[
https://issues.apache.org/jira/browse/MAPREDUCE-1568?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Scott Chen updated MAPREDUCE-1568:
----------------------------------
Attachment: MAPREDUCE-1568-v3.txt
Made the change based on Amareshwari's suggestions.
{quote}
Shall we rename the class BaseDir to something like BaseDirManager?
{quote}
That's a very good name. I have changed it.
{quote}
cacheStatus.baseDir is being accessed under global lock in
BaseDir.checkAndCleanup() method, every-where else it is accessed under
CacheStatus lock. I think it is no harm. Can you also check this once and add a
comment?
{quote}
I have made baseDir immutable
{quote}
Though not related to this issue, can you add a comment for
CacheStatus.refCount saying it should be always accessed under the global
cachedArchives lock.
{quote}
I have add comments on all fields in CacheStatus to explain how to access them.
{quote}
catch(InterruptedException) block in CleanupThread can do a return instead of
ignoring the exception.
I think cleanup thread should catch all Exceptions (not just IOException)
because thread should never crash, similar to CleanupQueue.
{quote}
Now it catches all exceptions, logs them and keeps running.
{quote}
Test case can call baseDir.checkAndCleanup() inline to avoid timing issues,
what do you think?
{quote}
I feel that it is better to test the whole CleanupThread. That way we can make
sure the thread actually does the right thing.
> TrackerDistributedCacheManager should clean up cache in a background thread
> ---------------------------------------------------------------------------
>
> Key: MAPREDUCE-1568
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1568
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Affects Versions: 0.22.0
> Reporter: Scott Chen
> Assignee: Scott Chen
> Fix For: 0.22.0
>
> Attachments: MAPREDUCE-1568-v2.1.txt, MAPREDUCE-1568-v2.txt,
> MAPREDUCE-1568-v3.txt, MAPREDUCE-1568.txt
>
>
> Right now the TrackerDistributedCacheManager do the clean up with the
> following code path:
> {code}
> TaskRunner.run() ->
> TrackerDistributedCacheManager.setup() ->
> TrackerDistributedCacheManager.getLocalCache() ->
> TrackerDistributedCacheManager.deleteCache()
> {code}
> The deletion of the cache files can take a long time and it should not be
> done by a task. We suggest that there should be a separate thread checking
> and clean up the cache files.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
