Greetings,
while working on implementing disk metrics for Solaris, I found the following
comment on the Linux implementation ([EMAIL PROTECTED]:1360) which
looked suspicious :
/* We report in GB = 1e9 bytes. */
sadly, in Solaris at least (and also in Linux, NetBSD and OpenBSD AFAIK) the
On Fri, Dec 15, 2006 at 11:06:48AM -0800, matt massie wrote:
i'm happy to open subversion access to any developers on this list who
are blocked waiting for someone to drop in a patch.
I got this bug/patch open to implement disk metrics for Solaris :
On Mon, Dec 18, 2006 at 02:55:29PM +0100, Erich Focht wrote:
On Monday 18 December 2006 11:28, Martin Knoblauch wrote:
no objections from my side. I am using it on several clusters without
any serious problems.
any objections to changing the part_max_used metric to be limited to RW
Greetings,
is there any reason why for the bootstrap we are still using a 1.5.8 version
of ltmain.sh that is almost 2 years old? and if that is the case, where did
that came from, as i couldn't find a match from the package with that same
version number from sourceware.
FWIW, I'd been doing
On Wed, Dec 27, 2006 at 03:12:51AM -0800, Martin Knoblauch wrote:
What I now need is the output from 2.4 based configs. Only multi-core
and/or HT-enabled systems actually.
AFAIK in vanilla kernel 2.4 there is no way to tell between virtual and real
CPUs from /proc/cpuinfo as can be seen by
On Fri, Dec 22, 2006 at 08:05:02AM -0800, Martin Knoblauch wrote:
Hi Folks,
in order to fix bz#84 for Linux.
http://bugzilla.ganglia.info/cgi-bin/bugzilla/show_bug.cgi?id=84
I think that the fix for this bug should actually include adding 2 more
metrics, as the problem as stated isn't
Greetings,
There seem to be some platforms (Debian (*) and SuSE being the ones that
are more relevant IMHO) that are maintaining their own forks of ganglia based
on 2.5.7, while some others (Gentoo, Fedora, NetBSD, FreeBSD and Solaris being
the ones that are more relevant IMHO) that are currently
On Wed, Dec 27, 2006 at 09:45:35AM -0800, Martin Knoblauch wrote:
I used NetBSD as a base from my port (as it is the closest), sadly
they are not that similar as to just work with the other source
as you can see by the diff.
Understand. Btw. you should check the use of the strings NetBSD
On Thu, Jan 04, 2007 at 10:59:01AM -0500, Vladimir Vuksan wrote:
I was wondering if there may be some value in using libstatgrab in Ganglia
for grabbing some of the metrics. I realize Ganglia has most (or all) of
the metrics in libgrabstat but may be a wise choice for the future since
On Fri, Jan 05, 2007 at 01:22:34AM -0800, Martin Knoblauch wrote:
--- Carlo Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
FWIW, I'd been doing `autoreconf --verbose --force --install` for the
last
days with the FC6 provided autotool packages :
autoconf 2.59
automake 1.9.6
guys-
i just went in an cleanup all the tags and branches in subversion.
great work, and definitely something we needed if we are to keep also the 2.x
branch for ganglia an some sanity as I mentioned before.
[EMAIL PROTECTED]:~/Code/Svn/ganglia/tags$ ls
gexec-0.2.1
On Sat, Jan 06, 2007 at 07:36:29PM -0600, Carlo Marcelo Arenas Belon wrote:
wasn't really sure either about the accuracy of the 2.5.x tags, as I'd been so
far unsuccessful trying to match the 2.5.7 tag with the corresponding update
version in trunk
after going up and down in the versions
On Fri, Jan 05, 2007 at 04:20:37PM -, [EMAIL PROTECTED] wrote:
1) Install a full cygwin system (specified on their downloader app),
don't use the default install or any other subset.
Why?, all you really should need is the basic stuff (going by default) and add
to it the compilers, make
On Fri, Jan 05, 2007 at 05:23:47PM -, [EMAIL PROTECTED] wrote:
Also remember about the cygwin agent build, which also processes
from cygwin's /proc.
the cpu_num metric was hardcoded to return 1 until 3.0.5, so that one will
change regardless, but this time it will show the right number of
On Fri, Jan 05, 2007 at 04:59:54PM -, [EMAIL PROTECTED] wrote:
Unfortunately it is cygwin itself (and not ganglia) that causes problems
by not imlpementing all the metrics that it should in /proc.
well, for Windows there is no concept of load average and therefore
the load_* metrics are
SNIP
I also have a simple patch to implement cygwin cpu count that
I will forward tonight. I remember a thread about this a while ago and
I'm surprised that cygwin/gmetric.c/cpu_num_func() looks like the
original one.
