I'm submitting this fast-track for Sowmini Varadhan. The release
binding is Minor and new zonecfg(1M) properties are Committed.
Summary:
-------
This case proposes a solution for
6944327 need to support address and defrouter resources for
exclusive-IP zones
Problem Description
--------------------
Typical zone deployments that exist today use shared-IP zones to run
applications and services like Apache or Weblogic in the contained
environment provided by the shared-IP zone. In these use-cases, the
Administrator in the global zone has full control over the networking
resources used by the non-global zone. In the common case, networking
is simply configured by specifying the IP interface, IP addresses and,
optionally, the default routers from zonecfg(1m). The configuration
resources thus supplied are then applied for the non-global zone when
it is booted, and the non-global zone itself may not modify any of
these configuration parameters.
However, there is no such simple configuration mechanism for the
simple networking use-case in place for exclusive IP zones, which must
be configured through sysidcfg, ifconfig, and an assortment of other
methods, all of which are not controllable from the global zone, and
may not be managed through zonecfg(1m).
The addition of many new virtualization and resource management
features to Solaris (PSARC 2006/357, PSARC 2005/132 and associated
cases) makes Exclusive-IP zones a cleaner and more powerful Zone model
than shared-IP zones. Thus, bridging any gaps in the configuration
methods for the simple use cases between shared- and exlusive-IP zones
is important to ease the transition of shared-IP customer
configurations to exclusive-IP.
In addition, there are many ongoing projects in the "Zones Networking"
effort [ZONES-NET] to facilitate the consolidation of existing Solaris
10 host installations as Solaris 10 Containers. These efforts would
leverage from the ability to specify networking resource values for
address and default router uniformly for zones using zonecfg(1m).
This case proposes support in zonecfg(1m) for 'address' and
'defrouter' properties in the 'net' resource for exclusive-IP zones.
The semantics of the values will be the same as with the shared-IP
zone definitions that exist today ([PSARC/2002/174], [PSARC/2003/621],
[PSARC/2008/057])
When the global-zone specifies the 'address' for an interface via
zonecfg(1m), the non-global zone may not use any other addresses for
the specified IP Version on that interface. The address information
provided via zonecfg(1m) will be used to set up Layer-3 protection
[PSARC/2009/436] for the non-global zone during zone-boot to filter
out all other addresses for the selected interface. For instance,
when zonecfg(1m) has been used in the global-zone to set one or more
IPv4 addresses on an interface, an attempt to set an IPv4 address on
the interface that is outside the globally defined set will encounter
the EPERM failure. Thus ifconfig(1m), ipadm(1m), and associated
ioctls will receive this error if they are used within the non-global
zone to set addresses that are not in the set that is permitted from
the global-zone, and attempts by the non-global zone to turn on
forwarding on the interface will also encounter EPERM. Note that IPv4
and IPv6 are considered as independent resources, so that
specification of an IPv4 address via zonecfg(1m) does not place any
constraints on permissible IPv6 addresses, and vice-versa.
Attempts to boot a zone that has already been configured for IP, or
has previously customized values for Layer-3 protection
[PSARC/2009/436] will fail.
Implementation Overview:
------------------------
A brief overview of the implementation is provided here. Note that all
interfaces between zoneadmd(1m) and the kernel/zonecfg(1m), are
Private interfaces, subject to change in the future.
When the non-global zone is booted, zoneadmd(1m) will store the
information specified by the 'address' and 'defrouter' properties as
nvlists in the kernel following a mechanism similar to that in use
today for other zonecfg(1m) resources such as the "physical"
datalink. At the same time, the zone boot process will also ensure
'ip-nospoof' protection for the datalink with the specified addresses
used as input to the 'allowed-ips' property. The stored nvlist
information for 'address' and 'defrouter' will be retrieved and
re-applied in the non-global zone by the daemon associated with the
ip-interface-management service before any other IP configuration is
applied.
The GLDv3 layer communicates the set of "allowed-ips" to the IP layer
through DLPI notifications. The notification sent is a DL_NOTIFY_IND
message of (Project Private) type DL_NOTE_ALLOWED_IPS, sent to the IP
clients that have registered for this notification. The payload of
the DL_NOTE_ALLOWED_IPS message is the mac_protect_t data-structure
introduced by PSARC 2009/436. Receipt of this notification enables
the IP layer to track the current set of permitted IP addresses per
interface, so that the IP layer can return informative error
diagnostics if an attempt is made to enable forwarding or set
addresses on the interface that are outside the permitted set.
All configuration state, including datalink properties like
protection, allowed-ips and zone, and associated nvlist information,
that is created during zone boot will be reset to the default values
when the zone is halted.
Relationship to other configuration mechanisms
----------------------------------------------
sysidcfg:
The Solaris Install team is currently working in conjunction with
the Network Configuration team [INSTALL] on the implementation of a
framework to replace sysidcfg(4) that will provide configuration
profiles to be processed by startup scripts when a system
boots. However this configuration information is only applied on
the first boot of the zone, and unlike zonecfg, may not be easily
modified in the global zone across reboot. Moreover, there is no
mechanism in place to prevent the non-global zone from modifying
the configuration provided via profiles.
As mentioned earlier, the global-zone is the authoritative source
of IP configuration information for the "secure container"
use-case. This implies that attempts to use boot profiles in the
non-global zone to add addresses other than those mandated in the
global-zone will fail with EINVAL erors.
