API visibility change: unhide allowing NetworkSpecifier
to be an arbitrary object.
Bug: 27533960
Bug: 36053921
Bug: 36275276
Test: builds and runs
Change-Id: I1d1705cca7ece077ef8d7c674c62d5369fedbb03
Tests various combinations of empty / non-empty NetworkSpecifiers
and changing specifiers on the fly.
Test: ConnectivityServiceTest passes
Bug: 27533960
Change-Id: Ibd6b4efc861dc51689e46f3882ba859223411bf0
This patch simplifies wakelock management by acknowledging that only one
acquisition at most is in flight at any time. This allows to remove the
serial number associated with wakelock acquisition and to avoid double
release when regaining a default network.
Example of $ adb shell dumpsys connectivity logs:
NetTransition WakeLock activity (most recent first):
03-31 00:15:13.816 - RELEASE (EVENT_CLEAR_NET_TRANSITION_WAKELOCK)
03-31 00:14:59.216 - ACQUIRE for NetworkAgentInfo [WIFI () - 101]
03-31 00:09:05.799 - RELEASE (EVENT_EXPIRE_NET_TRANSITION_WAKELOCK)
03-31 00:08:05.738 - ACQUIRE for NetworkAgentInfo [WIFI () - 100]
Test: watched wakelock section $ adb shell dumpsys connectivity while
turning on and off Wifi (with no other network).
Bug: 36703718
Change-Id: I899b0816c0e41b3991d9540e9b9a604914ff673a
This patch adds wakelock acquire and release activity to
ConnectivityService logs in dumpsys. All of the last 20 acquire and
release operations are logged.
This patch also:
- does a bit of cleanup to related dumpsys logs
- unify pretty printing of internal event ids to their names.
- delete mInetLog which was obsolete and not used.
Example of logs:
NetTransition WakeLock activity (most recent first):
03-30 02:15:40.628 - RELEASE 1 by EVENT_EXPIRE_NET_TRANSITION_WAKELOCK: not released (serial number was 2)
03-30 02:15:09.965 - ACQUIRE 2 for NetworkAgentInfo [WIFI () - 101]
03-30 02:15:01.877 - RELEASE 1 by EVENT_CLEAR_NET_TRANSITION_WAKELOCK: released
03-30 02:14:40.598 - ACQUIRE 1 for NetworkAgentInfo [WIFI () - 100]
Test: inspected result of $ adb shell dumpsys connectivity
Bug: 36703718
Change-Id: Ia8e2016fc512ee798ef0adc3f612d1f6c33d4870
Test: ConnectivityServiceTest passes on ryu on internal master
Test: ConnectivityServiceTest passes on bullhead
Change-Id: If94102c7df0257ea9e69e72b07a685ae3c2c4022
In order to avoid app breakage, we are putting back the system
properties and using selinux to disallow access from apps
targeting the O SDK or above.
This CL is cherry-picked from internal commit
aefac392ea, but with a different
commit description.
Bug: 33308258
Bug: 33807046
Bug: 34028616
Bug: 34115651
Test: system properties are readable as root and shell
Change-Id: I8d51e8e0a620d581c4251fb2f3c1d0a813b929be
This cleanup helps declutter ConnectivityService, and encapsulates the
always-on setting inside of Vpn instead of spreading it across two
classes.
In particular having the save code in one file and the load code in
another file was weird and I apologise for that.
Added a SystemServices wrapper for Settings.Secure and PendingIntent
calls to decouple some of the global state nastiness and make it
testable without forcing ConnectivityService to drive the load/save.
Test: runtest -x tests/net/java/com/android/server/ConnectivityServiceTest.java
Test: runtest -x tests/net/java/com/android/server/connectivity/VpnTest.java
Bug: 33159037
Change-Id: Ie2adb1c377adfcef0a5900dc866e6118f451b265
This avoids requiring the MANAGE_USERS permission in this function.
Bug: 32671528
Test: Manual. Also seeking unit test guidance from reviewer.
Change-Id: I841e721013b0e4b6db34d629a1e97b3cd54cd73b
- regroup together static final class fields, final instance fields, and
mutable instance fields.
