409 lines
16 KiB
Python
409 lines
16 KiB
Python
# Copyright (c) 2013 Mirantis Inc.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
|
# implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
from savanna import conductor
|
|
from savanna import context
|
|
from savanna.openstack.common import log as logging
|
|
from savanna.openstack.common import uuidutils
|
|
from savanna.plugins.general import exceptions as ex
|
|
from savanna.plugins.general import utils
|
|
from savanna.plugins import provisioning as p
|
|
from savanna.plugins.vanilla import config_helper as c_helper
|
|
from savanna.plugins.vanilla import run_scripts as run
|
|
from savanna.plugins.vanilla import scaling as sc
|
|
from savanna.utils import files as f
|
|
from savanna.utils import remote
|
|
|
|
|
|
conductor = conductor.API
|
|
LOG = logging.getLogger(__name__)
|
|
|
|
|
|
class VanillaProvider(p.ProvisioningPluginBase):
|
|
def __init__(self):
|
|
self.processes = {
|
|
"HDFS": ["namenode", "datanode", "secondarynamenode"],
|
|
"MapReduce": ["tasktracker", "jobtracker"],
|
|
"JobFlow": ["oozie"],
|
|
"Hive": ["hiveserver"]
|
|
}
|
|
|
|
def get_plugin_opts(self):
|
|
return []
|
|
|
|
def setup(self, conf):
|
|
self.conf = conf
|
|
|
|
def get_title(self):
|
|
return "Vanilla Apache Hadoop"
|
|
|
|
def get_description(self):
|
|
return (
|
|
"This plugin provides an ability to launch vanilla Apache Hadoop "
|
|
"1.2.1 cluster without any management consoles. Also it can "
|
|
"deploy Oozie 3.3.2")
|
|
|
|
def get_versions(self):
|
|
return ['1.2.1']
|
|
|
|
def get_configs(self, hadoop_version):
|
|
return c_helper.get_plugin_configs()
|
|
|
|
def get_hive_config_path(self):
|
|
return "/opt/hive/conf/hive-site.xml"
|
|
|
|
def get_hdfs_user(self):
|
|
return 'hadoop'
|
|
|
|
def get_node_processes(self, hadoop_version):
|
|
return self.processes
|
|
|
|
def validate(self, cluster):
|
|
nn_count = sum([ng.count for ng
|
|
in utils.get_node_groups(cluster, "namenode")])
|
|
if nn_count != 1:
|
|
raise ex.NotSingleNameNodeException(nn_count)
|
|
|
|
jt_count = sum([ng.count for ng
|
|
in utils.get_node_groups(cluster, "jobtracker")])
|
|
|
|
if jt_count not in [0, 1]:
|
|
raise ex.NotSingleJobTrackerException(jt_count)
|
|
|
|
oozie_count = sum([ng.count for ng
|
|
in utils.get_node_groups(cluster, "oozie")])
|
|
|
|
if oozie_count not in [0, 1]:
|
|
raise ex.NotSingleOozieException(oozie_count)
|
|
|
|
hive_count = sum([ng.count for ng
|
|
in utils.get_node_groups(cluster, "hiveserver")])
|
|
if jt_count == 0:
|
|
|
|
tt_count = sum([ng.count for ng
|
|
in utils.get_node_groups(cluster, "tasktracker")])
|
|
if tt_count > 0:
|
|
raise ex.TaskTrackersWithoutJobTracker()
|
|
|
|
if oozie_count > 0:
|
|
raise ex.OozieWithoutJobTracker()
|
|
|
|
if hive_count > 0:
|
|
raise ex.HiveWithoutJobTracker()
|
|
|
|
if hive_count not in [0, 1]:
|
|
raise ex.NotSingleHiveException(hive_count)
|
|
|
|
