Page 3
Brocade Communications Systems, Incorporated Corporate Headquarters Asia-Pacific Headquarters Brocade Communications Systems, Inc. Brocade Communications Singapore Pte. Ltd. 1745 Technology Drive 9 Raffles Place San Jose, CA 95110 #59-02 Republic Plaza 1 Tel: 1-408-333-8000 Singapore 048619 Fax: 1-408-333-8101 Tel: +65-6538-4700 Email: info@brocade.com Fax: +65-6538-0302 Email: apac-info@brocade.com European and Latin American Headquarters...
Page 4
Title Publication Number Summary of Changes Date Fabric Watch Administrator’s Guide 53-0000524-06 Renamed book. Combined March 2005 the Introduction and Concepts chapters into a single chapter. Added support for Brocade 200E, Brocade 3014, and Brocade 48000. Fabric Watch Administrator’s Guide 53-1000047-01 Updates to support Fabric November 2005...
About This Document This document is an administrator’s guide to help you use the Fabric Watch product to monitor and improve fabric health. This document is specific to Brocade Fabric OS version 5.3.0 and all switches running Fabric OS v5.3.0. For a list of supported hardware for this release, see “Supported Hardware and Software”...
Supported Hardware and Software This document is specific to Brocade Fabric OS version 5.3.0 and all switches running Fabric OS version 5.3.0, including: • Brocade 200E • Brocade 3250 • Brocade 3850 • Brocade 3900 • Brocade 4100 • Brocade 4900 •...
For further information, see the release notes. Document Conventions This section describes text formatting conventions and important notices formats. ORMATTING The narrative-text formatting conventions that are used in this document are as follows: bold text Identifies command names Identifies the names of user-manipulated GUI elements Identifies keywords and operands Identifies text to enter at the GUI or CLI italic text...
ERMS For definitions specific to Brocade and Fibre Channel, see the Brocade Glossary. For definitions of SAN-specific terms, visit the Storage Networking Industry Association online dictionary at http://www.snia.org/education/dictionary. Additional Information This section lists additional Brocade and industry-specific documentation that may be helpful. ROCADE ESOURCES The following related documentation is provided on the Brocade Documentation CD-ROM and on...
Getting Technical Help Contact your switch support supplier for hardware, firmware, and software support, including product repairs and part ordering. To expedite your call, have the following information available: 1. General Information Technical Support contract number, if applicable Switch model Switch operating system version Error numbers and messages received supportSave command output...
Introduction to fabric health Fabric Watch lets you define how often to measure each switch and fabric element and to specify notification thresholds. Whenever fabric elements exceed these thresholds, Fabric Watch automatically provides notification using several methods, including e-mail messages, SNMP traps, and log entries.
Fabric Watch components Although the concept of fabric health initially seems fairly simple, it can be a deep and complex topic due to the number of factors that are involved. One of the more obvious criteria for fabric health is the condition of the network hardware. A switch or port failure could easily prevent data packets from reaching their destination.
Fabric Watch components Table 1 describes the classes into which Fabric Watch groups all switch and fabric elements. TABLE 1 Product Name classes Class Description Environment Includes information about the physical environment in which the switch resides and the internal environment of the switch. For example, an Environment-class alarm alerts you to problems or potential problems with temperature, fans, and power.
Fabric Watch components Environment class areas Table 2 lists and describes the Product Name areas in the Environment class. TABLE 2 Environment class areas Area Description Refers to the speed of the fans inside the switch, in revolutions per minute. It is important that the fans spin quickly enough to keep the ambient temperature from rising to levels at which switch damage might occur.
Fabric Watch components TABLE 3 Fabric class areas (Continued) Area Description Loss of E_Port Tracks the number of times that an E_Port or VE_Port goes down. E_Ports and VE_Ports go down each time you remove a cable or an SFP (where there are SFP failures or transient errors).
Fabric Watch components Performance Monitor class areas Table 5 lists Product Name areas in the Performance Monitor class and describes each area. TABLE 5 Performance Monitor class areas Area Indicates Customer Define Relies on performance monitor telnet commands. For more information on this area, see the Fabric OS Command Reference.
Fabric Watch components Port class areas Table 6 lists and describes the Product Name areas in the port class. NOTE Fabric Watch monitors and reports the status of physical and virtual FC ports. Physical GbE ports and ISCSI ports are not monitored and are not included in the Port Class area. TABLE 6 Port class areas Area...
Fabric Watch components Resource class area Table 7 describes the Product Name resource class area. TABLE 7 Resource class area Area Description Flash Monitor Monitors the compact flash space available by calculating the percentage of flash space consumed and comparing it with the configured high threshold value. Security class areas Table 8 lists Product Name areas in the security class and describes what each area indicates.