Oops, my bad, I'd been using a patched version of cpu_num_func for a while
On Sat, Sep 29, 2007 at 10:30:31AM -0700, Bernard Li wrote:
Hi guys:
Ganglia 3.0.5 is ready, I have prepared the release notes here:
noticed the tag in the morning and had been busy doing packages already, as I
really expected another RC with the verified fixes for MacOS to use for
platform
On Tue, Oct 02, 2007 at 11:32:48PM -0700, Bernard Li wrote:
Anybody who have access to Cygwin want to test this patch and provide
feedback?
http://bugzilla.ganglia.info/cgi-bin/bugzilla/show_bug.cgi?id=154
took the bug, the provided patch conflicts with the fix I proposed, tested
and
On Thu, Oct 04, 2007 at 10:47:41AM -0500, Jim Rowan wrote:
Bernard Li wrote:
Dear all:
As you all know, 3.0.5 has just been released -- time to work on 3.1.0! :-)
I thought I would start a wishlist for 3.1.0 where we can brainstorm
for ideas to go into our next release.
I would
On Fri, Oct 05, 2007 at 11:30:39AM -0400, Richard Mohr wrote:
On Fri, 2007-10-05 at 00:17 -0400, Wood, David wrote:
I was under the impression that lots of people experienced disk i/o
issues with gmetad. I was also under the impression that the generally
accepted work around for this
On Fri, Oct 26, 2007 at 06:45:59PM +0400, Kirill Nikonorov wrote:
where to look in sources to switch off removing dead hosts after
Ganglia daemons restart. I need to keep the view of my cluster even
with dead nodes.
there is no way in the source to switch off removing dead hosts at shutdown,
On Wed, Oct 24, 2007 at 12:58:22PM +0100, richard grevis wrote:
unfortunately for you, this is a bug, and the bug lies
in cygwin rather than gmond as I remember it.
More than a bug, it is a limitation because for the point of view of gmond
it is limited to the ammount of memory that cygwin can
On Wed, Oct 24, 2007 at 05:20:59PM +0530, Deepak CHAUHAN wrote:
Hello All,
I have installed ganglia for windows 2003 server but its not showing RAM
more than 2 GB and server is having 16 GB of RAM.Pls. help me in resolving
the issue.
Can you see if the attached patch fixes the issue?
On Wed, Oct 31, 2007 at 04:29:21PM -0600, Brad Nicholes wrote:
-* Remove the 4T limit on ganglia metric results
-* Modify all byte count metric to 8 bytes ints
moving to 8 byte ints won't solve the problem, will only push it further and
will make the memory utilization for gmond and disk
On Tue, Nov 06, 2007 at 04:59:59PM -0800, Bernard Li wrote:
Got a quick question about multicpu.conf -- by default, this is
installed with the RPM and enabled, should it be?
Because apparently, the kernel which comes with CentOS 4 does not
support it, as I get the following syslog message
On Tue, Nov 06, 2007 at 11:45:37PM -0800, Bernard Li wrote:
attached a fix that was tested working in my athlon64 laptop running gentoo
linux 2007.0 with and without powernow-k8. a similar fix will need to be
done
for the multicpu code unless changes in r815 are reverted and then we go
On Thu, Nov 08, 2007 at 09:20:26AM -0800, Bernard Li wrote:
Anybody else have comments on this?
http://fedoraproject.org/wiki/Packaging/Python
Carlo
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through
On Fri, Nov 09, 2007 at 02:55:16PM -0700, Brad Nicholes wrote:
Bernard Li [EMAIL PROTECTED] wrote:
On 11/9/07, Carlo Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
http://fedoraproject.org/wiki/Packaging/Python
If it's okay with you, I will follow the instructions in the wiki docs
On Thu, Nov 29, 2007 at 09:56:14AM -0500, Jesse Becker wrote:
I've been trying to build the trunk version of Ganglia on an OpenBSD 4.1
box recently, and it isn't for the faint of heart.
sadly true, building trunk isn't that easy if you are not using Fedora,
indeed I didn't even though that
On Thu, Nov 29, 2007 at 10:26:57AM -0800, Bernard Li wrote:
In this case, static builds (with the updated dependencies from the
tarball) should be able to take advantage of the new features?
