BpfMap class supports AutoCloseable interface which closes
file descriptor only in try-exit. BpfMap class doesn't close
fds while the object is released.
Change the timing of opening and closing bpf map file descriptors
to clat is starting and stoping.
Moreover, the reason that manual close BPF map file descriptors is
as follows. Just don't rely on that GC releasing to close the file
descriptors even if class BpfMap supports close file descriptor in
finalize(). If the interfaces are added and removed quickly, too
many unclosed file descriptors may cause unexpected problems.
Bug: 230880517
Test: manual test
Steps:
1. Connect to IPv6 only wifi (GoogleGuest) and mobile data
2. Check that map fds are appeared:
/proc/$(system_server_pid)/fd/$(bpf_map_fd)
$ adb shell ps | grep system_server
system 1929 825 20311224 730060 do_epoll_wait 0 S system_server
$ adb shell ls -all proc/1929/fd | grep bpf-map
.. system system 64 2022-05-05 13:36:42 .. 331 -> anon_inode:bpf-map
.. system system 64 2022-05-05 13:36:42 .. 348 -> anon_inode:bpf-map
3. Check the clat maps are added.
$ adb shell dumpsys connectivity
NetworkAgentInfo{network{105} handle{454377263117} ni{WIFI ..
Nat464Xlat:
..
Forwarding rules:
BPF ingress map: iif nat64Prefix v6Addr -> v4Addr oif
47 /64:ff9b::/96 /2a00:79e1:abc:6f02:f182:6c29:ab56:9961 -> /192.0.0.4 62
BPF egress map: iif v4Addr -> v6Addr nat64Prefix oif
62 /192.0.0.4 -> /2a00:79e1:abc:6f02:f182:6c29:ab56:9961 /64:ff9b::/96 47 ether
NetworkAgentInfo{network{106} handle{458672230413} ni{MOBILE[LTE] ..
Nat464Xlat:
<not start>
4. Disconnect from wifi
5. Check that map fds are disappeared:
/proc/$(system_server_pid)/fd/$(bpf_map_fd)
$ adb shell ls -all proc/1929/fd | grep bpf-map
(fd 331 and 348 were not found)
Change-Id: I60c0301bf00beae5cf5ab3535c6a3da68a2a4a9b
isMangedProfile returns true for managed profiles.
But enterprise device can be fully managed like device owner.
Hence check specifically if request is coming on fully managed
device.
Bug: 226966328
Bug: 231071836
Test: ran DevicePolicyManager CTS and ConnectivityServiceTest
Change-Id: I7827466bd61e24ba9c36c3a2e25043257e2ed602
* changes:
Test VpnManager event for CATEGORY_EVENT_NETWORK_ERROR
Test VpnManager event for CATEGORY_EVENT_IKE_ERROR
Test VpnManager event for CATEGORY_EVENT_ALWAYS_ON_STATE_CHANGED
Test VpnManager event for CATEGORY_EVENT_DEACTIVATED_BY_USER
Lucas Linauthored and Gerrit Code Reviewcommited 2022-04-29 08:13:51 +00:00
(this is safe because on pre-T none of these maps and programs are mainlined
and thus safe to access from mainline code anyway)
Test: TreeHugger, manual
Bug: 218408035
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: I23e565d665247f33e084978890a1ee8ffe0fe568
Maciej Żenczykowskicommited 2022-04-23 09:03:06 +00:00
- For clatd start and stop, use ClatdCoordinator on T+ and Netd on S-
- Fix the unit test for T+ and S- devices
Note that mokito.verify(.., times(1)) is replaced by verify(..)
because times(1) is the default and can be omitted.
See verify in mockito/src/main/java/org/mockito/Mockito.java
Note that this commit needs to be merged with aosp/1956072.
