• Hp Procurve Discovery Tool
• Hp Procurve Switches
• Hp Procurve Switch Configuration Guide

Hewlett Packard Enterprise Support Center HPE Support Center. The HPE Network Discovery Tool provides small and midsize businesses with fast visibility of HPE Networking and other devices on the network. This free, Windows-based tool makes it easier for organizations to complete device setup and simplify ongoing device management, reducing the need for paper-based systems.

ProCurve Manager Plus version 2.0 includes a fee-based upgrade product to ProCurve Manager Plus 1.x. There are five products available for purchase in this release.

• J8778A ProCurve Manager Plus 2.0 - 100 device license
• J9009A ProCurve Manager Plus 2.0 - unlimited device license
• J8991A ProCurve Manager Plus 2.0 Upgrade to 1.x - 100 device license
• J8779A ProCurve Manager Plus 2.0 Upgrade to 1.x - unlimited device license
• J8477A ProCurve Identity Driven Manager 1.0

HP ProCurve Manager Plus is a Windows-based network management solution shipped with all manageable HP ProCurve devices. It provides all of the tools you need to effectively manage your network: mapping and polling capabilities, device auto-discovery and topology, tools for device configuration and management, monitoring network traffic,. HP Procurve Networking Initial Setup Note: HP Procurve Networking Configuration will be done via the CLI and assumes you have a console connection to the switch and are in configure mode. Enter configure mode by typeing configure.

This compressed, ZIP-format file contains ProCurve Manager 2.0 and a 30-day free trial of ProCurve Manager Plus version 2.0, and ProCurve Identity Driven Manager version 1.0.

Alcohol 120 full version for windows 7 64. These release notes include information on the following:

NOTE: These Release Notes are applicable at the date of the ProCurve Manager Version 2.0 Release. Please check the ProCurve Support Web site for more recent information.

Below is a summary of the new features that are available in ProCurve Manager (PCM) with this release. Please refer to the ProCurve Manager Network Administrator's Guide for a full description on the use of these features:

PCM Features:

• A new Import/Export tool lets you import Device and Subnet data from a .CSV (comma delimited) file into PCM, or export Device and Subnet data from PCM to a .CSV file, for use in other applications.
• SNMP V3 and SSH are now supported by PCM for device access and management, and for communications between PCM and ProCurve devices.
• The PCM automatic update feature keeps your system updated with the latest revisions and device driver support for PCM software.
• When new ProCurve devices are added to the network, PCM's new Automatic Registration feature can collect the necessary data and automatically register the devices with ProCurve support.

PCM Plus Features:

• PCM now includes a Find Node and Node-to-Node path trace feature to help you quickly and easily locate devices on the network.
• The PCM Plus Configuration manager now includes the ability to create a Device configuration 'Template' for each device type in the network. A Configuration Template policy can be used to automatically apply templates to new devices when they are found in the network, thus simplifying configuration and management as your network expands.
• PCM Plus now supports use of Event Driven Policies; that is, you can create a Policy to be launched when a specific alert (event) is received. Then configure the Alert to execute the Policy automatically based on the defined parameters in PCM.
• With the Network Consistency: Network Analyzer wizard you can check to make sure that the network connections between devices are configured correctly.
• The new Configurable Integration Platform included in PCM Plus can be used to:
• Create and manage 'User-defined devices,' that is other ProCurve or non-ProCurve devices not found through auto-discovery.
• Create user-defined 'Actions' and 'Triggers' to launch 3rd-party applications from within the PCM GUI.
• Receive and process traps, and log events for user-defined devices and to log events from OV-NNM.

Once you install PCM or PCM Plus Version 2.0, you cannot revert to the previously installed version. If you are uncertain if you want to upgrade to the 2.0 Version, it is best to install it on a system that does not have any ealier versions of PCM or PCM+ installed.

Minimum System Requirements

• 2 GHz Intel Pentium IV or equivalent
• 1 GB RAM
• 1 GB free hard disk space

Supported Operating Systems:

• MS Windows 2000
• MS Windows XP Professional (SP 1 or greater)
• MS Windows 2000 Server, Advanced Server or Professional with SP 4 or better
• MS Windows 2003 Server

Additional processing power and disk space may be required for larger networks and to support extensive traffic monitoring. Also, the online help feature in PCM is only supported on MS Internet Explorer web browser.