No, you need to be linked with a dynamically loaded apr and had dlopen support
to be able to do that
Greetings,
while preparing a bootstrap environment which could be used to generate
maintenance releases for the 3.0.x branch noticed we'd been using different
versions for autotools that the ones detailed in README.SVNusers (which
match the ones used by 2.5.7), as shown by the following table :
On Wed, Jan 02, 2008 at 10:34:37PM -0800, Bernard Li wrote:
On 12/31/07, Carlo Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
while preparing a bootstrap environment which could be used to generate
maintenance releases for the 3.0.x branch noticed we'd been using different
versions
On Thu, Jan 10, 2008 at 09:48:15AM -0800, Bernard Li wrote:
Hi Carlo:
(moved to ganglia-developers)
On 1/10/08, Carlo Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
can we make a newer snapshot, or even better an automatic snapshot per
commit
(we don't have a commit for day, so
On Tue, Jan 15, 2008 at 09:08:03PM -0500, Jesse Becker wrote:
This is a re-post of a patch I sent back in December (originally based
on someone else's code).
Committed revision 921, thanks
Carlo
-
This SF.net email is
On Thu, Jan 17, 2008 at 09:48:11AM -0500, Jesse Becker wrote:
On Jan 17, 2008 12:44 AM, Carlo Marcelo Arenas Belon
svn diff -c-921 | patch -p0 svn merge -c810
This actually reverted three lines too many, and broke the enlarging
of optional graphs.
Which shouldn't be the case
On Mon, Feb 04, 2008 at 01:02:03PM -0800, Bernard Li wrote:
BTW, anything that can be done on the Fedora 8 side? Would it be
possible to downgrade rrdtool...?
the web frontend used to look for rrdtool in the gmetad root tree by
default as shown in the comment in web/conf.php for the RRDTOOL
On Tue, Jan 22, 2008 at 04:17:07PM +0530, Kumar Vaibhav wrote:
I am using ganglia-3.0.5 on a woodcrest processor cluster. and I see
that after running for weeks the memory consumption of the gmond process
is something about 400 MB.
did you check what was the size 1 hour after all gmond
On Tue, Feb 12, 2008 at 04:57:25PM -0800, Bernard Li wrote:
Guys:
I plan to check-in the following patch and release 3.0.7:
can we use the same exact code already committed in trunk (including spaces
and other details) so that there are no unnecessary divergences in the
maintenance branch?
On Wed, Feb 13, 2008 at 05:00:56AM -0600, Carlo Marcelo Arenas Belon wrote:
On Tue, Feb 12, 2008 at 04:57:25PM -0800, Bernard Li wrote:
Guys:
I plan to check-in the following patch and release 3.0.7:
can we use the same exact code already committed in trunk (including spaces
and other
Greetings,
while looking at a diff between the 3.0.x branch and trunk for the web
frontend noticed the following snippet (htmlentities is missing in 3.0.x) :
@@ -331,7 +330,7 @@
header (Pragma: no-cache); // HTTP/1.0
if ($debug) {
header (Content-type:
On Wed, Feb 13, 2008 at 10:44:50AM -0800, Bernard Li wrote:
So what's the latest? Can I check in my original patch to the branch?
the original patch should work (if the spaces/tab is correct for the changes
to get_context.php just like it was done in revision 964).
Anything else needs to
On Tue, Feb 19, 2008 at 08:31:22PM +0100, Ulf Lange wrote:
Hi,
I don' t want to get on your nerves, but can somebody checkin the
patches from Micheal(bugid 146)?
