|
|
This document includes the following sections:
The DistributedDirector Enhancements for Cisco IOS Release 11.1(28)IA feature consists of the following three new features for the Cisco DistributedDirector:
Enhanced Fault Tolerance with Multiple Resource Records
Prior to this enhancement, DistributedDirector would return a single Resource Record (RR) in each DNS response. This was normally sufficient, but for some applications, server failover will occur more rapidly when applications are provided IP addresses of multiple servers.
The Enhanced Fault Tolerance with Multiple Resource Records feature enables DistributedDirector to return multiple RR. The number of RR returned in a single reply is configurable. The default number of RR returned is one.
Critical Event Recording with Syslog
The Critical Event Recording with Syslog feature enables DistributedDirector to log critical events by way of the the industry-standard syslog system. Server state is logged, providing a useful log of when servers are considered up or down. Additionally, the server selection process may be logged. In both cases, the logging priority level is "informational."
Enhanced Server Verification with Multiple Port Connect Test
Prior to this enhancement, DistributedDirector could evaluate server status by performing a TCP connect test to a single port. The Enhanced Server Verification with Multiple Port Connect Test feature allows multiple connect ports to be specified. If any one of the connect tests fail, the server is considered down.
The new features provided in Cisco IOS Release 11.1(28)IA help make networks using DistributedDirector more robust. These features ensure that applications have more useful information, perform better server verification and allow administrators to better track DistributedDirector. In particular:
Enhanced Fault Tolerance with Multiple Resource Records
Configuring DistributedDirector to return a large number of records could reduce the benefit of using DistributedDirector to select the best server.
Critical Event Recording with Syslog
Extensive syslog output is provided when logging server selection. Therefore, this feature should not be used when a heavy request load is expected.
For more information on the Cisco DistributedDirector, see the following documents, which are located on CCO at http://www.cisco.com/univercd/cc/td/doc/product/iaabu/distrdir/index.htm:
Standards
None
MIBs
None
For descriptions of supported MIBs and how to use MIBs, see the Cisco MIB web site on CCO at http://www.cisco.com/public/sw-center/netmgmt/cmtk/mibs.shtml.
RFCs
None
See the following sections for configuration tasks for the DistributedDirector Enhancements for Cisco IOS Release 11.1(28)IA feature. Each task in the list indicates if the task is optional or required.
To configure Enhanced Fault Tolerance with Multiple Resource Records on the DistributedDirector for a host name, use the following command in global configuration mode:
| Command | Purpose |
|---|---|
Router(config)# ip director host hostname [a | mx] multiple integer
| Specifies how many RR the DistributedDirector will return for each DNS response. |
To configure Critical Event Recording with Syslog on the DistributedDirector for a host name, use the following commands in global configuration mode:
| Command | Purpose |
|---|---|
Router(config)# ip director host hostname [a | mx]
logging
| Configures the DistributedDirector to send to syslog the DNS request and response information, and the server selection process of this host. The logging priority level is "informational." |
To configure Enhanced Server Verification with Multiple Port Connect Tests on the DistributedDirector, use the following commands in global configuration mode:
| Command | Purpose |
|---|---|
Router(config)# ip director host hostname [a | mx]
connect port-1 [interval] connection-interval
Router(config)# ip director host hostname [a | mx]
connect port-2 [interval] connection-interval ...
Router(config)# ip director host hostname [a | mx]
connect port-N [interval] connection-interval
| Specifies how often the DistributedDirector will verify that the remote servers are still active. When you configure multiple ip director host hostname connect commands for the same hostname but with different port numbers, the DistributedDirector will verify that all of the ports are accessible. If any of the ports are not accessible, the host will be considered down. |
To enable the DistributedDirector to verify that a server is available, use the ip director host connect global configuration command. The DistributedDirector redirects clients only to servers that respond. To turn off connection parameters, use the no form of this command.
ip director host hostname [a | mx] connect port [interval] connection-interval
Syntax Description
hostname The name of the host that maps to one or more IP addresses. Do not use an IP address. a (Optional) Indicates the configuration is for processing DNS address (A) queries for the specified host name. If no query type is specified, a is set by default. mx (Optional) Indicates the configuration is for processing Mail eXchange (MX) queries for the specified host name. port The port number to which the distributed servers are configured. interval (Optional) Configures the connection interval to be a time in seconds instead of minutes. connection-interval The time in minutes (or seconds, if the interval keyword is used) that elapses between availability checks.
Defaults
No connection parameter is set.
The default query type is a.
