[
https://issues.apache.org/jira/browse/OAK-7798?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Julian Reschke reopened OAK-7798:
---------------------------------
Getting the following test failure on Windows:
{noformat}
[ERROR] Tests run: 14, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 2.082
s <<< FAILURE! - in org.apache.jackrabbit.oak.plugins.blob.CachingDataStoreTest
[ERROR]
loadCacheErrorDirectTemp(org.apache.jackrabbit.oak.plugins.blob.CachingDataStoreTest)
Time elapsed: 0.053 s <<< FAILURE!
java.lang.AssertionError: expected:<1> but was:<0>
at
org.apache.jackrabbit.oak.plugins.blob.CachingDataStoreTest.loadDirectBackendTemp(CachingDataStoreTest.java:192)
at
org.apache.jackrabbit.oak.plugins.blob.CachingDataStoreTest.loadCacheErrorDirectTemp(CachingDataStoreTest.java:142)
{noformat}
Note that whether this happens might depend on the parent path; I can reproduce
this with a path of:
"/projects/apache/jackrabbit/jackrabbit-dev/target/jackrabbit-oak-1.9.9/svn/jackrabbit-oak-1.9.9"
> Return stream for downloaded files rather than directly from backend
> ---------------------------------------------------------------------
>
> Key: OAK-7798
> URL: https://issues.apache.org/jira/browse/OAK-7798
> Project: Jackrabbit Oak
> Issue Type: Bug
> Components: blob-plugins
> Reporter: Amit Jain
> Assignee: Amit Jain
> Priority: Major
> Labels: candidate_oak_1_6, candidate_oak_1_8
> Fix For: 1.10, 1.9.9
>
>
> When the \{{AbstractSharedCachingDataStore#getRecordIfStored}} is called and
> if the cache load returns an error from backend then the stream is directly
> retrieved from backend and returned. This stream should first be downloaded
> to the temp dir configured and then stream returned over that. If the
> consuming applications don't close the stream then there's chance of
> connection pool to the backend getting starved.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)