A note provides a tip, guidance, or advice, emphasizes important information, or provides a reference to related information. ATTENTION An Attention statement indicates potential damage to hardware or data.
What’s new in this document Other industry resources For additional resource information, visit the Technical Committee T11 website. This website provides interface standards for high-performance and mass storage applications for Fibre Channel, storage management, and other applications: http://www.t11.org For information about the Fibre Channel industry, visit the Fibre Channel Industry Association website: http://www.fibrechannel.
What’s new in this document If you cannot use the licenseIdShow command because the switch is inoperable, you can get the WWN from the same place as the serial number, except for the Brocade DCX. For the Brocade DCX, access the numbers on the WWN cards by removing the Brocade logo plate at the top of the nonport side of the chassis. Document feedback Quality is our first concern at Brocade and we have made every effort to ensure the accuracy and completeness of this document.
What’s new in this document xxii Fabric Watch Administrator’s Guide 53-1002153-01
Chapter 1 Fabric Watch In this chapter • Fabric health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 • Fabric Watch overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 • Role-based access control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 • Fabric Watch licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 Fabric Watch overview Fabric Watch overview Fabric Watch is an optional storage area network (SAN) health monitor that allows you to enable each switch to constantly monitor its SAN fabric for potential faults and automatically alerts you to problems long before they become costly failures. Fabric Watch tracks a variety of SAN fabric elements and events. Monitoring fabric-wide events, ports, and environmental parameters enables early fault detection and isolation as well as performance measurement.
Reasons to customize Fabric Watch settings 1 Reasons to customize Fabric Watch settings Customization is recommended to achieve the following objectives: • • • • • • Selecting one or more event settings Selecting an appropriate message delivery method for critical and noncritical events Selecting appropriate thresholds and alarm levels relevant to each class element Defining the appropriate Time Base event triggering based on the class element traits Eliminating message delivery that has little or no pr
1 Reasons to customize Fabric Watch settings Event behavior configuration You must first use the fwSetToCustom command to switch from default to custom settings, and then use the advanced configuration options provided with the portThConfig, thConfig, and sysMonitor commands to configure event behavior, actions, and time bases at the port level. Time base configuration The time base specifies the time interval between two samples to be compared.
Class, area, and element hierarchy 1 Class, area, and element hierarchy Fabric elements and events are organized in a hierarchy by class, area, and element. There is a class, area, and element associated with every monitored behavior. Classes are the highest level in the system, subdivided into one or more areas. Areas contain one or more elements. An example of a very simple Class --> Area --> Element hierarchy follows.
1 Switch monitoring components The Performance Monitor class is divided into the following areas: • EE (end-to-end) Performance Monitor - Monitors RX and TX performance between two devices. • Filter Performance Monitor - Measures the number of frames transmitted through a port that match specific values in the first 64 bytes of the frame.
Switch monitoring components 1 If frame discard errors or any other configured areas exceed the currently effective threshold settings, the Fabric Watch daemon can take one or more of the following actions: • • • • Send an SNMP trap. Log a RASlog message. Send an E-mail alert. Fence the port, if port fencing is enabled (not applicable to E_Ports). Refer to “Port fencing” for more information.
1 Logical switch support System resource monitoring System resource monitoring enables you to monitor your system’s RAM, flash, memory, and CPU. You can use the sysMonitor command to perform the following tasks: • Configure thresholds for Fabric Watch event monitoring and reporting for the environment and resource classes. Environment thresholds enable temperature monitoring, and resource thresholds enable monitoring of flash memory.
Threshold monitoring using SNMP tables 1 Threshold monitoring using SNMP tables Understanding the components of SNMP makes it possible to use third-party tools to view, browse, and manipulate Brocade switch variables remotely. Every Brocade switch and director supports SNMP. When an event occurs and its severity level is at or below the set value, the Event Trap traps (swFabricWatchTrap), are sent to configured trap recipients.
1 Fabric Watch event settings Use the advanced configuration option provided with the portThConfig, thConfig, and sysMonitor commands to view and modify custom and default values for specified classes and areas in Fabric Watch. You can customize the information reported by Fabric Watch by configuring event behavior types, threshold values, time bases, and event settings. These area attributes are used to define and detect events in Fabric Watch. NOTE Event settings are non-persistent.
Fabric Watch notification types 1 Fabric Watch notification types Fabric Watch provides event notifications in several different formats to ensure that event details are accessible from all platforms and operating systems. In response to an event, Fabric Watch can record event data as any (or all) of the following alarm options. E-mail alert An e-mail alert sends information about a switch event to a specified e-mail address.
1 Fabric Watch audit messages You must configure the software to receive trap information from the network device. You must also configure the SNMP agent on the switch to send the trap to the management station. You can configure SNMP notifications using the snmpConfig command and you can configure notifications using Fabric Watch. For information on configuring the SNMP agent using the snmpConfig command, see the Fabric OS Command Reference.
