impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "David Knupp (Code Review)" <ger...@cloudera.org>
Subject [Impala-ASF-CR] IMPALA-4259: build Impala without any test cluster setup.
Date Wed, 12 Oct 2016 00:38:26 GMT
David Knupp has posted comments on this change.

Change subject: IMPALA-4259: build Impala without any test cluster setup.
......................................................................


Patch Set 1:

(2 comments)

http://gerrit.cloudera.org:8080/#/c/4685/1/buildall.sh
File buildall.sh:

PS1, Line 386: 5
> No idea, I just moved the preexisting code. 
If a more meaningful test could be substituted, that would be great. If not, then I think
I agree -- removing it would probably be better.


PS1, Line 424: start_test_cluster
> It is a little weird. I guess because the test or data loading scripts want
I don't want to hold up this review, since this is a clear improvement on what exists, so
consider this speculative. It would be nice if the outcome of the buildall.sh script were
simply a running Impala cluster, whether or not data gets loaded or tests get run -- e.g.,
perhaps by adding a line to call start-impala-cluster.py with no params if neither TESTDATA_ACTION
nor TESTS_ACTION are true.

In fact, you could uncouple that line from the "if [[ ${IMPALA_KERBERIZE} -eq 1 ]]" block,
and add it in after the fact. In other words:


  if [[ ${TESTDATA_ACTION} -eq 0  &&  ${TESTS_ACTION} -eq 0 ]]; then
    "${IMPALA_HOME}/bin/start-impala-cluster.py"
  fi


(My bash still kinda sucks, so that might not be the exact right syntax.)

This way, we won't wind up with this one oddball condition whereby if we run non-kerberized,
but without loading up data or running tests, we don't even bother starting up Impala. Otherwise,
the starting of Impala will be handled by the other scripts. 

(FWIW, from a cursory glance at those other scripts, it appears that they each employ some
basic logic to determine the startup params to use for Impala (e.g., log_dir location, cluster_size,
impalad_args, etc. etc.), and then independently call start-impala-cluster.py. So it actually
gets called more than once if you are both loading data and then also running tests.)


-- 
To view, visit http://gerrit.cloudera.org:8080/4685
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I429da7bc6681b16c07fe58bb3efac6d1a8579137
Gerrit-PatchSet: 1
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Tim Armstrong <tarmstrong@cloudera.com>
Gerrit-Reviewer: David Knupp <dknupp@cloudera.com>
Gerrit-Reviewer: Tim Armstrong <tarmstrong@cloudera.com>
Gerrit-HasComments: Yes

Mime
View raw message