[
https://issues.apache.org/jira/browse/GEODE-5587?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16581863#comment-16581863
]
ASF subversion and git services commented on GEODE-5587:
--------------------------------------------------------
Commit f0feda4c0032e5675a8d000571f57a9446888a54 in geode's branch
refs/heads/develop from [~jens.deppe]
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=f0feda4 ]
GEODE-5587: Randomize HTTP service working directory (#2333)
> Randomize internal http service working directory
> -------------------------------------------------
>
> Key: GEODE-5587
> URL: https://issues.apache.org/jira/browse/GEODE-5587
> Project: Geode
> Issue Type: Bug
> Components: pulse, rest (admin), rest (dev)
> Reporter: Jens Deppe
> Assignee: Jens Deppe
> Priority: Major
> Labels: pull-request-available
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When starting the http service a given webapp is extracted into a working
> directory of the form: {{<host>_<port>_<context>}}. As part of shutdown this
> directory is cleaned up. However the directory is only completely removed
> when the JVM exits.
> Occasionally subsequent tests will request to use the same port. This results
> in the same working directory being used. On Windows this will fail as the
> directory does not let itself be cleaned up before the webapp is redeployed
> to the same location as before.
> We can randomize the working directory a bit more by adding a short sequence
> of random chars as part of the directory name.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