Data values 1 Important information related to event classes is also tracked and made available. For example, you can track changes from an external source by the user name, IP address, or type of management interface used to access the switch. NOTE Audit messages are generated for port fencing configuration changes, whether port fencing is enabled or disabled. You can set up an external host to receive Audit messages so you can easily monitor unexpected changes.
1 14 Fabric Watch support in Access Gateway mode Fabric Watch Administrator’s Guide 53-1002153-01
Chapter 2 Fabric Watch Thresholds In this chapter • Threshold values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Time bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Threshold triggers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Fabric Watch alarm behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 Threshold values In-between buffer values The below high threshold is the term used to configure “in between” buffer values, as shown in Figure 1. In this example, the high threshold value is 5 and the buffer value is 1. Therefore, the “in-between” boundary value is 4. Enter the portThConfig command using the following parameters.
Time bases 2 Time bases Time bases specify the time interval between two samples to be compared. You can set the time base to day (samples are compared once a day), hour (samples are compared once an hour), minute (samples are compared every minute), or second (samples are compared every second). This configurable field affects the comparison of sensor-based data with user-defined threshold values.
2 Threshold triggers Figure 3 shows a sample graph of data obtained by Fabric Watch (the type of data is irrelevant to the example). A high threshold of 2 is specified to trigger an event. A time base of minute is defined. An event occurs only if the rate of change in the specific interval (one minute in this example) is across the threshold boundary. It should be either higher than the high threshold limit or lower than the low threshold limit.
Fabric Watch alarm behavior 2 Above event trigger Set the Above event trigger for an element that requires only high threshold monitoring. In the Above event trigger, Fabric Watch triggers an event immediately after the data value becomes greater than the high threshold. Define a buffer zone within the operational limit of an area to suppress multiple events when the counter value goes above the high threshold and fluctuates around it.
2 20 Fabric Watch alarm behavior Fabric Watch Administrator’s Guide 53-1002153-01
Chapter Fabric Watch Threshold Components 3 In this chapter • Fabric Watch classes, areas, and elements . . . . . . . . . . . . . . . . . . . . . . . . . 21 Fabric Watch classes, areas, and elements Fabric Watch uses a hierarchical organization to track the network device information it monitors. There is a class, area, and element associated with every monitored behavior. Classes are the highest level in the system, subdivided into one or more areas. Areas contain one or more elements.
3 Fabric Watch classes, areas, and elements • Chapter 8, “System Monitoring” The Resource class and Environment class areas and actions are configured using the sysMonitor command. The FRU class actions are configured using the fwFruCfg command Elements Fabric Watch defines an element as any fabric or switch component that the software monitors. Within each area, the number of elements is equivalent to the number of components being monitored.
Fabric Watch classes, areas, and elements TABLE 1 3 Fabric Watch classes (Continued) Class Description Port Enables you to set additional thresholds specific to different types of ports. The Port class is made up of the following sub-classes: • E_Port class—Represents ports connected to another switch. • FOP_Port class — Represents fabric or fabric loop ports that are made of optical fiber. • FCU_Port class — Represents fabric or fabric loop ports that are made of copper.
3 24 Fabric Watch classes, areas, and elements Fabric Watch Administrator’s Guide 53-1002153-01
Chapter Fabric Watch Activation 4 In this chapter • Interfaces for activating Fabric Watch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Interfaces for activating Fabric Watch This section provides a brief overview of the available user interfaces for activating Fabric Watch. Further details about Fabric Watch operations for each interface are provided later in this guide.
4 Interfaces for activating Fabric Watch If the Fabric Watch license is not listed, continue to step 4; otherwise, you are ready to use Fabric Watch. 4. Enter the license key with the licenseAdd key command, where key is the Fabric Watch license key. License keys are case-sensitive, so type the license key exactly as it appears. switch:admin> licenseadd "R9cQ9RcbddUAdRAX" 5. Enter the licenseShow command to verify successful activation.
Interfaces for activating Fabric Watch 4 In Figure 5, the MIB browser populated the left side of the screen with a MIB tree that you can navigate. 3. Open Web Tools and select Tasks > Manage > Switch Admin. 4. Click Show Advanced Mode. 5. On the SNMP tab, enter the IP address of the trap receiver and the severity level, and click Apply. NOTE The severity level must be informational (4) in order to forward threshold alerts. 6.
4 Interfaces for activating Fabric Watch Trap Recipient's IP address in dot notation: Community (rw): [OrigEquipMfr] Trap Recipient's IP address in dot notation: Community (rw): [private] Trap Recipient's IP address in dot notation: Community (ro): [public] Trap Recipient's IP address in dot notation: Trap recipient Severity level : (0..
