pcp
[Top] [All Lists]

Re: pmcd.openfds Metric (was: [pcp] NSS/NSPR Testing Status)

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: pmcd.openfds Metric (was: [pcp] NSS/NSPR Testing Status)
From: "Frank Ch. Eigler" <fche@xxxxxxxxxx>
Date: Mon, 3 Dec 2012 16:21:50 -0500
Cc: Dave Brolley <brolley@xxxxxxxxxx>, Nathan Scott <nathans@xxxxxxxxxx>, pcp@xxxxxxxxxxx
In-reply-to: <1354482125.15639.14.camel@xxxxxxxxxxxxxxxxxxxxxxx>
References: <591870624.35407484.1354151497586.JavaMail.root@xxxxxxxxxx> <50B8DB53.20500@xxxxxxxxxx> <1354482125.15639.14.camel@xxxxxxxxxxxxxxxxxxxxxxx>
User-agent: Mutt/1.4.2.2i
Hi -

On Mon, Dec 03, 2012 at 08:02:05AM +1100, Ken McDonell wrote:
> [...]
> As far as I can recall (which is not far when you get to my age), the
> original motivation for this was to _cheaply_ collect a metric that
> would detect an obscure file descriptor leak problem.
> [...]
> I don't think iteration with dup() meets the "cheap" requirement.

It's not cheap, but it is thorough. :)

> So provided you have closure of all the places open() and friends are
> called (as the current code does) [...]

Do we really?  Third-party PMDA DSO's are unknown, and so may be some
of the fd operations used by various libraries our own DSO's link in
(e.g. the new systemd one! :-).

- FChE

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