df134df901
Enabled by a new > 0 integer config value, "servers_per_port" in the [DEFAULT] config section for object-server and/or replication server configs. The setting's integer value determines how many different object-server workers handle requests for any single unique local port in the ring. In this mode, the parent swift-object-server process continues to run as the original user (i.e. root if low-port binding is required), binds to all ports as defined in the ring, and forks off the specified number of workers per listen socket. The child, per-port servers drop privileges and behave pretty much how object-server workers always have, except that because the ring has unique ports per disk, the object-servers will only be handling requests for a single disk. The parent process detects dead servers and restarts them (with the correct listen socket), starts missing servers when an updated ring file is found with a device on the server with a new port, and kills extraneous servers when their port is found to no longer be in the ring. The ring files are stat'ed at most every "ring_check_interval" seconds, as configured in the object-server config (same default of 15s). Immediately stopping all swift-object-worker processes still works by sending the parent a SIGTERM. Likewise, a SIGHUP to the parent process still causes the parent process to close all listen sockets and exit, allowing existing children to finish serving their existing requests. The drop_privileges helper function now has an optional param to suppress the setsid() call, which otherwise screws up the child workers' process management. The class method RingData.load() can be told to only load the ring metadata (i.e. everything except replica2part2dev_id) with the optional kwarg, header_only=True. This is used to keep the parent and all forked off workers from unnecessarily having full copies of all storage policy rings in memory. A new helper class, swift.common.storage_policy.BindPortsCache, provides a method to return a set of all device ports in all rings for the server on which it is instantiated (identified by its set of IP addresses). The BindPortsCache instance will track mtimes of ring files, so they are not opened more frequently than necessary. This patch includes enhancements to the probe tests and object-replicator/object-reconstructor config plumbing to allow the probe tests to work correctly both in the "normal" config (same IP but unique ports for each SAIO "server") and a server-per-port setup where each SAIO "server" must have a unique IP address and unique port per disk within each "server". The main probe tests only work with 4 servers and 4 disks, but you can see the difference in the rings for the EC probe tests where there are 2 disks per server for a total of 8 disks. Specifically, swift.common.ring.utils.is_local_device() will ignore the ports when the "my_port" argument is None. Then, object-replicator and object-reconstructor both set self.bind_port to None if server_per_port is enabled. Bonus improvement for IPv6 addresses in is_local_device(). This PR for vagrant-swift-all-in-one will aid in testing this patch: https://github.com/swiftstack/vagrant-swift-all-in-one/pull/16/ Also allow SAIO to answer is_local_device() better; common SAIO setups have multiple "servers" all on the same host with different ports for the different "servers" (which happen to match the IPs specified in the rings for the devices on each of those "servers"). However, you can configure the SAIO to have different localhost IP addresses (e.g. 127.0.0.1, 127.0.0.2, etc.) in the ring and in the servers' config files' bind_ip setting. This new whataremyips() implementation combined with a little plumbing allows is_local_device() to accurately answer, even on an SAIO. In the default case (an unspecified bind_ip defaults to '0.0.0.0') as well as an explict "bind to everything" like '0.0.0.0' or '::', whataremyips() behaves as it always has, returning all IP addresses for the server. Also updated probe tests to handle each "server" in the SAIO having a unique IP address. For some (noisy) benchmarks that show servers_per_port=X is at least as good as the same number of "normal" workers: https://gist.github.com/dbishop/c214f89ca708a6b1624a#file-summary-md Benchmarks showing the benefits of I/O isolation with a small number of slow disks: https://gist.github.com/dbishop/fd0ab067babdecfb07ca#file-results-md If you were wondering what the overhead of threads_per_disk looks like: https://gist.github.com/dbishop/1d14755fedc86a161718#file-tabular_results-md DocImpact Change-Id: I2239a4000b41a7e7cc53465ce794af49d44796c6
