TcpKeepaliveController is the only user of
KeepalivePacketDataUtil.fromStableParcelable. Because of
fromStableParcelable, networkstack-client needs to depend on
net-utils-framework-commonm, which pulls a lot of unnecessary classes.
This is particularly problematic considering that networkstack-client
may need to be redistributed as a prebuilt.
Move the method to TcpKeepaliveController, simplifying dependencies.
This also shows that fromStableParcelable could be removed altogether
(or moved to tests) if TcpKeepaliveController built a
TcpKeepalivePacketData class directly.
Test: atest ConnectivityCoverageTests
Change-Id: I554318f6bcd07c73d153598a0231e9fcaf912e90
NetworkTemplate.buildTemplateMobileWithRatType() supports
to build a template with metered filter, so a mobile network
is metered or non-metered should be tested.
This commit also rename buildMobile3gState to buildMobileState
because originally this method uses NetworkInfo to simulate the
network type. The NetworkInfo related codes were removed.
Currently, setMobileRatTypeAndWaitForIdle is used to simulate
the network type in test.
Ignore-AOSP-First: The parent of this change contains a CL will
conflict with internal tree so ignore AOSP first.
Bug: 183776809
Test: this
Change-Id: I8e54e2acb06db81fb0eaf330c4fd7cb391b01ff3
NetworkIdentity has adapted NET_CAPABILITY_TEMPORARILY_NOT_METERED
network capability so add a test case for it.
Ignore-AOSP-First: The parent of this change contains a CL will
conflict with internal tree so ignore AOSP first.
Bug: 183776809
Test: this
Change-Id: I2ca2a8b4db664b76a5a5ee82fcff451efd87c8ff
The method buildTemplateMobileWithRatType is updated to take
metered as a parameter so the callers should be updated to
the new version.
Bug: 183776809
Test: atest FrameworksNetTest
Change-Id: I859514b93705f9ee4f9ae4b0bcddcb61005a78ff
Merged-In: I859514b93705f9ee4f9ae4b0bcddcb61005a78ff
Instead of crashing when parceling the NetworkInfo object,
crash at the time the bad call is made.
Bug: 145972387
Test: FrameworksNetTests
Change-Id: If8b5fd3d7b800c97211bcd16c9a8c5812708d4ab
Now, VPN will set underlying networks into NetworkCapabilities
directly. So the declaredUnderlyingNetworks can also be set
directly when creating a NetworkAgentInfo.
Bug: 191918368
Test: atest FrameworksNetTests:ConnectivityServiceTest
Change-Id: I507072d00ae1eb0c391e5261ab93e359b9c4cb5c
In coverage tests this seems to randomly fail, which suggests
some delay. Have a constant for timeout and increase its value
significantly where it makes no functional difference.
Test: FrameworksNetTests
Change-Id: I035d865f01688daf3bce30c5130ce550fa84b885
"when" is not thread-safe, as it relies on global state to find which
mock was called with the method to mock in its parameters.
This causes flakes where non-test code that interacts with the mock may
be called from another thread between the contents of "when" and the
"when" method another thread, causing UnfinishedStubbingExceptions or
other test errors. In particular
ConnectivityService#getNetworkCapabilitiesInternal was seen to be
wrongly identified as a mocking site for Dependencies.
Replace all usages with doReturn().when(mock).method() syntax, which has
better thread safety since global state is not necessary to tie the
mock, mocked method and return value.
Bug: 195626111
Test: atest ConnectivityServiceTest
Change-Id: I57c5ffb3b3f799fc59c3af4ccb323fb5d6794fad
The registered listeners are added into internal ArrayMap as a
reference but are not removed when the listeners are
unregistered. The actions for registration should be done in
pairs.
Test: atest FrameworksNetTests
Change-Id: Id9e674f5104d1471dd81224b6a271a8a92172e34
Upon changing the default SIM card, the radio will create a
new connection to the new subscription. If that subscription
works correctly, the stack will prefer it to the old one as
the new subscription will be marked with a Primary policy
flag it its score.
Normally, at this point the old network lingers to give apps
an opportunity to gracefully migrate their connections. But
with some radios, this may have a dramatic effect on the
performance of the new connection.
This patch introduces a flag so that devices with such radios
can be marked. In this case the stack will move to a degraded
mode and eschew the grace delay for apps and give them a hard
break instead, so that the new network can reach a good
performance immediately. Apps with existing connections will
suffer a worse experience.
If there is a request that can only be served by the old
connection, still keep it, as arguably the user still
expects their MMS be sent on the old connection, even if the
new connection doesn't work well until it's done.