Interfaces for activating Fabric Watch 4 10. Expand the tree on the left to find the Fabric Watch OID information. To find the OID, navigate the following hierarchy: SW-MIB. bcsi. commDev; fibrechannel, fcSwitch, sw, swFWSystem. Fabric Watch displays a screen similiar to the one shown in Figure 6. FIGURE 6 Example OID tree 11. Obtain the specific identifier for the element that will be modified.
4 Interfaces for activating Fabric Watch Activating Fabric Watch using Web Tools You can open Web Tools on any workstation with a compatible Web browser installed. 1. Open the Web browser and type the IP address of the device in the Address field: http://10.77.77.77 or https://10.77.77.77 2. Press Enter. A browser window opens to open Web Tools. A Login dialog box opens. 3. Enter your username and password. 4. Select a switch from the Fabric Tree and log in if necessary. 5.
Chapter 5 Fabric Watch Configuration In this chapter • Fabric Watch configuration tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Setting Fabric Watch custom and default values . . . . . . . . . . . . . . . . . . . . . • E-mail notification configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Notification configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5 Fabric Watch configuration tasks TABLE 2 32 Fabric Watch configuration tasks (Continued) Configuration task Command Location of procedure Set the following parameters for port monitoring: • Port type • Area type • Time base • Threshold level • Trigger (boundary level) • Action (notification type) • Buffer • Port fencing portThConfig portFencing “Port Monitoring” on page 57 Set the port persistence time fwSet --port -persistence “Setting the port persistence time” on page 69 Configure port fe
Setting Fabric Watch custom and default values 5 Setting Fabric Watch custom and default values Use the following commands to switch between custom and default values. These commands reset all thresholds for all classes: • fwSetToCustom - Sets the boundary and alarm level to custom. • fwSetToDefault - Restores the boundary and alarm level to the default. E-mail notification configuration In environments where it is critical that you are notified about errors quickly, you can use e-mail notifications.
5 E-mail notification configuration The Config Show menu lists each class for which you can provide a separate e-mail address. 2. Enter the number corresponding to the class for which the e-mail configuration should be displayed. Fabric Watch displays e-mail alert information such as: Mail Recipient Information ____________________________________ Email Alert = enabled Mail Recipient = sysadmin@mycompany.com The system returns to the main fwMailCfg menu. Disabling an e-mail alert 1.
E-mail notification configuration 5 Sending a test e-mail message 1. Enter 4 in the fwMailCfg menu to test the e-mail configuration for a specific class. The Config Show menu displays. 2. Select a class to test. If the e-mail configuration for the class is complete, the following confirmation message displays: Email has been sent If the e-mail configuration for the class is not complete, the following error message displays: Email has not been sent.
5 Notification configuration The following message displays: Setting 192.168.39.118 as Relay Host.. 4. Enter the Domain Name (example: Brocade.com). Displaying the relay host configuration 1. Enter 6 in the fwMailCfg menu to display the relay host configuration menu. 1 2 3 4 Display Relay Host configuration Set Relay Host IP Remove Relay Host configuration Quit 2. Enter 1 to display the configuration. Removing the relay host configuration 1.
Notification configuration 5 2. Enter the following command if you decide not to have notifications sent. switch:admin> fwAlarmsFilterSet 0 The 0 option turns the alarm notification off. All notifications are suppressed when alarm notifications are turned off, except for the Environment class and Resource class. 3. Verify or view your current alarm notifications by using the fwAlarmsFilterShow command.
5 38 Notification configuration Fabric Watch Administrator’s Guide 53-1002153-01
Chapter 6 Fabric, Security, SFP, and Performance Monitoring In this chapter • Fabric monitoring guidelines and default settings . . . . . . . . . . . . . . . . . . . . • Security monitoring guidelines and default settings. . . . . . . . . . . . . . . . . . . • SFP monitoring guidelines and default settings . . . . . . . . . . . . . . . . . . . . . . • Performance monitoring guidelines and default settings. . . . . . . . . . . . . . . • thConfig command . . . . . . . . . . . . . . . . . . . . . . . . . .
6 Fabric monitoring guidelines and default settings TABLE 3 Fabric class areas (Continued) Area Description Segmentation changes (SC) Tracks the cumulative number of segmentation changes. Segmentation changes occur because of one of the following: • Zone conflicts. • Incompatible link parameters. During E_Port and VE_Port initialization, ports exchange link parameters, and incompatible parameters result in segmentation. This is a rare event. • Domain conflicts.
Fabric monitoring guidelines and default settings 6 Fabric class default settings Table 4 provides default settings for areas in the Fabric class.
6 Security monitoring guidelines and default settings Security monitoring guidelines and default settings The Security class monitors all attempts to breach your SAN security, helping you fine-tune your security measures. Security class areas Table 5 lists Product Name areas in the Security class and describes what each area indicates. Although it is recommended that you leave the entire Security class in its default state (no alerts), you can configure the Security class using the thConfig command.
