Linux (32 and 64bit).
There is no hurry though, I have collectd installed on machines locally and
it provides io graphs by default. I am just curious because this is one of
basic things about a computer system and I find it weird that ganglia does
not have it included in official distribution
Thanks a lot!
b.
On 2 February 2010 19:28, Sid Stuart sidstu...@gmail.com wrote:
I've written a Python module for Linux that tracks disk io metrics. It is
attached.
--
SOLARIS 10 is the OS for Data Centers -
Hi all!
Is it a known bug that if gmond on head node is restarted, then other
nodes do not reconnect and stop sending data until their gmonds are
also restarted?
This happens by using unicast udp channel, I didn't try multicast.
Thanks,
b.
gmond does not reconnect if gmond at head node
is restarted
Hello, try this in gmond.conf:
globals {
(...)
send_metadata_interval = 600
}
Louis
2010/9/30 Bostjan Skufca bost...@a2o.si:
Hi all!
Is it a known bug that if gmond on head node is restarted, then other
nodes do
Why don't you try starting it directly, with debugging and not in daemonized
mode: /path/to/gmond -f -d10
You will see the error probably. Otherwise check your logs (I know you
know:).
b.
PS: I believe this subsys error comes from your distro's service management
system (not from gmond
I was just about to ask what this error (in subject) means but then I
discovered the solution. The same script for accessing gmetad's interactive
port, which worked fine on another machine (identically configured gmetad),
was causing this error message.
The solution is: when one connects to
no other log entries). It should report something about connection being
reset by peer or something.
If you need even more details, please do ask.
TC,
b.
On 5 November 2010 00:20, Bernard Li bern...@vanhpc.org wrote:
Hi Bostjan:
On Wed, Nov 3, 2010 at 6:14 PM, Bostjan Skufca bost...@a2o.si wrote
It definitely is unclear.
I, for one, did have a bit (large bit:) of a problem with this. If
only faq would say ...or when graphs are not updated or something
similar.
b.
On 17 November 2010 22:36, Cameron L. Spitzer cspit...@nvidia.com wrote:
Just out of curiosity, I followed the link in
We are not using TemplatePower anymore, and have moved to using the
Dwoo package instead.
I would be glad to know this before I started making customizations to
frontend, just 3 weeks ago. I know, my fault, I did not ask about it
:)
Anyway, I seriously hope that Dwoo supports 'if' statement in
, 2010 at 8:26 AM, Bostjan Skufca bost...@a2o.si wrote:
I would be glad to know this before I started making customizations to
frontend, just 3 weeks ago. I know, my fault, I did not ask about it
:)
Sorry for not making the announcement more obvious, but the switch
between TemplatePower to Dwoo
I agree, to the point.
I have my RRAs defined with avg, min and max values and full
resolution is not necessary for 2 whole weeks because of that. Minimum
and maximum values get preserved and I can still see when spikes or
dips have occured.
But you have to modify the frontend's graphing code to
deleted my old rrds.
On 17 December 2010 04:30, Robin Humble robin.humble+gang...@anu.edu.au wrote:
On Fri, Dec 17, 2010 at 02:31:01AM +0100, Bostjan Skufca wrote:
I have my RRAs defined with avg, min and max values and full
resolution is not necessary for 2 whole weeks because of that. Minimum
, Bostjan Skufca bost...@a2o.si wrote:
Hi all (and Bernard),
I've just created a patch that enables correct data collection for
network interfaces on certain systems.
The systems that exhibited missing data symptom have had the
following characteristics:
- 32bit OS
- network interface
2011 01:27, Bernard Li bern...@vanhpc.org wrote:
Hi Bostjan:
On Mon, Mar 21, 2011 at 5:16 PM, Bostjan Skufca bost...@a2o.si wrote:
Heh, my first reaction was He must be joking... :)
Anyway, done. However rebasing patches with SVN is major PITA and you
should consider yourself lucky that I
Hi,
occasionally I notice huge spikes in network graphs in ganglia (petabytes
per second or so). Not sure whether those are caused by gmond restarts or
network interface byte counter overflows or something else.
Is someone else also seeing similar behaviour? Running latest ganglia
(3.1.7).
b.
Mar 2011 20:23:31 +0200, Bostjan Skufca bost...@a2o.si wrote:
Hi,
occasionally I notice huge spikes in network graphs in ganglia
(petabytes
per second or so). Not sure whether those are caused by gmond restarts
or
network interface byte counter overflows or something else.
Is someone
around since 2003. You just have
to live with it, or try to fix it yourself.
