Github user shivaram commented on a diff in the pull request:
https://github.com/apache/spark/pull/4901#discussion_r26010373
--- Diff: ec2/spark_ec2.py ---
@@ -872,9 +890,16 @@ def deploy_files(conn, root_dir, opts, master_nodes,
slave_nodes, modules):
if "." in opts.spark_version:
# Pre-built Spark deploy
spark_v = get_validate_spark_version(opts.spark_version,
opts.spark_git_repo)
+ tachyon_v = get_tachyon_version(spark_v)
else:
# Spark-only custom deploy
spark_v = "%s|%s" % (opts.spark_git_repo, opts.spark_version)
+ if opts.tachyon_version is None:
--- End diff --
I see your point, but as @nchammas mentioned only some fraction of Spark
usage scenarios involve using Tachyon. If one just wants to say run the
`master` branch to do some spark-perf benchmarks expecting the user to set a
Tachyon version seems like a unnecessary burden.
If we don't want to use a default tachyon version I'd vote for just not
starting tachyon in cases where a spark git hash is used. (i.e. remove tachyon
from the list of modules and print a info message)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