Fabric Watch components TABLE 8 Security class areas (Continued) Area Indicates SLAP Bad Packets A SLAP (Switch Link Authentication Protocol) bad packets failure occurs when the switch receives a bad SLAP packet. Bad SLAP packets include unexpected packets and packets with incorrect transmission IDs. SLAP Failures A SLAP failure occurs when packets try to pass from a nonsecure switch to a secure fabric.
Configuring events Subclasses are a minor exception to the preceding mapping rule. Subclasses, such as E_Ports, contain areas with elements equivalent to the number of valid entries. Within the same example used thus far in this section, in a 64-port switch in which eight ports are connected to another switch, each area within the E_Port class would contain eight elements.
Configuring events DATA VALUES A data value represents three aspects of a fabric: counter value, measured value, or state value. Data values are updated by Fabric Watch approximately every six seconds, an interval that you cannot change. Counter value is the total number of times that a given event has occurred. For each monitored event during the time period, the value is incremental.
Configuring events Buffer values Figure 1 shows an example in which each time a signal crosses the high limit, an event occurs. The arrows indicate the points at which the event criteria is met. In this case, there is a great deal of fluctuation.
Page 30
Configuring events Figure 3 shows a high limit of 65 degrees Celsius placed on a counter measuring temperature. During each sample period, Fabric Watch measures the temperature and compares it against the high threshold. If the measured temperature exceeds the high threshold, it triggers an event. FIGURE 3 Time base set to none Specifying a time base...
Page 31
Configuring events As illustrated on the tenth sample, the counter value changes from 0 to 1; hence calculated rate of change is 1 per minute. At the thirteenth sample, the rate of change is 2 per minute. The rate of change must be at least 3 per minute to exceed the event-triggering requirement of 2, which is met on the eighteenth sample.
Configuring events At the tenth sample, the rate of change is one per minute. At the fourteenth, twenty-first, and twenty-fifth sample, the rate of change remains equal to the high threshold of 2. In this case, Fabric Watch does not trigger an event even though the absolute value of the counter reaches 4, which is well above the high threshold.
Configuring events the high threshold (event 1 in Figure 6). When the data value becomes less than the high threshold and buffer value, Fabric Watch triggers a second event (Event 2) to indicate that it has returned to normal operation. The second event will not be triggered until the counter value falls below the high threshold and buffer values.
Port persistence In-between event triggers Fabric Watch event triggers are usually set to notify the user of a warning or failure condition, but there is an exception. You can define the In-Between event trigger to receive a notification of fault recovery.
Notification methods Fabric Watch uses port persistence for a port event that requires the transition of the port into a marginal status. Fabric Watch does not record any event until the event persists for a length of time equal to the port persistence time. If the port returns to normal boundaries before the port persistence time elapses, Fabric Watch does not record any event.
Notification methods • Class, area, and index number of the threshold that the counter crossed • Event type • Value of the counter that exceeded the threshold • State of the element that triggered the alarm • Source of the trap The trap stores event information but does not actively send alerts.
Switch policies ASSIGNING NOTIFICATION METHODS Specify the particular notification method that you want Fabric Watch to use by assigning it a value. Table 10 shows the numerical values for each notification method. TABLE 10 Numerical values of notification methods Notification method Assigned value Error log entry SNMP trap...
Chapter Activating and Accessing Fabric Watch In this chapter • Activating Fabric Watch ..... . . 23 • Accessing Fabric Watch ..... . . 25 Activating Fabric Watch Fabric Watch must be activated on each switch individually before use.
Activating Fabric Watch If the Fabric Watch license is not listed, continue to step 3; otherwise, you are ready to use Fabric Watch. 3. Type licenseAdd “key”, where key is the Fabric Watch license key. License keys are case-sensitive, so type the license key exactly as it appears. switch:admin>...
Accessing Fabric Watch 8. Click the Add button on the License tab. The Add License dialog displays. 9. Type the license key in the License Key: field and click Add License. This activates Fabric Watch. Accessing Fabric Watch This section provides a brief overview of the available user interfaces. Further details about Fabric Watch operations for each interface are provided later in this guide.
Accessing Fabric Watch 1. Launch the Switch View for your switch, as described in “Activating with Web Tools” page 24. 2. To access the Fabric Watch View, click the Watch button in this portion of the screen, which appears as follows: SNMP-BASED ENTERPRISE MANAGERS Use SNMP-based enterprise managers to: •...
