You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When using a multi-DC environment, all the TM nodes are configured in the GW nodes to consume events from any of the TM nodes.
[apim.throttling] ### Update URLs and credentials to connect TM nodes
username= "************"
password= ""
service_url = "https://apitmanager*****************:443/services/"
throttle_decision_endpoints = ["tcp://ip_address_1:5672","tcp://ip_address_2:5672","tcp://ip_address_3:5672","tcp://ip_address_4:5672","tcp://ip_address_5:5672","tcp://ip_address_6:5672"]
So the GWs are allowed to connect to any of the TM from any of the DCs over a failover of the connection. But once a failover happens, if the GW is connected to a TM not in the same DC as the GW, the GW server will not come back to the TM of the same DC even if it is available after a while(after the connection is healthy).
There is a requirement to prioritize the failover to the same DC and they are searching for the options to achieve this.
As per our understanding, OOB it is not possible to achieve this. Is it possible to improve the code base to achieve such behaviour?
Suggested Improvement
If the failover happens to a TM from a different DC, it is expected to connect to the same DC when the connection is available.
If the above 1st is not possible, is there any option that we can go through the URL from the 1st configured in the above list after the next failover is triggered? Since we are recommending configuring the 1st two URLs to point to the same DC, hope we can achieve the priority.
Version
3.2.0
The text was updated successfully, but these errors were encountered:
Current Limitation
Hi All,
When using a multi-DC environment, all the TM nodes are configured in the GW nodes to consume events from any of the TM nodes.
[apim.throttling] ### Update URLs and credentials to connect TM nodes
username= "************"
password= ""
service_url = "https://apitmanager*****************:443/services/"
throttle_decision_endpoints = ["tcp://ip_address_1:5672","tcp://ip_address_2:5672","tcp://ip_address_3:5672","tcp://ip_address_4:5672","tcp://ip_address_5:5672","tcp://ip_address_6:5672"]
So the GWs are allowed to connect to any of the TM from any of the DCs over a failover of the connection. But once a failover happens, if the GW is connected to a TM not in the same DC as the GW, the GW server will not come back to the TM of the same DC even if it is available after a while(after the connection is healthy).
There is a requirement to prioritize the failover to the same DC and they are searching for the options to achieve this.
As per our understanding, OOB it is not possible to achieve this. Is it possible to improve the code base to achieve such behaviour?
Suggested Improvement
Version
3.2.0
The text was updated successfully, but these errors were encountered: