How does Accelerated Passivation Timeout help quiescing
Need help in understanding bit more about quiescing. I have gone through almost all the documents that are available but still have the following questions.
In case of Slow drain quiescing, at first node to be quiesced is taken out of load balancer rotation. Now, as the quiesce process begins Pega platform looks for the accelerated passivation timeout settings. Say suppose, the accelerated passivation timeout is set to 5 seconds (Default).
1. Does this mean that after 5 seconds from start of quiescing all the existing requestor informations will be taken out of JVM memory and stored in the passivation store? If so, why the minimum value is 5 seconds why not less than 5 seconds?
2. What will happen to all the requests coming from those existing user sessions within 5 seconds considering session affinity?
3. If we consider that session affinity or stickiness is maintained by Cookie, who does the cookie invalidation, when and how so that the requests coming after 5 seconds i.e. after passivation can be served as the new request and can be routed to another node by the load balancer?
4. What are the advantages or disadvantages of maintaining a large value of accelerated passivation timeout?
Please correct me if I am wrong.
I have gone through the following link as well still not very clear:
Need help in understanding bit more about quiescing. I have gone through almost all the documents that are available but still have the following questions.
In case of Slow drain quiescing, at first node to be quiesced is taken out of load balancer rotation. Now, as the quiesce process begins Pega platform looks for the accelerated passivation timeout settings. Say suppose, the accelerated passivation timeout is set to 5 seconds (Default).
1. Does this mean that after 5 seconds from start of quiescing all the existing requestor informations will be taken out of JVM memory and stored in the passivation store? If so, why the minimum value is 5 seconds why not less than 5 seconds?
2. What will happen to all the requests coming from those existing user sessions within 5 seconds considering session affinity?
3. If we consider that session affinity or stickiness is maintained by Cookie, who does the cookie invalidation, when and how so that the requests coming after 5 seconds i.e. after passivation can be served as the new request and can be routed to another node by the load balancer?
4. What are the advantages or disadvantages of maintaining a large value of accelerated passivation timeout?
Please correct me if I am wrong.
I have gone through the following link as well still not very clear:
Thanks in advance for the information!!
***Edited by Moderator Marissa to update Platform Capability tags****