Minimum System Requirements

• 1 GHz Intel Pentium IV or equivalent
• 1 GB RAM
• 250 MB free hard disk space

Recommended System Requirements

• 2 GHz Intel Pentium IV or equivalent
• 2 GB RAM
• 500 MB free hard disk space

Supported Operating Systems:

• MS Windows 2000
• MS Windows XP Professional (SP 1 or greater)
• MS Windows 2000 Server, Advanced Server or Professional with SP 4 or better
• MS Windows 2003 Server

NNM Versions that should be installed prior to PCM 2.0 install

• NNM 6.4 on Win 2000 Professional, Windows 2000 Server, Windows 2000 Advanced Server, Windows XP
• NNM 7.01 on Win 2000 Professional, Windows 2000 Server, Windows 2000 Advanced Server, Windows XP
• NNM 7.5 on Win 2000 Professional, Windows 2000 Server, Windows 2000 Advanced Server, Windows XP, Windows 2003 Standard edition and Windows 2003 Enterprise edition

Additional processing power and disk space may be required for larger networks and to support extensive traffic monitoring. Also, the online help feature in PCM is only supported on MS Internet Explorer web browser.

One Network Management Program per Computer

Because of resource conflict that will occur if you have multiple network management tools running on the same computer, for example during discovery of the network devices, make sure you uninstall any other network management programs from your computer before installing ProCurve Manager or ProCurve Manager Plus.

Note : The above is not applicable when PCM 2.0 is installed on systems running HP Openview NNM.

Working With Multi-homed Systems:

A multi-homed system is a server or PC that has more than one IP address. Generally this is achieved by installing more than one network card in the system, but there are other ways that a system can be multi-homed. Here are a few of the situations that meet this definition:

• A system with two or more network adaptors.
• A system with a traditional ethernet network adaptor, plus a wireless adaptor.
• A system with only one network adaptor, but that is running some network tunneling software such as a VPN client. Generally what happens in this situation, is that the system appears to have two network interfaces (each with its own IP address). But in reality the system only has one physical adaptor, and the VPN client software emulates a second adaptor (while using the original adaptor under the covers).

When ProCurve Manager (either client or server) starts up, it attaches itself to the primary network interface. All network traffic between the client and server will be directed to the selected network interface. For example, if the ProCurve Manager client application attaches itself to the 166.3.4.5 interface, and the ProCurve Manager server is running on the 15.255.120.* network, there is no way that the client will ever connect successfully to the server.

To resolve this problem PCM has a configuration file that you can change to correct this situation. To setup this file, follow these steps:

1. Find the commIpAddr.txt file. This file exists in the config directory, so for example, for the client this file exists in: C:Program FilesHewlett-PackardPNMclientconfig . For the PCM server, you need to create a text file (with Notepad or similar application) and name it 'commIpAddr.txt' and place it in the C:Program FilesHewlett-PackardPNMserverconfig directory.
2. Edit the file with a text-based editor (such as Notepad or Wordpad), and enter the IP address of the interface you want the application to attach to. For example for the network illustrated above, you would add the entry ' 15.255.120.25 ' (without the quotes) in the first line of the file. More than one IP address can be used, but each IP Address entry must be on a separate line.
3. Save your changes.
4. Restart the application. If this is the ProCurve Manager client, just restart the application. If this is the ProCurve Manager server, you must restart the PCM services (HP ProCurve -Datastore, -Network Manager Server, and -Traffic Launch Service) from the Services control panel.

This issue is also described in the ProCurve Manager Getting Started Guide that comes with the CD package.

Adding Remote Client Stations

When you install ProCurve Manager Plus, the server and client functions are installed on the computer. You can also install the client function on any number of other computers in your network that have network access to the server computer.

To install the client function on another computer, simply start a web browser, such as Microsoft Internet Explorer, and for the URL type in the IP address of the server computer followed by a colon and the port ID 8040. For example, if the IP address of the server computer is 10.15.20.25, then you would enter 'http://10.15.20.25:8040' on the web browser address line. That will launch the client installation wizard and step you through the installation process.

