-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Rick Cobb:
We had the same problem with gmond 3.0.4 on Solaris 10 / x86. As far
as we were able to debug, it's a bug in Solaris itself, and
particularly with the interaction between IGMPv3 support in the kernel
and switches that only do
Hi everyone,
Ok I got my Ganglia monitor up and working, and it was pulling results
from the localhost.
So I enable the hadoop-metrics.properties and made the appropriate
changes so that it pointed at me ganglia box.
I made a data_source in the gmetad.conf file, and attached the two
test
Ok.
I just ran a 'gstat --all'
And only one host comes up, just the localhost.
So there is something missing.
any ideas?
-John
On Nov 17, 2009, at 9:22 AM, John Martyniak wrote:
Hi everyone,
Ok I got my Ganglia monitor up and working, and it was pulling
results from the localhost.
So the udp_recv_channel in the gmond.conf file is as follows:
udp_recv_channel {
mcast_join = 239.2.11.71
bind = 239.2.11.71
port = 8649
}
if I change that to the ip address of the monitoring master machine, I
get an error that it can't join the cast or something
On Tue, 17 Nov 2009, John Martyniak wrote:
It should pretty much work out of the box John. Does your network not
allow multicasting?
So the udp_recv_channel in the gmond.conf file is as follows:
udp_recv_channel {
mcast_join = 239.2.11.71
bind = 239.2.11.71
port
It should, I don't restrict anything, and I have the firewalls turned
off on those two machines.
It is on a private network that I use NAT through my router to get to
the outside world. But that shouldn't matter because all of the
machine can get out to the internet.
-John
On Nov 17,
On Tue, 17 Nov 2009, John Martyniak wrote:
Are the monitored nodes on the same side as the monitoring node? If
not you might have to explicitly turn on mulicasting in the router.
Depends on the router.
It should, I don't restrict anything, and I have the firewalls turned off on
those
Yes they are all in the same subnet, all attached to the same switch.
monitor is: 10.1.1.25
the two devices are 10.1.1.128, 10.1.1.129
I tried the telnet test also: from each of the machines that are
monitored, I ran telnet 10.1.1.25 8649, and received the XML file.
-John
On Nov 17, 2009,
Beginner quetion:
how do I run gmetad in -d mode? I have been using /etc/rc.d/init.d/
gmetad start|stop|restart
-John
On Nov 17, 2009, at 11:07 AM, Chris Johnson wrote:
On Tue, 17 Nov 2009, John Martyniak wrote:
And they all are configured with the same grid name? Another
thing to
Hi Chris:
On Mon, Nov 16, 2009 at 11:37 AM, Chris Johnson
john...@nmr.mgh.harvard.edu wrote:
So I installed php-gd. Still just says Pie Chart though.
Anything I should do? Any logs to look at?
Have you tried re-starting apache? ;-)
Cheers,
Bernard
Hi Brad:
On Mon, Nov 16, 2009 at 1:06 PM, Brad Nicholes bnicho...@novell.com wrote:
I think I have all of the wiki page links fixed up. Especially on the
installation and configuration page. I also fixed up some links to the misc.
documents about Ganglia and monitoring. If anyone
How do I set the grid name?
Because these are hadoop machines so I used the following
configuration parameters in my hadoop-metrics.properties files:
dfs.class=org.apache.hadoop.metrics.ganglia.GangliaContext
dfs.period=10
dfs.serve...@ganglia@:8649
I too have been bangin my head on this for a few weeks. After much googling
i cannot seem to find the answer, so i hope someone (developer maybe) can
help.
I was successfully using ganglia 2.5 and 3.0.x. At some point i upgraded to
3.1.x and things went sour. I've even tried to revert back to
On 11/17/2009 at 10:04 AM, in message
b1eec58d0911170904r2f2613ads9244341a82b85...@mail.gmail.com, Ryan Robertson
89esp...@gmail.com wrote:
I too have been bangin my head on this for a few weeks. After much googling
i cannot seem to find the answer, so i hope someone (developer maybe) can
When I run it with gmetad --debug=5:
I get the following:
[r...@monitor ~]# gmetad --debug=5
Going to run as user nobody
Sources are ...
