I've pushed a quick fix to the attached branch. I just changed it to only create one global SQL_DIR, which may or may not be the right solution:
* Is global state related to the SQL cache okay?
* Should LoggerheadConfig be instantiated tons of times in the first place? I know it will be at least twice (serve-branches, BranchesFromFileSystemServer), so either that should be fixed or this bug matters anyway.
I've pushed a quick fix to the attached branch. I just changed it to only create one global SQL_DIR, which may or may not be the right solution:
* Is global state related to the SQL cache okay?
* Should LoggerheadConfig be instantiated tons of times in the first place? I know it will be at least twice (serve-branches, BranchesFromFil eSystemServer) , so either that should be fixed or this bug matters anyway.