[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [HTCondor-users] BOSCO question
- Date: Tue, 05 Sep 2017 11:08:39 -0500
- From: Greg Thain <gthain@xxxxxxxxxxx>
- Subject: Re: [HTCondor-users] BOSCO question
On 09/05/2017 09:50 AM, Zhuo Zhang wrote:
Thanks for the reply. But unfortunately flocking is not what I am
looking for. Jobs from pool A flock to pool B will have to go through
machine A first, which is still a single point of failure.
Ah, if you need a high-availability solution, there are a number of
solutions. In your case, I know that your condor jobs are part of a
production system that generates jobs from outside the condor system,
and high availability needs to be designed in that context.
For a single schedd, perhaps the easiest way to high availability is to
run on top of a virtual machine solution with live migration.