[Ganglia-developers] configure should exit with an error when it fails

I finally upgraded the FreeBSD port last week and bumped into the extra library dependencies in newer versions. On some user machines I had trouble figuring out what went wrong because they were getting to the build stage without a Makefile. The FreeBSD ports infrastructure should have prevented

Re: [Ganglia-developers] the sysconfdir issue

On Sun, Feb 07, 2010 at 01:30:43PM +, Daniel Pocock wrote: > > We've just been discussing the sysconfdir issue while here at FOSDEM > > For Ganglia <= 3.1.2, > - the configure --help output tells people they can use --sysconfdir to > set the location of gmond.conf > - the configure --help out

Re: [Ganglia-developers] [RFC] two step gmond initialization

[Resend with correct sender address] From: Brooks Davis Date: Fri, 11 Dec 2009 13:31:22 -0600 To: Carlo Marcelo Arenas Belon Cc: Brad Nicholes , ganglia-developers@lists.sourceforge.net Subject: Re: [Ganglia-developers] [RFC] two step gmond initialization On Fri, Dec 11, 2009 at 04:56

Re: [Ganglia-developers] [RFC] two step gmond initialization

[Resend to list with right from address] On Sat, Dec 12, 2009 at 10:00:04AM -0800, Bernard Li wrote: > Hi Daniel: > > On Sat, Dec 12, 2009 at 5:59 AM, Daniel Pocock wrote: > > > I've just posted the issue on d...@apr.apache.org > > Thanks! > > Here's the post for others who would like to moni

Re: [Ganglia-developers] updating to a more recent libtool version for the bootstrap?

On Mon, Jan 08, 2007 at 02:29:43AM -0800, Martin Knoblauch wrote: > --- Carlo Marcelo Arenas Belon <[EMAIL PROTECTED]> wrote: > > > I agree with you though that we should better come up with a version > > that is > > well supported for all development environments (*), and I presume > > that the >

Re: [Ganglia-developers] CPU Frequency Scaling and a patch

On Thu, Aug 31, 2006 at 07:25:08PM -0600, Ian Cunningham wrote: > I have filed bug 114 > . I > have also provided a patch. > > The jist of the bug is that we have these new machines, with new > versions of redhat, that scale the

Re: [Ganglia-developers] apr, expat, confuse asshipped with ganglia

The FreeBSD ports collection has up to date versions of all three. I think the library tarball is the right way to go. -- Brooks On Wed, Aug 16, 2006 at 03:40:41PM -0700, Bernard Li wrote: > Martin: > > I'll let others respond to the thread, but I'd just like to point out that > there is a fal

Re: [Ganglia-developers] apr, expat, confuse asshipped with ganglia

On Wed, Aug 16, 2006 at 09:50:19PM +0100, Stu Teasdale wrote: > On Wed, Aug 16, 2006 at 04:18:46PM -0400, Jarod Wilson wrote: > > > Stu, how about for Debian...? > > > > From what I can see, its available in Debian repos. > They're 2.5.x packages unfortunately. I have 3.0.3 packages, but a > coui

Re: [Ganglia-developers] apr, expat, confuse as shipped with ganglia

Throwing away all the external junk would not cause problems for any BSDs since we have perfectly good package systems. IMO, the lack of usable package systems on some platforms should not be our problem. -- Brooks On Sat, Aug 12, 2006 at 03:41:01PM -0700, Bernard Li wrote: > Hi all: > > In di

Re: [Ganglia-developers] Migrate to SVN

On Thu, May 04, 2006 at 01:46:16PM -0700, Bernard Li wrote: > Cool. > > Martin, what are your thoughts? > > If you're okay with it, should I go ahead and start the migration or did > you want to do it? > > Anybody else? Brooks? I'd say go for it. I haven't used it much yet because none of the

Re: [Ganglia-developers] RE: [Ganglia-general] New (final?) tarball for ganglia-3.0.3

I'm seeing reasonable load under FreeBSD 6.0 with 217 nodes. I wouldn't expect ~100 nodes to be very functional on a machine with anything by amazingly fast disks. With ~150 I was seeing 300+ transactions per second which rendered my boxes with 10K RPM SCSI disks in a hardware RAID5 unusable much

Re: [Ganglia-developers] apr, expat, confuse as shipped with ganglia

On Wed, Feb 22, 2006 at 05:07:08PM -0800, Martin Knoblauch wrote: > Hi Matt, > > you may have seen the recent report about problems in apr code which > are solved in later versions. This opens the question how to handle > this for Ganglia. > > Are the versions of apr, expat and confuse shipped