I tried to merge Michael metrics changes into the current implementation in
trunk, but it would be far easier if they will be
On Wed, Feb 20, 2008 at 12:18:41PM -0800, Bernard Li wrote:
On 2/20/08, Martin Knoblauch [EMAIL PROTECTED] wrote:
I just committed the fix for the spoofing leak from Martin Hicks. Can you
run a [final] snapshot for 3.0.7? I have something brewing to fix the
petabyte/sec spikes that
On Sat, Feb 23, 2008 at 01:13:44PM -0500, Jesse Becker wrote:
On Sat, Feb 23, 2008 at 1:07 PM, Carlo Marcelo Arenas Belon
[EMAIL PROTECTED] wrote:
anyone knows what happened here?
It's probably a sever-side issue. I can duplicate the problem, and
there are a number of people having
Use a local perfstat_cpu_total_t instead of the global cpu_total_buffer.
Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED]
---
trunk/monitor-core/libmetrics/aix/metrics.c | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/trunk/monitor-core/libmetrics/aix
add error checking and a local struct instead of the cpu_total_buffer global
using unknown in case of failure to avoid overflow and for consistency
with other platforms
Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED]
---
trunk/monitor-core/libmetrics/aix/metrics.c |9
-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED]
---
trunk/monitor-core/libmetrics/aix/metrics.c | 90 +++---
1 files changed, 52 insertions(+), 38 deletions(-)
diff --git a/trunk/monitor-core/libmetrics/aix/metrics.c
b/trunk/monitor-core/libmetrics/aix/metrics.c
index
merge fixes into one single function for boottime and creates a global
a global boottime is needed for previous patches so this might need to me
moved up.
Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED]
---
trunk/monitor-core/libmetrics/aix/metrics.c | 32
The biggest of all patches, but the less likely to generate any problems
It could be skipped, but then the series will have 13 patches, and you know
what they say about that number and it makes the code nicer to read and
easier to extend.
Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL
On Mon, Feb 25, 2008 at 10:56:33AM -0600, Martin Hicks wrote:
Ah okay. I don't see those warnings. Thanks for the update.
Committed revision 1003.
Carlo
-
This SF.net email is sponsored by: Microsoft
Defy all
On Fri, Feb 29, 2008 at 01:37:29PM +, Stu Teasdale wrote:
Not 100% true but I am having to keep 3.0 releases out of debian for now
for a variety of reasons. I think the worst of the licence problems are
cured in current HEAD, but confirmation that I've not missed anything
would be
On Sun, Mar 02, 2008 at 01:34:35PM -0600, Michael Sternberg wrote:
Here's a simple patch for gmond/gmond.c to chop domain names off the
ganglia web interface
why not doing the change in the web interface then?
Carlo
-
Greetings,
after removing expat from srclib, I'd seen the following problems with the
current implementation of --enable-static-build :
1) needed for bootstrap, otherwise the package generated will have an empty
srclib (this is most likely a bug and not documented in README.svnusers)
2) makes
On Mon, Mar 03, 2008 at 11:16:10AM -0700, Brad Nicholes wrote:
On 3/3/2008 at 1:14 AM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
this means IMHO that supporting static builds without srclib won't be
possible
(specially considering that apr
On Sat, Mar 08, 2008 at 10:21:01AM +0800, Cherife Li wrote:
make[4]: *** No rule to make target `../../lib/.libs/libganglia.a',
needed by `test-metrics'. Stop.
make[4]: Leaving directory `/root/ganglia-3.0.7/libmetrics/tests'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory
On Wed, Mar 12, 2008 at 02:20:45PM -0600, Brad Nicholes wrote:
I think I am going to see some more clarification on this. --enable-static
doesn't seem to fit exactly with what we want to do. According to the
automake docs (http://sourceware.org/autobook/autobook/autobook_85.html),
(and optionally installed)
libconfuse static library (the default) using --with-libconfuse.
Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED]
---
trunk/monitor-core/README.WIN |2 +-
trunk/monitor-core/configure.in| 38 +---
trunk
that cygwin won't be able to generate working binaries for ganglia
because of DSO linking problems in cygwin as reported in :
http://www.mail-archive.com/ganglia-developers@lists.sourceforge.net/msg03375.html
Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED]
---
trunk/monitor-core
On Thu, Mar 13, 2008 at 11:39:48AM -0600, Brad Nicholes wrote:
I think I have everything working now using your patches.
got a fair number of conflicts when merging them back with the most
interesting ones being :
1) I rearranged some of the values for GCFLAGS/GLDADD/GLDFLAGS out of the if
On Thu, Mar 13, 2008 at 01:42:05PM -0600, Brad Nicholes wrote:
I think that with the removal of the srclib directory from the SVN trunk
repository, we have completed everything that we thought needed to be done
before creating the 3.1.x stable branch.