Accessing Fabric Watch 3. Begin a telnet session with the switch and type the agtCfgSet command. For information on the agtCfgSet command, see the Fabric OS Command Reference. 4. Type the IP address for the switch into the Host field. Type the community into the Community field.
Page 46
Configuring Fabric Watch thresholds 1. Log in to the switch as the administrator. 2. Type fwConfigure at the command prompt. 3. The fwConfigure menu displays, as shown in Figure swd77:admin> fwconfigure : Environment class : SFP class : Port class : Fabric class : E-Port class : F/FL Port (Optical) class...
Configuring Fabric Watch thresholds STEP 2: CONFIGURE THRESHOLDS After you have identified and selected the appropriate class and areas, you can configure thresholds for those classes and areas. If you want a basic configuration, accept the default configuration settings. Unless you want to accept the basic (default) configuration, or first disable, enable, or refresh all existing thresholds, proceed to option 4, advanced configuration.
Page 52
Configuring Fabric Watch thresholds 222 eportRXPerf222 Triggered 223 eportRXPerf223 Triggered Threshold boundary level is setat : Default DefaultCustom Unit Percentage(%) Percentage(%) Time base minuteminute High BufSize FIGURE 15 Customizing high threshold boundary for RXPerformance Figure 16 shows how to change the threshold boundary level to custom so that the new custom value of 80 is the new trigger point.
Page 54
Configuring Fabric Watch thresholds 1. Add the numbers beside each state (for the states you want to include). The values for the states are: Errlog - 1 SnmpTrap - 2 PortLogLock - 4 RapiTrap - 8 EmailAlert - 16 2. Enter the total at the prompt. : change behavior type 11 : change threshold alarm level : change behavior interval...
Page 55
Configuring Fabric Watch thresholds Above InBetween FIGURE 18 Change above alarm Figure 19 shows how to select the custom settings for the threshold alarm level for the RXPerformance area. The options are either to accept the default settings or provide custom settings.
Page 56
Configuring Fabric Watch thresholds : change custom buffer 18 : return to previous page : apply threshold boundary changes 10 : cancel threshold boundary changes Select choice => : (1..18) [18]16 FIGURE 20 Applying threshold alarm changes NOTE To ensure that your alarm setting configuration is in effect, remember to change the alarm level to Custom and then apply the alarm settings.
Page 57
Configuring Fabric Watch thresholds Table 14 describes the 18 customization options displayed at the end of the Advanced Configuration menu. With the exception of the last option, which exits advanced configuration mode, each option has similar behavior. For each option, one or two lines will appear, prompting you to accept the new setting information, and, after the information has been provided, the entire screen will refresh to display the updated information.
Configuring notifications TABLE 14 Advanced Configuration options (Continued) Option Effect Input information change below alarm Changes the notification method for The required notification methods below event occurrences for this method, but only affects the custom column. change inBetween alarm Changes the notification method for The required notification methods inBetween event occurrences for this method, but only affects the custom...
Configuring notifications • “Configuring SNMP notifications” on page 43 • “Configuring API notifications” on page 43 • “Configuring Port Log Lock actions” on page 44 • “Configuring E-mail notifications” on page 44 CONFIGURING ALARM NOTIFICATIONS When you use alarm notifications, error messages are sent to designated locations such as an error log, SNMP trap view, or e-mail.
Configuring notifications CONFIGURING PORT LOG LOCK ACTIONS Port Log Lock freezes in time the port log dump output if an event is triggered. See “Step 3: Configure alarms” on page 37 for details about configuring port log lock actions. Chapter 1, “Fabric Watch Concepts” for more details about port log lock.
Page 61
Configuring notifications Mail Recipient Information ____________________________________ Email Alert = enabled Mail Recipient = sysadmin@mycompany.com The system returns to the main fwMailCfg menu. Disable e-mail alert 1. Type 2 in the fwMailCfg menu (shown in Figure 21) to disable e-mail alerts for a specific class. The Config Show menu (shown in Figure 22) displays.
Configuring switch status policy Check Mail configuration for Environment class! The e-mail address specified in the mail configuration receives a test e-mail message. The system returns to the main fwMailCfg menu. Set recipient mail address for e-mail alert 1. Type 5 in the fwMailCfg menu (shown in Figure 21) to specify the recipient to whom Fabric Watch should send the e-mail alert for a class.
Configuring switch status policy • Marginal One or more components are triggering a Warning alarm. • Down One or more contributors have failed. Status events are integrated into Brocade Web Tools and Fabric Manager so that if the overall status of your switch is Healthy, the switch color is green. If the overall switch status is Marginal, then the switch color is yellow.
