pcp
[Top] [All Lists]

Re: [pcp] pcp updates - pmdapapi update

To: Nathan Scott <nathans@xxxxxxxxxx>
Subject: Re: [pcp] pcp updates - pmdapapi update
From: Lukas Berk <lberk@xxxxxxxxxx>
Date: Wed, 19 Nov 2014 18:51:56 -0500
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <50063157.14443613.1415955185726.JavaMail.zimbra@xxxxxxxxxx> (Nathan Scott's message of "Fri, 14 Nov 2014 03:53:05 -0500 (EST)")
References: <87oascow3f.fsf@xxxxxxxxxx> <50063157.14443613.1415955185726.JavaMail.zimbra@xxxxxxxxxx>
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux)
Hi,

Nathan Scott <nathans@xxxxxxxxxx> writes:
[...]
> This is looking pretty good - I've updated some minor stuff (please do
> a review) but otherwise its all merged now.  There's a few things that
> would be good to still get in that I didn't tackle...

Thanks for merging this. I've worked on the comments below.

> - I'm not following why the (static) metrictab has all those entries
>   for event counters still?  Can those be removed now?

Yes it can, fixed on lberk/papi.

> - The permissions_check() changed to only look for uid==0 - should we
>   drop all references to gid* elsewhere?  I think so (slims down some
>   data structures, simplifies the attributes callback, etc)

I've done this as well on lberk/papi.

> - As per earlier mail, there's no error handling for auto-fu counters.
>   Simplest will be, I think, to give refresh_metric() a parameter that
>   indicates whether its caller is going to ignore the return code, and
>   in that case, handle_papi_error() should really log any errors into
>   the pmdapapi.log file.  In the pmStore-enable/disable-metrics case,
>   these need not be logged (the client tool will get notified about an
>   error in that case, which is better - keep that pmDebug diagnostic
>   for both cases though).

I believe the patch that Frank proposed for this has been merged
upstream.  The rest of the issues have been addressed in my latest pcp
updates mail[1].  

Thanks for reviewing and merging my changes!

Cheers,

Lukas

[1] - http://www.pcp.io/pipermail/pcp/2014-November/006012.html

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