diff options
Diffstat (limited to 'qa/workunits')
| -rwxr-xr-x | qa/workunits/mon/mon-stretch-mode-5-mons-8-osds.sh | 72 | ||||
| -rwxr-xr-x | qa/workunits/nvmeof/basic_tests.sh | 10 | ||||
| -rwxr-xr-x | qa/workunits/nvmeof/fio_test.sh | 9 | ||||
| -rwxr-xr-x | qa/workunits/nvmeof/scalability_test.sh | 51 | ||||
| -rwxr-xr-x | qa/workunits/nvmeof/setup_subsystem.sh | 17 | ||||
| -rwxr-xr-x | qa/workunits/rbd/cli_generic.sh | 5 |
6 files changed, 140 insertions, 24 deletions
diff --git a/qa/workunits/mon/mon-stretch-mode-5-mons-8-osds.sh b/qa/workunits/mon/mon-stretch-mode-5-mons-8-osds.sh new file mode 100755 index 00000000000..827fb0a0b13 --- /dev/null +++ b/qa/workunits/mon/mon-stretch-mode-5-mons-8-osds.sh @@ -0,0 +1,72 @@ +#!/bin/bash -ex + +# A bash script for setting up stretch mode with 5 monitors and 8 OSDs. + +NUM_OSDS_UP=$(ceph osd df | grep "up" | wc -l) + +if [ $NUM_OSDS_UP -lt 8 ]; then + echo "test requires at least 8 OSDs up and running" + exit 1 +fi + +# ensure election strategy is set to "connectivity" +# See https://tracker.ceph.com/issues/69107 +ceph mon set election_strategy connectivity + +for dc in dc1 dc2 + do + ceph osd crush add-bucket $dc datacenter + ceph osd crush move $dc root=default + done + +ceph osd crush add-bucket host01 host +ceph osd crush add-bucket host02 host +ceph osd crush add-bucket host03 host +ceph osd crush add-bucket host04 host + +ceph osd crush move host01 datacenter=dc1 +ceph osd crush move host02 datacenter=dc1 +ceph osd crush move host03 datacenter=dc2 +ceph osd crush move host04 datacenter=dc2 + +ceph osd crush move osd.0 host=host01 +ceph osd crush move osd.1 host=host01 +ceph osd crush move osd.2 host=host02 +ceph osd crush move osd.3 host=host02 +ceph osd crush move osd.4 host=host03 +ceph osd crush move osd.5 host=host03 +ceph osd crush move osd.6 host=host04 +ceph osd crush move osd.7 host=host04 + +# set location for monitors +ceph mon set_location a datacenter=dc1 host=host01 +ceph mon set_location b datacenter=dc1 host=host02 +ceph mon set_location c datacenter=dc2 host=host03 +ceph mon set_location d datacenter=dc2 host=host04 + +# set location for tiebreaker monitor +ceph mon set_location e datacenter=dc3 host=host05 + +# remove the current host from crush map +hostname=$(hostname -s) +ceph osd crush remove $hostname +# create a new crush rule with stretch rule +ceph osd getcrushmap > crushmap +crushtool --decompile crushmap > crushmap.txt +sed 's/^# end crush map$//' crushmap.txt > crushmap_modified.txt +cat >> crushmap_modified.txt << EOF +rule stretch_rule { + id 2 + type replicated + step take default + step choose firstn 2 type datacenter + step chooseleaf firstn 2 type host + step emit +} +# end crush map +EOF + +crushtool --compile crushmap_modified.txt -o crushmap.bin +ceph osd setcrushmap -i crushmap.bin + +ceph mon enable_stretch_mode e stretch_rule datacenter diff --git a/qa/workunits/nvmeof/basic_tests.sh b/qa/workunits/nvmeof/basic_tests.sh index dc6fd1669da..9e7a1f5134e 100755 --- a/qa/workunits/nvmeof/basic_tests.sh +++ b/qa/workunits/nvmeof/basic_tests.sh @@ -38,8 +38,10 @@ disconnect_all() { connect_all() { sudo nvme connect-all --traddr=$NVMEOF_DEFAULT_GATEWAY_IP_ADDRESS --transport=tcp -l 3600 sleep 5 - output=$(sudo nvme list --output-format=json) - if ! echo "$output" | grep -q "$SPDK_CONTROLLER"; then + expected_devices_count=$1 + actual_devices=$(sudo nvme list --output-format=json | jq -r ".Devices[].Subsystems[] | select(.Controllers | all(.ModelNumber == \"$SPDK_CONTROLLER\")) | .Namespaces[].NameSpace" | wc -l) + if [ "$actual_devices" -ne "$expected_devices_count" ]; then + sudo nvme list --output-format=json return 1 fi } @@ -72,11 +74,13 @@ test_run connect test_run list_subsys 1 test_run disconnect_all test_run list_subsys 0 -test_run connect_all +devices_count=$(( $NVMEOF_NAMESPACES_COUNT * $NVMEOF_SUBSYSTEMS_COUNT )) +test_run connect_all $devices_count gateways_count=$(( $(echo "$NVMEOF_GATEWAY_IP_ADDRESSES" | tr -cd ',' | wc -c) + 1 )) multipath_count=$(( $gateways_count * $NVMEOF_SUBSYSTEMS_COUNT)) test_run list_subsys $multipath_count + echo "-------------Test Summary-------------" echo "[nvmeof] All nvmeof basic tests passed!" diff --git a/qa/workunits/nvmeof/fio_test.sh b/qa/workunits/nvmeof/fio_test.sh index 57d355a6318..f7f783afc67 100755 --- a/qa/workunits/nvmeof/fio_test.sh +++ b/qa/workunits/nvmeof/fio_test.sh @@ -5,6 +5,7 @@ sudo yum -y install sysstat namespace_range_start= namespace_range_end= +random_devices_count= rbd_iostat=false while [[ $# -gt 0 ]]; do @@ -17,6 +18,10 @@ while [[ $# -gt 0 ]]; do namespace_range_end=$2 shift 2 ;; + --random_devices) + random_devices_count=$2 + shift 2 + ;; --rbd_iostat) rbd_iostat=true shift @@ -29,7 +34,7 @@ done fio_file=$(mktemp -t nvmeof-fio-XXXX) all_drives_list=$(sudo nvme list --output-format=json | - jq -r '.Devices | sort_by(.NameSpace) | .[] | select(.ModelNumber == "Ceph bdev Controller") | .DevicePath') + jq -r '.Devices[].Subsystems[] | select(.Controllers | all(.ModelNumber == "Ceph bdev Controller")) | .Namespaces | sort_by(.NSID) | .[] | .NameSpace') # When the script is passed --start_ns and --end_ns (example: `nvmeof_fio_test.sh --start_ns 1 --end_ns 3`), # then fio runs on namespaces only in the defined range (which is 1 to 3 here). @@ -37,6 +42,8 @@ all_drives_list=$(sudo nvme list --output-format=json | # run on first 3 namespaces here. if [ "$namespace_range_start" ] || [ "$namespace_range_end" ]; then selected_drives=$(echo "${all_drives_list[@]}" | sed -n "${namespace_range_start},${namespace_range_end}p") +elif [ "$random_devices_count" ]; then + selected_drives=$(echo "${all_drives_list[@]}" | shuf -n $random_devices_count) else selected_drives="${all_drives_list[@]}" fi diff --git a/qa/workunits/nvmeof/scalability_test.sh b/qa/workunits/nvmeof/scalability_test.sh index 5a26b6284f7..8ede4b7eda2 100755 --- a/qa/workunits/nvmeof/scalability_test.sh +++ b/qa/workunits/nvmeof/scalability_test.sh @@ -3,37 +3,64 @@ GATEWAYS=$1 # exmaple "nvmeof.a,nvmeof.b" DELAY="${SCALING_DELAYS:-50}" +POOL="${RBD_POOL:-mypool}" +GROUP="${NVMEOF_GROUP:-mygroup0}" +source /etc/ceph/nvmeof.env if [ -z "$GATEWAYS" ]; then echo "At least one gateway needs to be defined for scalability test" exit 1 fi -pip3 install yq - status_checks() { - ceph nvme-gw show mypool '' - ceph orch ls - ceph orch ps - ceph -s + expected_count=$1 + + output=$(ceph nvme-gw show $POOL $GROUP) + nvme_show=$(echo $output | grep -o '"AVAILABLE"' | wc -l) + if [ "$nvme_show" -ne "$expected_count" ]; then + return 1 + fi + + orch_ls=$(ceph orch ls) + if ! echo "$orch_ls" | grep -q "$expected_count/$expected_count"; then + return 1 + fi + + output=$(ceph orch ps --service-name nvmeof.$POOL.