Agree (sorta), as I was expecting
On Fri, Mar 14, 2008 at 09:04:54AM -0700, Witham, Timothy D wrote:
I haven't used 3.1 yet but I see memory changes to float. That is good.
However, old 3.0 gmonds could still be reporting integers. Are you sure
that gmetad can successfully get one correct total from two data types?
this as
On Sat, Mar 15, 2008 at 04:08:25PM -0600, Brad Nicholes wrote:
[EMAIL PROTECTED], Bernard Li
[EMAIL PROTECTED] wrote:
./bootstrap ./configure --with-gmetad make dist-gzip
After I run the command, I get a tarball and a lot of cruft in the
checkout directory. Even after I deleted
On Fri, Mar 14, 2008 at 10:04:09PM -0500, Carlo Marcelo Arenas Belon wrote:
On Fri, Mar 14, 2008 at 05:33:14PM -0700, Bernard Li wrote:
On 3/14/08, Bernard Li [EMAIL PROTECTED] wrote:
Got the following error w/ make:
make[4]: Entering directory `/root/code/ganglia/libmetrics/tests
On Sun, Mar 16, 2008 at 12:18:05PM -0400, Jesse Becker wrote:
Now that srclib/ has been yanked, I tried building a clean copy of
trunk on my OpenBSD box, and hit a few snags.
First, running ./bootstrap fails:
haven't tried bootstrap ('cause bootstrapping only works well in linux anyway)
but
On Sun, Mar 16, 2008 at 06:27:00PM -0400, Jesse Becker wrote:
So then the build process for trunk is basically broken for non-Linux systems?
it always was; before it used to work reliably only in fedora or centos AFAIK
which is what we had been using to do bootstraps and releases for all the
On Sun, Mar 16, 2008 at 09:16:01PM -0400, Jesse Becker wrote:
On Sun, Mar 16, 2008 at 8:26 PM, Carlo Marcelo Arenas Belon
[EMAIL PROTECTED] wrote:
On Sun, Mar 16, 2008 at 06:27:00PM -0400, Jesse Becker wrote:
So then the build process for trunk is basically broken for non-Linux
On Mon, Mar 17, 2008 at 03:39:05PM -0600, Brad Nicholes wrote:
In other words, the 3.1.x gmetad and frontend is able to acquire data from a
3.0.x cluster and a 3.0.x gmetad and frontend is able to acquire data from
3.1.x cluster. So the cluster by cluster migration from 3.0.x to 3.1.x should
Greetings,
is there any good reason why building the python module wrapper is not
enabled by default?, there will always of course the option of disabling it
using --disable-python at configure time but will had the following advantages
compared to the current situation :
1) make dist* will
On Tue, Mar 18, 2008 at 07:04:10PM -0700, Bernard Li wrote:
On 3/18/08, Carlo Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
is there any good reason why building the python module wrapper is not
enabled by default?, there will always of course the option of disabling it
using
On Tue, Mar 18, 2008 at 12:11:44PM -0700, Witham, Timothy D wrote:
Well, the patch I attached to bug#128 *does* convert all ints to floats
when doing sums, but it has not yet been applied to trunk as of 1078.
Part of it committed in revision 1083
Maybe you need me to enter a new bug for
On Wed, Mar 19, 2008 at 09:11:28AM -0600, Brad Nicholes wrote:
Carlo,
Params is actually a valid configuration directive. There are two ways to
pass in parameters into a gmond module. One way is to use a raw string with
the Params directive (mod_python uses this for the module path
On Wed, Mar 19, 2008 at 04:58:10PM -0700, Bernard Li wrote:
On 3/19/08, Bernard Li [EMAIL PROTECTED] wrote:
Should I add scoreboard.h to EXTRA_DIST??
This, however, does not install scoreboard.h to /usr/include which the
RPM spec expects to -- please advise.
Committed revision 1092
On Wed, Mar 19, 2008 at 01:35:26PM -0400, Jesse Becker wrote:
Guess I need to remove my copy of cool-stuff then.
