[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [HTCondor-users] HTCondor in an OpenStack instance
- Date: Tue, 9 Aug 2022 13:56:10 -0500 (CDT)
- From: Todd L Miller <tlmiller@xxxxxxxxxxx>
- Subject: Re: [HTCondor-users] HTCondor in an OpenStack instance
Right now, I just want to be able to start up a pool on an m6i.xlarge
instance to run my own testing or teach researchers how to use HTCondor.
Ah. If you're not trying to spin up or tear down the pool from
HTCondor, and you're not (presently) interested in dynamically expanding
said pool with OpenStack instances, I would recommend just spinning up a
VM (or three -- make sure they share a subnet and configure their
firewalls to allow inbound port 9618*) and pretending they're physical
infrastructure for purposes of following the manual's usual instructions
(use get_htcondor unless you have a good reason).
It can be helpful to split the exec node away from the submit node
when teaching because that will force the user to be "good" about
specifying file transfer. On the other hand, a "mini" (all-in-one)
installation on a VM could easily be turned into an AMI-equivalent and
might be easier for a learner to start self-service, if that's something
you want them to be able to do.
*: and port 22 for SSH, of course.