Hi Val, 

Yes, I think it's reasonable to handle "update_timestamp" before it goes
into Ignite cache instead.

As for "locking" however, we are relying on Hibernate's support for
optimistic locking -
https://docs.jboss.org/hibernate/orm/4.0/devguide/en-US/html/ch05.html

Basically when Hibernate saves an object, it will increment an integer on
it. For example, if you have 2 people trying to update an object at the same
time:
Person A reads myObj (locking = 0)
Person B reads myObj (locking = 0)
Person A updates and commits myObj (Hibernate verifies locking = 0, and then
increments it to 1)
Person B updates and commits myObj (Hibernate verifies locking = 1, it's
not, so this update is rolled back)




--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/CacheStore-implementation-using-Hibernate-tp1423p1452.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Reply via email to