Note that the profiles defined in [INSTALL] may still be applied to
interfaces for which 'net' resource properties have not been
defined in the global-zone, i.e., this proposal is fully
backward-compatible.
svc:/network/physical:{default, nwam}
Configuration specified via zonecfg(1m) in the global-zone
overrides any information specified via /etc/hostname.intf,
ipadm(1m) or nwam. Thus if the IP address information specified by
these other methods is not a subset of the globally managed
information, attempts to apply it within the non-global zone will
encounter an error.
DHCP:
It is assumed that when IP resources for an IP interface have been
set up in the global-zone using zonecfg(1m), then that interface
has been marked for static IP configuration only, and attempts to
send out outgoing DHCP DISCOVER packets will be dropped in
GLDv3. Thus attempts to start DHCP on these interfaces in the
non-global zone will time out.
Examples
--------
r...@gz# zonecfg -z tz1
zonecfg:tz1> info
zonename: tz1
zonepath: /rpool/zones/tz1
brand: ipkg
autoboot: false
bootargs:
pool:
limitpriv:
scheduling-class:
ip-type: exclusive
hostid:
net:
address: 11.1.1.1/24
physical: vnic0
defrouter: 11.1.1.2
net:
address: taddr2
physical: vnic0
defrouter not specified
net:
address not specified
physical: vnic1
defrouter not specified
In this example, the zone 'tz1' has been assigned 2 interfaces: vnic0
and vnic1.
The vnic1 interface does not have any IP address or default router
assigned to it, so that it may be configured in any desired way from
within the zone using ifconfig(1m), ipadm(1m), NWAM etc.
The vnic0 interface has 2 addresses assigned: 11.1.1.1/24 (with
default router 11.1.1.2 on the 11.1.1.0/24 subnet) and the address
'taddr2' whose numeric value and netmask will be resolved when the
zone boots. Thus, if, at the time of zone boot, the global-zone is
also configured as:
r...@gz# egrep 'hosts|netmask' /etc/nsswitch.conf
hosts: files nis dns
netmasks: files nis
r...@gz# grep taddr2 /etc/hosts
12.2.3.4 taddr2
r...@gz# grep 12 /etc/netmasks
12.2.0.0 255.255.0.0
tz1 will be booted as
r...@gz# zoneadm -z tz1 boot
r...@gz# zlogin tz1 ipadm show-addr
ADDROBJ TYPE STATE ADDR
vnic0/? static ok 11.1.1.1/24
vnic0/? static ok 12.2.3.4/16
lo0/v4 static ok 127.0.0.1/8
lo0/v6 static ok ::1/128
r...@gz# zlogin tz1 netstat -rn
Routing Table: IPv4
Destination Gateway Flags Ref Use
Interface
-------------------- -------------------- ----- ----- ----------
---------
default 11.1.1.2 UG 2 2
vnic0
11.1.1.0 11.1.1.1 U 3 2
vnic0
12.2.0.0 12.2.3.4 U 2 0
vnic0
127.0.0.1 127.0.0.1 UH 2 0
lo0
Routing Table: IPv6
Destination/Mask Gateway Flags Ref Use
If
--------------------------- ---------------------- ----- ---
------- -----
::1 ::1 UH 2
0 lo0
Proposed man page changes:
--------------------------
--- zonecfg.1m.orig Thu Apr 22 11:22:12 2010
+++ zonecfg.1m.new Thu Apr 22 11:27:48 2010
@@ -676,8 +676,13 @@
for the interface should be specified here.
For an exclusive-IP zone, the physical property must be
- set and the address and default router properties cannot
- be set.
+ set and the address and default router properties may
+ be set. The interface specifed for the physical property
+ must not be in use in the global zone. If an address and
+ default router are specified via zonecfg(1m), these will be
+ applied to the interface when the non-global zone is booted,
+ and the non-global zone will not be able to apply any other
+ addresses to that interface.
zhadum(866)% diff -wub dladm.1m.txt dladm.1m.txt.new
--- dladm.1m.txt Mon May 10 15:24:49 2010
+++ dladm.1m.txt.new Mon May 10 15:28:09 2010
@@ -4176,7 +4176,17 @@
option's link layer address must match the
link's MAC address.
+ When a datalink has been protected by setting allowed-ips
+ to a set of one or more IPv4 addresses, any attempts to
+ configure IPv4 addresses that are not in this set will
+ fail with an EPERM error being returned to the user.
+ Moreover, the interface may not be used for forwarding
+ IPv4 packets, and attempts to set the ipadm(1m) forwarding
+ property on the interface will encounter EPERM error.
+ Analogous constraints on address configuration and
+ forwarding settings also apply for IPv6 packets.
+
dhcp-nospoof
DHCP client ID (DUID for DHCPv6) and hardware
References:
----------
[PSARC/2002/174] Virtualization and Namespace Isolation in Solaris, aka
"Zones"
[PSARC/2003/621] Zone Administration Updates
[PSARC/2008/057] Default Route For Zones
[PSARC/2009/436] Anti-spoofing Link Protection
[INSTALL]
http://solaris-networking.sfbay/index.php/Network_Configuration:install
[ZONES-NET] http://solaris-networking.sfbay/index.php/Zones_Networking
_______________________________________________
opensolaris-arc mailing list
[email protected]
