In upgrading, we would like to leave the old environment in a usable condition.
As for the database, we plan to continue using the one we are currently using.
What other options do we have?
The following options are possible
(1) Build a new environment and change the DB connection from the old environment to the new environment.
(2) Upgrade the currently running pega and restart the container to switch from the old environment to the new environment.
Am I correct in understanding that this method will not maintain the old environment because the DB is not consistent?
(iii) Upgrade and create a new one on the new worker node. Leave the old one on the old worker node.
Am I correct in understanding that this method cannot be used in this case because DB consistency cannot be maintained?
***Edited by Moderator Rupashree to add Capability tags***
Almost all is possible. in-place, out-place, just new rules schema, new rules & data schemas.
Yet, at a point in time, if you've got a new data schema, you'll need to stop your system and to sync your old data schema with your new data schema.
And then, let's say it worked fine, during your validation/testing time or your upgraded env, you should also avoid to process any real data because you kept your old env as a backup and if you want to back switch to previous environment (rollback) then both your data schema must remain aligned.
Managing the rule schema is pretty easy, well, let's say manageable. But, you really need to consider the consistency of your data in the data schema.
Based on this feasibility and complexity, you might revise your upgrade approach/plan.
I personally prefer to perform upgrade on the real system, it avoids changing internal references or setup.
This approach requires a proper backup plan and a well defined/controlled/tested restore procedure if needed