aurora-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bill Farner <wfar...@apache.org>
Subject Re: Basic Question : Sandbox disk Space Reclaimed
Date Tue, 24 Nov 2015 00:46:25 GMT
You'll want to drill into why the task failed.  This usually means the
process launched for you exited non-zero.  Since the sandbox has been
deleted (as indicated by the final state transition), you won't be able to
look at stdout/stderr from the process(es), but that's likely what you'll
be after.  If you launch the task again, does it fail again?

On Sun, Nov 22, 2015 at 8:29 AM, Tameem Ahmed <tameem.ahmed@hotmail.com>
wrote:

> Hi Team,
>            I am putting in baby steps to learn aurora, please excuse me if
> I am asking very silly/basic questions.
>            Please do recommend if there is any other DL/Group who can help
> us learn little more about the system.
>            I have a job for which here is the status :
>
>            INFO] Active Tasks (0)           INFO] Inactive Tasks (4)
>            INFO] role: root, env: staging, name: app_1328, shard: 1,
> status: SANDBOX_DELETED on stage2.qa.com           cpus: 0.01, ram: 500
> MB, disk: 1024 MB     failure count: 1 (max 1)                 2015-11-20
> 12:10:01 PENDING: None            2015-11-20 12:10:01 ASSIGNED: None
>       2015-11-20 12:10:03 STARTING: Initializing sandbox.
>  2015-11-20 12:10:04 RUNNING: None               2015-11-20 12:13:20
> FAILED: Task failed.                2015-11-21 04:07:41 SANDBOX_DELETED:
> Sandbox disk space reclaimed.
>          What can I do the put the job back in to Active, tried all
> possibilities.   Made sure the slave machine is running both mesos-slave
> and observer, have enough disk space.
>          Master is running quite a few jobs, no issues.
>          What can do to make it running again.
> Thanks,Tameem
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message