IMPALA-4519: increase timeout in TestFragmentLifecycle

Increase the timeout to over 120s to match datastream_sender_timeout_ms.
This should avoid spurious test failures if we are unlucky and a sender
gets stuck waiting for a receiver fragment that will never start.

Testing:
Ran the test in a loop for a while to flush out any flakiness.

Change-Id: I9fe6e6c74538d0747e3eeb578cf0518494ff10c8
Reviewed-on: http://gerrit.cloudera.org:8080/5244
Tested-by: Impala Public Jenkins
Reviewed-by: Tim Armstrong <[email protected]>


Project: http://git-wip-us.apache.org/repos/asf/incubator-impala/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-impala/commit/e2cde13a
Tree: http://git-wip-us.apache.org/repos/asf/incubator-impala/tree/e2cde13a
Diff: http://git-wip-us.apache.org/repos/asf/incubator-impala/diff/e2cde13a

Branch: refs/heads/master
Commit: e2cde13a2b1b8c70a8f2d2d1d0853848de1f7d77
Parents: f2ce30e
Author: Tim Armstrong <[email protected]>
Authored: Mon Nov 28 10:34:05 2016 -0800
Committer: Tim Armstrong <[email protected]>
Committed: Tue Nov 29 01:46:05 2016 +0000

----------------------------------------------------------------------
 tests/query_test/test_lifecycle.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/incubator-impala/blob/e2cde13a/tests/query_test/test_lifecycle.py
----------------------------------------------------------------------
diff --git a/tests/query_test/test_lifecycle.py 
b/tests/query_test/test_lifecycle.py
index e1269d6..2bd4476 100644
--- a/tests/query_test/test_lifecycle.py
+++ b/tests/query_test/test_lifecycle.py
@@ -61,8 +61,9 @@ class TestFragmentLifecycle(ImpalaTestSuite):
       pass
 
     for v in verifiers:
-      # Long timeout required because fragments may be blocked while sending 
data, default
-      # timeout is 60s before they wake up and cancel themselves.
+      # Long timeout required because fragments may be blocked while sending 
data. The
+      # default value of --datastream_sender_timeout_ms is 120s before they 
wake up and
+      # cancel themselves.
       #
       # TODO: Fix when we have cancellable RPCs.
-      v.wait_for_metric(self.IN_FLIGHT_FRAGMENTS, 0, timeout=65)
+      v.wait_for_metric(self.IN_FLIGHT_FRAGMENTS, 0, timeout=125)

Reply via email to