def update_infra(self, cluster):
|
|
pass
|
|
|
|
def configure_cluster(self, cluster):
|
|
self._push_configs_to_nodes(cluster)
|
|
self._write_hadoop_user_keys(utils.get_instances(cluster),
|
|
cluster.management_private_key,
|
|
cluster.management_public_key)
|
|
|
|
def start_cluster(self, cluster):
|
|
nn_instance = utils.get_namenode(cluster)
|
|
datanodes = utils.get_datanodes(cluster)
|
|
jt_instance = utils.get_jobtracker(cluster)
|
|
tasktrackers = utils.get_tasktrackers(cluster)
|
|
oozie = utils.get_oozie(cluster)
|
|
hive_server = utils.get_hiveserver(cluster)
|
|
|
|
with remote.get_remote(nn_instance) as r:
|
|
run.format_namenode(r)
|
|
run.start_process(r, "namenode")
|
|
|
|
snns = utils.get_secondarynamenodes(cluster)
|
|
if snns:
|
|
for snn in snns:
|
|
run.start_process(remote.get_remote(snn), "secondarynamenode")
|
|
for dn in datanodes:
|
|
run.start_process(remote.get_remote(dn), "datanode")
|
|
LOG.info("HDFS service at '%s' has been started",
|
|
nn_instance.hostname)
|
|
|
|
if jt_instance:
|
|
run.start_process(remote.get_remote(jt_instance), "jobtracker")
|
|
for tt in tasktrackers:
|
|
run.start_process(remote.get_remote(tt), "tasktracker")
|
|
LOG.info("MapReduce service at '%s' has been started",
|
|
jt_instance.hostname)
|
|
|
|
if oozie:
|
|
with remote.get_remote(oozie) as r:
|
|
if c_helper.is_mysql_enable(cluster):
|
|
run.mysql_start(r, oozie)
|
|
run.oozie_create_db(r)
|
|
run.oozie_share_lib(r, nn_instance.hostname)
|
|
run.start_oozie(r)
|
|
LOG.info("Oozie service at '%s' has been started",
|
|
nn_instance.hostname)
|
|
|
|
if hive_server:
|
|
with remote.get_remote(nn_instance) as r:
|
|
run.hive_create_warehouse_dir(r)
|
|
if c_helper.is_mysql_enable(cluster):
|
|
with remote.get_remote(hive_server) as h:
|
|
if not oozie or hive_server.hostname != oozie.hostname:
|
|
run.mysql_start(h, hive_server)
|
|
run.hive_create_db(h)
|
|
run.hive_metastore_start(h)
|
|
LOG.info("Hive Metastore server at %s has been started",
|
|
hive_server.hostname)
|
|
|
|
LOG.info('Cluster %s has been started successfully' % cluster.name)
|
|
self._set_cluster_info(cluster)
|
|
|
|
def _extract_configs_to_extra(self, cluster, passwd_hive_mysql):
|
|
nn = utils.get_namenode(cluster)
|
|
jt = utils.get_jobtracker(cluster)
|
|
oozie = utils.get_oozie(cluster)
|
|
hive = utils.get_hiveserver(cluster)
|
|
|
|
extra = dict()
|
|
for ng in cluster.node_groups:
|
|
extra[ng.id] = {
|
|
'xml': c_helper.generate_xml_configs(ng.configuration,
|
|
ng.storage_paths,
|
|
nn.hostname,
|
|
jt.hostname
|
|
if jt else None,
|
|
oozie.hostname
|
|
if oozie else None,
|
|
hive.hostname
|
|
if hive else None,
|
|
passwd_hive_mysql),
|
|
'setup_script': c_helper.generate_setup_script(
|
|
ng.storage_paths,
|
|
c_helper.extract_environment_confs(ng.configuration),
|
|
append_oozie=(
|
|
oozie is not None and oozie.node_group.id == ng.id)
|
|
)
|
|
}
|
|
return extra
|
|
|
|
