giraph-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eli Reisman (JIRA)" <>
Subject [jira] [Commented] (GIRAPH-730) GiraphApplicationMaster race condition in resource loading
Date Sat, 24 Aug 2013 04:07:52 GMT


Eli Reisman commented on GIRAPH-730:

Hey Chuan than makes a lot of sense, thank you. I think the part I was confused about is that
I don't think getTaskResourceMap is called asynchronously, I thought the first call (the one
that actually initializes the map) was returned and completed before the first thread is launched,
and the remaining calls would be reading an immutable object (not by declaration but by convention
only) so would be essentially thread safe.

If you guys are sure I'm wrong thats good enough for me, we can commit this. I have not been
keeping up with the mailing list as I should -- have you noticed anyone else reproducing this
problem? Did this patch solve the problem for you?

If so, let me know and we should move forward with it. Good catch!

> GiraphApplicationMaster race condition in resource loading
> ----------------------------------------------------------
>                 Key: GIRAPH-730
>                 URL:
>             Project: Giraph
>          Issue Type: Bug
>    Affects Versions: 1.0.0
>         Environment: Giraph with Yarn
>            Reporter: Chuan Lei
>            Assignee: Chuan Lei
>         Attachments: GIRAPH-730.v1.patch
> In, getTaskResourceMap function is not multi-thread safe,
which causes the application master fail to distribute the resources (jar, configuration file,
etc.) to each container.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message