Re: [Ganglia-developers] Query From-To

2007-04-24 Thread Richard.Grevis
user base is experienced with ganglia, some are reducing their poll rate (Some users want capacity trending only, say). Richard Grevis Production Architecture Barclays Capital, Canary Wharf, London, E14 4BB *DDI : +44 (0) 20 7773 4915 * richard.grevis

Re: [Ganglia-developers] Query From-To

2007-04-24 Thread Richard.Grevis
, some are reducing their poll rate (Some users want capacity trending only, say). Richard Grevis Production Architecture Barclays Capital, Canary Wharf, London, E14 4BB *DDI : +44 (0) 20 7773 4915 * richard.grevis -Original Message- From

Re: [Ganglia-developers] A native windows gmond

2007-02-08 Thread Richard.Grevis
I thought that the top level license was BSD. Richard Grevis Production Architecture Barclays Capital -Original Message- From: matt massie [mailto:[EMAIL PROTECTED] Sent: 07 February 2007 19:49 To: Ian Cunningham Cc: Grevis, Richard: IT (LDN);

[Ganglia-developers] A native windows gmond

2007-02-07 Thread Richard.Grevis
All, a swiss consultancy has implemented a native windows gmond, and the binaries are in the public domain and free. Follow the trail here: http://aprconsulting.ch/product.htm I believe they are also offering ganglia consulting, support, and customisations for a fee. This daemon is much better

[Ganglia-developers] From/To time spec patch.

2007-02-03 Thread Richard.Grevis
All guru types, as you know I kind of volunteered to fold this mod back into the src tree, but while the way I did it was useful for me (and easy), it may not be what others want. What I did was to have typeable fields as well as the dropdowns,

Re: [Ganglia-developers] Query From-To

2007-01-19 Thread Richard.Grevis
). Richard Grevis Production Architecture Barclays Capital, Canary Wharf, London, E14 4BB *DDI : +44 (0) 20 7773 4915 * richard.grevis -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Martin Knoblauch Sent: 16 January 2007 09:44

Re: [Ganglia-developers] Query From-To

2007-01-18 Thread Richard.Grevis
their poll rate (Some users want capacity trending only, say). Richard Grevis Production Architecture Barclays Capital, Canary Wharf, London, E14 4BB *DDI : +44 (0) 20 7773 4915 * richard.grevis -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Martin

Re: [Ganglia-developers] Query From-To

2007-01-15 Thread Richard.Grevis
Giuliano, and also Matt, Martin, Alex, Bernard, Ben and other guru types, From/to like this perhaps? - http://www.aouk83.dsl.pipex.com/ganglia-toplevel.png and http://www.aouk83.dsl.pipex.com/ganglia-from-to.png http://www.aouk83.dsl.pipex.com/ These are only screen shots. The from/to can be

Re: [Ganglia-developers] Ganglia for Windows help

2007-01-09 Thread Richard.Grevis
7773 4915 * richard.grevis -Original Message- From: Carlo Marcelo Arenas Belon [mailto:[EMAIL PROTECTED] Sent: 08 January 2007 03:22 To: Grevis, Richard: IT (LDN) Cc: [EMAIL PROTECTED]; ganglia-developers@lists.sourceforge.net Subject: Re: [Ganglia-developers] Ganglia for Windows

Re: [Ganglia-developers] [Ganglia-general] Windows port issues

2007-01-08 Thread Richard.Grevis
Architecture Barclays Capital, Canary Wharf, London, E14 4BB *DDI : +44 (0) 20 7773 4915 * richard.grevis -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Martin Knoblauch Sent: 04 January 2007 13:43 To: Vladimir Cc: ganglia-developers

[Ganglia-developers] Windows versions of gmond/gmetric

2007-01-05 Thread Richard.Grevis
7773 4915 * richard.grevis For more information about Barclays Capital, please visit our web site at http://www.barcap.com. Internet communications are not secure and therefore the Barclays Group does not accept legal

Re: [Ganglia-developers] Ugly Font rendering in rrdtool generated images

2006-11-06 Thread Richard.Grevis
Hi, surely it is the bounding box size being wrong in the IMG part of the html, e.g. IMG border=0 HEIGHT=160 WIDTH=395 ALT=Nano MEM SRC=./graph.php?g=mem_reportz=mediumc=Nanom=r=hours=descendinghc= 4st=1162807383 If the dimensions given above differ from the image itself, this would cause

Re: [Ganglia-developers] Ugly Font rendering in rrdtool generatedimages

2006-11-06 Thread Richard.Grevis
* richard.grevis -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Martin Knoblauch Sent: 03 November 2006 18:03 To: Ian Cunningham Cc: ganglia-developers@lists.sourceforge.net Subject: Re: [Ganglia-developers] Ugly Font rendering in rrdtool

