Travis currently has some real trouble with OSx. We make it optional
so that we can cancel out without affecting overall test result.
This helps both TravisCI as well as us.
We can (and should) roll this back once the Travis issues are solved.
We now know it does not work, and will not fix it immediately. So
running it is just a waste of ressource.
[skip ci] - no changes other than travis config
The previous settings did not make much sense (same set of source
files checked) and also made us violate our Travis permitted runtime.
So rather than splitting the test into 2 VMs, we just remove one of
the static analyzer runs.
This is necessary as we hit the Travis max runtime limit per VM,
so we need to duplicate the tests. This is done via
--enable-testbench1 and --enable-testbench2 which we than use
in different VMs.
This PR also includes updates to the Travis scripts so that we
use the new capability. We have only duplicated Travis VMs where
acutally necessary -- we may need to do more of this in the future.
closes https://github.com/rsyslog/rsyslog/issues/2196
The PostgreSQL output module was woefully out-of-date the following
list is changes made to update the module to current Rsyslog standards.
- allow for v6 configuration syntax
- configurable ports
- support transactional interface
- push db connection into workers (libpq is threadsafe)
- enable module testing on travis
- ensure configuration syntax backwards compatibility
- formatting around postgres core templating
- use new test conventions
- add new configuration syntax test
- add valgrind tests for new and old syntax
- add threading tests
- add action queue long running tests
- add action queue valgrind test
We execute different code pathes under cron jobs. It doesn't make sense
to redo what we already did before a merge. Right now, cron builds a
Coverity scan submission and uploads it. This should be set to run
once a day.
closes https://github.com/rsyslog/rsyslog/issues/2037
We tried to disable this to prevent unnecessary Travis runs after
merging changes to master, however, disabling it also disabled
checking of PRs, so we cannot do this. Thus undoing this change.
this was recommended by Travis support for cases where
more logic is needed inside the script portion. Among others,
this permits to abort a build as soon a failure happens.
use "strong" warnings for the regular build, but use much
relaxed ones for the analyzer run (because otherwise it seems
to abort due to a bug in static analyzer)