$GROUP) + orch_ps=$(echo $output | grep -o 'running' | wc -l) + if [ "$orch_ps" -ne "$expected_count" ]; then + return 1 + fi + + ceph_status=$(ceph -s) + if ! echo "$ceph_status" | grep -q "HEALTH_OK"; then + return 1 + fi } +total_gateways_count=$(( $(echo "$NVMEOF_GATEWAY_IP_ADDRESSES" | tr -cd ',' | wc -c) + 1 )) +scaled_down_gateways_count=$(( total_gateways_count - $(echo "$GATEWAYS" | tr -cd ',' | wc -c) - 1 )) + echo "[nvmeof.scale] Setting up config to remove gateways ${GATEWAYS}" +ceph orch ls --service-name nvmeof.$POOL.$GROUP --export > /tmp/nvmeof-gw.yaml ceph orch ls nvmeof --export > /tmp/nvmeof-gw.yaml cat /tmp/nvmeof-gw.yaml -yq "del(.placement.hosts[] | select(. | test(\".*($(echo $GATEWAYS | sed 's/,/|/g'))\")))" /tmp/nvmeof-gw.yaml > /tmp/nvmeof-gw-new.yaml + +pattern=$(echo $GATEWAYS | sed 's/,/\\|/g') +sed "/$pattern/d" /tmp/nvmeof-gw.yaml > /tmp/nvmeof-gw-new.yaml cat /tmp/nvmeof-gw-new.yaml echo "[nvmeof.scale] Starting scale testing by removing ${GATEWAYS}" -status_checks -ceph orch rm nvmeof.mypool && sleep 20 # temp workaround +status_checks $total_gateways_count ceph orch apply -i /tmp/nvmeof-gw-new.yaml # downscale +ceph orch redeploy nvmeof.$POOL.$GROUP sleep $DELAY -status_checks -ceph orch rm nvmeof.mypool && sleep 20 # temp workaround +status_checks $scaled_down_gateways_count +echo "[nvmeof.scale] Downscale complete - removed gateways (${GATEWAYS}); now scaling back up" ceph orch apply -i /tmp/nvmeof-gw.yaml #upscale +ceph orch redeploy nvmeof.$POOL.$GROUP sleep $DELAY -status_checks +status_checks $total_gateways_count echo "[nvmeof.scale] Scale testing passed for ${GATEWAYS}" diff --git a/qa/workunits/nvmeof/setup_subsystem.sh b/qa/workunits/nvmeof/setup_subsystem.sh index cc4024323eb..b573647b1e3 100755 --- a/qa/workunits/nvmeof/setup_subsystem.sh +++ b/qa/workunits/nvmeof/setup_subsystem.sh @@ -26,14 +26,21 @@ list_subsystems () { done } +list_namespaces () { + for i in $(seq 1 $NVMEOF_SUBSYSTEMS_COUNT); do + subsystem_nqn="${NVMEOF_SUBSYSTEMS_PREFIX}${i}" + sudo podman run -it $NVMEOF_CLI_IMAGE --server-address $NVMEOF_DEFAULT_GATEWAY_IP_ADDRESS --server-port $NVMEOF_SRPORT --format plain namespace list --subsystem $subsystem_nqn + done +} + +echo "[nvmeof] Starting subsystem setup..." + # add all subsystems for i in $(seq 1 $NVMEOF_SUBSYSTEMS_COUNT); do subsystem_nqn="${NVMEOF_SUBSYSTEMS_PREFIX}${i}" sudo podman run -it $NVMEOF_CLI_IMAGE --server-address $NVMEOF_DEFAULT_GATEWAY_IP_ADDRESS --server-port $NVMEOF_SRPORT subsystem add --subsystem $subsystem_nqn --no-group-append done -list_subsystems - # add all gateway listeners for i in "${!gateway_ips[@]}" do @@ -65,11 +72,5 @@ done list_subsystems -# list namespaces -for i in $(seq 1 $NVMEOF_SUBSYSTEMS_COUNT); do - subsystem_nqn="${NVMEOF_SUBSYSTEMS_PREFIX}${i}" - sudo podman run -it $NVMEOF_CLI_IMAGE --server-address $NVMEOF_DEFAULT_GATEWAY_IP_ADDRESS --server-port $NVMEOF_SRPORT --format plain namespace list --subsystem $subsystem_nqn -done - echo "[nvmeof] Subsystem setup done" diff --git a/qa/workunits/rbd/cli_generic.sh b/qa/workunits/rbd/cli_generic.sh index 2aa27d3d655..0ceb9ff54cf 100755 --- a/qa/workunits/rbd/cli_generic.sh +++ b/qa/workunits/rbd/cli_generic.sh @@ -914,6 +914,11 @@ test_namespace() { rbd group create rbd/test1/group1 rbd group image add rbd/test1/group1 rbd/test1/image1 + rbd group image add --group-pool rbd --group-namespace test1 --group group1 \ + --image-pool rbd --image-namespace test1 --image image2 + rbd group image rm --group-pool rbd --group-namespace test1 --group group1 \ + --image-pool rbd --image-namespace test1 --image image1 + rbd group image rm rbd/test1/group1 rbd/test1/image2 rbd group rm rbd/test1/group1 rbd trash move rbd/test1/image1 |