Configuring FRUs STEP 2: IMPLEMENT YOUR SWITCH STATUS POLICY After planning and defining your switch status policy, enter the switchStatusPolicySet command to configure each policy. Each policy has two parameters that can be configured: Marginal and Down. Set the number of units Marginal or Down based on your system requirements for each policy/parameter.
Configuring Fabric Watch using Web Tools Note that the value 0 for a parameter means that it is NOT used in the calculation Configurable Alarm States are: Absent-1, Inserted-2, On-4, Off-8, Faulty-16 Configurable Alarm Actions are: Errlog-1, E-mail-16 Slot Alarm State: (0..31) [31] Slot Alarm Action: (0..17) [1] Power Supply Alarm State: (0..31) [0] Power Supply Alarm Action: (0..17) [0]...
Page 66
Configuring Fabric Watch using SNMP To configure Fabric Watch using SNMP 1. Open a MIB browser. 2. Load the appropriate MIB files. First, load the Brocade common MIB file (BRCD_v5_0.mib), followed by the Brocade software MIB file (SW_v5_3.mib). If this is successful, the system displays a screen similar to Figure FIGURE 24...
Page 67
Configuring Fabric Watch using SNMP SW-EXTTRAP (yes, y, no, n): [no] swd77:admin> FIGURE 25 Enabling Fabric Watch traps in SNMP swd77:admin> agtcfgset Customizing MIB-II system variables ... At each prompt, do one of the following: o <Return> to accept current value, o enter the appropriate new value, o <Control-D>...
Page 68
Configuring Fabric Watch using SNMP 4. Enter the IP address for the switch in the Host field in the MIB browser. Enter the community string in the Community field. To perform set operations, enter the write community into the Write Community field. 5.
Page 69
Configuring Fabric Watch using SNMP Any changes related to Fabric Watch, like changing the status of the Temperature sensor, (snapshot of whose details has been provided) traps, will generate traps. To view trap details, click Trap Details. Figure 28 displays the Trap Detail view. FIGURE 28 Trap Detail view 6.
Page 70
Configuring Fabric Watch using SNMP Fabric Watch displays a screen similar to the one shown in Figure FIGURE 29 Example OID tree Obtain the specific identifier for the element that will be modified. To get the identifier, click the swFwThresholdTable and swFwThresholdEntry directory, and run a get operation on swFwName.
Page 71
Configuring Fabric Watch using SNMP For detailed descriptions of the SNMP fields in both telnet and Web Tools, see the Fabric OS MIB Reference. FIGURE 30 Example swFwName screen In this example, 83.1 is numeric identifier for the element referenced as resFlash000. 8.
Chapter Generating Fabric Watch Reports In this chapter • Types of Fabric Watch reports ....57 • Viewing Fabric Watch reports ....62 Types of Fabric Watch reports You can run reporting commands in Fabric Watch to get instant access to switch information.
Page 74
Types of Fabric Watch reports 1/10 1/11 1/12 1/13 1/14 1/15 2/10 2/11 2/12 2/13 2/14 2/15 3/10 Table 16 describes the fields in the SAM report. TABLE 16 SAM report information Heading Meaning Total Up Time (Percent) The percentage of time that the port is active Total Down Time (Percent) The percentage of time that the port experiences faults Down Occurrence (Count)
Types of Fabric Watch reports SWITCH HEALTH REPORT The switch health report lists • Current health of each port, based on the currently configured policy settings. • High-level state of the switch as well as power supplies, fans, and temperatures. •...
Types of Fabric Watch reports Example: Switch status policy report The current overall switch status policy parameters: Down Marginal ---------------------------------- PowerSupplies Temperatures Fans Blade Flash MarginalPorts FaultyPorts MissingSFPs 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.
Page 77
Types of Fabric Watch reports HEALTHY 000:00 HEALTHY 000:00 HEALTHY 000:00 HEALTHY 000:00 HEALTHY 000:00 HEALTHY 000:00 HEALTHY 000:00 HEALTHY 000:00 HEALTHY 061:19 HEALTHY 061:19 HEALTHY 061:19 HEALTHY 003:37 HEALTHY 002:48 HEALTHY 061:19 HEALTHY 061:19 HEALTHY 061:19 NOTE Output of the Port Detail Report depends on the ports that belong to the current Admin Domain context.
Viewing Fabric Watch reports Viewing Fabric Watch reports You can view Fabric Watch reports in the following ways: • “Viewing Fabric Watch reports using telnet,” next • “Viewing Fabric Watch reports using Web Tools” on page 62 VIEWING FABRIC WATCH REPORTS USING TELNET To view Fabric Watch reports using telnet, start a telnet session and enter the command (from Table 18) corresponding to the report you want to view.
