[
https://issues.apache.org/jira/browse/JCR-314?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12488896
]
Oliver Zeigermann commented on JCR-314:
---------------------------------------
Any idea how this could be achieved?
If you use blocking locks, it is very hard to avoid deadlocks. Of course you
can use an ordered sequence of locks for all requests to avoid deadlocks, but
that would mean you need a list of all resources to be locked before you even
start doing anything.
After my experience with the Slide server I can only recommend non blocking
locks. Such locks would make a request fail if it is not able to acquire all
locks necessay instead of letting it wait. This would require a notion of a
transaction, however.
> Fine grained locking in SharedItemStateManager
> ----------------------------------------------
>
> Key: JCR-314
> URL: https://issues.apache.org/jira/browse/JCR-314
> Project: Jackrabbit
> Issue Type: Improvement
> Components: core
> Affects Versions: 0.9, 1.0, 1.0.1, 1.1, 1.1.1, 1.2.1, 1.2.2, 1.2.3
> Reporter: Marcel Reutegger
>
> The SharedItemStateManager (SISM) currently uses a simple read-write lock to
> ensure data consistency. Store operations to the PersistenceManager (PM) are
> effectively serialized.
> We should think about more sophisticated locking to allow concurrent writes
> on the PM.
> One possible approach:
> If a transaction is currently storing data in a PM a second transaction may
> check if the set of changes does not intersect with the first transaction. If
> that is the case it can safely store its data in the PM.
> This fine grained locking must also be respected when reading from the SISM.
> A read request for an item that is currently being stored must be blocked
> until the store is finished.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