If you have multiple ProCurve Manager servers in the network, when you install a remote client you will be prompted to select the server to which you want the client to attach. This server will be used each time the client program is launched. You can change the server that is being accessed by selecting the 'ProCurve Manager Server Discovery' option that was included when you installed the client. From the Windows 'Start' menu, select Programs ->ProCurve Manager ->ProCurve Manager Server Discovery

For PCM 2.0 (PCM-NNM version), the PCM Remote Client can only be installed on machines that have NNM Remote console installed. Once installed, the client will always connect to the server attached to the NNM remote console.

Configuring Client/Server Access Permissions

The ProCurve Manager server maintains a list of authorized clients that are permitted to log into the server. By default, when the ProCurve Manager server is installed, the only client allowed to log in is the client on the same system as the server--that is, no remote servers are allowed. This can be a problem for customers who are unaware of this security feature, because they will try to install remote clients using the browser, and will be unable to connect to the server after completing the client installation.

There are two files associated with ProCurve Manager client/server security that can easily be configured to allow access to any set of actual or potential clients. There are two ways that this file can be configured, depending on what you know about the clients that need to connect.

The access.txt file can be configured with a list of IP addresses specifying the clients that are authorized to log into the server. The file may contain as many addresses as needed, one IP address per line; or you may configure IP addresses with wildcards. DNS names are also allowed in the file, including DNS names with wildcards (this is useful for DHCP environments where a system's DNS name remains unchanged, although it's actual IP address may change from time to time). For example, below is an example of a valid access.txt file:

15.255.124.84
15.29.37.*
10.*.*.*
*.rose.hp.com
system1.hp.com

To add an entry, open the access.txt file, which can be found in the config directory ( C:Program FilesHewlett-PackardPNMserverconfig ). Be sure to edit the file using a text-based editor such as Notepad or Wordpad. Edit the file as necessary, one entry per line, then save it. It is NOT necessary to restart the server; the changes will take effect immediately.

There are situations where it is not possible to know ahead of time what IP address a potential client will have. This is particularly the case in situations where the client comes in through some sort of VPN, where the IP address of the client is assigned externally. To solve this problem it is possible to add client passwords to the access.txt file that correspond to specially configured clients. Note that even though you will be modifying the same access.txt file as for method 1 (above), the two mechanisms can freely co-exist--that is, the access.txt file can contain a combination of IP addresses and passwords. To enable password access for a particular client, follow these instructions:

1. First you must change an entry in the serverconfigTyphoonServer.cfgfile .
   This file is a text file and can be edited with Notepad or Wordpad. Look for the entry that reads
   'AUTHENTICATION=10 ', and change it to read ' AUTHENTICATION=100 '. Save the file and restart the server (listed as 'HP ProCurve Network Manager Server' in the services list).
2. Edit the access.txt file as described above, but instead of entering an IP address, just enter the selected password (on a line by itself). Save the file. It is not necessary to restart the server. For example, if we set the password to 'procurve':

procurve
*.rose.hp.com
system1.hp.com

1. On the client (the client must already be installed), you must edit the riptide.cfg file. This file already has several entries in it. You must add a line similar to the following:

PASSWORD = your password

Do not change any of the other entries in the file, as they are necessary for the correct operation of the client.

A sample Riptide.cfg file, once edited with the password 'procurve', would look like this:

LEASE_LENGTH = 40000
TRACING_PROPERTY_KEY = CoreServices.Main
MANUFACTURER = Hewlett-Packard
SERVICE_NAME = Typhoon
COMPONENT_DB = config/Components.prp
TRACING_DBFILE = config/Loggers.prp
NETWORK_DELAY = 25000
VERBOSE = true
PASSWORD = procurve

Once you have saved the riptide.cfg file, start the PCM Client and enter (select) the address of the PCM Server in the Direct address field of the 'Search for Servers' dialog. The client should now connect successfully to the server.

A trap is an alert sent by a device via UDP protocol to notify one or more hosts that something has occurred. A device may send a trap when a link on a port comes up, when a device has received an excessive amount of errors, or when a device has detected an excessive amount of broadcasts. In order to receive traps from a particular device, the switch must be configured with the host's address. This is accomplished in multiple ways, such as CLI, SNMP, and ProCurve Manager.

