Blog

A new version of the system tests front-end (0.7.3) is installed.

http://glast-ground.slac.stanford.edu/SystemTests/

See the release notes for more information.

A new version of the system tests front-end (0.7.2) is installed.

http://glast-ground.slac.stanford.edu/SystemTests/

(note the new URL, the old web007 URL still works, but please switch to using glast-ground as eventually the web007 access will go away).

See the release notes for more information.

A new version of the system tests front-end (0.7) is installed.

http://web007.slac.stanford.edu:8888/SystemTests/

See the release notes for more information.

A new version of the system tests front-end (0.6) is installed.

http://web007.slac.stanford.edu:8888/SystemTests/

See the release notes for more information.

Another new version of the system tests prototype front-end is now available at:

http://web007.slac.stanford.edu:8888/SystemTests

Thanks to Max Turri the statistics tab is now working, and it is now possible to switch between release tests and engineering model tests. In addition log-on is now supported, using Matt's CAS server. Currently you can only log-on if you are inside the SLAC firewall, and there are no extra features available to logged on users, but those issues will be addressed soon.

In addition a new Jira project has been set up, to track changes and feature/bug requests. You can access it at:

http://jira.slac.stanford.edu/secure/BrowseProject.jspa?id=10140

Another iteration of the prototype system tests iteration is available at

http://web007.slac.stanford.edu:8888/SystemTests/

The main change is that an initial implementation of plotting is now available.

There are now two prototype systems, one being put together by Tony and one by Matt. We are trying to coordinate to not duplicate too much effort, and will choose the best aspects of each system. They can be found at:

http://web007.slac.stanford.edu:8888/SystemTests/ http://glast-ground.slac.stanford.edu/systest

New features are being added to both systems for a few more days, then we will start the merge process, and aim to have a working system by 15th January.

A temporary page has been set up for prototyping ideas for the new system tests pages. It can be accessed at:

http://web007.slac.stanford.edu:8888/SystemTests

http://glast-ground.slac.stanford.edu/systest/

here's a first idea on what the new schema might look like:

Release Table:

SystemTest Table:

Metadata Table:

OutputFile Table:

Release Table contains an entry for each RM checked out package.
SystemTest table contains all the test done for a specific entry in Release Table
Metadata table contains all the metadata of a specific test in SystemTest
OutputFile contains the files written out for a specific SystemTest.

I'm not sure how the triggering of it will work. I am considering making it a script that's part of the RM set of scripts which get executed in order.

Believe it or not there is some documentation of the old system. It's at

http://www-glast.slac.stanford.edu/software/SysTests/

Tony's summary of the kick-off system tests meeting on 12-15-04.

• Would like history plots from database (e.g time history of releases, or scatter plots of one value vs another).
• Inline display of comments on histograms.

Summary of Julie's December 7 email outlining improvements to the existing ASP based system tests front end.

• Allow selection of which GlastRelease or EngineeringModel version to look at with only one submit button.
• Reformat page which lists the list of tests for a given release to make each line narrower, so that more tests can be viewed in the same screen (this could be done by altering the "plot histos get tuples" box).
• In the datafiles section, make a link to the log file and the merit tuple, and make the systest name lead people to the plot histos interface.
• Show people how to download the raw root files from this page (they are always in the same dir as the tuple).
• Allow the threshold value to change and recalc the pass/fail values on the fly.
• Add a button to change the test that decides on pass/fail (i.e. to use something other than KS).
• Disable the row ("gray it out") if the status of a system test is pending or failed.
• If the entry is "NA" in the histogram_file field, then there should not be a link to "plot histos".
• Show histos that fail when one clicks on the fail button.
• Allow user to specify number of plots per page in addition to the meta data that is already displayed.
• Make plots clickable so that clicking on one would lead you back to the single plot page where one could do more fiddling.
• Use the axis labels from the ROOT files.
• Display the values of the under/overflow bins.
• Display values of the statisical comparison test(s).
• Add facility to rebin the histos (into bigger bins), something akin to what you get with a TH1F::Rebin() in root.
• Allow user to flag histos as being different but deemed not a problem (so that we can move towards a method for declaring a release to have passed or failed the systests).