Skip to content

Fetch current OSD pool configuration over the API (#409) #343

Fetch current OSD pool configuration over the API (#409)

Fetch current OSD pool configuration over the API (#409) #343

Workflow file for this run

name: Tests
on:
- push
- pull_request
jobs:
build-microceph:
name: Build microceph snap
runs-on: ubuntu-22.04
env:
SNAPCRAFT_BUILD_ENVIRONMENT: "lxd"
steps:
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Install dependencies
run: |
~/actionutils.sh setup_lxd
sudo snap install snapcraft --classic
snap list
- name: Build snaps
run: snapcraft
- name: Upload snap artifact
if: always()
uses: actions/upload-artifact@v3
with:
name: snaps
path: "*.snap"
retention-days: 5
unit-tests:
name: Run Unit tests
runs-on: ubuntu-22.04
steps:
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Install dependencies
run: |
sudo add-apt-repository ppa:dqlite/dev -y --no-update
sudo apt-get update
sudo apt-get install -y libdqlite-dev
- name: Install Go (1.21)
uses: actions/setup-go@v5
with:
go-version: 1.21.x
- name: Run unit-tests
run: |
cd microceph
make check-unit
single-system-tests:
name: Single node with encryption
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install and setup
run: |
~/actionutils.sh install_microceph
set -uex
# Verify metadata.yaml
meta=/var/snap/microceph/current/conf/metadata.yaml
cat $meta
grep -q ceph-version $meta
# Verify health and auto crush rule
sudo microceph.ceph health | grep -q "OSD count 0 < osd_pool_default_size 3"
sudo microceph.ceph osd crush rule ls | grep -F microceph_auto_osd
- name: Add OSD with failure
run: |
set -eux
loop_file="$(sudo mktemp -p /mnt XXXX.img)"
sudo truncate -s 1G "${loop_file}"
loop_dev="$(sudo losetup --show -f "${loop_file}")"
minor="${loop_dev##/dev/loop}"
sudo mknod -m 0660 "/dev/sdi21" b 7 "${minor}"
set +e
rc=$(sudo microceph disk add --wipe "/dev/sdi21" --encrypt |& grep -c Failure)
if [[ $rc -eq 0 ]] ; then echo "FDE should fail without dmcrypt: $rc"; exit 1; fi
# actually creating additional loop devices is not required since this should fail in validation.
rc=$(sudo microceph disk add "/dev/sdi21" "/dev/sdi22" --wal-device "/dev/sdi23" |& grep -c "not supported for batch disk addition")
if [[ $rc -eq 0 ]] ; then echo "Batch disk addition should fail with wal device: $rc"; exit 1; fi
rc=$(sudo microceph disk add "/dev/sdi21" "/dev/sdi22" --db-device "/dev/sdi23" |& grep -c "not supported for batch disk addition")
if [[ $rc -eq 0 ]] ; then echo "Batch disk addition should fail with db device: $rc"; exit 1; fi
- name: Add OSDs
run: |
~/actionutils.sh add_encrypted_osds
~/actionutils.sh add_lvm_vol
- name: Enable RGW
run: ~/actionutils.sh enable_rgw
- name: Run system tests
run: |
set -eux
# List disks
sudo microceph disk list
# Show ceph's status
sudo microceph.ceph status
# Ceph status expectations for a single node cluster
test_single() {
local status="$1"
( echo "$status" | grep -qF "mon: 1 daemons" ) || { echo fail ; return ; }
( echo "$status" | grep -qE "mgr: .*active, " )|| { echo fail ; return ; }
( echo "$status" | grep -qF "osd: 3 osds" ) || { echo fail ; return ; }
( echo "$status" | grep -qF "rgw: 1 daemon" ) || { echo fail ; return ; }
echo ok
}
# Confirm ceph is healthy and services started
res=$( test_single "$( sudo microceph.ceph status )" )
[ $res = "ok" ] || { echo "single node status fails" ; exit 1 ; }