Re: [Ganglia-developers] gmetad umask trouble

On Mon, Oct 31, 2005 at 02:57:50AM -0800, Martin Knoblauch wrote: > Hi Seth, > > the comments came from Brooks. Hi Brooks, care to elaborate what > problems you see with the pidfile stuff from Seth? The big one is that if you use pidfile existence as the measure of the server being running, you

Re: [Ganglia-developers] Licensing Question

On Wed, Oct 26, 2005 at 06:42:01PM -0500, Brian Credeur wrote: > Hi All, > > I wanted to follow-up on a discussion thread back in mid-September on > licensing of 3.0.1. From that dicussion, there appears to be code from > a number of different licenses (BSD, GPL, LGPL, and Apache) and some > q

Re: [Ganglia-developers] licencing issues (was: 3.0.2 release)

On Tue, Sep 13, 2005 at 05:25:47PM -0400, michael chang wrote: > On 9/13/05, Stu Teasdale <[EMAIL PROTECTED]> wrote: > > > The killer is the GPL code linked into libmetrics. AIUI there's no legal > > way to use GPL code in a BSD licenced project, as there is no allowance > > in GPL for "derivative

Re: [Ganglia-developers] [patch] private cluster authenticate fix

On Sun, Sep 11, 2005 at 10:47:28AM -0400, michael chang wrote: > On 9/11/05, Jeong Bae Kim <[EMAIL PROTECTED]> wrote: > > Looks like a simple enough change to take the Apache basic authentication > > password from PHP built-in global array, _SERVER. > > >From http://www.php.net/manual/en/features.

Re: [Ganglia-developers] webfrontend fix

ge clusters/grids, > and are not obvious when used with smaller setups. That would be nice. -- Brooks > -Federico > > On Apr 27, 2005, at 1:42 PM, Brooks Davis wrote: > > >One issue with the webfront end that has annoyed me for a while is that > >the "Choose a No

[Ganglia-developers] webfrontend fix

One issue with the webfront end that has annoyed me for a while is that the "Choose a Node" dropdown is empty when showhosts is off. The following patch fixes that for me and decrease the differences between the two cases as well. Is there any reason this change is a bad idea? I don't notice any

Re: [Ganglia-developers] patches for ganglia on NetBSD

On Tue, Apr 12, 2005 at 10:07:35PM -0400, Jan Schaumann wrote: > Brooks Davis <[EMAIL PROTECTED]> wrote: > > On Sun, Apr 03, 2005 at 05:13:06PM -0400, Jan Schaumann wrote: > > > > I have patche available that port ganglia to NetBSD/i386. > > > Before there

Re: [Ganglia-developers] patches for ganglia on NetBSD

On Sun, Apr 03, 2005 at 05:13:06PM -0400, Jan Schaumann wrote: > Hi, > > I have patche available that port ganglia to NetBSD/i386. I sent these > patches to the maintainers in February right before the 3.0.0 release, > but I don't think they were included (neither 3.0.0, nor 3.0.1). Just > to ma

Re: [Ganglia-developers] Ganglia's linux sys_clock metric.

On Fri, Apr 01, 2005 at 04:45:50PM -0500, Jason A. Smith wrote: > A long time ago, someone noticed that ganglia's sys_clock metric was > basically the same as the gmond_started: > > http://sourceforge.net/mailarchive/message.php?msg_id=5320970 > > >From the 2.x sources it appears that this metric

[Ganglia-developers] thoughts on metrics

I've been thinking about the standard metrics in ganglia and I've run into a number of cases where the assumptions are either generally wrong or overly Linux specific. It would be good to think about ways to address these issues. There are three classes of metrics I am concerned with: cpu speed,

Re: [Ganglia-developers] 3.0.1 and java

On Wed, Mar 16, 2005 at 10:49:52AM -0800, Brooks Davis wrote: > On Tue, Mar 15, 2005 at 03:23:32PM +, Matt Massie wrote: > > guys- > > > > i just put what i believe will be the 3.0.1 release at > > http://matt-massie.com/ganglia/ > > > > i'm not go

Re: [Ganglia-developers] webfrontend version number

On Wed, Mar 16, 2005 at 02:38:56PM -0800, Bruce Schwartz wrote: > Hi -- > > I've just installed 3.0.0 on a cluster and noticed something funny > with the version reporting. The web page reports "Gmetad Web Frontend > version 2.5.7". Looking at the PHP sources I see that there is a > generated fi