Source: [Weive cluster, step 15] has 2 sources
10.1.1.129
10.1.1.130
xml listening on port 8651
interactive xml listening on port 8652
cleanup
Dear all:
Just a quick update -- I've talked to Matt and a few others and it
looks like late January would actually work best for everybody. So
right now let's set the date tentatively to the weekend of Jan 18,
2010.
Since I'm still gauging interest, for those of you who haven't
responded yet,
On Tue, 17 Nov 2009, Bernard Li wrote:
DOH! Thanks.
Hi Chris:
On Mon, Nov 16, 2009 at 11:37 AM, Chris Johnson
john...@nmr.mgh.harvard.edu wrote:
So I installed php-gd. Still just says Pie Chart though.
Anything I should do? Any logs to look at?
Have you tried re-starting
Ahh yes, i knew there was one other telnet snippet question. I am able to
telnet to localhost 8652 and feed it
/?filter=summary
I get outputthe output scrolled off the screen, but you get the idea
that it's returning...
--snip-
/METRICS
METRICS NAME=swap_total SUM=2019320 NUM=1
Sounds to me like it could be a file permissions problems then. Is your apache
server able to access the rrd files and/or port 8652?
On 11/17/2009 at 1:00 PM, in message
0016e64c2536e598710478969...@google.com, 89esp...@gmail.com wrote:
Ahh yes, i knew there was one other telnet snippet
I'm not sure if this is related to your issue, but it seems possibly
related... Last summer, with Ganglia 3.1.0, I found that bind
either does not work in a multicast _recv_channel, only in
_send_channel ... or the other way 'round. I forget which it was, but
it certainly did not work in one of
I was wondering if it is possible and if so how to add conditional
statements in Ganglia Web templates. What I am after is that I have some
custom consolidated reports like the ones from here
http://vuksan.com/linux/ganglia/#Apache_Traffic_Stats
Currently I modified the template to include the
So do the ip address need to be real ip addresses that are in the
multi-cast IP? It is currently set to 239.2.11.71, which isn't a real
ip address on my network, does it need to be?
I tried changing the hadoop-metrics.properties to that value and it
did not have any results. gmetad
Yes. We would see the traffic on other machines, but we would not see
multicast traffic coming into the machine we were using to aggregate
metrics. Restarting gmond would get the traffic flowing back in.
-- ReC
On Nov 17, 2009, at 4:17 AM, River Tarnell wrote:
-BEGIN PGP SIGNED
rrd dir and subdirs are owned by nobody.
ls -ld /var/lib/ganglia/rrds
drwxr-xr-x 7 nobody nobody 4096 May 28 2008 /var/lib/ganglia/rrds
ls -l /var/lib/ganglia/rrds
drwxr-xr-x 7 nobody root 4096 Sep 28 15:36 595
drwxr-xr-x 2 nobody root 4096 Sep 23 10:04 __SummaryInfo__
drwxr-xr-x 33 nobody root
Has anybody else had any trouble running nutch 0.19.2 with Ganglia
3.1.3?
I was surfing through Jira and it seems that there where some issues
but they have been resolved.
Any thoughts would be helpful.
Thank you,
-John
John Martyniak
President/CEO
Before Dawn Solutions, Inc.
9457 S.
try this command
#gstat --all -i a_hostname_in_cluster
Chifeng
On Tue, Nov 17, 2009 at 11:02 PM, John Martyniak
j...@beforedawnsolutions.com wrote:
Ok.
I just ran a 'gstat --all'
And only one host comes up, just the localhost.
So there is something missing.
any ideas?
-John
On Nov
26 matches
Mail list logo
