Repository navigation
JPA Storage Ignoring Cache Configuration / 20.0.3 #16809
Unanswered
AKermodeBear
asked this question in
Q&A
Replies: 1 comment 1 reply
|
Did you have a solution for it @AKermodeBear? |
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
- 8000 li>
Uh oh!
There was an error while loading. Please reload this page.
Hoping someone can help me out; it may be a misunderstanding from the documentation I've been able to find.
I currently have a HA cluster running on top of MariaDB/Remote Infinispan, but I would like to replace MariaDB with CRDB. The problem I'm running into is this:
JPA needs to be enabled for Keycloak to properly talk to CockroachDB. However, when --storage=jpa is set, it seems to completely ignore cache and cache-config-file configurations. Even when configuring Keycloak to look for a cache configuration file that doesn't exist, it starts right up. So there seems to be no attempt at all to interact with the cache configuration file.
If I attempt to start Keycloak without --storage=jpa, then Keycloak seems to read the cache configuration file (and it properly complains if it isn't there / invalid, so I know it's at least being looked at), but it then attempts to execute some updates via Liquibase and it fails.
So I am wondering at this point if enabling JPA is intended to completely replace the Infinispan cache layer, I'm missing part of the documentation, or there's some non-obvious configuration issue that I've run into.
I did notice that there's the ability to configure hotrod storage, and then section off various parts of Keycloak's data to either hotrod or the database, but that isn't the caching layer.
I've looked at the Cache documentation, a few scattered articles on Keycloak and CRDB, the RedHat 3-part series on setting something like this up, etc., but it seems I've missed something along the way.
Any help would be appreciated. Thanks!
All reactions