Dell PowerEdge 4200 Series, Administrator'S Manual, Page: 174 / 440

174 | SRA 6.0 Administrator’s Guide
Note
It is important to ensure that the same member receives all cookies to keep the user
authenticated. However, for improved performance in certain situations, all backend
members may be able to accept the session cookies of all users. In this case, the
administrator may decide to turn off Session persistence. The Load Balancer will
then strictly adhere to the LB method and LB factors in distributing the load.
Step 7
To add a new member to the group, see “Adding New Members to a Load Balancing Group” on
page 175.
Configuring Probe Settings
To configure probe settings for this load balancing group in the Probe Settings section of the
Portals > Load Balancing screen:
Step 1
Select a Probe Method from the drop-down list. Options include:
•
HTTP/HTTPS GET – The Load Balancer sends a HTTP(S) GET request periodically (based
on the configured Probe interval) to see if the HTTP response status code is not greater
than or equal to 500 to ensure there are no Web server errors. This is the most reliable
method to determine if a Web server is alive. This method ignores SSL Certificate warnings
while probing.
•
TCP Connect – The Load Balancer completes a 3-way TCP handshake periodically to
monitor the health of a backend node.
•
ICMP Ping – The Load Balancer sends a simple ICMP Ping request to monitor if a backend
node is alive.
Step 2
In the Deactivate Member after field, enter the number of missed intervals required to fail the
node. The default value is 2.
Step 3
In the Reactivate Member after field, enter the number of successful intervals required to
reinstate the node as functional. The default value is 2.
Step 4
In the Display error page when there is no resource available to fail over text box, enter a
custom message or Web page to display in the event that all of the configured backend nodes
have failed. HTML formatting is allowed in this field.