When ProCurve Manager (server) starts up, it binds to port number 162. Port 162 is the port that all incoming traps arrive on. The problem arises when a previous process is already bound to that port, in which case ProCurve Manager will not be able to receive traps because the port is in use.

To resolve the problem, make sure no process is bound to port 162. Examples of applications that bind to port 162 are the Windows SNMP Trap Receiver Service*, TopTools, HP OpenView and MG-Soft MIB Browser Trap Ringer. In the event that a process was bound to port 162 when ProCurve Manager was started simply terminate the process and restart the ProCurve Manger (server). To restart the ProCurve Manager (server) in Windows 2k go to Control Panel->Administrative Tools-> Services. Double click on the HP ProCurve Network Manager Server, click the Stop button, and then click the Start button. In Windows XP/2003 go to Control Panel->Administrative Tools-> Services, double click on the HP ProCurve Network Manager Server, click the Stop button, and then click the Start button.

NOTE for PCM-NNM Version:

The above is not applicable for the PCM-NNM installation. PCM cannot bind to port 162 since it is already used by NNM.All device traps and application events will be displayed on the NNM Alarm browser.

For the ProCurve Wireless Access Point 420 devices, the PCM Switch Software Update utility will only work with switch software versions newer than 2.0.29

For the ProCurve Wireless Access Point 520wl devices, if you upgrade to version 2.4.5 of the switch software, the PCM Switch Software Upgrade utility will not allow you to 'downgrade' to an earlier version of the switch software.

AP420 running firmware prior to 2.1.0

• If the 'Communication Parameters in PCM' tool is used to modify the SNMP settings used to communicate with an AP420 running firmware prior to the 2.1.0 release, the device will become unreachable due to the tool causing PCM to use SNMPv2 for this device. An update will be available shortly, please contact ProCurve support for a release date.

AP420 upgrading to 2.1.0 firmware

• There is a known issue that causes PCM to not be able to upgrade the AP420 to the 2.1.0 firmware version. An update will be available shortly, please contact ProCurve support for a release date.

Problem scanning configurations on AP420 running 2.1.0

• Due to a change in the CLI syntax introduced in the 2.1.0 firmware PCM is unable to scan the configurations of AP420s running this software. This issue will also cause the Software Update tool to be unable to successfully update the firmware on these devices. Please contact ProCurve support for a release date.

If you use a 520wl device as the starting point for a managed subnet, Discovery will not work properly unless you manually add the subnet in PCM and then restart Discovery.

To process traps sent from 520wl devices, you must deselect the 'Unknown' option under the 'Select events to ignore' section in the global preferences for Events.

The 'syslog' feature in PCM does not work with 520wl devices.

The following are known issues when using PCM 2.0 with the ProCurve 9300 Series devices:

• Port based Policies cannot target a range of ports (e.g., A1-A5, etc.) on 93xx devices.
• LLDP-capable devices connected directly to 93xx devices may not map properly. Using a CDP-capable device between the 93xx and LLDP-capable device will resolve the problem.
• Support for use of SNMPv3 and SSH Keys on 9300 series devices is not included with PCM 2.0
  
• VLAN Management capabilities are not available on 94xx devices.

Installation

• PCM/PCM+ and Terminal Services are not supported on same server
• Un-installation of PCM 1.5 leaves the TLS_HPPNM process running.

• When adding subnets for Discovery in Global Preferences, there is no verification of IP addresses. If an invalid IP address is used, the subnet will not be discovered.
• If you have a very large network with many subnets, PCM performance can be improved by increasing the discovery interval, and in particular increasing the interval for VLAN discovery.
• If you have a large network, and are using a Configuration Scan Policy while Discovery and Traffic Monitor are running, it may cause the PCM management server to 'hang' or lose its connection to the 'Typhoon Server'. It is best to use a phased method (use separate policies at staggered intervals to scan by subnets or device group) rather than a single policy. Alternately, you can stop the Discovey and Traffic Monitor processes while the configuration scan is being done.

Network Maps

• Wide Area Network (WAN) links are not supported (displayed) in the PCM network maps.
• Path Trace does not work properly in a meshed environment.
• If the device Display Name preferences are changed, you need to refresh (redraw) the network map in order to view the new device display name.

