When I tried to use hail 0.2, it says:
Py4JJavaError: An error occurred while calling z:is.hail.HailContext.apply.
: is.hail.utils.HailException: This Hail JAR was compiled for Spark 2.2.0, cannot run with Spark 2.4.0.
But https://hail.is/docs/0.2/getting_started.html says " For all methods other than using pip , you will additionally need Spark 2.4.x."
When I change the spark version to 2.2.0, it shows the following message in the end:
Java gateway process exited before sending its port number
What should I do?
Ah, this is sort of a docs versioning problem. Last week we switched from building against 2.2.0 to 2.4.0, but haven’t deployed a new version to PyPI that uses 2.4.0.
How did you install Hail?
After days of trying, I finally successfully installed and initiated hail. But one more error occured…
ERROR:py4j.java_gateway:An error occurred while trying to connect to the Java server (127.0.0.1:36348)
Did you install hail using
pip or did you compile a JAR yourself?
Can you run the following and copy paste the entire output to here?
python -m pip show hail
python -m pip show pyspark
python -c 'import hail as hl; hl.balding_nichols_model(3, 100, 100)._force_count_rows()'
Then run this:
ls -rt hail*.log | tail -n 1
That should print a filename. Can you please attach that file here?
Finally run this:
ls -lrt *hs_err_pid* | tail -n 1
If that outputs a filename, can you please attach that file here?
With that information, we will be able to diagnose your issue.