Re: [Ganglia-developers] 3.0.1 and java

On Tue, Mar 15, 2005 at 03:23:32PM +, Matt Massie wrote: > guys- > > i just put what i believe will be the 3.0.1 release at > http://matt-massie.com/ganglia/ > > i'm not going to release it until we've tested it here longer and you > guys find it works well for you. > > 3.0.1. > > fixes a

Re: [Ganglia-developers] libmetrics

On Fri, Mar 04, 2005 at 11:12:59AM -0600, Luke Kanies wrote: > > The only real changes I made, other than to libmetrics.{c,h}, was to > move the metric_init() function to the end of each metrics.c file (so > that it is aware of each of the metrics functions) and call new_metric() > for each metric

Re: [Ganglia-developers] Dynamic linking for packaging

On Thu, Mar 03, 2005 at 09:05:34AM +, Stu Teasdale wrote: > Hi all, > > I'm currently working on ganglia 3.0.0 packages for Debian and have a > small query. Debian general policy states that external libraries should > be dynamically linked to executables, and the dependency system ensures

Re: [Ganglia-developers] version number

On Thu, Feb 03, 2005 at 12:27:57AM -0800, Martin Knoblauch wrote: > > --- Brooks Davis <[EMAIL PROTECTED]> wrote: > > > On Tue, Feb 01, 2005 at 09:43:29PM -0800, Matt Massie wrote: > > > guys- > > > > > > i'd like to have an informal vote. p

Re: [Ganglia-developers] version number

On Tue, Feb 01, 2005 at 09:43:29PM -0800, Matt Massie wrote: > guys- > > i'd like to have an informal vote. please email whether you think this > release should be 2.6.0 or 3.0.0. You already know my position, but let me expand of my reasoning. In my opinion, version numbers are sign posts for

Re: [Ganglia-developers] are we there yet?

On Tue, Feb 01, 2005 at 05:52:34PM +, Matt Massie wrote: > guys- > > i just created what i think will be the 2.6.0 release at > http://matt-massie.com/ganglia/2.6.0/ I'll beat on it a bit tonight and tomarrow. I really do think bumping the major number was a better idea... > i've also updat

Re: [Ganglia-developers] complete snapshot?

On Mon, Jan 31, 2005 at 05:13:23PM +, Matt Massie wrote: > i've just built what i think it a complete snapshot (meaning all > features are working as they should). > > this latest snapshot adds > - support for "ttl" on multicast udp_send_channel definitions > - fixed a minor bug in cygwin met

Re: [Ganglia-developers] new snapshot with complete darwin metrics support

On Wed, Jan 26, 2005 at 01:15:02PM +, Matt Massie wrote: > http://matt-massie.com/ganglia/ganglia-2.6.0.200501261221.tar.gz This one hangs on freebsd with mute=true. -- Brooks -- Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8

[Ganglia-developers] non-portable wildcard binds

The tcp code in ganglia makes the non-portable assumption that wildcard binds will forward packets between the IPv4 and IPv6 stacks. The RFCs require this, but none of the BSDs except for Darwin support this because KAME feels it is a bad idea. See the "Interaction between IPv4/v6 sockets" sectio

[Ganglia-developers] lost CVS history

Is there a backup of the CVS repository prior to the last round of cleanups? The history of a number of files was lost in the process and some of this history could be important. In particular, the metrics reporting code is usually cribbed from all sorts of places and it would be good to have som

[Ganglia-developers] 2.5.8 on FreeBSD

FYI, I've verified that the latest ganglia snapshot works on FreeBSD 5 (actually 6-CURRENT, but they are mostly the same at the moment.) I did notice one small bug. If /etc/gmetad.conf doens't exist, you get a nice warning followed by a core dump instead of an exit. -- Brooks -- Any statement

Re: [Ganglia-developers] Re: [Ganglia-general] gmond error on Solaris8/SPARK

On Wed, Nov 24, 2004 at 11:49:23AM +, Matt Massie wrote: > i think it's good practice that when an application daemonizes that it > close all open file descriptors that it inherits from the parent... not > just stdin, stdout, stderr. it's also important that (at least on > freebsd) that the me

Re: [Ganglia-developers] Re: [Ganglia-general] gmond error on Solaris8/SPARK

On Tue, Nov 23, 2004 at 10:15:58AM +, Matt Massie wrote: > thanks for submitting this bug fix but i'm not sure what problem it > solves. the code that you have submitted below would not close all file > descriptor after fork but rather would only close 0,1 and 2 > (stdin,stdout,stderr). are y