Event Management

• Series 93xx traps are not automatically received by the PCM Event Browser. The Agent IP address that is embedded in the SNMP PDU is not the IP address from which the device was discovered. To resolve the problem, run the following command from the CLI:
  snmp-server trap source ve 200

• Traps will not be generated until modification of Thresholds in RMON Manager is done.
• Syslog trims entries after 1000 events based on severity, so it trims the events with the lowest severity first.
• Incomplete syslog message format for ProCurve 2800 and 5300 series devices

• If the speed of a connection being monitored in the Traffic Manager changes, errors will occur in the utilization rate display, due to the original configuration data cached in PCM. To correct the problem, delete the device from PCM, then use Manual Discovery to add the device back to PCM. This resets the traffic monitoring to match the current settings on the device.
• On a remote PCM client, the traffic graph and gauges flash correct data then quickly gray out. This occurs when the PCM remote client’s system clock is not synchronized with the server’s system clock. Make sure the PCM remote client’s system clock is synchronized with the server’s system clock (within 5 seconds).

VLAN Management

• When deleting VLANs directly on the switch, a full discovery cycle must run before changes are displayed in the PCM user interface.
• Creating VLANs in PCM may take several seconds to take effect on switch. To speed up the process, restart Discovery.
• If you create a VLAN directly on the switch (not using PCM), PCM will discover the VLAN at the next discovery cycle. To speed up the process, restart Discovery.

PCM Management Server Services

• PCM Management Server Services must be manually restarted if a network connection is lost.

Configuration Management

• Configuration labels cannot be removed until the configuration is deleted.
• When entering IP addresses for devices in PCM, they are not always verified. If an invalid IP address is entered for a device in PCM, then PCM will be unable to communicate with the device.
• Any change made to a device configuration file will cause the Configuration File change indicator to be displayed, even if it is only to the first line (date and version number) of the file and there are no actual configuation changes.
• When RADIUS authentication is configured on a switch, PCM does not support use of CLI. Only local password authentication and TACACS is supported. If you use RADIUS on a switch, the CLI commands, and CLI communications Parameters test will not work.

Network Settings

• The address used for SOCKs proxy in PCM Global Preferences for Network Settings are not validated. If an invalid SOCKS host or port is entered, devices will be unreachable in PCM.
• If the specified SOCKs server goes down, or it's network connection is lost, devices will be unreachable in PCM.

PCM-NNM synchronization component:

• Devices deleted in PCM will not be deleted from NNM.
• Changing subnets from Managed to Unmanaged on PCM will not be reflected in NNM.
• If SNMP community names are changed on NNM, the change will be not be reflected in PCM until after the next SNMP synchronization cycle. To speed up this process use the manual SNMP synchronization process.
  

© Copyright 2003, 2005 Hewlett-Packard Development Company, L.P.
All Rights Reserved.

This document contains information which is protected by copyright. Reproduction, adaptation, or translation without prior permission is prohibited, except as allowed under the copyright laws.

Publication Number

August, 2005
Edition 2.0-b

Part #5990-8848

Trademark Credits

Microsoft, Windows, Windows 95, and Microsoft Windows NT are registered trademarks of Microsoft Corporation. Internet Explorer is a trademark of Microsoft Corporation.
Ethernet is a registered trademark of Xerox Corporation.
Netscape is a registered trademark of Netscape Corporation.
Cisco® is a trademark of Cisco Systems, Inc.

PCM and PCM Plus uses two unmodified Open Source packages. The full source code and licenses to these packages can be found on the PCM distribution CD in the OpenSourcePackages directory. These packages are:
1) JDesktop Integration Components. http://javadesktop.org/articles/jdic/index.html
2) JRadiusClient. http://jradius-client.sourceforge.net/

The following applies to both of these packages:

This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.

This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.

You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

Disclaimer

The information contained in this document is subject to change without notice.

HEWLETT-PACKARD COMPANY MAKES NO WARRANTY OF ANY KIND WITH REGARD TO THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. Hewlett-Packard shall not be liable for errors contained herein or for incidental or consequential damages in connection with the furnishing, performance, or use of this material.

Hewlett-Packard assumes no responsibility for the use or reliability of its software on equipment that is not furnished by Hewlett-Packard.

Warranty

See the Customer Support/Warranty booklet included with the product.