# Check health after restart
sudo snap stop microceph
sudo snap start microceph
for i in $(seq 1 16); do
status=$( sudo microceph.ceph status )
echo "$status"
res=$( test_single "$status" )
if [ $res = "ok" ] ; then
echo "Single tests pass"
break
else
echo "Single tests fail, retry $i/16"
sleep 15
fi
done
sleep 1
pgrep ceph-osd || { echo "No ceph-osd process found" ; exit 1; }
- name: Exercise RGW
run: ~/actionutils.sh testrgw
- name: Disable RGW
run: ~/actionutils.sh disable_rgw
- name: Enable RGW with SSL enabled
run: ~/actionutils.sh enable_rgw_ssl
- name: Exercise RGW with SSL enabled
run: ~/actionutils.sh testrgw_ssl
- name: Test Cluster Config
run: |
set -eux
cip=$(ip -4 -j route | jq -r '.[] | select(.dst | contains("default")) | .prefsrc' | tr -d '[:space:]')
# pre config set timestamp for service age
ts=$(sudo systemctl show --property ActiveEnterTimestampMonotonic snap.microceph.osd.service | cut -d= -f2)
# set config
sudo microceph cluster config set cluster_network $cip/8 --wait
# post config set timestamp for service age
ts2=$(sudo systemctl show --property ActiveEnterTimestampMonotonic snap.microceph.osd.service | cut -d= -f2)
# Check config output
output=$(sudo microceph cluster config get cluster_network | grep -cim1 'cluster_network')
if [[ $output -lt 1 ]] ; then echo "config check failed: $output"; exit 1; fi
# Check service restarted
if [ $ts2 -lt $ts ]; then echo "config check failed: TS1: $ts2 TS2: $ts3"; exit 1; fi
# reset config
sudo microceph cluster config reset cluster_network --wait
# post config reset timestamp for service age
ts3=$(sudo systemctl show --property ActiveEnterTimestampMonotonic snap.microceph.osd.service | cut -d= -f2)
# Check service restarted
if [ $ts3 -lt $ts2 ]; then echo "config check failed: TS2: $ts2 TS3: $ts3"; exit 1; fi
- name: Test pool operations
run: |
set +e
sudo microceph.ceph osd pool create mypool
sudo microceph pool set-rf --size 1 ""
sudo microceph.ceph config get osd.1 osd_pool_default_size | fgrep -x "1"
sudo microceph pool list | fgrep "mypool" | fgrep -q " 3 "
sudo microceph pool set-rf --size 3 mypool
sudo microceph.ceph osd pool get mypool size | fgrep -x "size: 3"
sudo microceph pool list | fgrep "mypool" | fgrep -q " 3 "
sudo microceph pool set-rf --size 1 "*"
sudo microceph.ceph osd pool get mypool size | fgrep -x "size: 1"
sudo microceph pool list | fgrep "mypool" | fgrep -q " 1 "
- name: Test log operations
run: |
set +e
sudo microceph log set-level warning
output=$(sudo microceph log get-level)
if [[ "$output" != "3" ]] ; then echo "incorrect log level: $output"; exit 1; fi
- name: Print logs for failure
if: failure()
run: |
sudo snap logs microceph -n 1000
multi-node-tests:
name: Multi node testing
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install dependencies
run: ~/actionutils.sh setup_lxd
- name: Create containers with loopback devices
run: ~/actionutils.sh create_containers public
- name: Install local microceph snap
run: ~/actionutils.sh install_multinode
- name: Bootstrap
run: ~/actionutils.sh bootstrap_head public
- name: Setup cluster
run: ~/actionutils.sh cluster_nodes public
- name: Verify config
run: ~/actionutils.sh test_ceph_conf
- name: Add 2 OSDs
run: |
for c in node-wrk1 node-wrk2 ; do
~/actionutils.sh add_osd_to_node $c
done
lxc exec node-wrk0 -- sh -c "microceph.ceph -s"
- name: Exercise microceph status
