RE: [Ganglia-developers] Some updates to 2.5.8

[Adesanya, A. <ye...@sl...>]

OK,

Ignore that. Just realized that I have to run 'bootstrap'.

----
Yemi
=20

> -----Original Message-----
> From: gan...@li...=20
> [mailto:gan...@li...] On=20
> Behalf Of Adesanya, Adeyemi
> Sent: Wednesday, November 24, 2004 4:24 PM
> To: kn...@kn...; gan...@li...
> Cc: Matt Massie; Frederico Sacerdoti
> Subject: RE: [Ganglia-developers] Some updates to 2.5.8
>=20
>=20
> Hi.
>=20
> I just checked out the head of monitor-core from cvs. Where=20
> is the ./configure script?
>=20
> ----
> Yemi
> =20
>=20
> > -----Original Message-----
> > From: gan...@li...
> > [mailto:gan...@li...] On=20
> Behalf Of=20
> > Martin Knoblauch
> > Sent: Wednesday, November 24, 2004 4:17 AM
> > To: gan...@li...
> > Cc: Matt Massie; Frederico Sacerdoti
> > Subject: [Ganglia-developers] Some updates to 2.5.8
> >=20
> > Hi,
> >=20
> >  I just checked in some changes to libmetrics and gmond.
> >=20
> > - cpu_wio is now a metric exported by all platforms
> > - the networking metrics are back for solaris
> > - solaris now uses kstat instead of kvm
> > - consolidates cpu_intr, cpu_sintr for the platforms where=20
> they make=20
> > sense
> > - cleanup gmond/metrics.h. Removed stuff already in libmetrics.h
> >=20
> >  Basically we should now have a state which includes the=20
> changes that=20
> > were made post-2.5.7 together with the latest/greatest=20
> metrics files.
> >=20
> >  Compiled and tested on:
> >=20
> > - Linux-2.6.9 (Fedrora FC2, gcc-3.3.3, libc 2.3.3)
> > - Solaris 8 (Sparc9, gcc-3.3.1). Yemi, have a go !!!
> >=20
> >  Comments from other platforms are very welcome.
> >=20
> > One Observation: when running gmond-2.5.8 version together with a=20
> > 2.5.4 version of gmond on two Linux systems on the same=20
> mcast channel,=20
> > I see some confusion. Doing a "telnet localhost 8649" on the 2.5.8=20
> > system shows the 2.5.4 system reportin "cpu_wio", which is=20
> nonsense.=20
> > Also, the metrics values of the 2.5.4 system are wrong. It=20
> looks like=20
> > some correct values are stuffed into the wrong metrics. Is=20
> it possible=20
> > that the correctness of the reporting is dependant on both gmonds=20
> > "know" the same number and order of metrices?
> > Does this mean that one should only run the same version of=20
> gmond? :-)
> >=20
> > Another observation: gmond/gmond.c still has one=20
> arch-dependency via=20
> > #ifdefs. Should we try to get that into libmetrics.h?
> >=20
> >  One wish: would it be possible to make some more metrics "global"
> > (like the networking and cpu_intr/sintr stuff). Architectures that=20
> > don't support them would just report 0 (I would volunteer=20
> to add the=20
> > stubs :-). My rationale for this is to get to a more=20
> consolidated set=20
> > of core metrics and just have the real weird stuff arch-dependant.=20
> > That would make grouping different system types easier. Just a=20
> > thought.
> >=20
> > Cheers
> > Martin
> >=20
> > =3D=3D=3D=3D=3D
> > ------------------------------------------------------
> > Martin Knoblauch
> > email: k n o b i AT knobisoft DOT de
> > www:   http://www.knobisoft.de
> >=20
> >=20
> > -------------------------------------------------------
> > SF email is sponsored by - The IT Product Guide Read honest=20
> & candid=20
> > reviews on hundreds of IT Products from real users.
> > Discover which products truly live up to the hype. Start=20
> reading now.=20
> > http://productguide.itmanagersjournal.com/
> > _______________________________________________
> > Ganglia-developers mailing list
> > Gan...@li...
> > https://lists.sourceforge.net/lists/listinfo/ganglia-developers
> >=20
>=20
>=20
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide Read honest &=20
> candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.=20
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> Ganglia-developers mailing list
> Gan...@li...
> https://lists.sourceforge.net/lists/listinfo/ganglia-developers
>=20