A copy of the specific warranty terms applicable to your Hewlett-Packard products and replacement parts can be obtained from your HP Sales and Service Office or authorized dealer.
Hewlett-Packard Company
8000 Foothills Blvd, m/s 5551
Roseville, CA 95747-5551
http://www.procurve.com

HP ProCurve Stack Management allows you to configure a group of switches to appear as one unit with one IP address. This provides a single point of IP management for all switches in the stack. All members of the stack must be on the same subnet; the stack cannot cross a router. There is no limit on the number of stacks you can create. Stacking allows intermediate devices that do not support stacking, which enables you to include devices that are some distance from the Commander.

A stack can have up to 16 switches. One switch must be the Commander. the other switches in the stack are Members.

Stacking benefits include:

• Reducing the number of IP addresses required for managing your switches
• The ability to add switches to your network without performing IP addressing or using special cabling
• Simplifying network management for small groups

Quick Scroll To:

Devices Supported

HP ProCurve Stack Management operates on these HP ProCurve Switches:

• 8000M, 4000M, 1600M, 2424M and 2400M with software update C.08.XX or later installed
• 2512 and 2524 (Series 2500)
• 4104GL and 4108GL (Series 4100GL)

Virtual Stacking Terms

• Stack - A group of switches consisting of a single Commander and up to fifteen Members.
• Commander Switch - The Commander is the single point of IP management for a stack. It is used to configure all the Member switches in the stack. It always has a switch number of zero (0). A Commander can belong to only one stack.
• Candidate Switch - A Candidate switch is a switch that is not currently a member of a stack but available for membership in a stack. An IP address is optional. It must have a status of UP.
• Member Switch - Any switch that is a member of a stack. A switch can be a member of only one stack at a time.
• Transmission Interval - The Transmission Interval is a regular interval that controls how often a Discovery packet is transmitted. Discovery packets for a switch will be transmitted on all forwarding ports every x seconds, the default value being 60 seconds. Initially, discovery packets are transmitted out all ports; after that any discovery packets received are transmitted only on ports from which other discovery packets have been recently received.
• Discovery Packet - A Discovery packet contains information about:
  • Device Type
  • Operating System
  • System name of the switch
  • Neighboring switch
  • Ingress and egress ports
• Stack Name - The Stack Name is any name you assign to this stack, and can include up to 15 characters with no spaces. You must assign a Stack Name.
• Auto Grab (Commander only) - When Auto Grab is set to On, the Commander will add to the stack any Candidate switches that have Auto Join set to On if the Candidate does not have a password. Auto Grab is disabled in the default configuration.
• Auto Join (Candidate only) - When a Candidate switch has Auto Join set to On (the default), it will be added automatically to the stack whose Commander discovers it first and has Auto Grab set to On. The password for the switch has to be disabled. Once a Candidate is added to a stack, either automatically or manually, it becomes a member of that stack and Auto Join is turned off. If Auto Join is set toOff, you can manually add the switch to the stack. See Adding a Member or Candidate to a Stack.
• Enabled - When a Candidate switch comes up it announces (using the Discovery Protocol) that it is available for membership in a stack. This is the default operation.
• Disabled - A switch is not available for membership in a stack. When a Commander switch sets stacking to disabled (or becomes a Candidate), and the Member switches are UP so that they can receive communications from the Commander, the Member switches become Candidates automatically. If they are not in communication with the Commander, they don't know they are not members of that stack anymore, and a conflict is displayed in the show all view of the Stack Management window.

Stacking When VLANs Exist

Each switch in the stack (Commander and Members) uses the Primary VLAN configured in that switch for stack links. In the factory-default configuration, the default VLAN is the Primary VLAN. All switches in the stack need to be interconnected through the Primary VLAN.

The VLAN ID for the Primary VLAN must be the same for all switches in the stack. The Primary VLAN name can differ among switches in the stack. For example, if the VLAN ID (VID) for the Primary VLAN in Switch A is '1' and the VID for the Primary VLAN in Commander Switch B is '5', then Switch A can't be a Candidate or a Member of Commander Switch B's stack, even though you may have connectivity to Switch A from Commander Switch B through other tagged VLANs.

If the Primary VLAN is untagged, then all switches that have the Primary VLAN untagged on a port providing a stacking link are eligible to be in the stack (if stacking is enabled on such switches).