Re: [Ganglia-developers] ganglia 3.0.0. note

It looks like we're OK if there aren't any ','s in there at least in tcsh. -- Brooks On Fri, Oct 15, 2004 at 04:37:29PM -0400, Rick Warner wrote: > { and } have an effect too. > "cp file{.dat,.bak}" means the same thing as "cp file.dat file.bak"

Re: [Ganglia-developers] ganglia 3.0.0. note

On Thu, Oct 14, 2004 at 03:42:20PM -0700, Matt Massie wrote: > i'm not married to the '[' and ']' delimiters. > > />country/US/>state/Illinois/ > /(country/US/(state/Illinois/)group/cpu/)metric/number > > works just as fine too. the first one is cool because the brackets ">" > point to the "thi

Re: [Ganglia-developers] libmetrics results

On Tue, Sep 21, 2004 at 11:19:10AM -0700, Matt Massie wrote: > SunOS x86-solaris1 5.9 Generic_112234-10 i86pc i386 i86pc Solaris > - libmetrics passed with flying colors. > > > FreeBSD x86-freebsd1 4.10-BETA FreeBSD 4.10-BETA > > - needed to add a few system headers but it compiled and the > the

Re: [Ganglia-developers] libmetrics

On Mon, Sep 20, 2004 at 06:14:39PM -0700, Matt Massie wrote: > guys- > > when you get a chance can you please download > http://matt-massie.com/ganglia/libmetrics-0.1.0.tar.gz > and try it on your favorite operating system?? > > i've tested this library on linux and solaris and i _think_ i made t

Re: [Ganglia-developers] Re: Ganglia interface problem when multiple interfaces use same IP address

On Fri, Sep 03, 2004 at 05:27:28PM +0800, John Lau wrote: > Hi, > > I have mention (in a post on 25/8) about the gmond will have problem in > selecting multicast interface if there are multiple interfaces using > same IP address. I create a patch to ganglia-monitor-core-2.5.6 to fix > the problem.

Re: [Ganglia-developers] 2.6.0 apr test

.. and that will make me happy. > > -matt > > On Wed, 2004-08-04 at 16:26, Brooks Davis wrote: > > On Wed, Aug 04, 2004 at 01:36:54PM -0700, Matt Massie wrote: > > > on second thought... scratch that idea. i have a better one. > > > > > > jus

Re: [Ganglia-developers] 2.6.0 test

On Thu, Jul 15, 2004 at 12:10:02PM -0700, Matt Massie wrote: > On Thu, 2004-07-15 at 10:29, Brooks Davis wrote: > > > Partial success on FreeBSD 5-CURRENT, unless this is supposed to be a v6 > > only test. It's only binding to :: not 127.0.0.1. See below for my tests. &g

Re: [Ganglia-developers] 2.6.0 test

On Wed, Jul 14, 2004 at 07:12:48PM -0700, Matt Massie wrote: > guys- > > i have a 2.6.0 test that i'd like you to run on your respective > operating systems. > > download http://matt-massie.com/ganglia/ganglia-apr-2.6.0.tar.gz > > && ./configure && make > > i think i'm going to scrap the libunp

Re: [Ganglia-developers] 2.6.0 status?

Could you please spend some more time on restoring functionality. I haven't got 2.6.x to work with FreeBSD yet. At a minimum there's something wrong with the code that finds interfaces. Unfortunatly, gmond isn't exactly an easy thing to debug and I've had poor luck figuring out what I need to inc

Re: [Ganglia-developers] Adding cpu_wio to universal metrics

On Thu, May 27, 2004 at 09:03:38AM -0700, Martin Knoblauch wrote: > > --- Brooks Davis <[EMAIL PROTECTED]> wrote: > > On Thu, May 27, 2004 at 08:03:28AM -0700, Martin Knoblauch wrote: > > > Hi, > > > > > > after receiving overwhelming feedback on my

Re: [Ganglia-developers] Adding cpu_wio to universal metrics

On Thu, May 27, 2004 at 08:03:28AM -0700, Martin Knoblauch wrote: > Hi, > > after receiving overwhelming feedback on my request for discussing the > addition of cpu_wio as a global metric :-) I just went ahead and did > it. Arg, apparently I should have read my mail more carefully. What is WIO

Re: [Ganglia-developers] Trying to build from CVS fails

On Wed, May 26, 2004 at 10:46:20AM -0700, Martin Knoblauch wrote: > > --- Brooks Davis <[EMAIL PROTECTED]> wrote: > > On Wed, May 26, 2004 at 01:02:19AM -0700, Martin Knoblauch wrote: > > > Hi, > > > > > > today I tried to build from the CVS tree.

