Fix: tests: health thread stall: only stop consumerd when required
authorJérémie Galarneau <jeremie.galarneau@efficios.com>
Thu, 22 Apr 2021 16:32:26 +0000 (12:32 -0400)
committerJérémie Galarneau <jeremie.galarneau@efficios.com>
Tue, 11 May 2021 17:29:39 +0000 (13:29 -0400)
Since a0f8e3109, stop_lttng_consumerd will report a failure when
there is no consumer daemon to kill. This fix ensures it is only
invoked for tests that launch a consumer daemon.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ifba6c0a87158d5cf20eeacb2b65fe049197d9b55

tests/regression/tools/health/test_health.sh

index 27571d05f737ff908f91f44eecde60c5996074aa..f7b0f4c2b0599e8bc473efb5249e62a153e22afd 100644 (file)
@@ -19,7 +19,7 @@ UST_EVENT_NAME="tp:tptest"
 KERNEL_EVENT_NAME="sched_switch"
 CHANNEL_NAME="testchan"
 HEALTH_CHECK_BIN="health_check"
-NUM_TESTS=106
+NUM_TESTS=99
 SLEEP_TIME=30
 
 source $TESTDIR/utils/utils.sh
@@ -137,7 +137,10 @@ function test_health
                # it may have been killed volountarily by testpoint.
                stop_lttng_relayd_notap $KILL_SIGNAL
        fi
-       stop_lttng_consumerd $KILL_SIGNAL
+
+       if [ ${test_consumerd} -eq 1 ]; then
+               stop_lttng_consumerd $KILL_SIGNAL
+       fi
        stop_lttng_sessiond $KILL_SIGNAL
 
        unset LTTNG_TESTPOINT_ENABLE
This page took 0.026075 seconds and 4 git commands to generate.