@Mathias: Thanks for the information. I guess this solution requires some serious preparation/tools on my side considering that my db has around 400M nodes and 800M relationships. I've seen some nice bug fixes & improvements in v2.2.x but if 2.2.0 is stable enough, may be I'll wait for the release of 2.3 and the upgrade of the store.
@Nigel: WRT my initial problem, I've tried a few fixes with v1.6.4. Things seemed better but I finally encountered the same problem after 24h of processing. Therefore, I've decided to upgrade to v2.0.7. It required a few adaptations of my code but nothing too terrible. :) Things seem better. The "retry pattern" (try ...except... rollback) seems to do its job. So far, the unique problem encountered was a socket timeout which ended my script and left unclosed connections on the server (same messages as described in OP). I'm currently testing the solution proposed in this StackOverflow post: http://stackoverflow.com/questions/28776140/py2neo-py2neo-packages-httpstream-http-socketerror-timed-out-execute-stream Is it still the recommended solution for this problem ? -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