Re: [Ganglia-developers] Trying to build from CVS fails

On Wed, May 26, 2004 at 01:02:19AM -0700, Martin Knoblauch wrote: > Hi, > > today I tried to build from the CVS tree. So I followed the > README.CVSusers file and tried "./bootstrap". Results are not really > encouraging :-( All versions of required tools (automake, autoconf, > libtool) are highe

Re: [Ganglia-developers] Looking forward to 2.6.0

On Tue, May 18, 2004 at 04:41:59PM -0700, Matt Massie wrote: > On Tue, 2004-05-18 at 16:29, Brooks Davis wrote: > > > gcc -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/local/include -D_REENTRANT -Wall -c > > gzio.c -MT gzio.lo -MD -MP -MF .deps/gzio.TPlo -fPIC -DPIC -o .libs/gzio.lo &g

Re: [Ganglia-developers] Looking forward to 2.6.0

On Tue, May 18, 2004 at 03:44:51PM -0700, Matt Massie wrote: > i'm concerned about the portability of 2.6.0. i think it's only been > test on linux and freebsd so far. we have eight or so oses out there > that use ganglia. It's worse then that actually. I haven't sucessfully run 2.6 on FreeBSD

Re: [Ganglia-developers] server.c accept() 2.6.0 issue on Darwin

On Wed, Apr 14, 2004 at 11:08:26PM -0400, Josh Durham wrote: > Well, I figured out the cause of the problem, but not too sure how to > fix it. > > IPv6, which is enabled by default in OS X, seems to be breaking some of > the stuff. I'm not sure how the libunp is behaving differently than > the

Re: [Ganglia-developers] Getting 2.6.0 Patches

On Fri, Apr 09, 2004 at 06:53:47PM -0400, Josh Durham wrote: > Two patches. The first can be easily applied, the second should be > tested on another platform. First one allows Ganglia to be compiled if > GEXEC_SUPPORTED == 0, and the second fixes a libtool problem with > multiple memory_pool

Re: [Ganglia-developers] compression

On Tue, Mar 23, 2004 at 09:46:07AM -0800, Matt Massie wrote: > brooks talk about using LZ compression to save the CPU costs associated > with zlib. does anyone have code to do that? What I was actually trying suggesting was looking at the research on XML compression. There are specific compres

Re: [Ganglia-developers] quick question

On Tue, Mar 16, 2004 at 12:28:58PM -0800, Matt Massie wrote: > what is the maximum size of an rrd that you would tolerate? what is a > reasonable size? it is currently 11948 bytes per metric and double that > for summary metrics. > > that means that a 128 node cluster monitoring 30 metrics each

Re: [Ganglia-developers] the multicast chains have been broken

On Thu, Feb 26, 2004 at 07:17:05PM -0800, Matt Massie wrote: > i think brooks comment on the compressed XML is right on target. it > doesn't make sense to make put zlib compression on a separate port. it > make more sense to have a web server port (and the old style simple tcp > socket as well).

[Ganglia-developers] hardcoded paths in ganglia

When building ganglia as a FreeBSD port, I apply a number of patches and run some sed scripts to deal with the fact that the following paths are hardcoded into ganglia: /etc/gmetad.conf /etc/gmond.conf /var/lib/ganglia/rrds The problem with this is that FreeBSD ports expect to find their config f

Re: [Ganglia-developers] 2.5.7 DNS fix

On Tue, Feb 24, 2004 at 12:10:35PM -0800, Matt Massie wrote: > if we simply key on ip address, it would cause other problems. when you > remap ip/hostname pairs in a network overhaul.. it's very possible to > have ip gets swapped (not just changed for a single host). in that case > the data forks

Re: [Ganglia-developers] 2.5.7 DNS fix

On Tue, Feb 24, 2004 at 09:06:10AM +0100, Leif Nixon wrote: > Brooks Davis <[EMAIL PROTECTED]> writes: > > > Since this will require a conversion script anyway, would it be > > possiable to move away from using hostnames in the data store entierly? > > It has the

Re: [Ganglia-developers] 2.5.7 DNS fix

On Fri, Feb 20, 2004 at 01:17:51PM -0800, Matt Massie wrote: > guys- > > i just updated the CVS and uploaded a new tarball to > http://matt-massie.com/ganglia/ > > i fixed the RRD case-sensitivity problem by forcing the RRD paths to be > lower-case in both gmetad and the web frontend. this chang

