pcp
[Top] [All Lists]

RE: [pcp] qa problems with the slurm pmda

To: "'Martins Innus'" <minnus@xxxxxxxxxxx>, "'Nathan Scott'" <nathans@xxxxxxxxxx>
Subject: RE: [pcp] qa problems with the slurm pmda
From: "Ken McDonell" <kenj@xxxxxxxxxxxxxxxx>
Date: Sat, 19 Mar 2016 06:52:53 +1100
Cc: "'PCP'" <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <56EC24A8.7000909@xxxxxxxxxxx>
References: <56EB37BC.2070102@xxxxxxxxxxxxxxxx> <291199781.32302600.1458257109207.JavaMail.zimbra@xxxxxxxxxx> <56EC24A8.7000909@xxxxxxxxxxx>
Thread-index: AQDxfOK2l3skjgxsArgv4XEXjdDXPwJNTNzWAbEg0Cyg/7E04A==
G'day Martins.

> -----Original Message-----
> From: Martins Innus [mailto:minnus@xxxxxxxxxxx]
> Sent: Saturday, 19 March 2016 2:54 AM
> To: Nathan Scott <nathans@xxxxxxxxxx>; Ken McDonell
> <kenj@xxxxxxxxxxxxxxxx>
> Cc: PCP <pcp@xxxxxxxxxxx>
> Subject: Re: [pcp] qa problems with the slurm pmda
> ...
> I added some more test data and stub code here:
> 
> https://github.com/ubccr/pcp/tree/slurm-qafix

Thanks, I'll pull that commit and see if it helps my failure case.

> ...
> As a separate note, is there a way to short circuit all of the
> initialization the the QA suite does when running a single check?  I
> guess in a "user beware" manner, allow things like
> 
>   ./check 1031
> 
> without creating all the initial pmlogger archives, etc that I know
> aren't needed for this test.  Not a huge deal, but it would be nice to
> save some time when you need to go through the build/install/test
> cycle a bunch of times before getting everything right.  Then you
> could do a full test as one final check.

./check -q 1031

will be the magic you're seeking.

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