SNMP Communities in Stacks

When a Candidate switch becomes a member of a stack, it automatically becomes a member of any SNMP community to which the Commander belongs. However, if a Member has its own IP addressing, it can belong to SNMP communities to which other switches in the stack (including the Commander) do not belong.

How Passwords are Used with Stacking

If the Candidate switch has a password, you must know the password of the Candidate switch in order to add it to the stack. After a Candidate switch is added to a stack, it uses the password of the Commander switch from that point forward. If there is no Manager password for the stack, the new Member will not have a Manager password either, even if it had a Manager's password before it became a member of the stack. If you change the Commander's passwords, the new passwords are propagated to all Members of the stack.

If you move a Member switch from one stack to another, the Member will acquire the password of the new Commander switch. You must know the password of the current Commander in order to move the Member from one stack to another stack.

Operator and Manager passwords are the same for the entire stack. You cannot change the password on a Member switch.

If you want to move more than one switch in an operation, the switches must have the same password. If not, you need to move them separately, because you will be prompted for the password.

If you remove a member from a stack and do not add it to another stack, the member retains the password it acquired while a stack member.

The Commander

Characteristics

• The Commander is the only device in the virtual stack that must have an IP address if you want to access it through the network.
• The Commander must be on the same subnet as its Members, but it does not need to be directly connected to its Member switches.
• There can only be one Commander per stack.
• The Commander operates as an SNMP gateway to Members for all SNMP communities configured in the Commander.

Creating a Commander

Note: It is recommended that you leave Auto Grab set to Off if you plan to have more than one stack in a subnet. This prevents a Commander switch from automatically adding Candidate switches with Auto Join set to On. You can add the Candidate switches to the stack manually using the Commander's Stack Management window.

1. Assign an IP address to the switch that will be a Commander.
2. Select the Configuration tab in the Status view of the device.
3. Click on the Stacking button. The Stacking page appears.
4. In the Stacking drop-down list, select Commander.
5. Change the Transmission Interval if you want a shorter or longer interval between discovery packets being sent. It is recommended that you keep the default setting.
6. You must assign a name to the new stack in the Stack Name field.
7. Set Auto Grab to On if you want the Commander to add any new discovered devices to the stack automatically. The devices to be added as Members must have Auto Join set to On.
8. Click on Apply Changes to save your settings. The stack is now created.
9. If you do not want to save your changes, click on Reset Changes.

Moving a Member from One Stack to Another

1. From the Web page of the Commander for the stack to which you want to add the Member, click on the Stack Management button. The Stack Management page appears.
2. Click on the show all radio button to display all Members of all stacks.
3. Select the Member from the list in the bottom window, then click Add to Stack.
4. If you are prompted for a password, enter the password of the former Commander.
5. The Member will have a new Commander and be a Member that Commander's stack.

Removing a Commander from a Stack

When you remove a Commander, the Members of that stack become Candidates for membership in another stack. The Commander password acquired by the Members is not removed when the Commander leaves the stack.

To remove a Commander switch from a stack:

1. Select the Configuration tab in the Status view of the Commander switch.
2. Click on the Stacking button. The Stacking page appears.
3. In the Stacking drop-down list, select Disable, Member, or Candidate. This replaces the Commander designation for that switch.
4. Click on Apply Changes.

Hp Procurve Discovery Tool

The Members

Characteristics

• Members (or Candidates) do not require an IP address, but they can have one.
• There are up to fifteen Members in a stack.
• All Members must be in the same subnet.
• Members are accessed from the Commander.
• Members assume the Commander's password.
• Members do not have to be directly connected to each other with specialized cabling.
• When a Candidate becomes a Member, its Auto Join parameter is automatically set to Off. The default setting for a Candidate is On.
• The switch number (SN) assigned to the new Member is the lowest unassigned number between 1 and 15.
• Member switches provide a way to configure a new stack Commander if the main Commander fails.
• Members belong to the same SNMP community as the Commander. To join a community that excludes the Commander, the Member must have its own IP address. Removal from the stack results in loss of membership in any community configured only for the Commander.

Adding a Candidate or Moving a Member Between Stacks

Note: Both the Commander and the Candidate or Member switch must be UP for the Add to succeed.