`svn update` should take care of that
What about creating a contrib directory? This is common in many
different projects, and frequently used for files such as this. I've
yet
On Wed, Mar 19, 2008 at 05:29:50PM -0700, Bernard Li wrote:
In configure.in, the way to determine GANGLIA_NANO_VERSION is to run
svnversion. This works in the SVN checkout, but would fail when tried
to be executed from the tarball (instead of the SVN revision, you get
exported).
or you get a
On Wed, Mar 19, 2008 at 03:09:01PM -0500, James Richardson wrote:
I am I correct to interpret this thread as a stable v3.1 branch will be
available any day now?
yes, with some alpha packages released for testing once we iron up all build
issues (hopefully this week or next) and in preparation
On Wed, Mar 19, 2008 at 11:55:51AM -0700, Bernard Li wrote:
they can't be committed as-is either because the assumptions made are
configurable and users will expect that they match (regardless of how many
notes about not being supported are added) but will be useful if ever make
is
On Thu, Mar 20, 2008 at 10:26:36AM -0700, Witham, Timothy D wrote:
Without this, folks would have to upgrade
all gmonds to 3.1 simultaneously or they would get incorrect totals.
I haven't yet validated this by testing the setup, but from my understanding
of the code, you should be able to
On Thu, Mar 20, 2008 at 10:26:31AM -0600, Brad Nicholes wrote:
1. Remove the pymodule configuration directive
+1
2. Add an optional type name to a module section to designate the interface
type.
module { } - #No designation, the default is C
module Python {} - #Python type module
On Thu, Mar 20, 2008 at 05:35:44PM -0600, Brad Nicholes wrote:
I just checked in the changes to remove the pymodule section directive and to
introduce the language directive.
great
I am still working on the documentation for the directive change and how to
write modules in C and python.
On Fri, Mar 21, 2008 at 08:38:20AM -0600, Brad Nicholes wrote:
Isn't that what a no-arch devel package should be doing? Maybe I am missing
what you are really talking about.
a gmond C module needs to include gm_mmn.h (for the MAGIC) and metric.h (for
the mmodule_struct), but both are gmond
On Thu, Mar 20, 2008 at 02:25:36PM -0700, Bernard Li wrote:
This has the added option of allowing the person doing the snapshot
release to add additional tags to the SVN version (eg.
GANGLIA_NANO_VERSION = 1090-perlmodules)
the ganglia version is used for the version for libganglia and
On Thu, Mar 27, 2008 at 08:07:49AM -0700, Witham, Timothy D wrote:
If we remove the conditional the lock should silently fail and so we
would get the same result. This would make the code a bit simpler and
I'm OK with that.
Committed revision 1138.
But it looks like the only way it can
On Thu, Mar 27, 2008 at 01:07:53PM -0700, Bernard Li wrote:
I actually like the format better as it tells the exact Ganglia
version number:
the man pages generated with help2man also include the version number
but were post edited to remove that, as they are meant to be
generic enough to be
On Thu, Mar 27, 2008 at 12:58:53PM -0700, Bernard Li wrote:
Hi guys:
On Thu, Mar 27, 2008 at 11:32 AM, Carlo Marcelo Arenas Belon
[EMAIL PROTECTED] wrote:
On Thu, Mar 27, 2008 at 08:07:49AM -0700, Witham, Timothy D wrote:
If we remove the conditional the lock should silently fail
On Fri, Mar 28, 2008 at 03:57:40PM -0600, Brad Nicholes wrote:
I new there was something that I was missing when I checked in the header
file changes. It appears that I broke the 32-bit builds. I am trying to get
the problem fixed and it has to do with adding -D_LARGEFILE64_SOURCE to the
On Fri, Mar 28, 2008 at 11:44:44PM -0600, Brad Nicholes wrote:
The problem is the typedef of g_val_t. Both libmetrics and the metric
modules need this typedef. Before the commits, the metric modules had to
include libmetrics.h to get this typedef. This would have required
