[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [HTCondor-users] Condor and Ganglia



Great, thanks for that info Tim. I’ll get in touch with our ganglia guru and make a start on this.

 

Just one quick question. We have multiple pools, all reporting to a condorview server.

Is there any point running condor_gangliad on the CM in each pool, as opposed to only

running it on the condorview server. We have >10,000 cores/slots in total over all the pools.

 

Thanks again.

 

Cheers

 

Greg

 

From: HTCondor-users [mailto:htcondor-users-bounces@xxxxxxxxxxx] On Behalf Of Tim Theisen
Sent: Wednesday, 8 October 2014 9:55 PM
To: htcondor-users@xxxxxxxxxxx
Subject: Re: [HTCondor-users] Condor and Ganglia

 

You need to run a single condor_gangliad in a HTCondor pool. So, having it run on the Central Manager is certainly one good option. The condor_gangliad queries the Collector periodically for metrics to report to ganglia.

The condor_gangliad is not a condor specific gmond. It is a daemon that periodically sends metrics to a gmond. So, it is more like daemon running gmetric commands than a gmond daemon.

You do need to have the ganglia software installed wherever the condor_gangliad daemon is running. It uses the ganglia shared libraries to send metrics to a ganglia gmond. If the ganglia shared libraries cannot be found or the entry points don't match up, then condor_gangliad falls back to using the gmetric command to send metrics. However, using the gmetric command is several orders of magnitude slower than using the shared libraries.

Since many ganglia installations use a unicast configuration, we recommend that you configure the condor_gangliad to send the metrics to the gmond where your gmetad is running.

...Tim

On 10/07/2014 08:27 PM, Greg.Hitchen@xxxxxxxx wrote:

I’m looking at integrating Condor and Ganglia within our setup.

 

1. Our organization already has Ganglia setup and going and monitoring ~6 compute clusters.

This is completely outside of the Condor pool/s.

 

2. I know close to zero about ganglia technical details.

 

3. I’ve had a read through some HTCondor docs/presentations/manuals re Ganglia.

 

4. I’m unsure of what exactly is needed.

 

I was envisaging only running condor_gangliad on the Central Manager/s running the Collector daemon.

 

Is condor_gangliad essentially a condor specific gmond daemon? Does ganglia software need to be installed

on the Collector machine as well? So that we’re running a gmond and condor_gangliad?

 

Thanks for any help/info.

 

Cheers

 

Greg




_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users
 
The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/



-- 
Tim Theisen
Release Manager
HTCondor & Open Science Grid
Center for High Throughput Computing
Department of Computer Sciences
University of Wisconsin - Madison
4261 Computer Sciences and Statistics
1210 W Dayton St
Madison, WI 53706-1685
+1 608 265 5736