iptables: fail to start ovs/linuxbridge agents on missing sysctl knobs

For new kernels (3.18+), bridge module is split into two pieces: bridge
and br_netfilter. The latter provides firewall support for bridged
traffic, as well as the following sysctl knobs:

* net.bridge.bridge-nf-call-arptables
* net.bridge.bridge-nf-call-ip6tables
* net.bridge.bridge-nf-call-iptables

Before kernel 3.18, any brctl command was loading the 'bridge' module
with the knobs, so at the moment where we reached iptables setup, they
were always available.

With new 3.18+ kernels, brctl still loads 'bridge' module, but not
br_netfilter. So bridge existance no longer guarantees us knobs'
presence. If we reach _enable_netfilter_for_bridges before the new
module is loaded, then the code will fail, triggering agent resync. It
will also fail to enable bridge firewalling on systems where it's
disabled by default (examples of those systems are most if not all Red
Hat/Fedora based systems), making security groups completely
ineffective.

Systems that don't override default settings for those knobs would work
fine except for this exception in the log file and agent resync. This is
because the first attempt to add a iptables rule using 'physdev' module
(-m physdev) will trigger the kernel module loading. In theory, we could
silently swallow missing knobs, and still operate correctly. But on
second thought, it's quite fragile to rely on that implicit module
loading. In the case where we can't detect whether firewall is enabled,
it's better to fail than hope for the best.

An alternative to the proposed path could be trying
to fix broken deployment, meaning we would need to load the missing
kernel module on agent startup. It's not even clear whether we can
assume the operation would be available to us. Even with that, adding a
rootwrap filter to allow loading code in the kernel sounds quite scary.
If we would follow the path, we would also hit an issue of
distinguishing between cases of built-in kernel module vs. modular one.
A complexity that is probably beyond what Neutron should fix.

The patch introduces a sanity check that would fail on missing
configuration knobs.

DocImpact: document the new deployment requirement in operations guide
UpgradeImpact: deployers relying on agents fixing wrong sysctl defaults
               will need to make sure bridge firewalling is enabled.
               Also, the kernel module providing sysctl knobs must be
               loaded before starting the agent, otherwise it will fail
               to start.

Changes made to this backport:
   neutron/agent/linux/iptables_firewall.py
       - removed deprecation warning when setting sysctl values to 1 as
         they are planned to be removed in Ocata
   neutron/cmd/sanity/checks.py
       - Re-implemented the flow to check only for presence of sysctl
         options instead of checking the values. Kernel options are set
         in runtime thus the values don't matter.

Depends-On: Id6bfd9595f0772a63d1096ef83ebbb6cd630fafd
Change-Id: I9137ea017624ac92a05f73863b77f9ee4681bbe7
Related-Bug: #1622914
(cherry picked from commit e83a44b96a)
This commit is contained in:
Ihar Hrachyshka 2016-09-15 21:48:10 +00:00 committed by Jakub Libosvar
parent 7295beec36
commit 4371a4f5cd
8 changed files with 72 additions and 21 deletions

View File

@ -8,20 +8,19 @@
[Filters]
# neutron/agent/linux/iptables_manager.py
# neutron/agent/linux/iptables_firewall.py
# "iptables-save", ...
iptables-save: CommandFilter, iptables-save, root
iptables-restore: CommandFilter, iptables-restore, root
ip6tables-save: CommandFilter, ip6tables-save, root
ip6tables-restore: CommandFilter, ip6tables-restore, root
# neutron/agent/linux/iptables_manager.py
# neutron/agent/linux/iptables_firewall.py
# "iptables", "-A", ...
iptables: CommandFilter, iptables, root
ip6tables: CommandFilter, ip6tables, root
# neutron/agent/linux/iptables_manager.py
# "sysctl", "-w", ...
# neutron/agent/linux/iptables_firewall.py
sysctl: CommandFilter, sysctl, root
# neutron/agent/linux/ip_conntrack.py

View File

@ -23,7 +23,7 @@ from oslo_log import log as logging
from oslo_utils import netutils
import six
from neutron._i18n import _LI
from neutron._i18n import _, _LI
from neutron.agent import firewall
from neutron.agent.linux import ip_conntrack
from neutron.agent.linux import ipset_manager
@ -109,15 +109,17 @@ class IptablesFirewallDriver(firewall.FirewallDriver):
# enabled by default or not (Ubuntu - yes, Redhat - no, for
# example).
LOG.debug("Enabling netfilter for bridges")
utils.execute(['sysctl', '-w',
'net.bridge.bridge-nf-call-arptables=1'],
run_as_root=True)
utils.execute(['sysctl', '-w',
'net.bridge.bridge-nf-call-ip6tables=1'],
run_as_root=True)
utils.execute(['sysctl', '-w',
'net.bridge.bridge-nf-call-iptables=1'],
run_as_root=True)
entries = utils.execute(['sysctl', '-N', 'net.bridge'],
run_as_root=True).splitlines()
for proto in ('arp', 'ip', 'ip6'):
knob = 'net.bridge.bridge-nf-call-%stables' % proto
if 'net.bridge.bridge-nf-call-%stables' % proto not in entries:
raise SystemExit(
_("sysctl value %s not present on this system.") % knob)
enabled = utils.execute(['sysctl', '-b', knob])
if enabled != '1':
utils.execute(
['sysctl', '-w', '%s=1' % knob], run_as_root=True)
@property
def ports(self):