Security monitoring guidelines and default settings TABLE 6 6 Security class area default settings Area Description Default threshold settings Default alarm settings Threshold state DCC violations (DV) Monitors DCC violations Unit: Violations Time Base: minute Low: 1 High: 2 Buffer: 0 Below: 0 Above: 3 Informative Out_of_range HTTP violations (HV) Monitors HTTP violations Unit: Violations Time Base: minute Low: 1 High: 2 Buffer: 0 Below: 0 Above: 3 Informative Out_of_range Illegal commands
6 Security monitoring guidelines and default settings TABLE 6 44 Security class area default settings (Continued) Area Description Default threshold settings Default alarm settings Threshold state Telnet violations (TV) Monitors Telnet violations Unit: Violations Time Base: minute Low: 1 High: 2 Buffer: 0 Below: 0 Above: 3 Informative Out_of_range TS out of sync (TS) Monitors instances in which the timestamp is out of sync Unit: Violations Time Base: minute Low: 1 High: 2 Buffer: 0 Below: 0
SFP monitoring guidelines and default settings 6 SFP monitoring guidelines and default settings The SFP class groups areas that monitor the physical aspects of SFPs. An SFP class alarm alerts you to an SFP malfunction fault. SFP performance monitoring is not supported on VE_Ports. When a port goes offline, the RXP and TXP area values of the SFP become zero.
6 SFP monitoring guidelines and default settings SFP monitoring default settings The SFP default settings are shown in Table 8. The default alarm configuration (log all alarms only to the error log) is sufficient. It is recommended that you do not allow alerts to go out as SNMP traps. If other Port class issues are reported, review the error log for any supporting data for SFP issues.
Performance monitoring guidelines and default settings 6 Performance monitoring guidelines and default settings Performance monitoring serves as a tuning tool. The Performance Monitor class groups areas that track the source and destination of traffic. Use the Performance Monitor class thresholds and alarms to determine traffic load and flow and to reallocate resources appropriately. NOTE Performance Monitoring is not supported on VE_Ports.
6 thConfig command Table 11 provides default settings for areas in the End-to-End Performance Monitor class.
thConfig command TABLE 12 6 Configuration options for thConfig command Class name Valid area types Threshold Threshold action Configuration recommendation Fabric ED - Number of E_Ports down FC - Fabric reconfiguration DC - Domain ID changes SC - Segmentation changes ZC - Zone changes FL - Fabric logins Default or Custom1 Default or Custom2 It is recommended that you leave the entire Fabric class in its default state (no alerts).
6 thConfig command thConfig command examples With the exception of setting thresholds for the RX area of an end-to-end (EE) performance monitor, which requires special licensing in Access Gateway mode, it is recommended that you use the default settings for these classes. Setting the high threshold of the RX area of an EE monitor The thConfig command provides the ability to monitor thresholds for frame monitoring and end-to-end (EE) performance on both Access Gateway (AG) switches and non-AG switches.
thConfig command 6 Fabric Watch also monitors the Brocade Quad SFP (QSFP) and, as with the 16 Gbps SFP, if configured thresholds are crossed, Fabric Watch generates an SNMP alarm, a RASlog message, and an e-mail alert for the following SFP areas: • • • • Current Voltage Temperature RXP NOTE On core blades, only the 16 Gbps QSFPs can be installed. Voltage and temperature monitoring using the QSFP A QSFP connects four ports of one core blade of a chassis to another core blade of a different chassis.
6 thConfig command Specifying the 16 Gbps SFP type You can use the sfpType operand to manage the actions and thresholds for the Current, Voltage, RXP, TXP, and Temperature areas of the 16 Gbps SFPs. If you do not provide the SFP type parameters, the existing thresholds and actions of the SFP class are changed to the default. SFP types for the 16 Gbps and QSFP SFPs are listed in Table 13.
6 thConfig command .
6 Recommended settings for Fabric, SFP, Performance, Security classes Displaying the number of 16 Gbps SFP operational hours To show the number of hours that the 16 Gbps SFP is operational, enter the thConfig command using the following parameter. Note that the only supported timebase for this area is none. switch:admin> thconfig --show sfp -area PWRONHRS -sfptype 16GSWL Displaying the SFP health information The sfpshow command displays the health of 10 and 16 Gbps SFPs and the QSFP.
6 Recommended settings for Fabric, SFP, Performance, Security classes TABLE 14 Recommended settings for Fabric, SFP, Performance, and Security classes (Continued) E=Error_Log, S=SNMP_Trap, P=Port_LOG_LOCK, M=EMAIL_ALERT, F=Port Fence Unit Time Base Low Thresh High Thresh Buffer Default Below Above X C None -10 85 3 X E E RX power (RXP) X uWatts None 0 5000 25 X E E TX power (TXP) X uWatts None 0 5000 25 X E E Current X mA None 0 50 1 X E E Voltage X uV N
6 56 Recommended settings for Fabric, SFP, Performance, Security classes Fabric Watch Administrator’s Guide 53-1002153-01
Chapter 7 Port Monitoring In this chapter • Port class areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Port class guidelines and default settings. . . . . . . . . . . . . . . . . . . . . . . . . . . • Port configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • portThConfig command procedures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Port fencing . . . . . . . . . . . . . . .