def decommission_nodes(self, cluster, instances):
|
|
tts = utils.get_tasktrackers(cluster)
|
|
dns = utils.get_datanodes(cluster)
|
|
decommission_dns = False
|
|
decommission_tts = False
|
|
|
|
for i in instances:
|
|
if 'datanode' in i.node_group.node_processes:
|
|
dns.remove(i)
|
|
decommission_dns = True
|
|
if 'tasktracker' in i.node_group.node_processes:
|
|
tts.remove(i)
|
|
decommission_tts = True
|
|
|
|
nn = utils.get_namenode(cluster)
|
|
jt = utils.get_jobtracker(cluster)
|
|
|
|
if decommission_tts:
|
|
sc.decommission_tt(jt, instances, tts)
|
|
if decommission_dns:
|
|
sc.decommission_dn(nn, instances, dns)
|
|
|
|
def validate_scaling(self, cluster, existing, additional):
|
|
self._validate_existing_ng_scaling(cluster, existing)
|
|
self._validate_additional_ng_scaling(cluster, additional)
|
|
|
|
def scale_cluster(self, cluster, instances):
|
|
self._push_configs_to_nodes(cluster, instances=instances)
|
|
self._write_hadoop_user_keys(instances,
|
|
cluster.management_private_key,
|
|
cluster.management_public_key)
|
|
run.refresh_nodes(remote.get_remote(
|
|
utils.get_namenode(cluster)), "dfsadmin")
|
|
jt = utils.get_jobtracker(cluster)
|
|
if jt:
|
|
run.refresh_nodes(remote.get_remote(jt), "mradmin")
|
|
|
|
for i in instances:
|
|
with remote.get_remote(i) as r:
|
|
if "datanode" in i.node_group.node_processes:
|
|
run.start_process(r, "datanode")
|
|
|
|
if "tasktracker" in i.node_group.node_processes:
|
|
run.start_process(r, "tasktracker")
|
|
|
|
def _push_configs_to_nodes(self, cluster, instances=None):
|
|
passwd_mysql = uuidutils.generate_uuid() \
|
|
if utils.get_hiveserver(cluster) else None
|
|
extra = self._extract_configs_to_extra(cluster, passwd_mysql)
|
|
|
|
if instances is None:
|
|
instances = utils.get_instances(cluster)
|
|
|
|
for inst in instances:
|
|
ng_extra = extra[inst.node_group.id]
|
|
files = {
|
|
'/etc/hadoop/core-site.xml': ng_extra['xml']['core-site'],
|
|
'/etc/hadoop/mapred-site.xml': ng_extra['xml']['mapred-site'],
|
|
'/etc/hadoop/hdfs-site.xml': ng_extra['xml']['hdfs-site'],
|
|
'/tmp/savanna-hadoop-init.sh': ng_extra['setup_script']
|
|
}
|
|
with remote.get_remote(inst) as r:
|
|
# TODO(aignatov): sudo chown is wrong solution. But it works.
|
|
r.execute_command(
|
|
'sudo chown -R $USER:$USER /etc/hadoop'
|
|
)
|
|
r.execute_command(
|
|
'sudo chown -R $USER:$USER /opt/oozie/conf'
|
|
)
|
|
r.write_files_to(files)
|
|
r.execute_command(
|
|
'sudo chmod 0500 /tmp/savanna-hadoop-init.sh'
|
|
)
|
|
r.execute_command(
|
|
'sudo /tmp/savanna-hadoop-init.sh '
|
|
'>> /tmp/savanna-hadoop-init.log 2>&1')
|
|
|
|
nn = utils.get_namenode(cluster)
|
|
jt = utils.get_jobtracker(cluster)
|
|
|
|
with remote.get_remote(nn) as r:
|
|
r.write_file_to('/etc/hadoop/dn.incl', utils.
|
|
generate_fqdn_host_names(
|
|
utils.get_datanodes(cluster)))
|
|
if jt:
|
|
with remote.get_remote(jt) as r:
|
|
r.write_file_to('/etc/hadoop/tt.incl', utils.