Re: [Ganglia-developers] ganglia fixes

On Mon, Feb 23, 2004 at 04:50:35PM -0800, Matt Massie wrote: > i just updated CVS and created a new 2.5.7 tarball > (http://matt-massie.com/ganglia/). > > i THINK i've fixed the include path madness. i removed all AM_CFLAGS > specified in Makefile.am and instead explicitly wrote out the paths. >

[Ganglia-developers] include path problems persist

While getting rid of libdnet solved one class of include problems, a new one has been created. Specificly, you can not build ganglia if it is already installed because we include and the include order (in the case of gmetad) is: -I. -I. -I.. ${CPPFLAGS} -I../lib -I../gmond Since CPPFLAGS will o

Re: [Ganglia-developers] CVS reports

On Wed, Feb 18, 2004 at 10:48:43AM -0800, Matt Massie wrote: > guys- > > i would like some feedback here. i have the CVS setup now to email me > all the checkins and diffs. would you like me to have it mail the > developers list? > > another option is to create a new mailing list, say "gangli

Re: [Ganglia-developers] estimating disk usage

On Mon, Feb 09, 2004 at 11:13:39AM -0800, Federico Sacerdoti wrote: > Yes, this is the case. The problem is you dont know how many nodes you > will have. What if a large number are added? Everyone's RRD disk > requirements will go up. I'll make that a system config paramater with a reasionable d

[Ganglia-developers] estimating disk usage

Does anyone know how to go about estimating ganglia disk requirements assuming one knows the number of nodes and the number of metrics per node? It looks like we use about 12k per-metric per-node plus some summary info. Is this alwasy the case? I'm thinking about this because I'm looking at addi

Re: [Ganglia-developers] ganglia and libdnet

On Wed, Feb 04, 2004 at 04:36:00PM -0800, Matt Massie wrote: > brooks- > > my main objective with ganglia is to make it as easy to install as > possible. the reason i put libdnet in the distribution was to prevent > requiring people to install libdnet as a prerequisite (especially since > i only

[Ganglia-developers] ganglia and libdnet

Ganglia currently uses an old copy of libdnet embeded in the lib directory. This works fine in general, but I recently heard from a FreeBSD user who was having problems with it. The issue was that he had a copy of libdnet installed in /usr/local and for some reason ganglia picked up the headers f

Re: [Ganglia-developers] 2.5.6

On Mon, Feb 02, 2004 at 05:06:57PM -0800, Matt Massie wrote: > i've built a new tarball and rpms which you can find at > http://matt-massie.com/ganglia/ > > i really want to get 2.5.6 out the door today or tomorrow if possible. > i made some changes to the automake scripts to remove a dependency

Re: [Ganglia-developers] 2.5.6

On Tue, Jan 20, 2004 at 11:18:07AM -0800, Federico Sacerdoti wrote: > Matt, > I just checked in the host_dmax code after reading your email. The four > files changed are: > o gmond.conf > o listen.c > o gmond_config.c > o gmond_config.h > > If our people from other OSs (not linux) would compi

Re: [Ganglia-developers] New disk-alive metric

On Thu, Jan 15, 2004 at 03:18:43PM +0100, Sander van Vliet wrote: > Can somebody tell me when ganglia becomes a non-root user because it might be > possible to implement create_mount_list() before that so we can access the > disks directly. If you take a look at main() in gmond.c you'll see that

Re: [Ganglia-developers] 2.5.6

On Thu, Jan 15, 2004 at 03:03:59PM +0800, Matt Massie wrote: > i'd like to release 2.5.6 sometime very soon. let me know if anyone has > any objections. brooks can you check that all the relevant freebsd > autoconf stuff is in the tarball? > > http://matt-massie.com/ganglia/ > > i'm 99% sure it

Re: [Ganglia-developers] New disk-alive metric

On Mon, Jan 12, 2004 at 09:30:12PM +0100, Sander van Vliet wrote: > I have attempted to create a new disk-alive metric. Currently it only > works on Linux. A short description of what it, disk_alive_func(), > does: > 1 Get a list of filesystems via /proc/mounts If you split this out into a separat

Re: [Ganglia-developers] new disk-alive metric needed

On Sun, Jan 11, 2004 at 06:51:06AM -0800, Federico Sacerdoti wrote: > Brooks, > Sorry for the late reply. I'm glad to get some BSD feedback on my design. I > am suprised that the OS lies about writing so much, but I guess the error > from flushing the dirty pages have no way to propagate to the wri

Re: [Ganglia-developers] (no subject)

