[
https://issues.apache.org/jira/browse/TINKERPOP-1714?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16080414#comment-16080414
]
ASF GitHub Bot commented on TINKERPOP-1714:
-------------------------------------------
GitHub user spmallette opened a pull request:
https://github.com/apache/tinkerpop/pull/658
TINKERPOP-1714 GremlinExecutor checks for timeout from time script submitted
https://issues.apache.org/jira/browse/TINKERPOP-1714
Prior to this change timeouts were considered from the time the script
started evaluation which probably isn't in line with user expectations but also
made it hard to clear the job queue on an overloaded server. This change also
has the benefit of simplifying the code as well as reducing the number of jobs
submitted to the various thread pools.
All tests pass with `docker/build.sh -t -n -i`
VOTE +1
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/apache/tinkerpop TINKERPOP-1714
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/tinkerpop/pull/658.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #658
----
commit ac19d5dec659a1a8f399d5a11a4812774eebeca7
Author: Stephen Mallette <[email protected]>
Date: 2017-07-06T15:49:21Z
TINKERPOP-1714 GremlinExecutor checks for timeout from time script submitted
Prior to this change timeouts were considered from the time the script
started evaluation which probably isn't in line with user expectations but also
made it hard to clear the job queue on an overloaded server.
----
> Gremlin Server scriptEvaluationTimeout should take into account request
> arrival time
> ------------------------------------------------------------------------------------
>
> Key: TINKERPOP-1714
> URL: https://issues.apache.org/jira/browse/TINKERPOP-1714
> Project: TinkerPop
> Issue Type: Improvement
> Components: server
> Affects Versions: 3.2.5
> Reporter: stephen mallette
> Assignee: stephen mallette
> Fix For: 3.2.6
>
>
> Currently the {{GremlinExecutor}} tries to timeout a eval after the job has
> started execution in the thread pool, when the likely user expectation is
> that it begin the timeout testing from the time the eval was invoked. While
> the current approach works (though not in line with what users probably are
> thinking is happening), an overloaded server using {{GremlinExecutor}} will
> have a more difficult time clearing the queue of jobs (if it can clear it at
> all).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