-Cameron
Bostjan Skufca wrote:
That really seems to be the case. Speaking out of my head now but it seems
that I only see this on HP DL3x0 with Broadcom Corporation NetXtreme II
BCM5708 Gigabit Ethernet (rev 12
...
--
Martin Knoblauch
email: k n o b i AT knobisoft DOT de
www: http://www.knobisoft.de
- Forwarded Message
*From:* Martin Knoblauch kn...@knobisoft.de
*To:* Bostjan Skufca bost...@a2o.si
*Sent:* Wed, March 30, 2011 11:42:12 AM
*Subject:* Re: [Ganglia-general] Network bytes spikes
I added disk metrics and gmetad just started summing it alltogether. Per
cluster and overall.
b.
On 31 March 2011 00:20, Michael Bravo mike.br...@gmail.com wrote:
Hello,
I have a somewhat noob-ish question, the answer to which I haven't
been able to find after some searching and reading. I
I adjusted frontend to suit my needs. Not particularly hard to do, but it's
a bit messy. Maybe you should look into Web Frontend 2.0.
b.
On 6 May 2011 11:22, Indranil C indran...@rediff.co.in wrote:
Hi All,
Is it possible to generate a graph with details of all hosts in a graph
in the
According to the code it only omits lo* and bond* interfaces but sums
data of all other interfaces and returns it as metric. I think you are the
victim of large counters on 32bit OS. Does gmond collect interface data
correctly if you bring it down and up again which resets it's counters?
If that
!~!~!!!
*Ron Cavallo*
*Sr. Director, Infrastructure*
Saks Fifth Avenue / Saks Direct
12 East 49th Street
New York, NY 10017
212-451-3807 (O)
212-940-5079 (fax)
646-315-0119(C)
*www.saks.com* http://www.saks.com/
--
*From:* Bostjan Skufca [mailto:bost...@a2o.si
I had choppy graphs in ganglia until I defined custom RRAs in gmetad.conf
and dumped/reimported the data.
b.
On 21 February 2012 19:20, Aidan Wong aidanw...@attinteractive.com wrote:
Hi,
I have a centralized gmetad collector that collects metrics for its own
cluster and also pulls
Hi,
there is a way to query gmond headnode for any data, I use it, it works
well.
My question is: is there a way to query local gmond instance for local
data, without it being a headnone?
Further explanation of context:
*Currently* data is being collected by gmetad, which is connecting to
Hi Alex!
On 22 October 2012 18:14, Alex Dean a...@crackpot.org wrote:
Hi Bostjan. Have you considered adding multiple head nodes to each unicast
cluster? This would remove the single points of failure you describe.
1. Each node in the cluster reports its metrics to 2 (or more) nodes.
2.
{
port = 8649
}
Then you can simply telnet localhost 8649 and get metrics just for that
host.
Vladimir
On Mon, 22 Oct 2012, Bostjan Skufca wrote:
Hi,
there is a way to query gmond headnode for any data, I use it, it works
well.
My question is: is there a way to query local gmond
It goes like this:
- Ganglia agents (gmond) push data (over UDP) to designated cluster
headnodes (same gmond daemon, only configured to accept UDP data).
- the most recent data is accumulated in memory on these headnodes
- gmetad must be configured to periodically pull data from one (or more) of
Ludmil,
do you have multiple headnodes? Do they receive data from all the
nodes? If yes, did you verify it (telnet to each headnode to port 8649
and count occurences of HOST... xml tag)?
b.
On 19 August 2015 at 12:01, Ludmil Stamboliyski l.stamboliy...@ucdn.com wrote:
Thank you Dave,
I've
are
present in xml and they get their field Reported changed on every run, so
I think gmond collector works correctly.
2015-08-19 14:48 GMT+03:00 Bostjan Skufca bost...@a2o.si:
Ludmil,
do you have multiple headnodes? Do they receive data from all the
nodes? If yes, did you verify it (telnet
If you need client only (not gmetad), maybe host_sflow is what you are
looking for?
http://www.sflow.net/
b.
On 16 November 2015 at 13:28, Keith Brown wrote:
> Hello,
>
> I am having a hard time trying to compile windows binaries using
> cygwin. Can the developers please
Hi Francis,
ganglia gives you two main pieces of information:
- actually used memory (blue graph)
- used + cached memory (green graph)
Whether you interpret cached memory as free or not, is up to you. Generally
it is free, but if you dump all caches and use them for processes, your
systems will
Hi Alexis,
this is my take on the issue:
Since each cluster has its own headnode(s) and clusters are not shared
among clients, when I need to provide access to metric charts to client, I
set up separate instance of gmetad+webui for given client, on their
infrastructure, and point it to their
32 matches
Mail list logo