 |
 |
Hello Daniel, Douglas, and Jacek, The last version of test/runTests.py (and associated python modules) is pushed on branch tickets/2014, some comment have been added to the related 2014 ticket, and this one now needs to be reviewed. (Even if it still need to be a bit rewritten on order to have a smart code) The tests run successfully for case01 small dataset, and also on an inconsistent extract of PT1.1 of 20 MB. Maybe *the fully-automatic loading part of this full-python procedure could be quite easily improved and used in order to automatically load/configure distributed Qserv cluster ?* If you think this is interesting, I propose to first clean and refactor the code, in order to have a data loading procedure easier to understand. Ticket is here. https://dev.lsstcorp.org/trac/ticket/2014#comment:7 Source code is here : https://dev.lsstcorp.org/cgit/LSST/DMS/qserv.git/tree/tests?h=tickets/2014 A quick an dirty user guide is in tests/readme.txt. We'll be here on friday 15 if needed. Thanks for your answer, Fabrice On 03/07/2013 06:21 PM, Jacek Becla wrote: > Emmanuel > >> do you think we have to switch to use qms >> in our tests ? Maybe we could integrate it >> later and validate our ticket as it >> is for the moment ? > > Note that qserv system will not take any advantage of > the new metadata system just yet, e.g., even if we > produce the metadata, the qserv will not use it. > I'm working on integrating the new metadata into qserv, > should have it done in ~2 weeks. So, I'd say it'd be > good if you'd start modifying the install/loader > scripts and integrate with the new metadata, but > don't break the existing code until we are sure that > qserv can use the new metadata. I suspect it will take > you few weeks too, so do start the work in a separate > branch. > > And sure, let's validate what you already have first! > I will look into it as soon as I find some time, hopefully > tomorrow. > > thanks, > Jacek > > ######################################################################## > Use REPLY-ALL to reply to list > > To unsubscribe from the QSERV-L list, click the following link: > https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1 ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the QSERV-L list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1