Bug: 212345928
Test: atest FrameworksNetTests
manual test
1. Connect to ipv6-only wifi.
2. Try IPv4 traffic.
$ ping 8.8.8.8
3. Check bpf entries are added
4. Disconnect from ipv6-only wifi.
5. Check bpf entries are removed
6. testipv4.com shows 10/10
Change-Id: I7dfda6eec19de94e4258971effcd8a1210542473
PermissionMonitor does not store any information regarding
traffic permission. It just checks permission from each package
and sends the result to netd. In addition, bpf can store traffic
permissions for appIds only because of memory limitations.
However, there is a mistaken permission overwriting when
different packages are installed on multi-user. Because
PermissionMonitor updates appIds traffic permissions to netd
from each user, so the permission result will depend on the
latest user state.
To fix the problem, save the traffic permissions by each user and
appId on PermissionMonitor. An appId has a permission if and
only if at least one uid in that appId has that permission.
bug: 224775316
Test: atest FrameworksNetTests CtsNetTestCases \
CtsNetTestCasesUpdateStatsPermission \
CtsNetTestCasesInternetPermission
Change-Id: Ieb01fb21f6201a58723eb7e6446b127627be2c0e
NsdService is using binder call to communicate with
mdnsresponder, so there is no one to use NativeDaemonConnector
now. Thus, remove NativeDaemonConnector relevant files.
Bug: 209894875
Test: atest FrameworksNetTests
Change-Id: Ic70d9417dc85c04c537e09d8f8586a762ac9a97b
paulhuauthored and Paul Hucommited 2022-04-12 07:35:31 +00:00
- Use MDns aidl to communicate with mdns service and register
event listener to receive callback.
- Remove all NDC relevant code on NsdService.
- Use MDns aidl on NsdServiceTest.
Bug: 209894875
Test: atest FrameworksNetTests CtsNetTestCases
Change-Id: I65929dee3838fef753396e86c665abd66b6fec81
The onError callback can use the same future as the onResult callback by
calling completeExceptionally.
This change is also removing the test for a specific exception message.
Testing for a specific exception message would mean that tests break if we
ever were to update them. If we decide that there is value to
distinguish between different types of
EthernetNetworkManagementException we should probably subclass it
internally.
Test: atest EthernetNetworkFactoryTest
Change-Id: Ib6f8cf79fc26e0f856c880cbe505b101f12aa187
As per the TODO, merge EthernetServiceTests into the larger
FrameworksNetTests suite.
Similarly to NetworkStats, NSD or IpSec tests, the tests are also marked
as "non-connectivity-module-test", where "module" actually refers to
modules being built for release (from an S-based branch) today. This is
necessary as the tests and associated code cannot build without T APIs.
Also add FrameworksNetTests to presubmit as non-connectivity-module
tests are not run in presubmit without this.
Test: TH
Merged-In: Id533cdb4ac184b963f570af299dea04754ba88e9
Change-Id: I9950fcb49fdc2217134a59e993941cbe5da0b556
This code is correct on userdebug builds, but it is dangerous on
user builds because proguard might strip out the POLICY_*
constants and that would lead to crashes.
For now just log a wtf if an invalid policy name is found. A
better solution would be to make MessageUtils robust to this
problem, e.g., by having it store the SparseArray internally and
providing getters that do not throw, instead of the current
behaviour that returns the SparseArray. That is left to a future
CL.
Fix: 227161413
Test: atest FullScoreTest
Change-Id: I68b69ee9dd84773018e62c9a8f43e754ae04c486
Lorenzo Colitticommited 2022-03-29 08:38:06 +00:00
Reference Sources.bp from Android.bp, so that they are used as build
files.
Sources.bp is separate from Android.bp as it was migrated from the
Android.bp files in the old code location. They can be merged into the
main Android.bp file (or removed by removing the filegroups) after the
migration is done.
