* changes:
Rename hasCarrierPrivilege... to isCarrierServiceUid...
Detect discrepancies in subIds
Allow carrier service UIDs to access their own WiFi networks.
Jean Chalardauthored and Gerrit Code Reviewcommited 2023-11-20 06:55:04 +00:00
This method tests whether the UID is the carrier service
UID, not whether the UID has carrier privilege.
Note that in normal operation, the carrier service UID always
has carrier privilege.
Test: NetworkAgentTest
Change-Id: I30f8814cc0419a2d6bdbe75fdc82156516eba026
This tests that the network providers are subject to appropriate
limitations as to what they can set as allowedUids – both in the
positive (can do) case, and the negative (can't do) case.
Test: this
Change-Id: I115e2a4bc02ddcd03ecf2f35130fcb0378da22bd
This reverts commit 30fb9b1ca4.
Reason for revert: Potential culprit for CtsLibcoreTestCases test failures b/311612650
Change-Id: Ie7de97159f92a174476a8fa8ea83e3e01914aea1
Michał Brzezińskiauthored and Gerrit Code Reviewcommited 2023-11-17 11:18:56 +00:00
As a specialized network, a network with PRIORITIZE_* is not
generally a good default network, and might cause unexpected
charges if allowed to become one.
Test: CSDefaultNetworkTests
Change-Id: I8f7184f80bb987da720bd067862bd7a546a38e26
Ideally the rules should be set up before onAvailable is
sent, but onLocalNetworkChanged should be sent after.
The previous code had CSLocalAgentTests#testForwardingRules
flaky because it was waiting for onAvailable and immediately
checking that the routing rules have been set up.
There are other ways to fix this flake (e.g. wait for
onLocalNetworkChange in the test or add a timeout on
the verify(netd) call) but this patch probably implements
the best production code behavior.
Bug: 309688089
Test: CSLocalAgentTests#testForwardingRules 1'000 times
Before this patch : flakes after 50~200 iterations
After this patch : no flake in 1'000 iterations
Change-Id: I4dcccaa3607c1e28aa89f268f0c721d87b1e9466
This is a preliminary change for filtering sockets that is not
in the uid ranges for automatic on/off keepalives. This commit
itself is a no-op change to pass the uid information to
AutomaticOnOffKeepaliveTracker.
Bug: 311119352
Test: atest FrameworksNetTests
Change-Id: I7d96e7a0d3f3054d1409de350420a24378b28cdb
SysUi needs to see connectivity diagnostic callbacks to display
"Checking for internet access..." -> "Connected" or "No internet access".
Bug: 307161342
Test: atest FrameworksNetTests, manually connect to a network without internet connectivity,
verify Internet Dialog doesn't get stuck showing
"Checking for internet access..."
Change-Id: I86e2b852ff9e5dfbfc37ba711f477166b0ac6835
Quang Anh Luongcommited 2023-11-15 11:06:42 +09:00
Currently, BatteryStatsService register netd callback and track radio
power state only for the default network.
So, battery stats are not accurate if cellular and Wi-Fi networks are
active at the same time.
aosp/2561090 updates BatteryStatsService not to register netd callback
on V+ devices.
This CL updates LegacyNetworkActivityTracker to call BatteryStats API
and update radio power state for networks including non default networks
on V+ devices.
Bug: 267870186
Bug: 279380356
Test: atest FrameworksNetTests
Change-Id: I826b3b4046cc7b8aef46c13300854eaf14b1b777
on V+ devices, NetworkActivityTracker adds idleTimer when the network is
first connected and removes idleTimer when the network is disconnected
so that activity tracker can receive netd
onInterfaceClassActivityChanged callback for multiple networks.
Bug: 267870186
Bug: 279380356
Test: atest FrameworksNetTests
Change-Id: I6f3a95c27e3e58f3f60c40065f562d00431a56b1
Following CL adds idleTimer for multiple networks and track activities
including non default network on V+
Current LegacyNetworkActivityTracker uses transport type as an idleTimer
label but this makes it difficult to have multiple idleTimers.
So, this CL updates LegacyNetworkActivityTracker to support using netId
as an idleTimer label.
Bug: 267870186
Bug: 279380356
Test: atest FrameworksNetTests
Change-Id: I0039f5624ae1d142dbacba53aa90ca9bf6d43599
A new firewall chain is needed to configure background network
restrictions for apps.
This change only adds the API stubs and traffic controller constants to
make the chain work. Policy changes using this chain will follow in
the framework code.
Test: atest CtsNetTestCases:ConnectivityManagerTest
Test: atest ConnectivityServiceTest
NO_IFTTT=The Lint rule along with the relevant code in Common.h is
being deleted in aosp/2819759
Bug: 304347838
Change-Id: I33e2db6671431f7c576fc931d9f96e684fc1e78a
Also add some toString() methods that were useful in debugging.
Test: new test in CSLocalAgentTests
Change-Id: Ife95815e39d92bbef84b1c5ea75a151882590d09
Avoid network agents trying to do things that the stack
does not support.