216 lines
9.0 KiB
Python
Executable File
216 lines
9.0 KiB
Python
Executable File
#!/usr/bin/python -u
|
|
# Copyright (c) 2010-2012 OpenStack Foundation
|
|
#
|
|
# 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 unittest import main
|
|
from uuid import uuid4
|
|
|
|
from swiftclient import client
|
|
|
|
from swift.common import direct_client
|
|
from swift.common.exceptions import ClientException
|
|
from swift.common.manager import Manager
|
|
from test.probe.common import kill_server, ReplProbeTest, start_server
|
|
|
|
|
|
class TestObjectHandoff(ReplProbeTest):
|
|
|
|
def test_main(self):
|
|
# Create container
|
|
container = 'container-%s' % uuid4()
|
|
client.put_container(self.url, self.token, container,
|
|
headers={'X-Storage-Policy':
|
|
self.policy.name})
|
|
|
|
# Kill one container/obj primary server
|
|
cpart, cnodes = self.container_ring.get_nodes(self.account, container)
|
|
cnode = cnodes[0]
|
|
obj = 'object-%s' % uuid4()
|
|
opart, onodes = self.object_ring.get_nodes(
|
|
self.account, container, obj)
|
|
onode = onodes[0]
|
|
kill_server((onode['ip'], onode['port']),
|
|
self.ipport2server, self.pids)
|
|
|
|
# Create container/obj (goes to two primary servers and one handoff)
|
|
client.put_object(self.url, self.token, container, obj, 'VERIFY')
|
|
odata = client.get_object(self.url, self.token, container, obj)[-1]
|
|
if odata != 'VERIFY':
|
|
raise Exception('Object GET did not return VERIFY, instead it '
|
|
'returned: %s' % repr(odata))
|
|
|
|
# Kill other two container/obj primary servers
|
|
# to ensure GET handoff works
|
|
for node in onodes[1:]:
|
|
kill_server((node['ip'], node['port']),
|
|
self.ipport2server, self.pids)
|
|
|
|
# Indirectly through proxy assert we can get container/obj
|
|
odata = client.get_object(self.url, self.token, container, obj)[-1]
|
|
if odata != 'VERIFY':
|
|
raise Exception('Object GET did not return VERIFY, instead it '
|
|
'returned: %s' % repr(odata))
|
|
|
|
# Restart those other two container/obj primary servers
|
|
for node in onodes[1:]:
|
|
start_server((node['ip'], node['port']),
|
|
self.ipport2server, self.pids)
|
|
|
|
# We've indirectly verified the handoff node has the container/object,
|
|
# but let's directly verify it.
|
|
another_onode = next(self.object_ring.get_more_nodes(opart))
|
|
odata = direct_client.direct_get_object(
|
|
another_onode, opart, self.account, container, obj, headers={
|
|
'X-Backend-Storage-Policy-Index': self.policy.idx})[-1]
|
|
if odata != 'VERIFY':
|
|
raise Exception('Direct object GET did not return VERIFY, instead '
|
|
'it returned: %s' % repr(odata))
|
|
|
|
# Assert container listing (via proxy and directly) has container/obj
|
|
objs = [o['name'] for o in
|
|
client.get_container(self.url, self.token, container)[1]]
|
|
if obj not in objs:
|
|
raise Exception('Container listing did not know about object')
|
|
for cnode in cnodes:
|
|
objs = [o['name'] for o in
|
|
direct_client.direct_get_container(
|
|
cnode, cpart, self.account, container)[1]]
|
|
if obj not in objs:
|
|
raise Exception(
|
|
'Container server %s:%s did not know about object' %
|
|
(cnode['ip'], cnode['port']))
|
|
|
|
# Bring the first container/obj primary server back up
|
|
start_server((onode['ip'], onode['port']),
|
|
self.ipport2server, self.pids)
|
|
|
|
# Assert that it doesn't have container/obj yet
|
|
try:
|
|
direct_client.direct_get_object(
|
|
onode, opart, self.account, container, obj, headers={
|
|
'X-Backend-Storage-Policy-Index': self.policy.idx})
|
|
