curator-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nick Hill (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CURATOR-294) Optimize TreeCache, fix possible concurrency issue
Date Fri, 29 Jan 2016 18:16:39 GMT

    [ https://issues.apache.org/jira/browse/CURATOR-294?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15123902#comment-15123902
] 

Nick Hill commented on CURATOR-294:
-----------------------------------

My other comment on the changes in the PR didn't show up here for some reason:

There's one public behaviour change I notice - a ChildData from a TreeCache might now not
be equal() to one from a PathChildrenCache where it was before. But that seems like a somewhat
obscure scenario.
Looking at this gave me another idea though:
Instead of having a private subclass, PathChildrenCache could use the new ChildData directly,
with the clearDataBytes() method changed to do a currentData.replace(data, new ChildData(data.getPath(),
data.getStat(), null).
I think this would mean only a one-line change to PCC is needed and seems generally simpler.
It would have the side effect of not clearing the data from ChildData objects that had been
previously obtained from the cache - but I think this is probably a good thing too?
WDYT?

> Optimize TreeCache, fix possible concurrency issue
> --------------------------------------------------
>
>                 Key: CURATOR-294
>                 URL: https://issues.apache.org/jira/browse/CURATOR-294
>             Project: Apache Curator
>          Issue Type: Improvement
>          Components: Recipes
>            Reporter: Nick Hill
>            Assignee: Scott Blum
>             Fix For: 3.0.1, 2.9.2
>
>
> Hi, I have been looking at the TreeCache impl and have some questions.
> It doesn't look right to me that there's separate atomic refs for a node's data and stat.
It seems the stat in a ChildData object obtained from getCurrentData() might not correspond
to the data that it's with. This could be problematic when doing conditional state changes
given assumptions about that data.
> An obvious and simple solution to this would be to have a single AtomicReference<ChildData>
field instead, which would have the additional significant benefit of eliminating ChildData
obj creation on every cache access. PathChildrenCache works this way, but my understanding
was that TreeCache is intended to be a (more flexible) replacement.
> Furthermore I'd propose that the data field of ChildData be just a final byte[] instead
of an AtomicReference. This would avoid needing two volatile reads to get to the data, and
mean that "sharing" these (per above) is a bit safer. The ChildData byte[] AtomicReference
is only used by PathChildrenCache.clearDataBytes() (not currently used by TreeCache at all),
and that capability could be easily maintained by having PathChildrenCache use it's own simple
subclass of ChildData containing the atomic reference.
> If similar capability were to be added to TreeCache, I'd suggest it would be better to
just replace the node's ChildData object with a copy that has the byte[] field nulled out
(but same stat ref).
> I'm fairly new to the code so apologies if there's something I've missed/misunderstood!
But if there is agreement, I'd also be happy to prepare a PR.
> Regards,
> Nick



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message