On Thu, Jan 01, 2004 at 04:25:17AM +0800, Matt Massie wrote: > if i hear that everything is ok.. i'll officially release 2.5.6. I just committed a change to configure.in which should fix FreeBSD ia64 support (there was a bogus catchall ia64 case in the match against the os, vendor, machine triplet

Re: [Ganglia-developers] new disk-alive metric needed

On Tue, Dec 23, 2003 at 11:14:37AM -0800, Robert Walsh wrote: > > I'd love to see some sort of ganglia integration with SMART, maybe via > > the smartmontools code (http://smartmontools.sourceforge.net/). One > > gotcha to watch out for is that a number of IBM GXP drives fail when > > they get SMA

Re: [Ganglia-developers] new disk-alive metric needed

On Mon, Dec 22, 2003 at 11:54:17PM -0800, Robert Walsh wrote: > > I'm pretty sure this doesn't provide the semantics you want. > > I've been experimenting with something similar, actually. I'm looking > at the SMART data from the drive and working from that. The advantage > with this is that it

Re: [Ganglia-developers] new disk-alive metric needed

On Mon, Dec 22, 2003 at 08:26:10PM -0800, Federico Sacerdoti wrote: > > This method takes advantage of the fact that most UNIX filesystems > flush dirty pages to disk at least once a minute. The idea is that > eventually we will detect that the disk cannot record new information > and tell the

[Ganglia-developers] Ganglia on TechTV

I took part in a short segment on TechTV about building clusters. The main reason for it was to promote FreeBSD, but we also plugged Ganglia, since it's the easiest way to get some graphical output from a cluster. There's a press release with links to that part of the show here: http://www.offmys

[Ganglia-developers] configure bug breaks new ia64 platforms

The configure.in file contains some ia64 specific code. The *ia64*) entry around line 332 breaks ia64 support on any existing plaform below that point in the case statement. Any objection to deleting that line? At this point, if you hit it, you can't compile. -- Brooks -- Any statement of the

Re: [Ganglia-developers] Metric platform table

On Thu, Nov 13, 2003 at 09:21:34PM -0800, matt massie wrote: > i think we should have a table showing what metrics are supported for > what platform. i've started the table to get the ball rolling but it is > not complete. the table has every metric listed in alphabetical order. > > when/if you

[Ganglia-developers] weird includes problem

The ganglia build has a weird problem with it's includes that can break the build when a previous version is installed. The build it set up such that $(prefix)/include is before the local ganglia directories which means that incompatible versions of ganglia headers are picked up before the system

Re: [Ganglia-developers] 2.5.5

On Wed, Nov 05, 2003 at 08:39:32PM -0800, matt massie wrote: > i have moved all the documentation into perlpod format. you'll see it in > a single file in ./ganglia.pod. docbook was overkill and i think it hurt > the documenation process. perlpod is super simple and can be converted > easily

Re: [Ganglia-developers] 2.5.5

On Thu, Nov 06, 2003 at 02:03:41PM -0800, matt massie wrote: > robert- > > nope. that wasn't intentional and i'm not sure how they got there. i > didn't need to touch the readdir.[c,h] files for the fix. do the newlines > cause any problems when compiling the source? It didn't cause problems

Re: [Ganglia-developers] 2.5.5

On Thu, Nov 06, 2003 at 11:00:25AM -0800, matt massie wrote: > i just made a new 2.5.5 distribution > http://matt-massie.com/ganglia/ > > let me know what you find. i'm pretty sure this distribution will be > complete and compile happily. This one looks OK to me. -- Brooks -- Any statement o

Re: [Ganglia-developers] Re: 2.5.5

On Wed, Nov 05, 2003 at 09:57:48PM -0700, Jim Prewett wrote: > On Wed, 5 Nov 2003, matt massie wrote: > > > i just created a new release of ganglia but i don't want to send it out > > until we do a little testing. > > > > you can find it at http://matt-massie.com/ganglia/ (for now). > > > > j

Re: [Ganglia-developers] Enhancements to FreeBSD support

On Mon, Oct 06, 2003 at 04:38:03PM -0700, Brooks Davis wrote: > Please find below a patch making proc_run report non-zeron values and > adding all the meterics provided by Linux. I've tested them on my > laptop runnng 5.1-CURRENT and a 4.9-PRERELEASE machine. This patch > brings

Re: [Ganglia-developers] Enhancements to FreeBSD support

On Mon, Oct 06, 2003 at 04:38:03PM -0700, Brooks Davis wrote: > Please find below a patch making proc_run report non-zeron values and > adding all the meterics provided by Linux. I've tested them on my > laptop runnng 5.1-CURRENT and a 4.9-PRERELEASE machine. This patch > brings

[Ganglia-developers] Enhancements to FreeBSD support

Please find below a patch making proc_run report non-zeron values and adding all the meterics provided by Linux. I've tested them on my laptop runnng 5.1-CURRENT and a 4.9-PRERELEASE machine. This patch brings the FreeBSD port to feature parity with Linux. FreeBSD users, please test/review this.

