pcp
[Top] [All Lists]

Re: pmmgr memory hog

To: "Frank Ch. Eigler" <fche@xxxxxxxxxx>
Subject: Re: pmmgr memory hog
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Thu, 5 Feb 2015 19:20:27 -0500 (EST)
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <20150205233642.GA1910@xxxxxxxxxx>
References: <23273355.372.1423022546978.JavaMail.rmckee@wsrmckee> <1838902881.6952157.1423024805273.JavaMail.zimbra@xxxxxxxxxx> <y0md25p63ji.fsf@xxxxxxxx> <1683339037.8571387.1423176489834.JavaMail.zimbra@xxxxxxxxxx> <20150205225822.GA28568@xxxxxxxxxx> <1790828395.8575235.1423177464966.JavaMail.zimbra@xxxxxxxxxx> <20150205232342.GB28568@xxxxxxxxxx> <20150205233642.GA1910@xxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: z9OwzESzDXoEX4d+EmV7NVkxIfUQSQ==
Thread-topic: pmmgr memory hog

----- Original Message -----
> [...] 
> A brief scanny run now produces actual valgrind leak reports ...

It could be this first one is part of the issue - does pmmgr create and
destroy contexts with attributes alot by default?  (if so, at what sort
of rates?  IOW, can we explain 1GB of memory this way?)

> ==17174==    by 0x4E6D346: __pmParseHostAttrsSpec (spec.c:937)
> ==17174==    by 0x4E49545: pmNewContext (context.c:468)
> ==17174==    by 0x116B8A: pmmgr_job_spec::compute_hostid(std::string const&)

This second one is fine - its a once-off thing, not on-going.  The QA
tests guard against reporting this (see the _run_valgrind shell code).

> ==17174==    by 0x4E70EA3: __pmConfig (config.c:218)
> ==17174==    by 0x4E71069: pmGetConfig (config.c:242)
> ==17174==    by 0x10DA53: main (pmmgr.cxx:1201)

cheers.

--
Nathan

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