Re: [Ganglia-developers] Q: Ganglia on MS Windows platform

2006-09-07 Thread Richard.Grevis
Vitaly, unfortunately the problem lies with cygwin rather than Ganglia. The windows agent is essentially the same as the Linux one (hence not too much development work) and cygwin was the way to make the code work. The author of this did not find out until afterwards how broken cygwins emulation

Re: [Ganglia-developers] 3.0.4 and srclib

2006-09-04 Thread Richard.Grevis
Matt, some comments below. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of matt massie Sent: 03 September 2006 08:36 To: ganglia-developers@lists.sourceforge.net Subject: [Ganglia-developers] 3.0.4 and srclib hey guys- i wanted to

Re: [Ganglia-developers] 3.0.4 and srclib

2006-09-04 Thread Richard.Grevis
Markus, your envionment certainly sounds like it is under control, but in mine, the engineered Linux build has packages that in some instances are really quite old. Old enough for the Ganglia build to barf. In some cases a fix would have been easy because the package was not there at all. And

Re: [Ganglia-developers] 3.0.4 and srclib

2006-09-04 Thread Richard.Grevis
Fair enough, although I did not realise you were talking about apr. I was talking about autoconf, freetype, libart, rrdtool etc, and all of those are installed in /opt/ganglia. kind regards, Richard -Original Message- From: Marcus Rueckert [mailto:[EMAIL PROTECTED] Sent: 04

Re: [Ganglia-developers] Ganglia-developers Digest, Vol 2, Issue 17

2006-07-31 Thread Richard.Grevis
Chuck, out of interest, when you say you use the gmetric interface, are you still doing XDR binary encoding for the core metrics, or are you having all the metrics as sets of strings (ala gmetric)? kind regards, richard -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL

Re: [Ganglia-developers] Linux/cygwin gmond metric poll rate.

2006-07-28 Thread Richard.Grevis
Martin, thanks for the reply. It would make sense that perhaps the code is a legacy of before when collection_groups were implemented (aside: collection groups are smart and obviously necessary, but did I realise this before seeing Ganglia? Nope...) Setting the intervals to 0 seems to work fine.

[Ganglia-developers] Linux/cygwin gmond metric poll rate.

2006-07-24 Thread Richard.Grevis
Guys, the code below is in the cygwin and linux metric.c files. typedef struct { uint32_t last_read; uint32_t thresh; char *name; char buffer[BUFFSIZE]; } timely_file; timely_file proc_stat= { 0, 15, /proc/stat }; timely_file

Re: [Ganglia-developers] api for data gathering using ganglia

2006-07-14 Thread Richard.Grevis
Arun, as an example, run nc localhost 8651 on your ganglia server. That is the grid level raw data that could be used for snapshot processing. If you netcatted directly to a cluster member (or headnode if you unicast), then that provides current values for the cluster. So even though a C API

RE: [Ganglia-developers] Host spoofing for SNMP

2006-05-09 Thread Richard.Grevis
Nah, don't worry. If no firewalls are present, UDP spoofing is trivial anyway, so you are not making the UDP gmond comms materially less secure. kind regards, richard -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Adeyemi Adesanya Sent: 08 May 2006

RE: [Ganglia-developers] Trigger on defined conditions

2006-04-24 Thread Richard.Grevis
Miguel, you may want to consider parsing the XML for all your clusters on your server (by connecting to localhost oprt 8651). This perl for example prints hosts that are down: #!/usr/bin/perl open (FD, nc localhost 8651 |) or die

[Ganglia-developers] Patch for clickable bigger graphs.

2006-04-20 Thread Richard.Grevis
All, the attached patch file can be applied to the php web code to enable you to click on the smaller graphs in the cluster and host views, and then get a bigger graph. It is based on the 3.0.3 release, but you prpbably patch the 3.0.2 release too, as little has changed. try: patch -p1

RE: [Ganglia-developers] Ganglia 3.0.3 status

2006-04-06 Thread Richard.Grevis
Hi Martin, I would be keen on a new tarball also. Will this tarball include Matt's fix to the XML truncation problem? (I think it is in CVS, so I assume yes). Cheers, Richard -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Martin Knoblauch Sent: 06

RE: [Ganglia-developers] XML stream from gmond made block

2006-04-05 Thread Richard.Grevis
Although looking at the fairly scary apr code, it rather looks like the function has different semantics between wnidows and UNIX - windows supporting blocking IO with a timeout, but the UNIX case kind of looks like blocking implies no timeout. hmm... - richard -Original Message- From:

RE: [Ganglia-developers] gmond UDP data coming from 2 NICs on a host

2006-03-29 Thread Richard.Grevis
Martin, yes, that is a possibility, but the downside is potentially having a unique gmond.conf for every host in cluster. And assuming you don't want to explicitly tag every packet with the hostname, then you would need to ensure that the my hostname is packet gets sent first, even then, if the

[Ganglia-developers] Truncated XML again

2006-03-22 Thread Richard.Grevis
Matt/Martin and all. I am finding that I am still getting occassional truncated XML from gmond, even after the EAGAIN patches to gmond.c. Interestingly, when the data was truncated, it ended with a /HOST tag. i.e. a host boundary. Looking at the code, I see this: snip /* Walk the host hash */

[Ganglia-developers] Possible bug in hosts up calculation when federating clusters.