7 Port class guidelines and default settings TABLE 15 Port class areas (Continued) Area Description Protocol errors (PE) The number of times a protocol error occurs on a port. Invalid state due to LRR on an online link. Occasionally these errors occur due to software glitches. Persistent errors occur due to hardware problems. Received packets (RXP) The percentage of maximum bandwidth consumed in packet receipts.
Port class guidelines and default settings 7 Port class default settings Table 16 provides default settings for areas in the Port class. TABLE 16 Port class default settings Area Description Default threshold settings Default alarm settings Threshold state Cyclic redundancy check (CRC) Monitors the number of CRC errors.
7 Port configuration TABLE 16 Port class default settings (Continued) Area Description Default threshold settings Default alarm settings Threshold state Transmitted packets (TXP) Monitors transmission rate, by percentage. Unit: Percentage (%) Time Base: minute Low: 0 High: 100 Buffer: 0 Below: 0 Above: 0 Informative Informative I Link reset Monitors the number of link resets sent by a given port (LR-Out) and received on a given port (LR-In).
portThConfig command procedures 7 Using the nosave command The nosave command prevents the configuration changes from being saved persistently. This option allows you to make and view changes without overwriting the saved configuration. CAUTION When you use --set with the --nosave option and the switch reboots, your changes are lost. portThConfig command procedures The following sections provides specific examples for the Port class.
7 portThConfig command procedures switch:admin> portthconfig --apply port -area crc -action cust -thresh_level custom 3. To display the port threshold configuration for the Port class and all areas: switch:admin>portthconfig --show port Configuring all physical ports to monitor for invalid transmission words Invalid transmission words (ITW) occur when a word does not transmit successfully, resulting in encoding errors. Invalid word messages usually indicate a hardware problem. 1.
portThConfig command procedures 7 E_Port subclass setting guidelines E_Port guidelines for the areas listed below represent a more aggressive approach in most areas, because failing or failed E_Ports in a large fabric can cause serious fabric-wide issues if not detected early. NOTE The E_Port class represents ports connected to another switch. • Area: Link Failure Count You want to be immediately notified if an E_Port loses a link, so set the alarm configuration to Changed for this area.
7 portThConfig command procedures • Area: Trunk Utilization Set the high boundary to 75 percent and the alarms to Above and In-Between conditions. These settings indicate if the 75 percent threshold is exceeded and for how long. • Areas: Primitive Sequence Protocol Error, State Changes, Utilization, Packet Loss Use the default settings. E_Port class default settings Table 17 provides default settings for areas in the E_Port class.
portThConfig command procedures TABLE 17 E_Port class default settings (Continued) Area Description Default threshold settings Default alarm settings Threshold state Sync loss Monitors the number of loss of synchronization errors. Unit: Errors Time Base: minute Low: 0 High: 500 Buffer: 50 Below: 0 Above: 0 Informative Out_of_range Packet loss (VE_Port only) The number of packets routed through a port exceeds the port bandwidth.
7 portThConfig command procedures FOP_Port and FCU_Port subclass setting guidelines FOP_Port and FCU_Port guidelines for the areas listed below represent a more aggressive approach in most areas. NOTE The settings in these subclasses include settings for the host bus adapter (HBA) ports as well as the storage ports.
portThConfig command procedures 7 • Area: Class 3 (C3) Discards Unlike the other areas, take a conservative approach for the C3 Discards area. Use the default settings and configure the alarms for Above. The goal is to locate issues with the device or its infrastructure, so monitor the data to help isolate issues. Port fencing is one of the recommended solutions for isolating issues. • Area: Trunk Utilization The Trunk Utilization area is new; therefore, recommended settings are not yet available.
7 portThConfig command procedures TABLE 18 68 FOP_Port and FCU_Port class default settings (Continued) Area Description Default threshold settings Default alarm settings Threshold state Signal loss Monitors the number of signal loss errors. Unit: Errors Time Base: minute Low: 0 High: 5 Buffer: 0 Below: 0 Above: 0 Informative Out_of_range Sync loss Monitors the number of loss of synchronization errors.
portThConfig command procedures 7 VE_Port class default settings Table 19 provides default settings (per minute) for areas in the VE_Port class. TABLE 19 VE_Port class default settings Area Description Default threshold settings Default alarm settings Threshold state Packet loss The number of packets routed through a port exceeds the port bandwidth. Unit: Errors Time Base: minute Low: 0 High: 10 Buffer: 0 Below: 0 Above: 0 Informative Out_of_range State changes (ST) Monitors state changes.
