Previously, they were only updated when underlying network set was
non-null.
This change also ensures that all the calls b/w ConnectivityService and
Vpn that leads to updating capabilities are on ConnectivityService
handler thread.
Additionally, it also ensures that capabilities are propagated after VPN
enters connected state.
This change also updates VPN capabilities inline from
ConnectivityService handler thread. Previously, there was an additional
loop where Vpn would update capabilities via NetworkAgent thru
AsyncChannel which posts back to CS handler thread, which could
potentially lead to delays in updating VPN capabilities.
Bug: 119129310
Bug: 118856062
Bug: 124268198
Test: atest FrameworksNetTests
Test: manual - verified VPNs capabilities are getting updated and
DownloadManager is working correctly.
(cherry picked from commit 273f900b3b)
Change-Id: Iae5f2024b19df04c31938815b52687781d016cde
Merged-In: Id0abc4d304bb096e92479a118168690ccce634ed
In handleUpdateLinkProperties(), it will always assign newLp
to nai first. Then, the copied newLp would add some configurations
ex: private dns/clatd. This updated newLp wouldn't be assigned back to
nai when linkproperties is not changed.
Bug: 113637648
Test: - build, flash, booted
- atest FrameworksNetTests
- run CtsNetTestCases
Change-Id: I9e25e46718e076d4afa784ee5e1d3abbe0f11911
Rather than use the crufty config.xml list of upstream transport types,
use ConnectivityService's notion of the default network for the upstream.
In cases where a DUN network is required and the default network is
currently a mobile network, look for a DUN network (code in Tethering
is currently responsible for requesting one).
Test: as follows
- built, flashed, booted
- runtest frameworks-net
- tethered via mobile, joined captive portal network, maintained
laptop access via mobile until captive passed (then used wifi)
- disabled client mode wifi, disabled mobile data, plugged in
ethernet adapter, observed connectivity via ethernet
Bug: 32163131
Bug: 62648872
Bug: 63282480
Bug: 109786760
Bug: 110118584
Bug: 110260419
Change-Id: I925b75994e31df8046f3ef9916a2457b4210485e
updateLinkProperties copied the linkProperties in networkAgent,
but the clat fix-up function did not copy stacked link into new one.
This caused an incorrect clat iface removal, and the IPv4 network
to be unreachable.
Bug: 80261579
Test: 1. ping after ipv6 only data call with 2g voice call on/off
2. runtest frameworks-net
3. run cts -m CtsNetTestCases
Change-Id: Ide47a3b0680dddfcf3e2e759a59b19aee3605050
VPNs are not driven by NetworkRequests, so there's no risk of a
capability change on a VPN causing a connect/teardown loop.
Bug: 80439912
Test: builds, boots
Change-Id: Ic4c489ccc9fb97551d1ef440766f6cf6f99522db
...as opposed to after the async channel finished disconnecting.
Bug: 78308259
Test: runtest frameworks-net
also used a device with this patch over the weekend and
tried all I could think of
Change-Id: I77ad6d97abb20815b801a794eaa9685acf2d1173
This is a pinpoint fix against the bug listed below. While a client
is synchronously reading the LinkProperties of a network, the
ConnectivityServiceThread is updating its properties. Make sure
that update is done atomically.
This is a stopgap countermeasure against a problem that is
pervasive with usage of LinkProperties, but fixing the problem
itself will happen later.
Bug: 80077223
Test: runtest frameworks-net
Change-Id: I9302f8fb5303cb39aa82691d4f6d7f38707a41fa
Support keeping IpClient logs around and dumping them
during dumpsys. Previously we got this benefit for
wifi by virtue of WifiStateMachine's long-lived nature.
Now that this is changing we need to be sure we have
logs, and this method gets us Ethernet logs as well.
Bug: 62476366
Bug: 77999594
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
- dumpsys connmetrics [ipclient] works
Change-Id: I1136a83de8097fdb4130debe1eaf689be7132fe5
ApfFilter maintains separate counters for each reason why a packet was
passed or dropped by the filter logic.
There's also a total which should match the individual counters,
*unless* the APF interpreter aborted execution early due to an illegal
instruction or an out-of-bounds access.
Test: both on APFv2 and APFv4-capable device:
runtest -x tests/net/java/android/net/ip/IpClientTest.java
runtest -x tests/net/java/android/net/apf/ApfTest.java
manual tests connected to an AP
Bug: 73804303
Change-Id: I54b17fcbb95dfaea5db975d282314ce73d79d6ec
Bug: 77737389
Test: runtest framework-net
new test don't pass without the main code change, but they
do with it
Change-Id: I0cd83a935ab0b349aa47e065b830e5a43ab9a091
Relies on events sent from netd in aosp/578162.
Test: Added tests to ConnectivityServiceTest. Added a new test
class DnsManagerTest. Built a simple app that appears to
receive onLinkProperties events correctly upon manual changes
to the private DNS settings on a Pixel.
Bug: 71828272
Merged-In: I1e6c54ba016f6a165a302bd135a29d9332aaa235
Merged-In: I7705412803fb9aa707a18ae5a1c50292e084d851
Change-Id: I3223c1285a73d5d531c5051ce70007857caa57e3
(cherry picked from commit f4977499c4)
Moves this out of ConnectivityService and into each NetworkMonitor
(where it's more self-contained).
Test: as follows
- builds, flashes, boots
- runtest frameworks-net passes
- manual testing with working and non-working hostnames behaves
somewhat (but not entirely) as expected, and not always quickly
Bug: 64133961
Bug: 72345192
Bug: 73872000
Bug: 77140445
Merged-In: I5dc90ecfe6f6f10967b7501645ad8e030cb38982
Merged-In: Ida4967d22f0781524f0f269e30e653b8ec867258
Change-Id: Ic4322af3cb49149f2d975cb31f54b2ac7927f907
(cherry picked from commit 076bd3d8e3)
This change comprises the following parts:
[1] android.net.dns.ResolvUtil, containing methods that encapsulate the
use of the high bit in netids used in DNS resolution contexts.
