Hi Lorenz, We base all functionality on docker, so we dockerise every new application we develop locally and want to ship/deploy to a server. Docker limitations could bother us, that's what I keep in mind too, but so far application errors still lead the way in debugging. It could be that the TX abort error is not explicit enough and is thrown as a general exception handler.... I'm not a java programmer, so I hope someone will have a look at this.
Regards, Erik Op ma 14 feb. 2022 om 10:41 schreef Lorenz Buehmann < buehm...@informatik.uni-leipzig.de>: > Hi > > On 14.02.22 09:26, Erik Bijsterbosch wrote: > > Hi, > > > > I want to resolve the transaction error I mentioned before in an earlier > > post/conversation. > > This question was cluttered too much with context to get noticed, I > guess. > > So here's a new attempt... > > > > After starting a (4.4.0 docker) fuseki server or a fuseki geosparql > server > > with inference enabled on my large dataset I get the following error > > message: > > > > fuseki_1 | Write transaction with no commit() or abort() > before > > end() - forced abort > > > > Inference seemed to work earlier on this dataset with my > > previous implementation and I assume now this is data related. > > What can I do to debug this? > > what was your previous implementation? What did you change? > > Do we already know if it works without Docker? > > > > > Regards, > > Erik > > >