run: |
set -uex
lxc exec node-wrk0 -- sh -c "sudo microceph status"
- name: Test failure domain scale up
run: |
set -uex
# We still have failure domain OSD
lxc exec node-wrk0 -- sh -c "sudo microceph.ceph config get mon osd_pool_default_crush_rule" | fgrep -x 1
# Add a 3rd OSD, should switch to host failure domain
~/actionutils.sh add_osd_to_node node-wrk0
~/actionutils.sh headexec wait_for_osds 3
sleep 1
lxc exec node-wrk0 -- sh -c "sudo microceph.ceph -s"
# Now default to host rule
lxc exec node-wrk0 -- sh -c "sudo microceph.ceph config get mon osd_pool_default_crush_rule" | fgrep -x 2
- name: Test 3 osds present
run: |
set -uex
lxc exec node-wrk0 -- sh -c "microceph.ceph -s" | egrep "osd: 3 osds: 3 up.*3 in"
- name: Test crush rules
run: |
set -uex
lxc exec node-wrk0 -- sh -c "microceph.ceph osd crush rule ls" | grep -F microceph_auto_host
lxc exec node-wrk0 -- sh -c "microceph.ceph osd pool ls detail" | grep -F "crush_rule 2"
- name: Add another OSD
run: |
~/actionutils.sh add_osd_to_node node-wrk3
~/actionutils.sh headexec wait_for_osds 4
- name: Remove OSD again
run: |
set -uex
lxc exec node-wrk0 -- sh -c "microceph disk remove 4"
lxc exec node-wrk0 -- sh -c "microceph.ceph -s" | egrep "osd: 3 osds: 3 up.*3 in"
- name: Test migrate services
run: |
set -uex
~/actionutils.sh test_migration node-wrk1 node-wrk3
- name: Enable services on wrk1
run: ~/actionutils.sh headexec enable_services node-wrk1
- name: Remove OSD again but don't scaledown crush rule.
run: |
set -uex
lxc exec node-wrk0 -- sh -c "microceph disk remove 3 --prohibit-crush-scaledown --bypass-safety-checks"
lxc exec node-wrk0 -- sh -c "microceph.ceph osd crush rule ls" | grep -F microceph_auto_host
- name: Re-Add OSD
run: |
set -uex
~/actionutils.sh add_osd_to_node node-wrk0
~/actionutils.sh headexec wait_for_osds 3
- name: Test remove node wrk3
run: |
set -uex
~/actionutils.sh headexec remove_node node-wrk3
if lxc exec node-wrk0 -- sh -c "microceph status" | grep -q "^- node-wrk3 " ; then
echo "Failed: node-wrk3 still present"
exit 1
fi
output=$(lxc exec node-wrk0 -- sh -c "microceph.ceph -s")
(echo $output | fgrep "mon: 3 daemons") ||
(echo $output | grep "mon: 4 daemons.*out of quorum: node-wrk3")
- name: Test client configurations
run: ~/actionutils.sh check_client_configs
- name: Print logs for failure
if: failure()
run: |
sudo snap logs microceph -n 1000
multi-node-tests-with-custom-microceph-ip:
name: Multi node testing with custom microceph IP
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install dependencies
run: ~/actionutils.sh setup_lxd
- name: Create containers with loopback devices
run: ~/actionutils.sh create_containers internal
- name: Install local microceph snap
run: ~/actionutils.sh install_multinode
- name: Bootstrap
run: ~/actionutils.sh bootstrap_head internal
- name: Setup cluster
run: ~/actionutils.sh cluster_nodes internal
- name: Add 3 OSDs
run: |
for c in node-wrk0 node-wrk1 node-wrk2 ; do
~/actionutils.sh add_osd_to_node $c
done
- name: Test 3 osds present
run: |
set -uex
~/actionutils.sh headexec wait_for_osds 3
lxc exec node-wrk0 -- sh -c "microceph.ceph -s"
- name: Print logs for failure
if: failure()
run: |
sudo snap logs microceph -n 1000
loop-file-tests:
name: Test with loopback file OSDs
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install and setup
run: ~/actionutils.sh install_microceph
