Changes the temperature atom to use an int instead of a float because
metrics and anomaly detection assume the data will be an int or a long.
Bug: b/74011562
Test: statsd temperature CTS test
Change-Id: I6e5845a40c28aefd30c2be9709c6f9de1783cf02
This atom logs time spent on cpu per frequency per uid.
On marlin, there are 27 frequency steps per uid (should be 50+ if
flatten to both cores)
On walleye, there are 52 frequency steps per uid.
So it easily goes to 6k data.
Soft limit set to 6k now.
Hard limit set to 10k to accommodate future processors
Memory impact: on marlin, start memory no config is 2,346K
after using this atom in one guage metric, it is 3,067K
Bug: 72505991
Test: cts test
Change-Id: I067a32e54e4a457fdf9a25911aa16030e893ef4d
Previously tried an optimization that results in corrupted proto
output. This changes to a safer approach of storing the snapshot data
in memory and only converting to proto output when the
ProtoOutputStream is provided.
Also fixes a security issue when trying to invoke triggerUidSnapshot
since we forgot to use SCS' permissions.
Test: Added a unit-test to verify output of StatsLogProcessor.
Bug: 76231867
Change-Id: Id410ce3505fda9d71caa71942ef3068b55872c66
We can increase the buffer of metrics we store in statsd memory, but
we still request the clients to call getData when the metrics memory
exceeds 128 KB (previously was 90% of 128 KB).
Bug: 76171061
Test: Test that unit-tests still pass.
Change-Id: I901545b364ed313af8c033ce9b40d3cfadb93213
Sanitizer is picking on the protobuf generated code. Need to figure out why
Disable them to unblock development first.
Bug: 76117312
Test: manual
Change-Id: I3f3a39dfe933f4c1b1cf4063d621392b66f0adea
Statsd clients may want to set an empty config temporarily, so it's
more convenient to allow them to set an empty config instead of
having to use the removeConfig and then having to remember to call
StatsManager#setDataFetchOperation.
Test: Added unit-tests and check they pass on marlin-eng.
Bug: 74997752
Change-Id: I2e762e5ec01e5a2c9a3469fb330b53fefbd734d6
ProtoOutputStream
- Specify the length of message to avoid libprotoutil from thinking that
we are trying to write bool
- We only attach the previous dump file to the upload file where config
key matches
- Store ConfigMetricsReport (instead of ConfigMetricsReportList) onto
disk
- Stop use stack after scope in StorageManager
- Migrate UidMap to use ProtoOutputStream and renaming variables to
prevent confusion
Bug: 74021554
Bug: 75968524
Test: manual test, statsd_test, CTS tests
Change-Id: Iedf52633d7f5b985f5a934a3fb5a0c3c3b2e7fd1
Previously, we wrote a log entry regardless of permission checks, so
the logging could be misleading. Now we only send the log to statsd
after verifying that this setting mutation is valid.
Test: Flashed onto marlin-eng and verified stats-log as expected.
Bug: 73493944
Change-Id: I2a8b052aa8c380ffc5d15caec089fffcdc5823f4
If the bucket size is lower than 1 min, the interval should be at least
1 min.
Bug: 74533948
Test: cts
Change-Id: Icb72579367cd12d3f475c7e2935b98c0ebc66419
Atom definitions for MobileConnectionStateChanged and
MobileRadioTechnologyChanged
Also cleans up batterystats.
Bug: b/72320589
Test: verified logs appear in adb logcat -b stats
Change-Id: I9feb258cf6dd4a8c8bf1cffc9566b5d0a851a9fa
The explicit resetStorage function in the subclass isn't needed.
It does exactly what its superclass does.
The previous check for !mAlarms.empty() was actually not needed, since
there can be legit cases where !mAlarms.empty() (namely, when the
mMostRecentBucket is moved forward initially - we may call reset, but we
still want the old alarms to remain).
Test: still pass
Change-Id: I3ffb6e97c02b75170c246f57d55ccf59cd368b9c
Fixes: 74607818
[memory]
statsd binary size from 664k -> 600k
memory usage 1978k -> 1813k (with no configs)
+ Avoid initialize any static map in statslog.h to avoid many copies of the map in each include.
- Do it in cpp so that it is initialized only in places that use them
[Uid annotation]
+ Uid annotation is needed for extracting uid from dimension for UidCpuPuller.
+ After the change, stand-alone uids don't need to be in field 1 anymore.
+ Also added exclusive bit annotation in AppDied
+ Currently only allow one uid field in an Atom. This is to keep things simple until
we find an exception.
Test: statsd_test
Bug: 73958484
Bug: 72129300
Change-Id: I8a916d5c00d5930e24ae7e0825a57dea19c0e744
Every time stopAlarm() is called, it should also,
right then, check to see if the alarm should actually
have already fired (but didn't due to AlarmManager lag).
Right now, the client needs to do this check separately,
but they always go together. Indeed, MaxDurationTracker
forgot to do the check, which is a bug. It would make
much more sense if the stopAlarm takes care of it for
them, to prevent such mistakes.
Bug: 75273733
Test: make statsd_test && adb sync data && adb shell data/nativetest64/statsd_test/statsd_test
Test: cts-tradefed run cts-dev -m CtsStatsdHostTestCases -t android.cts.statsd.alert
Change-Id: I689df13690df822090ac34b1171e948be1ad0d9f
Various fixes and improvements to statsd's anomaly detection.
Bug: 74607818
Test: make statsd_test && adb sync data && adb shell data/nativetest64/statsd_test/statsd_test
Change-Id: Ia67a8eb6da0ea9293f698949e1565f7f024a7cb9
We should fail right away if someone forgets to set this field
instead of setting default values since this may lead to hard to
explain bugs in the future (eg, why isn't systemui logging an atom
to a config).
Bug: 74608359
Test: Verified unit-tests still pass on marlin-eng.
Change-Id: Ibd8a6ccbc5cc8d2dfb8a1577c64bf9b49822a2c3
We don't need to parse the proto of uid map, so we use the
ProtoOutputStreame class to generate the binary form of the proto
output that's needed for parsing the uid map data.
Test: Verified unit-tests still pass.
Bug: 74010813
Change-Id: Ia2f7572f3b78bb6f7b60e8b14cf5d65428469ab6
