Question
Citigroup technology inc
US
Last activity: 4 Jan 2019 10:18 EST
How to open and save non-versioned rule instance across different nodes and servers
We have 22 nodes spanned across 7 servers. I wanted to know how to open a non versioned rule instance(data instance) from one server1 node1 and perform update and save it across all other remaining 21 nodes that is spanned across other 6 servers.
Need a solution or help/support. Is there any api to do it.
***Edited by Moderator Marissa to update platform capability tags****
Knowledge Expert
FR
You should be able to do it by regular saving instance on one node. Then your changes will be reflected on other nodes automatically. It doesn't matter if your rule/data versioned or not. From documentation (https://community.pega.com/sites/default/files/help_v717/concepts/concepts2/conceptscaching.htm):
In a multinode cluster, a rule recently updated by a user on one node can also be present (and stale) in the rule cache of another node. During system pulse processing, the Pega-RULES agent on each node clears such stale rules from the rule cache.
Tata consultancy services limited
US
Yeah thanks for clarification, but although system pulse running and the change done in one node is getting reflected in other 21 nodes. The problem is during the runtime the users are not seeing the updates. So pega suggested to save the instance on all 22 nodes again. So we kind of looking for feasibility of how to open a non versioned rule instance from of other nodes from single node and save it again. We are trying to build an utility. So we need to help how to achieve it.
Knowledge Expert
FR
So this is definitely a workaround suggested by Pega, not a right way of resolving the issue.
If you want to follow this workaround - then one of the ways can be - expose service with pzInsKey or other key as input parameter. Make the activity behind the service for resave rule/instance by key. Then call that service on each node (you can write another activity in Pega or just use external software, like Postman for calling).
But again - such issue should never happen on Pega cluster, otherwise - this is Pega bug.