Regression testing results.
Just a quick note for those eagerly awaiting new regression test results. There's now a "backup" set of results at the beta web site <http://beta.boost.org/development/testing.html>. The updating may be a bit rocky at first as I get the cronjobs all working correctly. Enjoy. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
Rene Rivera wrote:
Just a quick note for those eagerly awaiting new regression test results. There's now a "backup" set of results at the beta web site <http://beta.boost.org/development/testing.html>. The updating may be a bit rocky at first as I get the cronjobs all working correctly.
Thanks, Rene! That's very helpful. --Beman
Rene Rivera wrote:
Just a quick note for those eagerly awaiting new regression test results. There's now a "backup" set of results at the beta web site <http://beta.boost.org/development/testing.html>. The updating may be a bit rocky at first as I get the cronjobs all working correctly.
How often do you run the update? I've got a Ubuntu machine I'd be willing to dedicate to running rapid-cycle tests. But it doesn't make sense to cycle the tests more often than you are updating the results. Thanks, --Beman
Beman Dawes wrote:
Rene Rivera wrote:
Just a quick note for those eagerly awaiting new regression test results. There's now a "backup" set of results at the beta web site <http://beta.boost.org/development/testing.html>. The updating may be a bit rocky at first as I get the cronjobs all working correctly.
How often do you run the update?
I've got a Ubuntu machine I'd be willing to dedicate to running rapid-cycle tests. But it doesn't make sense to cycle the tests more often than you are updating the results.
The rate depends on how long regression testing takes plus how long building the results takes. The machine I run this on is the same Ubuntu machine I run the gcc-4.1.2 and gcc-4.2.1 incremental tests on. Currently, overall the cycle is about *two hours* as seen from <http://beta.boost.org:8081/Boost_HEAD/Dashboard/> (look at the RSI Droid entries). Looking at the timestamps on my log files it looks like the time is split evenly between the two parts. So my 1.6Ghz+2.5Gig machine builds the results in about an hour. Hence if we want faster result reporting we either move the result building to another equivalent machine, or move it to a faster machine :-) -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
Rene Rivera wrote:
Just a quick note for those eagerly awaiting new regression test results. There's now a "backup" set of results at the beta web site <http://beta.boost.org/development/testing.html>. The updating may be a bit rocky at first as I get the cronjobs all working correctly.
I realize it's beta and maybe a little slower than we would like, but maybe we should go ahead and point the main site here...at least we can see something new instead of Sept 3... Jeff
Jeff Garland wrote:
Rene Rivera wrote:
Just a quick note for those eagerly awaiting new regression test results. There's now a "backup" set of results at the beta web site <http://beta.boost.org/development/testing.html>. The updating may be a bit rocky at first as I get the cronjobs all working correctly.
I realize it's beta and maybe a little slower than we would like, but maybe we should go ahead and point the main site here...at least we can see something new instead of Sept 3...
Done. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
participants (3)
-
Beman Dawes -
Jeff Garland -
Rene Rivera