Re: [Ganglia-developers] linux magazine

On Tue, Aug 12, 2003 at 12:16:27PM -0700, steven wagner wrote: > Gexec hasn't really had a lot of development time thrown at it lately. > > Unless I missed a couple e-mails on this list. :) Last I heard it was > in an unmaintained but compatible state, with some moldy documentation. > > Hey, ho

Re: [Ganglia-developers] Ganglia monitor-core 2.5.4

On Wed, Jul 30, 2003 at 02:36:09PM -0700, Federico Sacerdoti wrote: > We are planning to release Ganglia 2.5.4 (monitor-core and webfrontend) > soon. > > This version contains a more efficient gmetad that supports the new > webfrontend structure: the load pie chart, faster page loading, > diffe

Re: [Ganglia-developers] gexec cvs

On Fri, Jul 25, 2003 at 01:55:52PM -0700, matt massie wrote: > brooks- > > gexec is not in CVS right now .. although .. once i had some time .. i > planned to incorporate gexec and it's friend into the g3 distribution. > it's a bit of a big project so i wasn't planning on making it a part of >

[Ganglia-developers] gexec cvs

Are gexec and friends available in a public CVS repository? I'm not seeing them in the sourceforge repo. I'm taking a shot at porting them to FreeBSD and that would make my life easier. -- Brooks -- Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7

Re: [Ganglia-developers] Hyperthreading patch for cpu_num on Linux

On Wed, Jul 23, 2003 at 02:59:25PM -0700, Robert Walsh wrote: > On Wed, 2003-07-23 at 14:22, Rick Mohr wrote: > > So it looks like the recipe should be: > > > > Is the ht flag set? If not, there is no hyperthreading. If yes, check > > for the physical id field. If it is not present, then assume

Re: [Ganglia-developers] Hyperthreading patch for cpu_num on Linux

On Wed, Jul 23, 2003 at 03:16:44PM -0400, Rick Mohr wrote: > > I believe you are right. I'm afraid that I just trusted the advice of > someone else when he said that the "ht" flag indicates > hyperthreading. And as I had no plain P4's to check this against, it > seemed to be the right way. > >

Re: [Ganglia-developers] Hyperthreading patch for cpu_num on Linux

On Wed, Jul 23, 2003 at 02:14:08PM -0400, Rick Mohr wrote: > > I believe someone mentioned a while back that Ganglia reported the wrong > number of CPUs for hyperthread enabled processors. I ran into the same > problem, so I created a quick and dirty patch for Linux. It simply checks > for the "

Re: [Ganglia-developers] $max_graphs = 0

On Mon, Jul 21, 2003 at 01:36:29PM -0700, Federico Sacerdoti wrote: > > On Monday, July 21, 2003, at 11:15 AM, Brooks Davis wrote: > > >>I've got to get the 2.5.4 gmetad and webfrontend out. It includes an > >>explicit option (a button on the webpage) to

Re: [Ganglia-developers] $max_graphs = 0

On Mon, Jul 21, 2003 at 10:43:49AM -0700, Federico Sacerdoti wrote: > > On Friday, July 18, 2003, at 04:07 PM, Brooks Davis wrote: > > >It would be nice if setting $max_graphs in the web frontend resulted > >in showing no graphs instead of all of them. If you change the tw

[Ganglia-developers] $max_graphs = 0

It would be nice if setting $max_graphs in the web frontend resulted in showing no graphs instead of all of them. If you change the two "$max_graphs > 0" lines to >= you can use -1 for all and have 0 be useful. I'm much happier looking at my status page now since my connection to the cluster is >2

Re: [Ganglia-developers] Re: G3 compiling FreeBSD problem

On Fri, Jul 18, 2003 at 12:14:08PM -0700, matt massie wrote: > > there is no need to upgrade the machine. i want ganglia to support > freebsd 4.x as well. i found a 4.7 box here and whipped up a new g3 > snapshot that will compile and run on freebsd 4.x. > > http://matt-massie.com/g3 > > ha

  1   2   >