Test: new test in this patch, and add relevant tests in both modes
also manually change the value of the flag and run
FrameworksNetTests and CtsNetTestCasesLatestSdk
Bug: 200226979
Change-Id: I4ace82f90e873bf06298cc689bb1d794ed5124bd
Add two test cases for matchesMobile and matchesMobileWildcard
Bug: 183776809
Test: FrameworksNetTests:NetworkTemplateTest
Change-Id: I14d31071655204fe6ad36c71f7f521640e01c96e
Before sForceAllNetworkTypes is removed, the network
type and meteredness will be ignored when matchesMobile
or matchesMobileWildcard is called.
After sForceAllNetworkTypes is removed, the matches
method should check the network type and the meteredness.
Thus, if the test data contains different type or it's
not metered should not be counted.
Bug: 183776809
Test: FrameworksNetTests
Change-Id: Ie7194495d26c0f5ef7a247733f43c64688626c67
Suspended network should be considered as temporary shortage of
connectivity of a connected network. Thus, it should not be
excluded from network state snapshots and causes data usage to
stop accounting or iptables rules to be removed on the interface
of the suspended network.
This change also address the naming confusion of default networks
parameter of expectNotifyNetworkStatus.
Test: atest ConnectivityServiceTest#testGetAllNetworkStateSnapshots
Bug: 196079981
Change-Id: I8096356f9a472fb1c1246fbdf3fd5f981387fb1c
- This is a test only change.
- Remove calling startMonitoring() in setup() because this method
should be called only once, so each test needs to call
startMonitoring() if the testing function is running on it.
- Return empty list instead of null when getting installed
pacakges in setup(). This can help test to have default user
MOCK_USER1 when they call startMonitoring().
Bug: 192431153
Test: atests FrameworksNetTests
Change-Id: Id2ffb056b378873c3ba6a8bb31b7dedb56ad6d46
- Some code are used many times, it's better to make them as
common code for reducing inconsist behavior.
- Also stop using Boolean to represent network permissions and
replace them with int value. Because using Boolean for
permission comparison is really complicated and bizarre.
- Use PERMISSION_* for netd network permission directly.
Bug: 189705071
Test: atests FrameworksNetTests
Change-Id: I2a4e298b9a01f4b2874ae68e9d9539a0ab4aff4c
The test flakes with UnfinishedStubbingException, but the root cause
could not be identified. Add some logging to try to get more information
when that happens.
Bug: 195626111
Test: atest ConnectivityServiceTest
Change-Id: If12c1ea809789148ca9251386e5ee6ca6d74ff74
Some uids should be app ids, correct them for avoiding confusion
and incorrect use.
Bug: 189705071
Test: atests FrameworksNetTests
Change-Id: I4a5930e5dc63b4d901e1567f8935ad7203866c89
- This is a test only change.
- Some methods are very similar and duplicated. So merge them to
improve readability and reduce code complexity.
- Stop spying PermissionMonitor.
Bug: 189705071
Test: atests FrameworksNetTests
Change-Id: I8ec17bd2d396c4d49dd8b64be85d89d0145f4c3c
The test may be run on S builds older than itself, in which case the
resource ID may not match. Query the resource ID dynamically instead of
using the constant.
IpSecAlgorithm is not updatable in S, which is why it can use
hidden/unstable platform resources, so the test could also be skipped
altogether on S. But considering that keeping it around is not very
costly yet, this change fixes it for running on S.
Bug: 197489543
Test: atest IpSecAlgorithmTest
Change-Id: I533eede3a360a8db9089e65e017b9d7453b35539
When the avoidBadWifi configuration is false and not overridden,
a WiFi network that was validated in the past but becomes
unvalidated needs to outscore a cell network that is validated.
This is happening correctly when the stack compares two networks.
However, when the stack compares an existing network to an offer
for a cellular network, the offer was automatically considered
not to yield. This would mean the stack would be requesting cell
out of the telephony factory, only for that network to lose to
WiFi and be discarded immediately, then recreated again etc.
When there is some other reason cell should be up (such as the
"mobile always on" setting being active), this would not be
visible because the cell network would have another reason not
to be torn down.
Have offers correctly account for the current value of the
configuration and setting. This has the ranking of the offer
lose against WiFi like the actual network loses, meaning the
offer is not needed.
This also requires updating the offers whenever the value of
the setting changes.
Test: new test for this, also ConnectivityServiceTest
Bug: 195441367
Change-Id: I4fe5de98bc15bcf9bbbe25c6c7c8a7ba382f8db7