Test: new test in this change
Change-Id: I3ffa1c3ddbb0b648b06862dc0a44f8525d358acc
This is needed for data stall detection mechanism in NetworkStack
to get the information about whether the network is blocked for
a given uid and conditions. Because the API will be called
frequently from NetworkStack to resolve all status for all uids
on the device, the API cannot call into the service which
creates IPC. Instead, the API need to directly access bpf maps
in the user process to retrieve the status. In this case the
user process is the network stack, the access control is provided
by linux file permission and selinux.
Test: atest FrameworksNetTests:android.net.connectivity.android.net.BpfNetMapsReaderTest
Test: atest FrameworksNetTests:android.net.connectivity.android.net.ConnectivityManagerTest
NO_IFTTT=Refactor only change for firewall chains definitions
Bug: 297836825
Change-Id: Iaf983b71ec98cbfe5152dcfade8a3120f938f135
Dns Resolver needs to know whether a network is metered when checking if
DNS requests will be blocked by Data Saver.
Bug: 288340533
Test: atest FrameworksNetTests
Change-Id: Ia5822dc522c766c7815680003c7ba275d15ccaff
The information is needed by modules who want to know whether a
specific UID is blocked by Data Saver feature.
1. Add a one-element map data_saver_enabled_map.
2. Update current data saver setting to the map.
Bug: 288340533
Test: atest FrameworksNetTests:android.net.connectivity.com.android.serv
er.BpfNetMapsTest
Test: atest bpf_existence_test
Change-Id: I981da4b569247c33cba2d365cb6f2691f673474e
In particular, this fixes a bug where a non-validated destroyed
network can't get disconnected after it's replaced by another
non-validated network because CS thinks it should keep it in
case it validates and beats the replacement network.
Test: new test for this : CSDestroyedNetworkTests
Change-Id: I5d7e413624d6fca28b06484e5369cd17e4a599a4
1. At present, the transportTypes in NetworkCapabilities have been
passed into DnsManager::updateTransportsForNetwork() as a parameter. In
the following CL, we also need to pass the 'metered' (also in
NetworkCapabilities) as a parameter to the function. Instead of passing
the members of NetworkCapabilities one by one, it is better to pass the
entire NetworkCapabilities.
2. Rename updateTransportsForNetwork() to
updateCapabilitiesForNetwork().
Bug: 288340533
Test: presubmit
Change-Id: I0966124f87b12c1d5a2eaee681885c3d5c7f74dc
(mostly to improve code coverage)
We can always revert if it ever turns out this is useful,
but I'd prefer for cgroup attach to be a once-at-boot
thing and handled either from bpfloader or the netd updatable
which are both directly C++ code.
Test: TreeHugger
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: I79b5ca8a47388cb6b9189234942e74ab6056aab9
Maciej Żenczykowskicommited 2023-10-25 17:40:35 +00:00
Revert submission 2799494-revert-2745215-npmsbpf-OBOPDXREUI
Reason for revert: Need to 1. Merge API first. 2. Wait for prebuilt. 3. Merge the caller.
Reverted changes: /q/submissionid:2799494-revert-2745215-npmsbpf-OBOPDXREUI
Change-Id: Icd5f3d40595a0d2a221b016cec70568bf3597c16
Revert submission 2745215-npmsbpf
DroidMonitor: Potential culprit for Bug 307256512 - verifying through ABTD before revert submission. This is part of the standard investigation process, and does not mean your CL will be reverted.
Reason for revert: 307256512
Reverted changes: /q/submissionid:2745215-npmsbpf
Change-Id: I27119a7484453d1369bbdf71ad2479edec7d664f
Jordan Silvaauthored and Gerrit Code Reviewcommited 2023-10-23 13:23:01 +00:00
Currently, data saver switch is controlled in NPMS, which
calls into NetworkManagementService and netd when switching
status. In netd, BandwidthController manipulates the
bw_data_saver chain to control the overall behavior.
However, this code are all platform implementation, which
are not updatable. In order to migrate data saver switch
from iptables to bpf, this API is needed for Connectivity to
change the implementation to directly write bpf in later
patches.
Test: atest CtsHostsideNetworkTests:com.android.cts.net.HostsideRestrictBackgroundNetworkTests
Test: atest FrameworksServicesTests:NetworkManagementServiceTest
Bug: 297836825
Change-Id: I71fa41ca739fef8e191fba91b02758ad5f732d5c
I0afdda023208c3f8620cb5b89add66448af596d7 added some assertions
to ensure NetworkAgentInfo can only be accessed from the handler
thread. This is good to eliminate any potential concurrent
accessing problem. Howver, dumpsys runs on binder thread, this
change breaks dumpsys functionality.
Thus, this change:
1. Copy Handler#runWithScissors() and related class
locally since they are @hide methods.
2. Run doDump on the handler thread and waits for return.
Test: adb shell dumpsys connectivity
Test: atest FrameworksNetTests:android.net.connectivity.com.android.server.HandlerUtilsTest
Fix: 303348620
Change-Id: Ic3a22f5d370bbf8c18b1972e86d695b6ad61bd0d
Jean Chalard
authored and
Gerrit Code Review
commited 
Michał Brzeziński
authored and
Chiachang Wang
authored and
Motomu Utsumi
authored and
Quang Luong
authored and
Suprabh Shukla
commited 
Junyu Lai
authored and
Ken Chen
commited 
Maciej Żenczykowski
commited 
Jordan Silva
authored and
Anton Kulakov
commited