7 Port fencing 1. Use the fwSet --port -persistence command to set the port persistence time. switch:admin> portthconfig --show [port_type] 2. Set the port persistence time. switch:admin> fwSet --port -persistence seconds Port fencing Port fencing monitors ports for erratic behavior and disables a port if specified error conditions are met. You can customize the thresholds and configure the ports to report errors for one or more areas using the portThConfig command.
Port fencing 7 The allowed threshold configuration settings are displayed on a per-class basis. FOP_Port class thresholds apply to the entire switch. You can set different thresholds for Storage and Host FOP_Ports if they are on different switches, based on the fabric configuration. Port fencing configuration You must configure port thresholds with the portThConfig command before you can enable port fencing using the portFencing command.
7 Port fencing Enabling port fencing 1. Connect to the switch and log in as admin. 2. Configure port thresholds. Information about how to configure port thresholds is detailed in Chapter 7, “Port Monitoring”. 3. Enter the portFencing --enable command.
Port fencing • • • • 7 Link Reset (Fabric OS only) Protocol Errors (M-EOS and Fabric OS) Security (M-EOS) State Changes (Fabric OS only) Refer to Chapter 5, “Device Configuration,” of the Data Center Fabric Manager User Manual for detailed instructions on how to add, edit, view, and remove thresholds.
7 Recommended port configuration settings Recommended port configuration settings Table 22 lists the recommended settings for physical port, E_Port, FOP_Port, and FCU_Port for both the host device and the storage device.
7 Recommended port configuration settings TABLE 22 Recommended configuration for the Port class (Continued) E=Error_Log, S=SNMP_Trap, P=Port_LOG_LOCK, M=EMAIL_ALERT, pf=Port Fence High Thresh Buffer 0 5 0 X E Trunk Util X Percentage Minute 0 75 0 X E FOP_Port and FCU_Port Link Loss X Errors Minute 0 15 0 X E,S Sync Loss X Errors Minute 0 45 0 X E,S HOST Signal Loss X Errors Minute 0 45 0 X E,S Notes Low Thresh Minute Above Time Base Errors Below Unit
7 76 Recommended port configuration settings Fabric Watch Administrator’s Guide 53-1002153-01
Chapter 8 System Monitoring In this chapter • Environment monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • Resource class settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • System monitoring using the sysMonitor command . . . . . . . . . . . . . . . . . . . • Recommended environment and resource monitoring settings . . . . . . . . . • Switch monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8 Environment monitoring Environment monitoring setting guidelines Use Environment Class default settings. Temperature settings are switch-dependent and there is no need to alter them. The default alarm configuration, sending alerts to the error log and SNMP, is sufficient. Environment class default settings Table 24 provides default Environment class settings for all switches. Check the appropriate hardware reference manual for differences in actual environmental requirements.
Resource class settings TABLE 24 8 Environment class default settings (Continued) Area Description Default threshold settings Default alarm settings Threshold state Below: 3 Above: 3 Informative Out-of-range Brocade 8000 Low: 0 High: 73 Buffer: 10 Brocade DCX Low: 0 High: 70 Buffer: 10 Brocade DCX-4S Low: 0 High: 75 Buffer: 10 Resource class settings The Resource class monitors flash memory. It calculates the amount of flash space consumed and compares it to a defined threshold.
8 System monitoring using the sysMonitor command System monitoring using the sysMonitor command Use the sysMonitor command to configure temperature and system resource settings at the chassis level. For detailed information about the sysMonitor command, refer to the Fabric OS Command Reference Manual. The following operations are supported by the sysMonitor command: • Configure thresholds for Fabric Watch event monitoring and reporting for the environment and resource classes.
Examples of the sysMonitor command 8 Examples of the sysMonitor command The following sections provide specific examples for the Environment class, CPU, and memory. Environment class settings Temperature settings are switch-dependent and there is no need to alter them. The default alarm configuration, sending alerts to the error log and SNMP, is sufficient. See “Environment monitoring setting guidelines” on page 78 for more information.
8 Examples of the sysMonitor command CPU and memory When configuring CPU monitoring, specify a value in the 1-100 range. When the CPU usage exceeds the limit, a Fabric Watch alert is triggered. The default CPU limit is 75 percent. When configuring memory, the limit specifies a usage limit as a percentage of available resources. When used to configure memory monitoring the limit value must be greater than the low limit and smaller than the high limit.
8 Recommended environment and resource monitoring settings Configuring the system memory usage monitoring threshold 1. Enter the sysMonitor command using the following parameters: switch:admin> sysmonitor --config mem -poll 10 -retry 1 -limit 20 -action snmp, raslog -high_limit 80 2. Apply the changes: switch:admin> sysmonitor --apply mem -action_level cust Recommended environment and resource monitoring settings Table 27 lists the recommended settings for Environment and Resource classes.
