No doubt many of you will have experienced the instability of Travis CI
builds, with PRs plagued by intermittent failures. I have been
investigating these failures for some time. These failures do not seem
to occur locally.
While I have no proof, I have direct and indirect evidence that the
failures are related to EPSG code lookup failures in the HSQL database.
Static analysis indicates that some failures can *only* occur after
disconnection from the HSQL database. It is not the authority disposer,
which runs only after 30 minutes. I can reproduce one failure by
deleting /tmp/Geotools, which contains the HSQL database files, during
the build.
I thought that something on the Travis CI workers is eating files in
/tmp. Perhaps some rogue cleanup service? Who knows. I tried using a
private java.io.tmpdir (-Djava.io.tmpdir=$HOME/tmp on the travis-tmpdir
branch) as a workaround but still saw failures.
We need to get to the bottom of these failures because they are harming
the usability of Travis CI.
Do we have a rogue cleanup service in our own infrastructure?
Kind regards,
--
Ben Caradoc-Davies <[email protected]>
Director
Transient Software Limited <https://transient.nz/>
New Zealand
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Geoserver-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
