Re: [HTCondor-users] CCB problems and hight daemon load

[Todd L Miller <tlmiller@xxxxxxxxxxx>]

> Also the Collector, Negotiator and Scheduler get up to a daemon load of
> 100% and condor_q /condor_status became slow. However the machines has
> free resources in memory and CPU.

	All three of those daemons are single-threaded, so this is not 
unexpected.

> The network between the VMs and the Collector looks stable. Our plan is
> to start additional Collectors with CCBs. Would that help? How much
> Collectors do we need and how we should configure our system?

	The configuration at the URL below worked for 5,000 nodes from 
EC2, but except for the "network connectivity" section and the third line 
of the execute node configuration (TCP_FORWARDING_HOST), it applies to 
your situation:

https://htcondor-wiki.cs.wisc.edu/index.cgi/wiki?p=HowToSetUpElasticComputeCloudPools

I'm not sure about how fast we were adding nodes when we tested that, but 
that may be another factor increasing how many additional collectors to 
run; see the section labelled 'Tiered Collectors'.

- ToddM