8 Switch monitoring Switch status policy planning Fabric Watch monitors the health of the switch under various classes. Table 28 lists the current overall switch status policy parameters in a switch and identifies the factors that affect their health. Note that not all switches use the listed monitors. NOTE The default setting for a MARGINAL state is 0, which prevents Fabric Watch from generating notifications due to missing power supplies.
FRU monitoring 8 The following example shows a switch status policy for temperature: Bad Temperatures contributing to DOWN status: (0..10) [0] 3 Bad Temperatures contributing to MARGINAL status: (0..10) [0] 1 The following example shows a switch status policy for fans: Bad Fans contributing to DOWN status: (0..3) [0] 2 Bad Fans contributing to MARGINAL status: (0..3) [0] 1 Switch status policies are saved in a nonvolatile memory, and therefore are persistent until changed.
8 FRU monitoring FRU class areas Table 29 lists Fabric Watch areas in the FRU class and describes each area. Possible states for all FRU-class areas are absent or removed, faulty, inserted, on, off, ready, and up. Configure the FRU class using the fwFruCfg command TABLE 29 FRU class areas Area Description Fan State of a fan has changed. Power supply State of a power supply has changed. Slot State of a slot has changed. WWN State of a WWN card has changed. SFP State of the SFP has changed.
FRU monitoring 8 Absent-1, Inserted-2, On-4, Off-8, Faulty-16 Configurable Alarm Actions are: Errlog-1, E-mail-16 Slot Alarm State: (0..31) [1] Slot Alarm Action: (0..17) [1] Power Supply Alarm State: (0..31) [1] Power Supply Alarm Action: (0..17) [1] Fan Alarm State: (0..31) [1] Fan Alarm Action:(0..17) [1] WWN Alarm State: (0..31) [1] WWN Alarm Action: (0..31) [1] SFP Alarm State: (0..31) [1] SFP Alarm Action:(0..
8 88 FRU monitoring Fabric Watch Administrator’s Guide 53-1002153-01
Chapter Fabric Watch Configuration Using Web Tools 9 In this chapter • Fabric Watch Configuration Using Web Tools . . . . . . . . . . . . . . . . . . . . . . . . 89 Fabric Watch Configuration Using Web Tools You can use Web Tools to define the following Fabric Watch configurations: • Configure custom threshold values on particular elements. • Place limits on the acceptable values of those elements and enable the custom limits (configure threshold boundaries).
9 Fabric Watch Configuration Using Web Tools NOTE Unless the switch is a member of the current Admin Domain context, Fabric Watch is view-only. FIGURE 7 Fabric Watch dialog box Fabric Watch Explorer, on the left side of the window, displays the available classes. Not all classes are available for all switches. The status bar at the bottom of the window provides you with a summary of recent actions, and the date and time the module was last updated.
Fabric Watch Configuration Using Web Tools 9 4. Click the Alarm Configuration tab. 5. Select SNMP Trap, RAS log, or both options. 6. Click Apply. Fabric Watch threshold configuration The Threshold Configuration tab enables you to configure event conditions. From this tab, you configure threshold traits, alarms, and e-mail configurations. NOTE Use the procedures in this section to configure threshold traits for all classes except for the FRU class.
9 Fabric Watch Configuration Using Web Tools 7. If necessary, select a time to record the event in the Time Base field. 8. Enter the lowest boundary of the normal zone in the Low Boundary field. 9. Enter the highest boundary of the normal zone in the High Boundary field. 10. Enter the size of the buffer zone in the Buffer Size field. 11. Click Apply.
Fabric Watch Configuration Using Web Tools 9 5. In the Activate Level area: • Click System Default to use the system default settings and proceed to step 7. or • Click Custom Defined to specify new settings and proceed to the next step. 6. Select the check box for the type of notification method you want to use for each event type. The available alarm actions are: • • • • 7. ERROR_LOG SNMP_TRAP PORT_LOG_LOCK EMAIL_ALERT Click Apply.
9 Fabric Watch Configuration Using Web Tools Configuring alarms for FRUs Configuration for the FRU class is different from configuration for the other classes. Because FRUs are not monitored through a threshold-based system, they have a simpler interface for configuration. For FRUs, you configure the states for which an event occurs, as described in the following procedure. 1. Open the Fabric Watch window. 2. Select the Threshold Configuration tab. 3. In Fabric Watch Explorer, select a FRU class. 4.
Fabric Watch Configuration Using Web Tools 9 Fabric Watch alarm information From Fabric Watch, you can view two types of reports: • Alarm notifications—Displays the alarms that occurred for a selected class or area. • Alarm configuration—Displays threshold and alarm configurations for a selected class or area.
9 Fabric Watch Configuration Using Web Tools To display the alarms page, perform the following steps. 1. Open the Fabric Watch window. 2. In Fabric Watch Explorer, select the class that you want to check for alarms. 3. Select the Alarm Notification tab. 4. In Area Selection, select the area that you want to check for alarms from the list. All alarms for that area display. E-mail notification You can be notified of an alarm condition through an e-mail alert.