[2] Updates to captive portal apps to call the ResolvUtil method that
enables DNS-over-TLS bypass for the captive portal app process.
Test: as follows
- builds
- flashes
- boots
- runtest frameworks-net passes
Bug: 64133961
Bug: 72345192
Merged-In: I0994b53d24ed25a2eb9e65429c61cf6fa87c7513
Merged-In: I4c49e23d8caa4d485df1c1d2f135a7282d439c0b
Change-Id: I2072c1f68d6978fa0d7e9d8693135a2c51bb0f87
(cherry picked from commit 95a39bec90)
Tethering currently wants access to complex isTetheringSupported
logic that is only available in ConnectivityService. Instead of
trying to access that via ConnectivityManager, pass this capability
in to Tethering directly, in the TetheringDependencies object.
Also:
- ConnectivityManager is only a source of static constants now,
so "import static" all the constants that are actually used.
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net works
- manual USB towards WiFi tethering works
Bug: 68951715
Merged-In: Ifa121b057f9959ddb980edc940327929e48ea973
Merged-In: Iad6358dc2f1d10b322d22ec90543adc50882962d
Change-Id: Ia64faaadefb4a5d84a50da98bdebd544b6fda101
(cherry picked from commit b9f6f7e668)
Useful for clients such as BatteryStats which currently rely
on NetworkStatsFactory. Data at that stage is incomplete as
it does not account for tethering, VT data and corresponding
464xlat corrections.
Test: runtest frameworks-net, CTS tests pass.
Bug: b/72107146
Merged-In: I31c5b9b4a7c6e72910152415894a137f000a5858
Merged-In: I2527d95000c7500c824ede70f87ecb38e21ed323
(cherry picked from aosp 6f124c455f)
Change-Id: Ie80f1bb21124241f3414f9be77aceac9a44ec6d1
This will let ConnectivityService send the right callbacks to the
relevant apps.
Test: manual with apps
runtest frameworks-net
cts
new tests for this functionality
Bug: 67408339
Change-Id: I6f08efd9e73c7e191f833d7f307a3bf4c9e2f0b4
Fix test breakages I caused when adding cell
support for NATT keepalives.
-Make the minimum keepalive interval a constant in
ConnectivityManager and use it in tests.
-Re-Disallow IPv6 Keepalives
Bug: 73327535
Test: 'runtest -x ConnectivityServiceTest' now passes
Change-Id: I5ec4367d250ee371014e65c897c3897a25a05e2d
This has no concrete impact on the behavior of ConnectivityService,
but in principle TRACK_DEFAULT requests should not be counted toward
requests that make a network foreground. It does not have an impact
because only VPNs could be affected by this, and VPNs are always in
the foreground by definition.
Test: runtest frameworks-net
Test: cts
Change-Id: Id2ae6b5c9d542fe168e64ed713b6ec0a04062c82
NOT_SUSPENDED and FOREGROUND are capabilities that need to
be public so as to reach feature parity with what information
can be gotten through the use of CONNECTIVITY_ACTION and
synchronous calls to ConnectivityManager. This change makes
them public, and wires up the NOT_SUSPENDED capability.
This deprecates in effect the old onSuspended and onResumed
callbacks, but these have never been public.
This also converts the onAvailable path from a multiple
binder call design to a simpler, single binder call. This
is only for internal convenience
Test: runtest frameworks-net
Test: cts
Test: also manual testing
Change-Id: I6ea524bb361ecef0569ea2f9006c1e516378bc25
Prior to this change ConnectivityManager used to patch in the UID
of the requesting app inside the NetworkCapabilities sent to it.
The rationale was that the app may not know what other apps may
use the network, so the view it should have of the network should
always say the network only applies to that app.
But this has an unfortunate side effect : apps can't match the
received network against a default NetworkCapabilities. Ostensibly
this only applies to the system because all involved calls are
@hide, but still : system code would get some NetworkCapabilities,
for example using networkCapabilitiesForType, and then try to
match the capabilities of an available network using
satisfiedByNetworkCapabilities. Because the passed network is
declared to only apply to one's own UID and the UIDs of the
NetworkCapabilities are set to null meaning "I need this network
to apply to all UIDs", the answer will be "false".
While this is WAI in a sense, it is very counter-intuitive that
code trying to match a network would be required to patch in its
own UIDs.
There are three ways of fixing this :
1. Require all apps to do the above. It's correct, but it's
cumbersome and counterintuitive. Multiple places in existing
code needs to be fixed, Tethering is an example.
2. Write the UIDs of the caller in any NetworkCapabilities object
that is created. This is not very practical, because it imposes
the converse requirement on all NetworkAgents, which would then
have to clear the UIDs before they send the capabilities to
ConnectivityService. All NetworkAgents need to be fixed.
3. Instead of sending an object with a list of one UID to apps,
send a null list. The drawback is that the networks nominally
look to apps like they apply to all apps. I argue this does
not matter ; what matters is that the UID lists do not leak.
Clients just see a null list of UIDs (and third party can't
even access them without using reflection). No other changes
are required besides this two-line patch.
This patch implements 3. I believe it is the saner approach, with
both the most intuitive behavior and the best backward compatibility
characteristics, as well as the easiest change.
This does not encroach on the future plans to make the actual
UID list available to apps with NETWORK_SETTINGS.
Test: runtest frameworks-net
Change-Id: I978d91197668119e051c24e1d04aafe1644a41cf