libmetrics.h
On Sat, Mar 29, 2008 at 05:29:15PM -0600, Brad Nicholes wrote:
On 3/29/2008 at 2:43 AM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
g_val_t is a public interface of libganglia and libganglia is used by the
modules so it should be IMHO moved
On Sun, Mar 30, 2008 at 10:56:31AM -0600, Brad Nicholes wrote:
On 3/29/2008 at 11:31 PM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
IMHO, even if libganglia uses apr internally having it also to require apr
headers might not be a good idea
On Mon, Mar 31, 2008 at 06:31:46PM -0600, Brad Nicholes wrote:
I have checked in the code that removes APR from all of the public header
files except for gm_metric.h which is needed by the modules. Everything else
should be APR-free. This includes libmetrics, gmetric, gmetad, gstat and
On Tue, Apr 01, 2008 at 04:44:26PM -0700, Bernard Li wrote:
Carlo, it looks like r1179 broke configure for me:
http://ganglia.svn.sourceforge.net/viewvc/ganglia/trunk/monitor-core/configure.in?r1=1179r2=1178pathrev=1179
This results in passing -L with no arguments:
Committed revision
On Tue, Apr 01, 2008 at 03:45:58PM -0600, Brad Nicholes wrote:
On 3/18/2008 at 1:58 AM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
On Mon, Mar 17, 2008 at 04:12:38PM -0600, Brad Nicholes wrote:
I just submitted the patches to remove the APR 1.2
On Tue, Apr 01, 2008 at 08:32:49AM -0600, Brad Nicholes wrote:
On 3/31/2008 at 9:48 PM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
On Mon, Mar 31, 2008 at 06:31:46PM -0600, Brad Nicholes wrote:
I have checked in the code that removes APR from all
On Tue, Apr 01, 2008 at 03:24:48PM -0700, Bernard Li wrote:
Since r1066, Carlo has deleted all the files inside the top level
config/ directory -- is there a point to keep this directory then?
Since it is auto-generated during bootstrap, I suggest we remove it.
Deleting config
Deleting
On Wed, Apr 02, 2008 at 10:08:04AM -0600, Brad Nicholes wrote:
On 4/1/2008 at 10:28 PM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
ok, but then if the application also has a need to access the metric data
(like gmond) then we will end up including
On Wed, Apr 02, 2008 at 03:25:23PM -0600, Brad Nicholes wrote:
On 4/2/2008 at 12:27 PM, in message
[EMAIL PROTECTED], Bernard Li
[EMAIL PROTECTED] wrote:
On Tue, Apr 1, 2008 at 7:28 PM, Carlo Marcelo Arenas Belon
[EMAIL PROTECTED] wrote:
1064carenas if test -f $libaprpath
On Wed, Apr 02, 2008 at 11:18:53AM -0700, Bernard Li wrote:
Hi Carlo:
On Tue, Apr 1, 2008 at 8:56 PM, Carlo Marcelo Arenas Belon
[EMAIL PROTECTED] wrote:
there is also the little inconvenience with the building of snapshots which
will need to be addressed since we are planning
On Wed, Apr 02, 2008 at 04:31:37PM -0600, Brad Nicholes wrote:
On 4/2/2008 at 2:43 PM, in message [EMAIL PROTECTED], Carlo
Marcelo Arenas Belon [EMAIL PROTECTED] wrote:
On Wed, Apr 02, 2008 at 10:08:04AM -0600, Brad Nicholes wrote:
On 4/1/2008 at 10:28 PM, in message [EMAIL PROTECTED
On Wed, Apr 02, 2008 at 05:40:04PM -0700, Gilad Raphaelli wrote:
Would that look something like this?
/opt/local/bin/gcc-mp-4.3 -dynamiclib ${wl}-flat_namespace ${wl}-undefined
${wl}suppress -o .libs/libganglia-3.1.0..0.0.0.dylib .libs/gm_protocol_xdr.o
.libs/become_a_nobody.o
On Wed, Apr 02, 2008 at 08:55:58PM -0700, Gilad Raphaelli wrote:
And some more problems. I'm getting this building gmond, gmetad, and
gmetric. gmond, for example, says:
/bin/sh ../libtool --tag=CC --mode=link /opt/local/bin/gcc-mp-4.3 -I../lib
-I../include/ -I../libmetrics
On Wed, Apr 02, 2008 at 09:31:35PM -0700, Witham, Timothy D wrote:
While looking at this, I think I see an area for efficiency improvement.
gmetad hierarchical setups has several areas for efficiency improvement.
this one (doing summaries from partial summaries in the leafs) being one
that was
1 - 100 of 385 matches
Mail list logo
