Commit 02f15260 authored by Karol Latecki's avatar Karol Latecki Committed by Jim Harris
Browse files

test/nvmf: lower number of io queues in connect_disconnect nightly test 



By default nvme-cli will try to connect to subsystem using
number of io queues equal to number of available cores.
In case of more powerful CPUs and HT enabled this results in
more time needed to allocate resources.

Using "-i" option for nvme connect allows us to control the
number of IO queues and avoid timeout issues when testing.

Fixes issue #927

Change-Id: Ibb9b89b2d8be84fba29360db0f781cc99ae5c5b1
Signed-off-by: default avatarKarol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466389


Reviewed-by: default avatarTomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: default avatarBen Walker <benjamin.walker@intel.com>
Reviewed-by: default avatarBroadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: default avatarSPDK CI Jenkins <sys_sgci@intel.com>
parent d40944a2

test/nvmf/target/connect_disconnect.sh

+2 −1
Original line number Diff line number Diff line
@@ -26,13 +26,14 @@ $rpc_py nvmf_subsystem_add_listener nqn.2016-06.io.spdk:cnode1 -t $TEST_TRANSPOR 


if [ $RUN_NIGHTLY -eq 1 ]; then

	num_iterations=200

	IO_QUEUES="-i 8"

else

	num_iterations=10

fi



set +x

for i in $(seq 1 $num_iterations); do

	nvme connect -t $TEST_TRANSPORT -n "nqn.2016-06.io.spdk:cnode1" -a "$NVMF_FIRST_TARGET_IP" -s "$NVMF_PORT"

	nvme connect -t $TEST_TRANSPORT -n "nqn.2016-06.io.spdk:cnode1" -a "$NVMF_FIRST_TARGET_IP" -s "$NVMF_PORT" "$IO_QUEUES"

	waitforblk "nvme0n1"

	nvme disconnect -n "nqn.2016-06.io.spdk:cnode1"

	waitforblk_disconnect "nvme0n1"