- name: Add loopback file OSDs
run: |
set -uex
sudo microceph disk add loop,1G,4
~/actionutils.sh wait_for_osds 4
sudo microceph.ceph -s
- name: Enable RGW
run: ~/actionutils.sh enable_rgw
- name: Exercise RGW
run: ~/actionutils.sh testrgw
- name: Remove OSD
run: |
set -uex
sudo microceph disk remove osd.1
~/actionutils.sh wait_for_osds 3
sudo microceph.ceph -s
- name: Print logs for failure
if: failure()
run: |
sudo snap logs microceph -n 1000
wal-db-tests:
name: Test WAL/DB device usage
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install and setup
run: ~/actionutils.sh install_microceph
- name: Add loopback file OSDs
run: |
set -uex
sudo microceph disk add loop,1G,3
~/actionutils.sh wait_for_osds 3
sudo microceph.ceph -s
- name: Add WAL/DB enabled OSD
run: |
set -uex
~/actionutils.sh create_loop_devices
sudo microceph disk list
sudo microceph disk add /dev/sdia --wal-device /dev/sdib --db-device /dev/sdic
~/actionutils.sh wait_for_osds 4
sudo microceph.ceph -s
- name: Check that disk list doesn't contain WAL/DB blockdevs
run: |
set -uex
sudo microceph disk list
for d in sdb sdc ; do
if sudo microceph disk list | grep -q /dev/$d ; then
echo "Error: found WAL/DB device: /dev/$d"
exit 1
fi
done
- name: Enable RGW
run: ~/actionutils.sh enable_rgw
- name: Exercise RGW
run: ~/actionutils.sh testrgw
- name: Print logs for failure
if: failure()
run: |
sudo snap logs microceph -n 1000
upgrade-quincy-tests:
name: Test quincy upgrades
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install dependencies
run: ~/actionutils.sh setup_lxd
- name: Create containers with loopback devices
run: ~/actionutils.sh create_containers public
- name: Install quincy stable from store
run: ~/actionutils.sh install_store quincy/stable
- name: Bootstrap
run: ~/actionutils.sh bootstrap_head
- name: Setup cluster
run: ~/actionutils.sh cluster_nodes
- name: Add 3 OSDs
run: |
for c in node-wrk0 node-wrk1 node-wrk2 ; do
~/actionutils.sh add_osd_to_node $c
done
~/actionutils.sh headexec wait_for_osds 3
- name: Enable RGW
run: ~/actionutils.sh headexec enable_rgw
- name: Exercise RGW
run: ~/actionutils.sh headexec testrgw
- name: Install local build
run: ~/actionutils.sh upgrade_multinode
- name: Wait until 3 OSDs are up
run: ~/actionutils.sh headexec wait_for_osds 3
- name: Exercise RGW again
run: ~/actionutils.sh headexec testrgw
- name: Exercise microceph status
run: |
set -uex
lxc exec node-wrk0 -- sh -c "sudo microceph status"
cluster-tests:
name: Test MicroCeph Cluster features.
runs-on: ubuntu-22.04
needs: build-microceph
steps:
- name: Download snap
uses: actions/download-artifact@v3
with:
name: snaps
path: /home/runner
- name: Checkout code
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Copy utils
run: cp tests/scripts/actionutils.sh $HOME
- name: Clear FORWARD firewall rules
run: ~/actionutils.sh cleaript
- name: Free disk
run: ~/actionutils.sh free_runner_disk
- name: Install and setup
run: ~/actionutils.sh install_microceph
- name: Add loopback file OSDs
run: |
set -uex
sudo microceph disk add loop,1G,3
~/actionutils.sh wait_for_osds 3
sudo microceph.ceph -s
- name: Enable RGW
run: ~/actionutils.sh enable_rgw
- name: Exercise RGW
run: ~/actionutils.sh testrgw
- name: Bombard MicroCeph with cluster configs
run: ~/actionutils.sh bombard_rgw_configs
- name: Exercise RGW again
run: ~/actionutils.sh testrgw "newFile"