[
https://issues.apache.org/jira/browse/OAK-3892?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15105334#comment-15105334
]
Julian Reschke commented on OAK-3892:
-------------------------------------
I wanted to make sure (on the source level) that we actually have the same Lock
implementations; this seemed to be the easiest way without relying on
out-of-band information.
> RDBDocumentStore: StripedNodeDocumentLocks - special case root?
> ---------------------------------------------------------------
>
> Key: OAK-3892
> URL: https://issues.apache.org/jira/browse/OAK-3892
> Project: Jackrabbit Oak
> Issue Type: Technical task
> Components: rdbmk
> Reporter: Julian Reschke
> Assignee: Julian Reschke
> Priority: Minor
> Attachments: stripes.diff
>
>
> {{StripedNodeDocumentLocks}} currently handles all document IDs the same way.
> I assume that we see a lot of updates on {{0:/}}, though. These updates would
> be blocked if another different ID fell into the same lock stripe (and was
> being updated).
> We currently have 4096, so the likelihood isn't that big; on the other hand,
> a change doesn't seem to be risky.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
