Posted: 29 Mar 2019 12:05 EDT Last activity: 4 Apr 2019 14:02 EDT
In Place Upgrade for 8.2
We are planning to upgrade our application which is built on Pega 7.3.1 to Pega 8.2.1. Our plan is to perform In place upgrade and while going through the upgrade document, it seems data schema upgrade part is missing (Please refer to the attachments below). Does in-place upgrade do not require running upgrade script for data schema separately? If you see content of out of place upgrade, it clearly mentions steps about data schema upgrade separately.
If you look into the below explanation which is provided by our SME about In-Place upgrade,
A “standard” upgrade is the traditional in-place upgrade process used for upgrades to 6.x and earlier, and a “recommended” upgrade is the new out-of-place upgrade process introduced in 7.1. Yes, in-place upgrades are simpler since tables do not have be migrated and there is no data schema upgrade step.
Thanks Mahesh for your reply but this now confuses me. As per upgrade guide, the only difference between in-place & out of place upgrade I could see is creating a separate rule schema which needs to be upgraded and then data schema needs to be upgraded via in-place manner. in both type of upgrade, we run the same script (upgrade.sh). so how come running the same during in place upgrade will take care of data schema upgrade automatically?
In the in-place upgrade you specify both the rules and data schema names in setDatabase.properties file and the upgrade.bat/sh will upgrade both together, that's the reason you don't need to perform dataOnly upgrade step again. If you read through the setDatabase.properties file and complete the top section(rules.schema.name and data.schema.name) and complete both if there is 1 rules and 1 data, if there is only 1 schema for both then use same name for both fields.
Thanks Mahesh! this is now clear. The last upgrade I did was out of place and during the rule schema upgrade i specified both rule schema & data schema name in property file and then upgraded data schema again separately with dataonly argument. I believe even in out of place upgrade data upgrade step is not required if we are specifying both rule & data schema name during rule schema upgrade. If my assumption is correct then i think it should be mentioned in the upgrade guide somewhere. Please correct me if I am wrong.
If you go through the "Performing an out-of-place upgrade with a single migration" section in your attached upgrade document you can upgrade only rules schema in step 4 by only specifying the rules schema related information and in the step 6 you will perform the upgrade of only data schema using --dataOnly option because the system needs to be down during the data schema upgrade that is the reason you perform it separately as i understand, but you can do it at a time also in the same step4.
If you read the step 1 in Upgrading the data schema section in page 39 its stated clearly as follows,
"If you have not already done so, configure the connection properties. Use your current data schema name for data.schema.name. Use the new rules schema name for rules.schema.name. If you have an optional customer data schema separate from the Pega data schema, enter the customerdata.schema.name."