Also remove references to the AIDL filegroups, as files are now merged
into the common framework/aidl-export directory which is used as include
directory for aidl source generation already, so they do not need to be
referenced separately. framework/aidl-export can contain AIDLs for S-
API classes that became updatable API in T (the contents of
framework-t), as they are the same as any other API class for the
purposes of AIDL code generation: it's OK for any S+ code to reference
S+ API classes.
Finally, fix reference to mByteStream in IpConfigStoreTest. It was
already wrong but did not break the build because the test was not being
built in the old location.
Bug: 222234190
Test: TH
Merged-In: I81893df9f327abb84f1561b2b33027a2d23a4d65
Merged-In: I67c703e3f7aa9d5787f032a79ed62e45412baf4f
Change-Id: I9093f9ae2f9d76e597f20de48680e459891755b4
Test: New tests in this patch
also FrameworksNetTests
Test: atest CtsNetTestCases CtsNetTestCasesLatestSdk
Change-Id: I68aa5201ad94531e5a40999816844faef5531525
Chalard Jeanauthored and Chiachang Wangcommited 2022-03-23 11:19:44 +00:00
This reverts commit 7569d51446.
Reason for revert: ag/17318348 moved the declaration of the protected-broadcasts from the telephony package to the platform's manifest, so a flag is no longer required when registering for these. In addition, ag/17307852 is syncing the behavior of unexported runtime receivers with that of manifest receivers by limiting the UIDs that can broadcast to these receivers to the root and system UIDs; with the RECEIVER_NOT_EXPORTED flag from this commit, these receivers would no longer be able to receive these broadcasts when sent from the telephony UID.
Bug: 225999840
Change-Id: Ifb15506434edabdfacf84c302881d0cf97d3add8
Michael Groovercommited 2022-03-23 02:04:57 +00:00
This is a preparation for accessing BPF program and map from existing
Java class. The new class ClatdTracker stores the clatd information
for forking clatd, accessing BPF maps and attaching BPF programs.
Using an object is easier to pass all required parameters and reuse
known resource.
Also refactor existing function and rename existing variables for this
commit.
Bug: 221213090
Test: atest FrameworksNetTests
Change-Id: I02ea889de89329c52710726fbcf1d2eac61707d7
This change allows callers with the NETWORK_STACK or
PERMISSION_NETWORK_STACK to receive administrator UIDs. This will be
used by the VCN to ensure that administrator UIDs are accurately
populated, ensuring that the ConnectivityDiagnostics are triggered for
the VCN.
Bug: 193204385
Test: atest ConnectivityServiceTest
Change-Id: I3ecfbef6847dcc78adac05d0229719c97e3a220c
In some scenarios, Wi-Fi connectivity is disrupted after roam. If an application reacts to network errors by calling reportNetworkConnectivity(false), validation will immediately fail and the device will switch to cell data.
Add a way to ignore these validation failures for up to 10 seconds after Wi-Fi roams.
Bug: 216567577
Test: new unit tests
Change-Id: I58de486533bcc7c6e32b23536ae15bccb779e394
* changes:
Fix the NPE thrown when starting Ethernet service in OS lower than T.
Remove the stub ethernet service from Connectivity module.
Update hidden API files for EthernetManager API move.
Provide a stub ethernet service to build in sc-mainline-prod.
Build ethernet framework source into framework-connectivity-tiramisu
Patrick Rohrauthored and Gerrit Code Reviewcommited 2022-03-16 16:03:30 +00:00
Hungming Chen
commited 
chiachangwang
authored and
Treehugger Robot
authored and
Gerrit Code Review
commited 
Vadim Caen
commited 
Sooraj Sasindran
authored and
Lucas Lin
authored and
Maciej Żenczykowski
commited 
Natasha Lee
authored and
paulhu
commited 
Patrick Rohr
authored and
Remi NGUYEN VAN
commited 
Michael Groover
authored and
Lorenzo Colitti
authored and
Chalard Jean
authored and
Benedict Wong
commited 
Martijn Coenen
authored and
he_won.hwang
commited