- uses correct Android naming conventions for instance fields.
- reorder imports correctly.
Test: no functional change.
Change-Id: If3d73bc28972a004c2ba3cc14d869a2a679abfce
Test: as follows
- built (bullhead)
- flashed
- booted
- runtest frameworks-net passes
- manual USB tethering toggling between WiFi and mobile
Bug: 32163131
Change-Id: I57edf5114b6361f320577c7870e40f8b3cdf74ce
The state that needs to be transferred includes:
- NetworkCapabilities
- LinkProperties
- whether the network is currently suspended
Additionally:
- Rename notifyNetworkCallback() to notifyNetworkAvailable()
in order to clarify its real function.
- fix previous copy/paste error in unittest
Test: as follows
- built (bullhead)
- flashed
- booted
- runtest frameworks-net passes
- USB tethering with mobile and Wi-Fi upstream toggling
Bug: 32163131
Change-Id: Ib4460bcd5d08863a9feac9e8ab41a238897bb3ea
Add (unhide) a public API which provides network requests with a
timeout. When timed-out the (newly unhidden) onUnavailable() callback
is triggered.
Note: this CL does not add a handler to the API to be consistent
with the existing APIs. There is a separate effort (b/32130437)
to update these APIs with Handlers.
Bug: 31399536
Test: unit tests and CTS (new)
Change-Id: I45ce9ada63372cb56937bb620bfbb7729e5e25d2
This patch adds overloaded version of registerDefaultNetworkCallback
registerNetworkCallback, and requestNetwork with an additional Handler
argument that is used for running the caller provided NetworkCallback.
It also clarifies the documentation of the existing methods that
implicitly uses the internal singleton ConnectivityThread about which
internal Handler is used for running NetworkCallbacks.
Test: build, flashed, booted device
Bug: 32130437
Change-Id: Iae15f81e47e2dc0355baf2f2c1679b77e56af299
The request network with timeout was originally created with a
check of max timeout against a constant of 100 minutes. However,
the API was not public and did not implement a timeout. Any users
were internal and never got any onUnavailable() callback (since
timeout never triggered).
There is no reason to have a max timeout so the constant is
remove.
Bug: 31399536
Test: unit tests and CTS of ConnectivityManager
Change-Id: Icbedfb4299d75b6a7e3e43720111531f1faafd06
ConnectivityManager static sCallbackHandler is referenced and directly
used in a way that is not ensuring its proper initialization.
This patch fixes this potential NPE by using getHandler() instead.
Also this patch changes sendRequestForNetwork's signature to only accept
the subtype CallbackHandler instead of Handler: without using
CallbackHandler the NetworkCallbacks are not triggered properly and
bookkeeping of sCallbacks does not happen. sendRequestForNetwork's
signature now makes this explicit.
This step prepares the addition of overloaded versions of
registerNetworkCallback and cie that takes custom Handlers.
Test: build, flashed, manually checked connectivity
Change-Id: I52e8a2cb5075e7aef7b35e30c9845cacba927d13
This static method returns a NetworkCapabilities instance with
transports and capabilities set according to the given legacy type.
Also:
- add NetworkRequest.Builder.setCapabilities(), to be able to use
the NetworkCapabilities instances returned from the above
- update UpstreamNetworkMonitor to make immediate use of this
Test: as follows
- build (bullhead)
- flashed
- booted
- runtest frameworks-net passes
- WiFi to DUN upstream tethering works
Bug: 32163131
Change-Id: Idfe1ddd2815c355cbf27cf29eb0e3de177de84e9
All the tests are now asynchronous. The only remaining uses of
Thread.sleep() are in the tests that check for NetworkRequest
timeouts.
Test: ConnectivityServiceTest passes 100 times in a row
Bug: 32561414
Change-Id: If420bd66c692a90d5031ee06a888a8cc3b4398a8
The calls in testAvoidBadWifi are no longer necessary now that
waitForIdle is reliable, and the calls in testPacketKeepalives
are replaced with a wait for the NetworkAgent to disconnect.
Test: ConnectivityServiceTest passes 100 times in a row.