Method A: From the Stack Commander

Note: You can Ctrl-click to add multiple switches to a stack if they have the same password. If you select multiple switches with different passwords and try to add them, the switches that are not added will appear in a dialog box. You can then try to add them using their password(s). Members will take the password of the Commander after they are added to the stack.

1. From the Web page of the Commander, click on the Stack Management button (above the web page tabs). The Stack Management page appears.
2. Click on the show candidates radio button to display the Candidate switches (switches not currently members of a stack), or click on the show all radio button to display all Candidates and Members of all stacks.
3. Click on the Candidate or Member switch in the bottom list that you want to add to the stack, then click on the Add to Stack button. You will be prompted for the password of the switch. Enter the password if it exists.
4. The Candidate or Member switch becomes a member of the stack for the selected Commander.

Method B: From a Candidate

1. Select the Configuration tab in the Status view of the device.
2. Click on the Stacking button. The Stacking page appears.
3. Select Member from the Stacking drop-down list.
4. Change the Transmission Interval if you want a shorter or longer interval between discovery packets being sent.
5. Choose a Commander for this Member in the Select Command Switch drop-down list.
6. Click on Apply Changes to save your settings.
7. If you do not want to save your changes, click on Reset Changes.

Removing a Member from a Stack

The Member switch must be UP when it is removed from a stack. This permits handshaking to occur between the Member and the Commander, and the Member automatically becomes a Candidate. If the Member is not UP when removed from the stack, the Commander cannot tell the Member that it is no longer a member of that stack. The Member switch still thinks it is part of the stack. The result is that the Member has a conflicting configuration.

To resolve a conflict with a Member switch, select the Stacking button for the Member and choose Candidate from the Stacking drop-down list.

When you remove a Member from the stack, its switch number (SN) becomes available for use by another switch that is added to the stack. Auto Join is set to No.

Method A: From the Stack Commander

1. From the Web page of the Commander, click on the Stack Management button. The Stack Management page appears.
2. Click on the Member switch that you want to remove from this stack.
3. Click on the Remove from Stack button. The Member is removed from this stack.

Method B: From a Member

1. Select the Configuration tab in the Status view of the device.
2. Click on the Stacking button. The Stacking page appears.
3. In the Stacking page select Candidate from the Stacking drop-down list.
4. The switch returns to being a Candidate.

The Stack Management Page

Stack management allows Commander switches to coordinate configuration information with other switches in the stack and maintain information about stack membership.

To access the Stack Management page,

Click on the Stack Management button at the top of the Commander page. The Stack Management page appears.

Information displayed:

• Switch Number (SN) - A number between one and fifteen, inclusive, that denotes a member of a stack. The Commander is always number zero (0).
• MAC Address - Physical address of the switch
• System Name - The name you have given the stack will appear here. The Member switches will display the system name appended with the Switch Number.
• Device Type - for example, HP 8000M.
• Status -Displays whether the switch is a Commander or Member, and if the switch is Up or Down.

The Stack Closeup Page

The Stack Closeup page displays the Closeup views for all the switches in the stack.

To access the Stack Closeup page:

• Click on the Stack Closeup button in the Commander page. The Closeup views of the devices in the stack appear.

The following information is displayed with each device:

• Switch number
• MAC Address
• System Name
• Device Type

Enabling Ports

1. Click on the ports you wish to enable, or click on the Select All Ports button if you want all ports enabled.
2. Click on the Enable Ports button.

Disabling Ports

1. Click on the ports you wish to disable, or click on the Select All Ports button if you want all ports disabled.
2. Click on the Disable Ports button.

The Stack Access Page

Hp Procurve Switches

The Stack Access page allows Web access to all Members of a stack.

To access the Stack Access page:

• Select a device from the Stack Access drop-down list at the top of the Commander page. The Stack Access page for the device appears.

Hp Procurve Switch Configuration Guide

The Stack Access page allows you to:

• Enable stacking for a switch
• Disable stacking for a switch.
• Change the Transmission Interval
• Change the Command Switch for a Member switch
• Edit the Stack Name (SN) for a stack
• Turn Auto Grab on or off for a Command switch
• Turn Auto Join on or off for a Member switch.

Back to Contents

Copyright © 2001-2002 by Hewlett-Packard Company