airajena commented on PR #5566: URL: https://github.com/apache/fineract/pull/5566#issuecomment-4013287818
> This feature is Ok to be turned off for this upgrade. For production, this variable turned off (which is the default setting in previous MariaDb versions) will introduce data I consistency. In Apache Fineract it has optimistic locking code and it must be upgraded. I think that a follow up Jira Ticket should be raised about turning on this variable that will help to fix the existing locking in Apache Fineract. Regards Victor El vie., 6 de marzo de 2026 12:05 p. m., Aira Jena ***@***.***> escribió: > […](#) > *airajena* left a comment ([apache/fineract#5566](https://github.com/apache/fineract/pull/5566)) <[#5566 (comment)](https://github.com/apache/fineract/pull/5566#issuecomment-4013201921)> Woah, fascinating. Would we then be using innodb_snapshot_isolation=OFF during tests and also recommending it for prod/deployments as well? If this is a change we'll be making everywhere, please write the dev list and see if anyone has input/feedback/advice. I don't have experience with this database setting. Actually for this PR, I treated it as a compatibility setting for the MariaDB 12 upgrade path, not as a claim that Fineract should globally prefer. I agree this deserves wider input if we intend to keep applying it outside CI/tests, I will explain this in detail in mailing list, let's see if we get a better way. There are really two possible directions. 1. Keep innodb_snapshot_isolation=OFF for MariaDB 12 so runtime behavior stays aligned with what Fineract has already been validated agains t. 2. Keep MariaDB 12 defaults and revisit the affected Fineract transaction/locking flows so they are explicitly correct under the new upgrade. — Reply to this email directly, view it on GitHub <[#5566 (comment)](https://github.com/apache/fineract/pull/5566#issuecomment-4013201921)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALD2ZAQDLKJ3H2KRIHWGAC34PMHQPAVCNFSM6AAAAACWDIBVR6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DAMJTGIYDCOJSGE> . You are receiving this because you were mentioned.Message ID: ***@***.***> @meonkeys @IOhacker hii, so should I open a discussion on mailing list or leave, now that we already have a direction to proceed -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
