kmb at mssl.ucl.ac.uk
Thu May 25 06:24:04 PDT 2006
Yep that is essentially what I was thinking something like a simple
callback or polling mechanism. The UWS spec does specify what are the
interface methods namely createJob, startJob, resumeJob, destroy, and
GetResourceProperty(for polling); so a client could easily interface
with this UWS and be able to call and poll the "job" which in this case
is some logging query that eventually goes to vospace. Though I have
not really thought about the JDL much in this scenario.
Yep e-mail might still be nice granted I would rather see this
machine/automated client that kicks off a bunch of logging
request/queries for harvest to all the many VO services give me "one"
nice e-mail that said here are the services that succeeded, here are the
ones that failed, and here is a URL to a really nice constructed graph
of all the VO activity :) versus the potential of a single e-mail
from each of the logging services. Unfortunately I think you could only
recommend that idea for clients not require it.
Eric Saunders wrote:
> Hi Kevin
> What you describe sounds to me like a simple callback mechanism. I
> think it's a good idea. As you say, it allows the client to
> dynamically poll the logging-enabled service for status information,
> which may be important if the client wishes to initiate other jobs or
> services which are strongly dependent on the state of that service.
> However, I think it may still be a good idea to send an explicit
> confirmation message from the service if/when processing is complete,
> for example.
> With regard to the 'UWS pattern' - as I understand it, this is simply
> an OO Factory pattern. Whether you choose to implement the logging
> service using a factory or through some other mechanism (e.g. a single
> collection or aggregate object representing the service) should make
> no difference to the external interface seen by the client. As far as
> the client is concerned, it makes a job request, receives a unique id
> which it uses to check on the status of the job... and that's all it
> knows, or cares.
> Eric Saunders
> eSTAR Project (http://www.estar.org.uk)
> Astrophysics Group
> University of Exeter
> On Thu, 25 May 2006, KevinBenson wrote:
>> On the last day of the IVOA meeting me and Guy Rixon were chatting
>> about the Logging and namely how it will need to do e-mail and how
>> users will check e-mails to see when it is complete or if an error
>> occurred. And if your a machine client you will need to monitor
>> e-mail and/or VOSpace to see when things are complete or error to do
>> things more automated. Also possibility of e-mail systems being down
>> even though rare could be a small/minor problem.
>> A suggestion was made; how about implement the UWS pattern with
>> Logging. What do others think about it?
>> It would mean that clients could do polling to see if a job is
>> complete or error happened, and possibly resume jobs if there was a
>> problem. It might seem a little heavy weight, but since we agreed we
>> need security, vospace, handle long running queries, and a requestid
>> which matches the resource identifier it would seem UWS idea might be
>> a good fit. Plus it is quite likely that libraries to handle UWS
>> will incorporate the security and vospace.
More information about the grid