On 22-05-15 12:28, Andrea Aime wrote:
Before I spend a lot of more time on this I wanted to check if you
guys had an easy solution. I noticed the commit adds a lot of
'lazy-init=false' to the bean definitions. Now the problem is
that this regards beans defined in the geofence
Hello Andrea and Jody,
There are two tests in geofence-server that are based on
GeoServerTestSupport to initialize the spring context.
They used to work fine but only since recently they started breaking,
and I traced it back to this commit you guys did:
On Fri, May 22, 2015 at 12:17 PM, Niels Charlier ni...@scitus.be wrote:
Hello Andrea and Jody,
There are two tests in geofence-server that are based on
GeoServerTestSupport to initialize the spring context.
They used to work fine but only since recently they started breaking, and
I traced
On Fri, May 22, 2015 at 1:23 PM, Niels Charlier ni...@scitus.be wrote:
For all clarity, with geofence modules I was talking about the actual
*geofence* backend modules themselves, not the community modules in
geoserver. Because the problem concerns beans declared in geofence itself,
not the
Okay, I found nothing on google but I have further debugged, as I had
expected it appears the problems are not caused by lazy initialization,
but by the following single line
On Fri, May 22, 2015 at 4:26 PM, Niels Charlier ni...@scitus.be wrote:
I would recommend removing that line altogether since it seems a
malfunction to me. Otherwise it must at least be made optional somehow
because at this point it is already too late to change when oneTimeSetUp is
run in my
On 22-05-15 16:28, Andrea Aime wrote:
On Fri, May 22, 2015 at 4:26 PM, Niels Charlier ni...@scitus.be
mailto:ni...@scitus.be wrote:
I would recommend removing that line altogether since it seems a
malfunction to me. Otherwise it must at least be made optional
somehow because at
