def test_target_master_and_proxy_failover(self, config, instance_data, expected_data):
instance = instance_data["modify_standard_instance"]
expected_object = baseCheckPoint(expected_data[instance["cacheInstanceClass"]],
instance["instance_password"])
client, _, instanceId = create_validate_instance(config, instance, expected_object)
shard_num = instance["target_shardNumber"]
resp = reset_class(config, instanceId, instance["target_cacheInstanceClass"], client, shard_num)
assertRespNotNone(resp)
expected_object = baseCheckPoint(expected_data[instance["target_cacheInstanceClass"]], instance["instance_password"])
expected_object.side = 1
expected_object.current_rs_type = "b"
expected_object.next_rs_type = "a"
# 等待spaceStatus变为DoingCopyfrom
for i in range(0, 1200):
resp_status = get_space_status(instanceId, config)
if resp_status == "DoingCopyfrom":
break
sleep(1)
# 触发master failover
next_rs_type = get_next_rs_type(instanceId, config)
redisNum = get_redis_num(instanceId, config, next_rs_type)
redisId = get_shard_id(redisNum, 1)[0]
replicasetName = instanceId + "-master-" + next_rs_type
dockerName = replicasetName + "-" + str(redisId)
oldRunTime = get_docker_running_time(config, instanceId, replicasetName, dockerName)
status = trigger_docker_failover("redis", config, instanceId, config["region"], docker_name=dockerName)
assert status == 200
# 触发proxy failover
proxyId = get_shard_id(get_proxy_num(instanceId, config), 1)[0]
proxyReplicasetName = instanceId + "-proxy"
proxyDockerName = proxyReplicasetName + "-" + str(proxyId)
proxyOldRunTime = get_docker_running_time(config, instanceId, proxyReplicasetName, proxyDockerName)
status = trigger_docker_failover("proxy", config, instanceId, config["region"],id=proxyId)
assert status == 200
# 等待redis failover结束
assert wait_docker_run_time_change(config, instanceId, oldRunTime, replicasetName, dockerName)
#等待proxy failover结束
assert wait_docker_run_time_change(config, instanceId, proxyOldRunTime, proxyReplicasetName, proxyDockerName)
# 手动调用任务恢复接口
print("please run recover task interface.For example:curl http://127.0.0.1:1818/reloadTask -d '{\"taskId\":\"$taskId\",\"isRollback\":false}'")
sleep(10)
for i in range(0, 3600):
if get_space_status(instanceId, config) == "Running":
break
sleep(1)
assert check_admin_proxy_redis_configmap(instanceId, config, expected_object, shard_num)
def test_source_slave_failover(self, config, instance_data, expected_data):
instance = instance_data["modify_cluster_instance"]
expected_object = baseCheckPoint(
expected_data[instance["cacheInstanceClass"]],
instance["instance_password"])
client, _, instanceId = create_validate_instance(
config, instance, expected_object)
shard_num = instance["target_shardNumber"]
resp = reset_class(config, instanceId,
instance["target_cacheInstanceClass"], client,
shard_num)
assertRespNotNone(resp)
expected_object = baseCheckPoint(
expected_data[instance["target_cacheInstanceClass"]],
instance["instance_password"])
expected_object.side = 1
expected_object.current_rs_type = "b"
expected_object.next_rs_type = "a"
# 等待resize开始
for i in range(0, 600):
resp_get_job = get_job(instanceId, config, str(resp.request_id))
if resp_get_job["code"] == 0:
break
sleep(1)
# 触发slave failover
current_rs_type = get_current_rs_type(instanceId, config)
redisNum = get_redis_num(instanceId, config, current_rs_type)
redisId = get_shard_id(redisNum, 1)[0]
replicasetName = instanceId + "-slave-" + current_rs_type
dockerName = replicasetName + "-" + str(redisId)
oldRunTime = get_docker_running_time(config, instanceId,
replicasetName, dockerName)
status = trigger_docker_failover("redis",
config,
instanceId,
config["region"],
docker_name=dockerName)
assert status == 200
assert wait_docker_run_time_change(config, instanceId, oldRunTime,
replicasetName, dockerName)
# 手动调用任务恢复接口
print(
"please run recover task interface.For example:curl http://127.0.0.1:1818/reloadTask -d '{\"taskId\":"
",\"isRollback\":false}'")
sleep(10)
for i in range(0, 3600):
if get_space_status(instanceId, config) == "Running":
break
sleep(1)
assert check_admin_proxy_redis_configmap(instanceId, config,
expected_object, shard_num)
def test_redis_slave_failover_notLocal(self, instance_data, config,
expected_data):
instances = instance_data["create_cluster_specified"]
expected_object = baseCheckPoint(
expected_data[instances[0]["cacheInstanceClass"]],
instances[0]["instance_password"])
client, _, instanceId = create_validate_instance(
config, instances[0], expected_object)
current_rs_type = get_current_rs_type(instanceId, config)
redisNum = get_redis_num(instanceId, config, current_rs_type)
redisId = get_shard_id(redisNum, 1)[0]
replicasetName = instanceId + "-slave-" + current_rs_type
dockerName = replicasetName + "-" + str(redisId)
oldRunTime = get_docker_running_time(config, instanceId,
replicasetName, dockerName)
oldIp = get_redis_ip(instanceId, config, replicasetName, dockerName)
status = trigger_docker_failover("redis",
config,
instanceId,
config["region"],
docker_name=dockerName,
changeIp=1)
