parallel running issue
I compiled a parallel udf and work very well in local PC.
When I ran it in cluster with parallel mode, it claimed a lot of warns at the first time. Then i built and load udf again in cluster(the warns still claimed).And then i read the same case for the second try, nothing claim~~everything seems ok!
So, is that ok? I heard if you run your case in parallel, you must make your "build path" as a "absolute path", or you must put your udf into every node your case use....... All of that I didn't do, but just build udf again and read case again, seems everything goes well! I really worry about there do have some error but just no claim to me.
ps, I use fluent13 version.
|All times are GMT -4. The time now is 15:09.|