View File

@ -228,6 +228,17 @@ def dhcp_release6_supported():
return True
def bridge_firewalling_enabled():
cmd = ['sysctl', '-N', 'net.bridge']
entries = agent_utils.execute(cmd, run_as_root=True)
for proto in ('arp', 'ip', 'ip6'):
knob = 'net.bridge.bridge-nf-call-%stables' % proto
if knob not in entries:
LOG.debug("sysctl value %s not present on this system.", knob)
return False
return True
class KeepalivedIPv6Test(object):
def __init__(self, ha_port, gw_port, gw_vip, default_gw):
self.ha_port = ha_port

View File

@ -256,6 +256,16 @@ def check_dhcp_release6():
return result
def check_bridge_firewalling_enabled():
result = checks.bridge_firewalling_enabled()
if not result:
LOG.error(_LE('Bridge firewalling is not enabled. It may be the case '
'that bridge and/or br_netfilter kernel modules are not '
'loaded. Alternatively, corresponding sysctl settings '
'may be overridden to disable it by default.'))
return result
# Define CLI opts to test specific features, with a callback for the test
OPTS = [
BoolOptCallback('ovs_vxlan', check_ovs_vxlan, default=False,
@ -298,6 +308,9 @@ OPTS = [
help=_('Check ip6tables installation')),
BoolOptCallback('dhcp_release6', check_dhcp_release6,
help=_('Check dhcp_release6 installation')),
BoolOptCallback('bridge_firewalling', check_bridge_firewalling_enabled,
help=_('Check bridge firewalling'),
default=False),
]
@ -343,6 +356,15 @@ def enable_tests_from_config():
if ('sriovnicswitch' in cfg.CONF.ml2.mechanism_drivers and
'qos' in cfg.CONF.ml2.extension_drivers):
cfg.CONF.set_default('vf_extended_management', True)
if cfg.CONF.SECURITYGROUP.firewall_driver in (
'iptables',
'iptables_hybrid',
('neutron.agent.linux.iptables_firewall.'
'IptablesFirewallDriver'),
('neutron.agent.linux.iptables_firewall.'
'OVSHybridIptablesFirewallDriver'),
):
cfg.CONF.set_default('bridge_firewalling', True)
def all_tests_passed():

View File

@ -85,3 +85,6 @@ class SanityTestCaseRoot(functional_base.BaseSudoTestCase):
def test_keepalived_ipv6_support(self):
checks.keepalived_ipv6_supported()
def test_bridge_firewalling_enabled(self):
checks.bridge_firewalling_enabled()

View File

@ -94,6 +94,8 @@ class BaseIptablesFirewallTestCase(base.BaseTestCase):
RAW_TABLE_OUTPUT.splitlines())
self.firewall = iptables_firewall.IptablesFirewallDriver()
self.firewall.iptables = self.iptables_inst
# don't mess with sysctl knobs in unit tests
self.firewall._enabled_netfilter_for_bridges = True
class IptablesFirewallTestCase(BaseIptablesFirewallTestCase):

View File

@ -2746,6 +2746,8 @@ class TestSecurityGroupAgentWithIptables(base.BaseTestCase):
context=None, plugin_rpc=self.rpc,
defer_refresh_firewall=defer_refresh_firewall)
self._enforce_order_in_firewall(self.agent.firewall)
# don't mess with sysctl knobs in unit tests
self.agent.firewall._enabled_netfilter_for_bridges = True
def _device(self, device, ip, mac_address, rule):
return {'device': device,
@ -2794,12 +2796,6 @@ class TestSecurityGroupAgentWithIptables(base.BaseTestCase):
self.assertThat(kwargs['process_input'],
matchers.MatchesRegex(expected_regex))
expected = ['net.bridge.bridge-nf-call-arptables=1',
'net.bridge.bridge-nf-call-ip6tables=1',
'net.bridge.bridge-nf-call-iptables=1']
for e in expected:
self.utils_exec.assert_any_call(['sysctl', '-w', e],
run_as_root=True)
self.assertEqual(exp_fw_sg_updated_call,
self.agent.firewall.security_group_updated.called)
@ -3123,6 +3119,8 @@ class TestSecurityGroupAgentWithOVSIptables(
context=None, plugin_rpc=self.rpc,
defer_refresh_firewall=defer_refresh_firewall)
self._enforce_order_in_firewall(self.agent.firewall)
# don't mess with sysctl knobs in unit tests
self.agent.firewall._enabled_netfilter_for_bridges = True
def test_prepare_remove_port(self):
self.rpc.security_group_rules_for_devices.return_value = self.devices1

View File

@ -0,0 +1,14 @@
---
deprecations:
- The iptables firewall driver will no longer enable bridge firewalling in
next versions of Neutron. If your distribution overrides the default
value for any of relevant sysctl settings
(``net.bridge.bridge-nf-call-arptables``,
``net.bridge.bridge-nf-call-ip6tables``, and
``net.bridge.bridge-nf-call-iptables``) then make sure you set them back
to upstream kernel default (``1``) using /etc/sysctl.conf or
/etc/sysctl.d/* configuration files.
upgrades:
- On newer Linux kernels (3.18+) you will need to load the ``br_netfilter``
kernel module before starting an Open vSwitch or Linuxbridge agent using
iptables based firewall. Otherwise the agent will fail to start.