NetBSD testbed testbed
The links at the bottom of this page lead to reports from
automated builds and tests of
NetBSD-current. These reports are generated using pre-release
versions of the "bracket" test fixture, which are tested
here before being put in production on the official
NetBSD testbed, babylon5.netbsd.org. In other words,
this could be considered a testbed for the testbed.
For most purposes, the official
babylon5.netbsd.org Test Run Logs
should be used in preference to those below.
An example of a case where the results below may be useful is
when it is suspected that a test failure on babylon5
is not caused by an actual NetBSD bug, but by some failure specific to that testbed instance;
double-checking against the independently generated results below may help
eliminate that possibility.
Some differences between these tests and those on babylon5 are:
Reports are available for the following configurations:
- These tests are run on a far less regular schedule than those on babylon5
- These reports may have new features that are not (yet) on babylon5
- These reports may have new bugs which will hopefully be caught here in time to not appear on babylon5
- There may be additional tests of non-default configurations that are not tested on babylon5, such as the i386-debug tests.
- The i386 test results here cover a longer period of historical data, going back to December 2008
- These tests may be hosted on a different version of NetBSD. Some historical data are from tests hosted on Linux.
- These tests may be run using a different version of qemu or, in the case of the "amd64-baremetal" tests, not using qemu at all.
- i386-debug, NetBSD-current/i386 built with MKDEBUG=YES and (since October 2013) MKX11=YES.
- amd64-baremetal, automated network installs and tests of NetBSD-current/amd64 on a physical AMD Athlon64 machine