except ClientException as err:
|
|
self.assertEquals(err.http_status, 404)
|
|
else:
|
|
self.fail("Expected ClientException but didn't get it")
|
|
|
|
# Run object replication, ensuring we run the handoff node last so it
|
|
# will remove its extra handoff partition
|
|
for node in onodes:
|
|
try:
|
|
port_num = node['replication_port']
|
|
except KeyError:
|
|
port_num = node['port']
|
|
node_id = (port_num - 6000) / 10
|
|
Manager(['object-replicator']).once(number=node_id)
|
|
try:
|
|
another_port_num = another_onode['replication_port']
|
|
except KeyError:
|
|
another_port_num = another_onode['port']
|
|
another_num = (another_port_num - 6000) / 10
|
|
Manager(['object-replicator']).once(number=another_num)
|
|
|
|
# Assert the first container/obj primary server now has container/obj
|
|
odata = direct_client.direct_get_object(
|
|
onode, opart, self.account, container, obj, headers={
|
|
'X-Backend-Storage-Policy-Index': self.policy.idx})[-1]
|
|
if odata != 'VERIFY':
|
|
raise Exception('Direct object GET did not return VERIFY, instead '
|
|
'it returned: %s' % repr(odata))
|
|
|
|
# Assert the handoff server no longer has container/obj
|
|
try:
|
|
direct_client.direct_get_object(
|
|
another_onode, opart, self.account, container, obj, headers={
|
|
'X-Backend-Storage-Policy-Index': self.policy.idx})
|
|
except ClientException as err:
|
|
self.assertEquals(err.http_status, 404)
|
|
else:
|
|
self.fail("Expected ClientException but didn't get it")
|
|
|
|
# Kill the first container/obj primary server again (we have two
|
|
# primaries and the handoff up now)
|
|
kill_server((onode['ip'], onode['port']),
|
|
self.ipport2server, self.pids)
|
|
|
|
# Delete container/obj
|
|
try:
|
|
client.delete_object(self.url, self.token, container, obj)
|
|
except client.ClientException as err:
|
|
if self.object_ring.replica_count > 2:
|
|
raise
|
|
# Object DELETE returning 503 for (404, 204)
|
|
# remove this with fix for
|
|
# https://bugs.launchpad.net/swift/+bug/1318375
|
|
self.assertEqual(503, err.http_status)
|
|
|
|
# Assert we can't head container/obj
|
|
try:
|
|
client.head_object(self.url, self.token, container, obj)
|
|
except client.ClientException as err:
|
|
self.assertEquals(err.http_status, 404)
|
|
else:
|
|
self.fail("Expected ClientException but didn't get it")
|
|
|
|
# Assert container/obj is not in the container listing, both indirectly
|
|
# and directly
|
|
objs = [o['name'] for o in
|
|
client.get_container(self.url, self.token, container)[1]]
|
|
if obj in objs:
|
|
raise Exception('Container listing still knew about object')
|
|
for cnode in cnodes:
|
|
objs = [o['name'] for o in
|
|
direct_client.direct_get_container(
|
|
cnode, cpart, self.account, container)[1]]
|
|
if obj in objs:
|
|
raise Exception(
|
|
'Container server %s:%s still knew about object' %
|
|
(cnode['ip'], cnode['port']))
|
|
|
|
# Restart the first container/obj primary server again
|
|
start_server((onode['ip'], onode['port']),
|
|
self.ipport2server, self.pids)
|
|
|
|
# Assert it still has container/obj
|
|
direct_client.direct_get_object(
|
|
onode, opart, self.account, container, obj, headers={
|
|
'X-Backend-Storage-Policy-Index': self.policy.idx})
|
|
|
|
# Run object replication, ensuring we run the handoff node last so it
|
|
# will remove its extra handoff partition
|
|
for node in onodes:
|
|
try:
|
|
port_num = node['replication_port']
|
|
except KeyError:
|
|
port_num = node['port']
|
|
node_id = (port_num - 6000) / 10
|
|
Manager(['object-replicator']).once(number=node_id)
|
|
another_node_id = (another_port_num - 6000) / 10
|
|
Manager(['object-replicator']).once(number=another_node_id)
|
|
|
|
# Assert primary node no longer has container/obj
|
|
try:
|
|
direct_client.direct_get_object(
|
|
another_onode, opart, self.account, container, obj, headers={
|
|
'X-Backend-Storage-Policy-Index': self.policy.idx})
|
|
except ClientException as err:
|
|
self.assertEquals(err.http_status, 404)
|
|
else:
|
|
self.fail("Expected ClientException but didn't get it")
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main()
|