Bit of a long-shot since my example is related to Storm not Hadoop, and my experience was five years ago, but this blog post might help:
http://derek.troywest.com/articles/trident-in-clojure and this related ticket: https://github.com/sorenmacbeth/marceline/issues/9 I could run a Storm topology locally, or on a single node, but not when deployed to a Storm cluster. It came down the the way Storm chops up topologies and serialises them across the network to run on remote nodes. There a few different ways of implementing interfaces in Clojure, and (at the time at least) they had different semantics in the serialise-reconstruct-and-run world. This may also be helpful, it has been fixed since I wrote the blog post: http://dev.clojure.org/jira/browse/CLJ-1208 Good luck! On Saturday, 5 January 2019 08:35:22 UTC+11, cloje wrote: > > I have a project that I'm trying to run on Hadoop in cluster mode. > > I have written and executed the code on my local machine and on a single > node on Hadoop. But when I try to executed the same jar file in cluster > mode with 5 executors, it bombs at a function and says, Attempting to call > unbound fn, '#my-func-name. > > Has anyone faced the same kind? > Was there a work around? > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