Command Modes
Global configuration
Command History
11.1(1)IA This command was introduced. 11.1(25)IA and 12.0(3)T The a and mx keywords were added. 11.1(28)IA and 12.1(3)T Enhanced Server Verification with Multiple Port Connect Test functionality was added.
Release
Modification
Usage Guidelines
When this parameter is configured, the DistributedDirector will attempt to create a TCP connection to each of the distributed servers on a configured port (for example, port 80 for HTTP servers) over the configured time interval. Servers that yield unsuccessful TCP connection attempts will be marked as unavailable. Following a failed TCP connection, the DistributedDirector uses a linear backoff algorithm to create subsequent TCP connections to the server to determine when it is again available. This algorithm is used to smoothly handle changes in server or network availability.
The initial connection trial to a server that is labeled as "up" is done three times in rapid succession. If no connection is successful, the percentage confidence that the server is down is set to 10 percent. The retry interval is calculated as the configured interval multiplied by the confidence percentage with a minimum of 1 minute. Each successive connection attempt is done once, and each time the attempt is unsuccessful the confidence percentage is incremented by 10 percent until it reaches 100 percent.
 |
Note Although TCP connection state information may take up to 4 minutes to be cleared, TCP connection timeouts usually occur within about 30 seconds. As a result, the minimum configurable TCP connection time interval on the DistributedDirector is 1 minute. The minimum retry interval in the linear-backoff algorithm of the DistributedDirector is also 1 minute. |
Beginning with Cisco IOS Release 11.1(28)IA, the ip director host hostname connect command supports the Enhanced Server Verification with Multiple Port Connect Test feature. When you enter several ip director host hostname connect commands to the same host name but with different port numbers, the DistributedDirector verifies that all of the ports are accessible. The DistributedDirector will only consider the server accessible if all of the ports are accessible.
Examples
The following example sets the connect interval to 5 minutes to the distributed servers on port 80 and port 90. The distributed servers will only be considered accessible if both port 80 and port 90 are accessible:
ip director host www.sleet.com connect 80 5
ip director host www.sleet.com connect 90 5
Related Commands
Configures the order in which the DistributedDirector considers metrics when picking a server.
Command
Description
To configure the DistributedDirectorto log critical events to syslog, use the ip director host logging global configuration command. To turn off metric priorities, use the no form of this command.
ip director host [a | mx] hostname logging
Syntax Description
hostname The name of the host that maps to one or more IP addresses. Do not use an IP address. a (Optional) Indicates the configuration is for processing DNS address (A) queries for the specified host name. If no query type is specified, a is set by default. mx (Optional) Indicates the configuration is for processing Mail eXchange (MX) queries for the specified host name. logging Specifies to syslog the DNS request and response, and the server selection process of this host. The logging priority level is "informational."
Defaults
Disabled
Command Modes
Global configuration
Command History
11.1(28)IA and 12.1(3)T This command was introduced.
Release
Modification
Usage Guidelines
The Critical Event Recording with Syslog feature gives the ability to examine DNS traffic and how servers are chosen. Server state is logged, providing a useful log of when servers are considered up or down. Additionally, the server selection process may be logged. In both cases, the logging priority level is "informational."
 |
Caution Extensive syslog output is provided when logging server selection. Therefore, this feature should not be used when a heavy request load is expected. |
Examples
The following example configures the DistributedDirector to log critical events for the host www.sleet.com:
ip director host www.sleet.com logging
To configure the number of Resource Records (RR) that the DistributedDirector will return for each DNS response, use the ip director host multiple global configuration command. To configure the DistributedDirector to only return the best RR for each DNS response, use the no form of this command.
ip director host [a | mx] hostname multiple integer
Syntax Description
hostname The name of the host that maps to one or more IP addresses. Do not use an IP address. a (Optional) Indicates the configuration is for processing DNS address (A) queries for the specified host name. If no query type is specified, a is set by default. mx (Optional) Indicates the configuration is for processing Mail eXchange (MX) queries for the specified host name. multiple Returns multiple A or MX servers, which are sorted according to the specified priorities of metrics. integer Number of servers that are returned.b
Defaults
Disabled
Command Modes
Global configuration
Command History
11.1(28)IA and 12.1(3)T This command was introduced.
Release
Modification
Usage Guidelines
Configuring DistributedDirector to return a large number of records could reduce the benefit of using DistributedDirector to select the best server.
Examples
The following example configures the DistributedDirector to return the 2 best servers:
ip director host www.sleet.com multiple 2
Posted: Wed May 9 11:01:51 PDT 2001
All contents are Copyright © 1992--2001 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.