After you create a target group for your Gateway Load Balancer, you can edit its target group attributes.
Target failoverWith target failover, you specify how the Gateway Load Balancer handles existing traffic flows after a target becomes unhealthy or when the target is deregistered. By default, the Gateway Load Balancer continues to send existing flows to the same target, even if the target has failed or is deregistered. You can manage these flows by either rehashing them (rebalance) or leaving them at the default state (no_rebalance).
The Gateway Load Balancer continues to send existing flows to failed or drained targets. If the Gateway Load Balancer cannot reach the target, the traffic is dropped.
However, new flows are sent to healthy targets. This is the default behavior.
The Gateway Load Balancer rehashes existing flows and sends them to healthy targets after the deregistration delay timeout.
For deregistered targets, the minimum time to failover will depend on the deregistration delay. The target is not marked as deregistered until deregistration delay is completed.
For unhealthy targets, the minimum time to failover will depend on the target group health check configuration (interval times threshold). This is the minimum time before which a target is flagged as unhealthy. After this time, the Gateway Load Balancer can take several minutes due to additional propagation time and TCP retransmission backoff before it reroutes new flows to healthy targets.
Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.
On the navigation pane, under Load Balancing, choose Target Groups.
Choose the name of the target group to open its details page.
On the Group details page, in the Attributes section, choose Edit.
On the Edit attributes page, change the value of Target failover as needed.
Choose Save changes.
Use the modify-target-group-attributes command, with the following key value pairs:
Key=target_failover.on_deregistration and Value= no_rebalance (default) or rebalance
Key=target_failover.on_unhealthy and Value= no_rebalance (default) or rebalance
Both attributes (target_failover.on_deregistration and target_failover.on_unhealthy) must have the same value.
When you deregister a target, the Gateway Load Balancer manages flows to that target as follows:
The Gateway Load Balancer stops sending new flows.
The Gateway Load Balancer handles existing flows based on the protocol:
TCP: Existing flows are closed if they are idle for more than 350 seconds.
Other protocols: Existing flows are closed if they are idle for more than 120 seconds.
To help drain existing flows, you can enable flow rebalancing for your target group. For more information, see Target failover.
A deregistered target shows that it is draining until the timeout expires. After the deregistration delay timeout expires, the target transitions to an unused state.
Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.
On the navigation pane, under Load Balancing, choose Target Groups.
Choose the name of the target group to open its details page.
On the Group details page, in the Attributes section, choose Edit.
On the Edit attributes page, change the value of Deregistration delay as needed.
Choose Save changes.
Use the modify-target-group-attributes command.
Flow stickinessBy default, the Gateway Load Balancer maintains stickiness of flows to a specific target appliance using 5-tuple (for TCP/UDP flows). 5-tuple includes source IP, source port, destination IP, destination port, and transport protocol. You can use the stickiness type attribute to modify the default (5-tuple) and choose either 3-tuple (source IP, destination IP, and transport protocol) or 2-tuple (source IP and destination IP).
Flow stickiness considerationsFlow stickiness is configured and applied at the target group level, and it applies to all traffic that goes to the target group.
2-tuple and 3-tuple flow stickiness are not supported when AWS Transit Gateway appliance mode is turned on. To use appliance mode on your AWS Transit Gateway, use 5-tuple flow stickiness on your Gateway Load Balancer
Flow stickiness can lead to uneven distribution of connections and flows, which can impact the availability of the target. It is recommended that you terminate or drain all existing flows before modifying the stickiness type of the target group.
Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.
On the navigation pane, under Load Balancing, choose Target Groups.
Choose the name of the target group to open its details page.
On the Group details page, in the Attributes section, choose Edit.
On the Edit attributes page, change the value of Flow stickiness as needed.
Choose Save changes.
Use the modify-target-group-attributes command with the stickiness.enabled and stickiness.type target group attributes.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4