Viewing Fabric Watch reports A report view window (shown in Figure 32) displays. FIGURE 32 Report View window The Web Tools report view window contains choices to view the switch health, port detail, or SAM report. NOTE The switch status policy report is not available through Web Tools. Viewing reports To view a report, click its name on the left of the report view window.
Page 80
Viewing Fabric Watch reports The menu, shown in Figure 33, displays. FIGURE 33 Action menu The following section explains the options in the Action menu. Refresh report Use this option to update the current report with the new data. Customize report Use this option to modify the report design.
Page 81
Viewing Fabric Watch reports Switch Health Port Detail If you select the Port Detail option, you can filter the ports for which the report is generated based on port number or port state. If you filter based on port number, you can enter either the port numbers (separated by commas) or the range of port numbers.
Environment class Environment class Table 19 provides default settings for areas in the Environment class. These defaults are hardware-dependent. Check the appropriate hardware reference manual for differences in environmental requirements. NOTE For the Brocade 200E, there is no fan default threshold because the fans are not monitored by Fabric Watch.
Page 85
Environment class TABLE 19 Environment class threshold defaults (Continued) Area Description Default threshold settings Default alarm settings Threshold state Power Supply Monitors power supply Unit: 1/0 (OK/FAULTY) Changed: 0 Informative condition Time Base: none Below: 3 Out_of_range Above: 3 In_range The default threshold In-Between: 0 Informative...
Fabric class Fabric class Table 20 provides default settings for areas in the Fabric class. These defaults are hardware-dependent. Check the appropriate Hardware Reference Manual for differences in environmental requirements. TABLE 20 Fabric class threshold defaults Area Description Default threshold settings Default alarm Threshold state settings...
Performance Monitor class Performance Monitor class Table 21 provides default settings for areas in the AL_PA Performance Monitor class. TABLE 21 AL_PA Performance Monitor class threshold defaults Area Description Default threshold settings Default alarm Threshold state settings AL_PA Invalid CRCs Monitors the number of Unit: Error(s) Changed: 0...
Port class Port class Table 24 provides default settings for areas in the Port class. TABLE 24 Port class threshold defaults Area Description Default threshold settings Default alarm settings Threshold state Invalid CRC Count Monitors the Unit: Error(s) Changed: 0 Informative number of CRC Time Base: minute...
Page 89
Port class Table 25 provides default settings for areas in the E-Port class. TABLE 25 E-Port class threshold defaults Area Description Default threshold settings Default alarm settings Threshold state Invalid CRC Count Monitors the number Unit: Error(s) Changed: 0 Informative of CRC errors Time Base: minute Below: 0...
Page 90
Port class Table 26 provides default settings for areas in the F/FL_Port class. TABLE 26 F/FL_Port class threshold defaults Area Description Default threshold settings Default alarm settings Threshold state Loss of Monitors the Unit: Error(s) Changed: 0 Informative Synchronization number of loss of Time Base: minute Below: 0 Informative...
Resource class Resource class Table 27 provides default settings for areas in the Resource class. TABLE 27 Resource class threshold defaults Area Description Default threshold settings Default alarm settings Threshold state Flash Monitors the Unit: Percentage(%) Changed: 0 Informative percentage of Time base: none Below: 3 Informative...
Page 92
Security class TABLE 28 Security class threshold defaults (Continued) Area Description Default threshold Default alarm Threshold state settings settings Invalid Certificates Monitors invalid Unit: Violation(s) Changed: 0 Informative certificates Time Base: minute Below: 0 Informative Low: 1 Above: 3 Out_of_range High: 2 In-Between: 0 In_range...
Appendix Basic Fabric Watch Configuration Guidelines A default Fabric Watch configuration is available for the purpose of saving setup time. As you gain familiarity with Advanced Fabric Watch features, they can be tailored to suit the fabric environment. The custom settings available in Fabric Watch provide an advanced user much needed flexibility of redefining boundary thresholds and alarm notification methods.
Profiles Profiles Brocade provides partial configuration files, or profiles, that help you configure Product Name in a way that is most appropriate to your particular SAN needs. To configure Product Name with a profile 1. Upload the configuration file to the host by typing configUpload. 2.
Page 100
fwclassinit fwconfigreload fwfrusfg RapiTrap fwmailcfg resource class area high and low thresholds security class areas setting time base to none SFP class areas SNMP capabilities in-between triggers SNMP trap installing Fabric Watch specifying a time base interface types switch event (error) log entry interpreting event messages switch policies switch status...