2006-03-21 Thread Richard.Grevis
All, when I had debugging turned on in gmetad, the daemon was announcing data sources as old, when they were not. Looking at the code, 3.0.2 or 3.0.3 gmetad/process_xml.c, line 821 or so, function startElement_GANGLIA_XML: snip if (xt-tag == VERSION_TAG) {

RE: [Ganglia-developers] Possible bug in hosts up calculation when federating clusters.

2006-03-21 Thread Richard.Grevis
Martin, no doubt you will see matt's mail soon, we are doing a bit of mail tag. The version I already tested had an explicit assignment to 0, because I was pretty sure it was needed. I will happily go with whatever code version you and Matt prefer. In the meantime I am patched and happy. kind

RE: [Ganglia-developers] Possible bug in hosts up calculation when federating clusters.

2006-03-21 Thread Richard.Grevis
Hi, just tested, yup - it works, and you are right, strcmp is required. Thanks. But I still think you need the else to set xmldata-old to 0. If not done, surely this will mean that the variable gets stuck on 1 for subsequent clusters/grids after the first old cluster is encountered. There is no

RE: [Ganglia-developers] RE: First prerelease of ganglia-3.0.3 ready for testing

2006-03-20 Thread Richard.Grevis
This was the kind of question I was going to ask - for the 500-750k loops, how long did this take in real time? maybe a nanosleep is the go, but I have not tested the call nuder cygwin. regards, richard -Original Message- From: Martin Knoblauch [mailto:[EMAIL PROTECTED] Sent: 17 March

RE: [Ganglia-developers] RE: First prerelease of ganglia-3.0.3 ready for testing

2006-03-16 Thread Richard.Grevis
Martin, yes, quite right - it would be a hard loop until EAGAIN went away. As I do 5 second polling rather than 20, I would prefer not to put a simple sleep in. One could always have a counter timeout, or a timeout after (say) 1-2 seconds of looping. In fact I am nervous now, so I think I will

RE: [Ganglia-developers] RE: First prerelease of ganglia-3.0.3 ready for testing

2006-03-15 Thread Richard.Grevis
Gee, I thought that was fixed with this patch: http://bugzilla.ganglia.info/cgi-bin/bugzilla/show_bug.cgi?id=50 Actually, looking at 3.0.3 gmond.c, it looks like the patch did not make it into the release - that's a shame. Even looking at the patch, it looks as if it is a partial fix, because

RE: [Ganglia-developers] (no subject)

2006-03-01 Thread Richard.Grevis
oops -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of James Mcininch Sent: 01 March 2006 14:40 To: ganglia-developers@lists.sourceforge.net Subject: [Ganglia-developers] (no subject) I want

RE: [Ganglia-developers] hyperthreading cpu_num patch.

2006-03-01 Thread Richard.Grevis

RE: {Spam?} [Ganglia-developers] Re: [Ganglia-general] Early termination of XML stream from a windows based agent.

2006-02-13 Thread Richard.Grevis
Thank you all very much, I am not alone after all. I will digest the info, apply the patch, and test the code. regards, Richard -Original Message- From: Ian Cunningham [mailto:[EMAIL PROTECTED] Sent: 10 February 2006 21:21 To: Jason A. Smith Cc: Grevis, Richard: IT (LDN); Ganglia

[Ganglia-developers] when not all node graphs of a cluster are displayed, and rollup graphs don't get updated.

2005-11-24 Thread Richard.Grevis
I had this problem for a cluster. Ganglia version 3.0.1 In the cluster view, only some hosts had graphs, ans the rollup displays either were not created at all, or were not updated. It turned out it was caused by two nodes in my cluster (different Ips), having the same name upon reverse lookup. a

RE: [Ganglia-developers] Re: [Ganglia-general] windows gmond client

2005-11-11 Thread Richard.Grevis
Thank you so very much for this matt. We have some windows experts - but I am not one of them. We have enough interest for there now to be an internal process finding the funding for the programming work (yes, I work in a bank), We have people for example that know how to use the xyz.dll direct