Bug: 32561414
Change-Id: Icbb161ca6e343bd14764a1c9ccfdd14b6cd6803f
Rename the class and ensure that all code related to bad wifi
avoidance has names that reflect its purpose. This will allow us
to reuse most of the the code for other purposes in future CLs.
Test: runtest frameworks-net passes
Bug: 34630278
Change-Id: Ida190e5a79e7a12f4eec07e1a3e4d9d138e83bcd
The current implementation of IdleableHandlerThread is based on
the assumption that MessageQueue#isIdle will return true iff the
message loop has finished processing its messages.
Unfortunately, this is incorrect: isIdle returns true iff are no
more messages waiting in the queue; thus, it will also return
true while it is processing the last message before going idle.
Instead of using idle handlers, take the simpler approach of
posting a runnable to the message loop and waiting for it to
be processed.
Test: see other CLs in this patch series
Bug: 31479480
Change-Id: Iae75781f067b762c8653a488a5e4d5ee0c789e01
Reverts half of commit 224743c6b7
Reverts parts of commit 4908ef3e42
Lockdown VPN still needs this to start properly on devices without FBE,
such as phones upgraded from M. Without listening to USER_PRESENT the
first unlock of keystore may get missed, since USER_UNLOCKED is sent too
soon to be useful.
Bug: 34587043
Change-Id: If7a2f5a3173b0f6856244a5f7904bf00c9a5b5e7
This patch removes testNotWaitingForIdleCausesRaceConditions() from
ConnectivityServiceTest because it is in nature flaky and prevents using
ConnectivityServiceTest as a patch presubmit check. Estimated failure
rate is 1/15 on Nexus 6p.
This patch also simplifies how ConnectivityServiceTest waits for
handlers to become idle by removing IdleableHandlerThread and turning it
into a signle static method.
Test: $ runtest frameworks-net
Bug: 31479480
Change-Id: I2d78709cbb61d5d01cd59cff326469417f73f1ab
This patch adjusts timeouts in ConnectivityServiceTest to fix tests
known to fail spruriously.
Test: ran the following tests 500 times without observing any failure
- testBackgroundNetworks()
- testMultipleLingering()
- testPacketKeepalives()
- testSatisfiedNetworkRequestDoesNotTriggerOnUnavailable()
- testSatisfiedThenLostNetworkRequestDoesNotTriggerOnUnavailable()
Bug: 32561414
Change-Id: I184fe7189aac768a65b927cc42eefaf8b1b3f930
Symptom : Infinite reboot
Reproduce step :
1. Set the Always-on VPN in M OS
2. OS upgrade from M to N
Reproduce frequency : 100%
Reason of issue :
https://android.googlesource.com/platform/frameworks/base/+/4908ef3
As you know, in M OS, Always-on VPN information is stored in keystore with encryted.
However, in N OS, there is no encryption when it put in keystore.
So, You deleted keystore check(locked/unlock) logic on ConnectivityService.
By this reason, when device upgrade to N OS(set Always-on VPN), it goes infinite boot.
(Cannot read old always-on vpn information untill device unlock.)
Solution :
I founded exception handling when this case as follows:
If getting Credentials.LOCKDOWN_VPN information has null value(catch the exception),
updateLockdownVpn returns false value.
Signed-off-by: SangJin Cha <sj.cha@lge.com>
Change-Id: I6fd980152440bb5248aab45e2f8fda448d3f6c7b
When network setting reset is started while always-on VPN is enabled,
application crash occurs with IllegalStateException in prepareVpn().
Therefore, it's necessary to also disable always-on VPN in factoryReset
before prepareVpn() when turning off VPN.
Bug: 29030858
Change-Id: I08c5406717dd7937d197c2e934a56d037477aafc
- CallbackInfo becomes a static class so that it can be instantiated
and looked at more easily than before.
- CallbackInfo gains hashCode() because it has equals().
- effectively final field are qualified final
- fixes bug when polling the next callback: poll() on a LinkedBlockingQueue
returns null when timing out, which was not failing the test cleanly.
Test: ConnectivityServiceTest passes
Bug: 32561414
Change-Id: I64d382a35fe83f21bded98e6906c1ec70a8f1085
