search

michellab / Cluster

This repository is used for tracking any issues regarding the cluster
2 stars 0 forks source link

Ganglia #procs #9

Open jmichel80 opened 9 years ago

jmichel80 commented 9 years ago

Why did the ganglia switched from reporting 324 available procs to 180 procs on Friday ?

All hosts appear up.

http://section6.chem.ed.ac.uk/ganglia/?r=week&cs=&ce=&c=section9&h=&tab=m&vn=&hide-hf=false&m=load_one&sh=1&z=small&hc=4&host_regex=&max_graphs=0&s=by+name

ppxasjsm commented 9 years ago

I have no idea. I'll try and find out.

-Toni

From: jmichel80 notifications@github.com Sent: 28 September 2015 12:00 To: michellab/Cluster Subject: [Cluster] Ganglia #procs (#9)

Why did the ganglia switched from reporting 324 available procs to 180 procs on Friday ?

All hosts appear up.

http://section6.chem.ed.ac.uk/ganglia/?r=week&cs=&ce=&c=section9&h=&tab=m&vn=&hide-hf=false&m=load_one&sh=1&z=small&hc=4&host_regex=&max_graphs=0&s=by+name

Reply to this email directly or view it on GitHubhttps://github.com/michellab/Cluster/issues/9.

The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

ppxasjsm commented 9 years ago

It looks like it isn't processors available, but loads/processor is the legend on the graphs. There should be 292 CPUs listed as online, so does not correspond to any of the data. Therefore I assume they must measure something different and I am not 100% sure what/how this is done. The spike might be due to the fact that slurm was actually disabled during that week. I can't think of any other explanation.

jmichel80 commented 9 years ago

ok, sounds harmless anyway.

Dr. Julien Michel, Royal Society University Research Fellow Room 263, School of Chemistry Joseph Black Building, University of Edinburgh David Brewster Road

Edinburgh

EH9 3FJ United Kingdom phone: +44 (0)131 650 4797

http://www.julienmichel.net/

On Mon, Sep 28, 2015 at 2:12 PM, ppxasjsm notifications@github.com wrote:

It looks like it isn't processors available, but loads/processor is the legend on the graphs. There should be 292 CPUs listed as online, so does not correspond to any of the data. Therefore I assume they must measure something different and I am not 100% sure what/how this is done. The spike might be due to the fact that slurm was actually disabled during that week. I can't think of any other explanation.

— Reply to this email directly or view it on GitHub https://github.com/michellab/Cluster/issues/9#issuecomment-143739604.

The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.