Fabric Watch Configuration Using Web Tools 9 To enable an e-mail alerts recipient, perform the following steps. 1. Open the Fabric Watch window. 2. Select the Email Configuration tab. 3. Select a FRU class in the Fabric Watch Explorer tree. 4. Click Enable. 5. Enter the e-mail address of the recipients in the Recipient Email Address field. Separate the e-mail addresses with commas. 6. Click Apply. 7. Repeat steps 3 through 6 for any additional FRU classes. 8.
9 98 Fabric Watch Configuration Using Web Tools Fabric Watch Administrator’s Guide 53-1002153-01
Chapter 10 Fabric Watch Reports In this chapter • Fabric Watch reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 • Switch Availability Monitor report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 • Switch Health report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 • Switch Status Policy report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 Switch Availability Monitor report Switch Availability Monitor report The Switch Availability Monitor (SAM) report lets you see the uptime and downtime for each port. It also enables you to check if a particular port is failing more often than the others. NOTE SAM report details do not display the health status of GbE ports. Fabric Watch only monitors and reports the status for physical and virtual FC ports. You can run reporting commands in Fabric Watch to get instant access to switch information.
Switch Health report 2/12 2/13 2/14 2/15 3/0 3/1 3/2 3/3 3/4 3/5 3/6 3/7 3/8 3/9 3/10 LB LB U LB T U U U U U U U U U VE 100 100 0 100 100 0 0 0 0 0 0 0 0 0 100 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10 0 0 100 0 0 100 100 100 100 100 100 100 100 100 0 Switch Health report The Switch Health report lists the following information: • Current health of each port, based on the currently-configured policy settings.
10 Switch Status Policy report The final portion of the report, detailing port health, is not available without a Fabric Watch license. Switch Status Policy report The Switch Status Policy report displays the current policy parameter. The following example of the switchStatusPolicyShow command output is for enterprise-class platforms such as the DCX Backbone. For modular switches, the switch status policy report also contains information on the WWN, Blade, and CP.
10 Port Detail report Port Detail report If the Switch Health report shows marginal throughput or decreased performance, use the Port Detail report to see statistics on each port. The Port Detail report is a Fabric Watch licensed product. You can also see port details by health. For example, you can see only healthy ports, only marginal ports, only faulty ports, or only offline ports. The following is an example of a Port Detail report.
10 Port Detail report 219 220 221 222 223 VE VE VE VE VE HEALTHY HEALTHY HEALTHY HEALTHY HEALTHY 003:37 002:48 061:19 061:19 061:19 - - - - - - - - - - - - - NOTE Output of the Port Detail report depends on the ports that belong to the current Admin Domain context. If a port does not belong to the current Admin Domain, nothing other than the port number is displayed for that port.
Index A C above event triggers, 19 access gateway mode, restrictions in Fabric Watch, 13 action configuration guidelines, 3 activating Fabric Watch, 25, 89 using a Telnet session, 25 using SNMP, 26 using Web Tools, 30 alarm behavior, 19 alarm configuration report for Fabric Watch, 95 alarm notification configuration, 36 alarms triggered, 10 alarms, Fabric Watch configuring, 92, 94 displaying, 95 enabling and disabling, 93 alerts configuration recommendations, 4 area environment class, 77 fabric class, 39
D F data values, 13 DCFM, port fencing configuration, 72 default settings E_Port, 64 environment class, 78 Fabric class, 41 FOP_Port and FCU_Port, 67 performance monitor class default settings, 47 port class, 59 VE_Port, 69 disabling Fabric Watch threshold alarms, 93 displaying alarms, Fabric Watch, 95 fabric class areas, 39 default settings, 41 fabric event monitoring, 5 Fabric health concepts, 1 fabric monitoring guidelines, 39 recommended settings, 54 setting guidelines, 40 Fabric Watch actions, 7 act
FCU_Port default settings, 67 FCU_Port setting guidelines, 66 FOP_Port default settings, 67 FOP_Port setting guidelines, 66 FRU alarms, configuring, 94 FRU class areas, 85 configuration, 85 recommended settings, 87 specifying triggers for alarms, 87 FRU configuration, 32 FRU monitoring, 85 I interface types, 25 invalid CRC area, configuring, 61 IP address, setting for notification, 35 L licenseAdd key command, 26 locked port log notification type, 12 M management information base (MIB), 9 memory configur
Q QSFP description of, 50 monitoring, 50 support for, 6 R RASlog notification type, 12 RASlog, generation when core blade is removed, 83 RBAC, permissions required for Fabric Watch, 2 relay host configuration displaying, 36 removing, 36 setting, 35 resource class area, 79 default settings, 79 recommended settings, 83 setting guidelines, 79 resource class area, 79 S security class areas, 42 security monitoring, 6 recommended settings, 54 setting time base, 17 settings, customizing, 3 SFP support for 10 Gbp