assert status == 200
assert wait_docker_run_time_change(config, instanceId, oldRunTime,
replicasetName, dockerName)
newIp = get_redis_ip(instanceId, config, replicasetName, dockerName)
assert oldIp != newIp
assert check_admin_proxy_redis_configmap(instanceId, config,
expected_object,
instances[0]["shardNumber"])
def test_multi_proxy_failover(self, config, instance_data, expected_data):
instances = instance_data["create_cluster_specified"]
expected_object = baseCheckPoint(
expected_data[instances[0]["cacheInstanceClass"]],
instances[0]["instance_password"])
client, _, instanceId = create_validate_instance(
config, instances[0], expected_object)
oldRunTimes = {}
proxyNum = get_proxy_num(instanceId, config)
proxyIds = get_shard_id(get_proxy_num(instanceId, config),
proxyNum / 2)
replicasetName = instanceId + "-proxy"
for id in proxyIds:
dockerName = replicasetName + "-" + str(id)
oldRunTimes[id] = get_docker_running_time(config, instanceId,
replicasetName,
dockerName)
status = trigger_docker_failover("proxy",
config,
instanceId,
config["region"],
id=id)
assert status == 200
for id in proxyIds:
dockerName = replicasetName + "-" + str(id)
assert wait_docker_run_time_change(config, instanceId,
oldRunTimes[id], replicasetName,
dockerName)
assert check_admin_proxy_configmap(instanceId, config, expected_object)
def test_proxy_failover_notLocal(self, config, instance_data,
expected_data):
instances = instance_data["create_cluster_specified"]
expected_object = baseCheckPoint(
expected_data[instances[0]["cacheInstanceClass"]],
instances[0]["instance_password"])
client, _, instanceId = create_validate_instance(
config, instances[0], expected_object)
proxyId = get_shard_id(get_proxy_num(instanceId, config), 1)[0]
oldIp = get_proxy_ip(instanceId, config, proxyId)
replicasetName = instanceId + "-proxy"
dockerName = replicasetName + "-" + str(proxyId)
oldRunTime = get_docker_running_time(config, instanceId,
replicasetName, dockerName)
status = trigger_docker_failover("proxy",
config,
instanceId,
config["region"],
id=proxyId,
changeIp=1)
assert status == 200
assert wait_docker_run_time_change(config, instanceId, oldRunTime,
replicasetName, dockerName)
newIp = get_proxy_ip(instanceId, config, proxyId)
assert oldIp != newIp
assert check_admin_proxy_configmap(instanceId, config, expected_object)
def test_shard_failover(self, instance_data, config, expected_data):
instances = instance_data["create_cluster_specified"]
expected_object = baseCheckPoint(
expected_data[instances[0]["cacheInstanceClass"]],
instances[0]["instance_password"])
client, _, instanceId = create_validate_instance(
config, instances[0], expected_object)
current_rs_type = get_current_rs_type(instanceId, config)
redisNum = get_redis_num(instanceId, config, current_rs_type)
redisId = get_shard_id(redisNum, 1)[0]
replicasetMaster = instanceId + "-master-" + current_rs_type
replicasetSlave = instanceId + "-slave-" + current_rs_type
masterName = replicasetMaster + "-" + str(redisId)
slaveName = replicasetSlave + "-" + str(redisId)
oldMasterRunTime = get_docker_running_time(config, instanceId,
replicasetMaster,
masterName)
oldSlaveRunTime = get_docker_running_time(config, instanceId,
replicasetSlave, slaveName)
status = trigger_docker_failover("redis",
config,
instanceId,
config["region"],
docker_name=masterName)
assert status == 200
status = trigger_docker_failover("redis",
config,
instanceId,
config["region"],
docker_name=slaveName)
assert status == 200
assert wait_docker_run_time_change(config, instanceId,
oldMasterRunTime, replicasetMaster,
masterName)
assert wait_docker_run_time_change(config, instanceId, oldSlaveRunTime,
replicasetSlave, slaveName)
# 等待手动运维恢复正常
for i in range(0, 3600):
if check_topo(instanceId, config) == 0:
break
sleep(1)
sleep(30)
assert check_admin_proxy_redis_configmap(instanceId, config,
expected_object,
instances[0]["shardNumber"])
def test_multi_redis_master_failover(self, instance_data, config,
expected_data):
instances = instance_data["create_cluster_specified"]
expected_object = baseCheckPoint(
expected_data[instances[0]["cacheInstanceClass"]],
instances[0]["instance_password"])
client, _, instanceId = create_validate_instance(
config, instances[0], expected_object)
current_rs_type = get_current_rs_type(instanceId, config)
redisNum = get_redis_num(instanceId, config, current_rs_type)
redisIds = get_shard_id(redisNum, 2)
oldRunTimes = {}
replicasetName = instanceId + "-master-" + current_rs_type
for id in redisIds:
dockerName = replicasetName + "-" + str(id)
oldRunTimes[id] = get_docker_running_time(config, instanceId,
replicasetName,
dockerName)
status = trigger_docker_failover("redis",
config,
instanceId,
config["region"],
docker_name=dockerName)
assert status == 200
for id in redisIds:
dockerName = replicasetName + "-" + str(id)
assert wait_docker_run_time_change(config, instanceId,
oldRunTimes[id], replicasetName,
dockerName)
sleep(30)
assert check_admin_proxy_redis_configmap(instanceId, config,
expected_object,
instances[0]["shardNumber"])