|
|
generate_fqdn_host_names(
|
|
utils.get_tasktrackers(cluster)))
|
|
|
|
oozie = utils.get_oozie(cluster)
|
|
if oozie:
|
|
with remote.get_remote(oozie) as r:
|
|
r.write_file_to('/opt/oozie/conf/oozie-site.xml',
|
|
extra[oozie.node_group.id]
|
|
['xml']['oozie-site'])
|
|
if c_helper.is_mysql_enable(cluster):
|
|
sql_script = f.get_file_text(
|
|
'plugins/vanilla/resources/create_oozie_db.sql')
|
|
files = {
|
|
'/tmp/create_oozie_db.sql': sql_script
|
|
}
|
|
remote.get_remote(oozie).write_files_to(files)
|
|
|
|
hive_server = utils.get_hiveserver(cluster)
|
|
if hive_server:
|
|
ng_extra = extra[hive_server.node_group.id]
|
|
files = {
|
|
'/opt/hive/conf/hive-site.xml':
|
|
ng_extra['xml']['hive-site']
|
|
}
|
|
if c_helper.is_mysql_enable(cluster):
|
|
sql_script = f.get_file_text(
|
|
'plugins/vanilla/resources/create_hive_db.sql'
|
|
)
|
|
sql_script = sql_script.replace('pass',
|
|
passwd_mysql)
|
|
files.update({'/tmp/create_hive_db.sql': sql_script})
|
|
remote.get_remote(hive_server).write_files_to(files)
|
|
|
|
def _set_cluster_info(self, cluster):
|
|
nn = utils.get_namenode(cluster)
|
|
jt = utils.get_jobtracker(cluster)
|
|
oozie = utils.get_oozie(cluster)
|
|
info = {}
|
|
|
|
if jt:
|
|
info['MapReduce'] = {
|
|
'Web UI': 'http://%s:50030' % jt.management_ip
|
|
}
|
|
|
|
if nn:
|
|
info['HDFS'] = {
|
|
'Web UI': 'http://%s:50070' % nn.management_ip
|
|
}
|
|
|
|
if oozie:
|
|
info['JobFlow'] = {
|
|
'Oozie': 'http://%s:11000' % oozie.management_ip
|
|
}
|
|
|
|
ctx = context.ctx()
|
|
conductor.cluster_update(ctx, cluster, {'info': info})
|
|
|
|
def _write_hadoop_user_keys(self, instances, private_key, public_key):
|
|
files = {
|
|
'id_rsa': private_key,
|
|
'authorized_keys': public_key
|
|
}
|
|
|
|
mv_cmd = 'sudo mkdir -p /home/hadoop/.ssh/; ' \
|
|
'sudo mv id_rsa authorized_keys /home/hadoop/.ssh ; ' \
|
|
'sudo chown -R hadoop:hadoop /home/hadoop/.ssh; ' \
|
|
'sudo chmod 600 /home/hadoop/.ssh/{id_rsa,authorized_keys}'
|
|
|
|
for instance in instances:
|
|
with remote.get_remote(instance) as r:
|
|
r.write_files_to(files)
|
|
r.execute_command(mv_cmd)
|
|
|
|
def _get_scalable_processes(self):
|
|
return ["datanode", "tasktracker"]
|
|
|
|
def _get_by_id(self, lst, id):
|
|
for obj in lst:
|
|
if obj.id == id:
|
|
return obj
|
|
|
|
return None
|
|
|
|
def _validate_additional_ng_scaling(self, cluster, additional):
|
|
jt = utils.get_jobtracker(cluster)
|
|
scalable_processes = self._get_scalable_processes()
|
|
|
|
for ng_id in additional:
|
|
ng = self._get_by_id(cluster.node_groups, ng_id)
|
|
if not set(ng.node_processes).issubset(scalable_processes):
|
|
raise ex.NodeGroupCannotBeScaled(
|
|
ng.name, "Vanilla plugin cannot scale nodegroup"
|
|
" with processes: " +
|
|
' '.join(ng.node_processes))
|
|
if not jt and 'tasktracker' in ng.node_processes:
|
|
raise ex.NodeGroupCannotBeScaled(
|
|
ng.name, "Vanilla plugin cannot scale node group with "
|
|
"processes which have no master-processes run "
|
|
"in cluster")
|
|
|
|
def _validate_existing_ng_scaling(self, cluster, existing):
|
|
scalable_processes = self._get_scalable_processes()
|
|
dn_to_delete = 0
|
|
for ng in cluster.node_groups:
|
|
if ng.id in existing:
|
|
if ng.count > existing[ng.id] and "datanode" in \
|
|
ng.node_processes:
|
|
dn_to_delete += ng.count - existing[ng.id]
|
|
if not set(ng.node_processes).issubset(scalable_processes):
|
|
raise ex.NodeGroupCannotBeScaled(
|
|
ng.name, "Vanilla plugin cannot scale nodegroup"
|
|
" with processes: " +
|
|
' '.join(ng.node_processes))
|
|
|
|
dn_amount = len(utils.get_datanodes(cluster))
|
|
rep_factor = c_helper.determine_cluster_config(cluster, 'HDFS',
|
|
"dfs.replication")
|
|
|
|
if dn_to_delete > 0 and dn_amount - dn_to_delete < rep_factor:
|
|
raise ex.ClusterCannotBeScaled(
|
|
cluster.name, "Vanilla plugin cannot shrink cluster because "
|
|
"it would be not enough nodes for replicas "
|
|
"(replication factor is %s)" % rep_factor)
|