Hello,
I'm running Ganglia on a couple of cluster (CentOS5 x86_64) and
haven't been able to get any readings for the network on either
cluster. All other graphs work just fine. One of the clusters is
Dell machines and the other is just a bunch of pizza-boxes. All I see
is a flat line at the bottom of the network graph with a very severe
spike (~600m) every now and again. I'm running the following version
of Ganglia (i386) on all machines:
]# rpm -qa | grep gan
ganglia-gmond-3.1.1-1
libganglia-3_1_0-3.1.1-1
ganglia-web-3.1.1-1
ganglia-gmetad-3.1.1-1
When I just start gmond it seems to get data from the network just
fine. After that, as the program runs, the values turn to zero.
Running gmond in debug mode I get the following (extra stuff has been
stripped out).
<snip>
metric 'bytes_out' being collected now
********** bytes_out: 3.481266
metric 'bytes_out' has value_threshold 4096.000000
metric 'bytes_in' being collected now
********** bytes_in: 3.481266
metric 'bytes_in' has value_threshold 4096.000000
metric 'pkts_in' being collected now
********** pkts_in: 0.174656
metric 'pkts_in' has value_threshold 256.000000
metric 'pkts_out' being collected now
********** pkts_out: 0.167840
metric 'pkts_out' has value_threshold 256.000000
</snip>
<snip>
metric 'bytes_out' being collected now
********** bytes_out: 0.000000
metric 'bytes_out' has value_threshold 4096.000000
metric 'bytes_in' being collected now
********** bytes_in: 0.000000
metric 'bytes_in' has value_threshold 4096.000000
metric 'pkts_in' being collected now
********** pkts_in: 859.684706
metric 'pkts_in' has value_threshold 256.000000
metric 'pkts_out' being collected now
********** pkts_out: 285.055977
metric 'pkts_out' has value_threshold 256.000000
metric 'disk_free' being collected now
</snip>
<snip>
metric 'bytes_out' being collected now
********** bytes_out: 0.000000
metric 'bytes_out' has value_threshold 4096.000000
metric 'bytes_in' being collected now
********** bytes_in: 0.000000
metric 'bytes_in' has value_threshold 4096.000000
metric 'pkts_in' being collected now
********** pkts_in: 289.710952
metric 'pkts_in' has value_threshold 256.000000
metric 'pkts_out' being collected now
********** pkts_out: 103.687930
metric 'pkts_out' has value_threshold 256.000000
</snip>
There doesn't seem to be any errors:
]# grep bytes_in gmond.log | grep errors
sent message 'bytes_in' of length 52 with 0 errors
sent message 'bytes_in' of length 52 with 0 errors
sent message 'bytes_in' of length 52 with 0 errors
]# grep bytes_out gmond.log | grep errors
sent message 'bytes_out' of length 56 with 0 errors
sent message 'bytes_out' of length 56 with 0 errors
sent message 'bytes_out' of length 56 with 0 errors
Any help would be greatly appreciated.
-Cesar Delgado
Systems/Network Admin @ Rapleaf
email: [email protected]
-------------------------
We're looking for an Amazing Software Engineers (+ interns):
http://business.rapleaf.com/careers.html
The Rapleaf Bailout Plan - Send a qualified referral (resume) and we
will award you with $10,007 bailout package if we hire that person.
------------------------------------------------------------------------------
Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
software. With Adobe AIR, Ajax developers can use existing skills and code to
build responsive, highly engaging applications that combine the power of local
resources and data with the reach of the web. Download the Adobe AIR SDK and
Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
_______________________________________________
Ganglia-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ganglia-general
