ssh-based job communications?

[Hilary Oliver]

Does anyone use the old ssh-based job communications mechanism in Cylc (by which "cylc message" uses ssh to re-invoke itself on the suite host, instead of sending the job status message directly from the job host)?

This is currently supported in case you can't do direct job status messaging from job host to suite host, and you don't want to track job status by one-way polling from the suite host, BUT (unlikely?) you can ssh back from job host to suite host.

We'd like to remove support for this in cylc-8.

Hilary

[Scott Wales]

Hi Hilary,

It looks like this is what we use at NCI

$ cylc get-global-config 

...

[hosts]

    [[localhost]]

        task communication method = ssh
        ...

    [[raijin.*]]

        task communication method = ssh
        ...

Cheers, Scott

Scott Wales | Computational Modelling Systems Specialist

ARC Centre of Excellence for Climate Extremes

School of Earth Sciences

The University of Melbourne

Ph: +61-3-8344-6907  Email: scott...@unimelb.edu.au ORCID

www.climateextremes.org.au

---

From: cy...@googlegroups.com <cy...@googlegroups.com> on behalf of Hilary Oliver <hilary....@gmail.com>
Sent: Wednesday, 6 March 2019 2:02:38 PM
To: cy...@googlegroups.com
Subject: [cylc-dev] ssh-based job communications?

 

--

---
You received this message because you are subscribed to the Google Groups "cylc" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cylc+uns...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Hilary Oliver]

Hi Scott,

OK, next question - do you absolutely need to be using it?!

Hilary

[Hilary Oliver]

(discussion ongoing here: https://github.com/cylc/cylc/issues/2975)

[Mike Bedington]

We also are using this at PML as we couldn't get direct messaging to work with our cluster. We didn't try very hard for more elegant solutions once this worked though so we can probably work out a way without if we take another look.

[Hilary Oliver]

Thanks for the response.  We're considering how to proceed - you can follow on the GitHub Issue: https://github.com/cylc/cylc/issues/2975

On Thu, 7 Mar 2019 at 20:25, Mike Bedington <m.bed...@gmail.com> wrote:

We also are using this at PML as we couldn't get direct messaging to work with our cluster. We didn't try very hard for more elegant solutions once this worked though so we can probably work out a way without if we take another look.

[Declan Valters]

Hi Hilary,

We use this method with many of the Edinburgh Uni servers as I
couldn't seem to get direct messaging configured to work properly. I
guess we could use one-way polling as a fall back...

[Hilary Oliver]

Hi Declan,

Roger that.  Looks like we'll need to keep it, or replace with ZeroMQ over ssh...

Hilary