DomGarguilo commented on issue #637: URL: https://github.com/apache/accumulo/issues/637#issuecomment-756902064
It seems there could be two ways this issue could be handled. In the [zookeeper recipes](https://zookeeper.apache.org/doc/r3.5.8/recipes.html#sc_recipes_GuidNote) which was linked earlier, it does not explicitly say what should happen to a node once its path is recovered. It seems like the path to the recovered node could either be returned, or the node itself could be deleted and node creation restarted. In curator, it seems that when a node in this situation is found, it is scheduled to be deleted and an error is thrown ([heres the code](https://github.com/apache/curator/blob/95f147f8a864375744a566af3d550659f512728b/curator-framework/src/main/java/org/apache/curator/framework/imps/CreateBuilderImpl.java#L589)). I'm not sure which would work best however if the node is deleted with an exception thrown, this might work well within the retry loop that is used to create nodes in accumulo. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
