pcp
[Top] [All Lists]

Re: [pcp] pmdaslurm

To: Martins Innus <minnus@xxxxxxxxxxx>
Subject: Re: [pcp] pmdaslurm
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Thu, 17 Sep 2015 22:53:42 -0400 (EDT)
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <55F2E230.2090700@xxxxxxxxxxx>
References: <55F2E230.2090700@xxxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: 9W9YucwOWJhFpJnBmSYl04zyt321PA==
Thread-topic: pmdaslurm
Hi Martins,

----- Original Message -----
> Hi,
> 
>      Here is a proposed pmda for the Slurm scheduler
> (http://slurm.schedmd.com/).  Its gone through several iterations,
> mostly because the Slurm api can take a long time to return information
> on a heavily loaded system and this was causing the pmda to be killed by
> pmcd.  I'm new to perl threading, but this seems to be working fairly
> well so far.  I implemented a very basic stub of the Slurm api to allow
> QA to run anywhere.

Looks good to me - very clean code.

> For now it just shows information on jobs that are running on the node,
> my plan is to add more metrics in the future.  There are a few stubs and
> comments to that affect in the code, mostly as reminders to myself.  Let
> me know if you want those removed.

No problem here, test passed first time too - all merged now.

cheers.

--
Nathan

<Prev in Thread] Current Thread [Next in Thread>