This fixes the documentation for UploadService and the README file:
* stop mentioning Chrome OS as we forked the code.
* update the histogram declaration workflow. It is not tied to Chrome's
histograms.xml file anymore.
* update the architecture explanation of metricsd. We split
metrics_daemon into two daemons and are now using binder to log
metrics.
* convert README to markdown to make it prettier when viewed in gitiles.
Bug: 26314417
Change-Id: I1e492f1211c1784e65dd4d3e473bb9aacefc3b5d
/data/misc/metricsd only needs to be accessed by metricsd.
/data/misc/metrics_collector only needs to be accessed by
metrics_collector.
/data/misc/metrics needs to be writable by metrics_collector
and readable by metricsd. Make it 0750 metrics_collector:system
and keep running metricsd as group 'system' to allow reading.
Bug: 26337609
Change-Id: I169573cc547f154daa8b65c5f13bdc20964a3f2c
Callers will not guarantee that they can or will ratelimit, we need to
retain the ability to blacklist snet_event_log as a result.
This reverts commit 6aa21b225d.
Bug: 26178938
Change-Id: Ibf47d2e23a84c56f5f72d02312c698df7ff2b601
This allows you to flash to a slot other than the current one
without needing to enter the name of the slots.
Change-Id: I6bf8c29817be0a29b1abb005f1e72056cee126df
IMetricsCollectorService implementation moved from libmetricscollectorservice
to metrics_collector executable, which means it needs to be compiled into test
main as well.
Bug: None
Change-Id: I29b0a3ae6164d2256366d61c3a3639d22b943a69
No longer needed to trampoline between RTTI and non-RTTI compiles.
Move the service impl from libmetricscollector service to the
metrics_collector executable, now that RTTI is no longer used (and the impl
is not used by binder clients).
Other fixups, including:
* Move service registration and binder watcher to metrics_collector main.
* Compile libmetricscollectorservice with clang just because.
Bug: 26292405
Change-Id: Iae8115c13b203ff03d49bf7678d5eb3ede30e855
-metrics_collector and metricsd gather and upload device metrics.
-webservd is part of the Weave (Brillo remote setup and management) stack.
Bug: 26337609
Change-Id: Icf6ce75e64293adbe773f2ee65e80629f2eeb737
Bad comment advise in LogBuffer.cpp results in partners failing to
considering using ro.logd.size to set the platform buffer size
default.
NB: It is not good practice to increase the log buffer size to deal
with logspam, as increases will result in logd scale issues getting
closer to hitting the background cgroup cpu cap. Once we hit that
cap, logd spirals, pruning old entries slower than the incoming log
entries. logd.writer will take 100% cpu.
Change-Id: If4a7a74f300d078eeaed0ffd3eb3fd77d1f9fe90
Without -s argument, adb wait-for-device will make a call to
acquire_one_transport() with sinfo->serial.c_str() == "".
Waiting for acquire_one_transport() to be reworked to use std::string
rather than const char * for serial, work around this by passing NULL
when sinfo->serial is not initialized in host_service_to_socket().
Change-Id: Ifef53e0e82850364f5029ee36560376f2a4a5104
Signed-off-by: Leo Sartre <leox.sartre@intel.com>
In libraries relying on binder it is useful to be able to obtain an
existing copy of BinderWrapper or create one if it doesn't exist.
This allows to abstract the actual RPC (binder) from clients who have
no other dependencies on this RPC.
BUG: 23782171
Change-Id: Ie775d3d8ab83d75e38abc7e1385eb39a363555ef
There is no longer a reliance on RTTI in libbrillo, so disable RTTI
in the rest of Brillo codebase.
Bug: 26292405
Change-Id: I2751a2ac973a74741a82f5cd949fc0adaf2449a6
Dangerous bridge to cross to whitelist, who is special, who is not?
Rationalized as these events are used to catch exploits on platform.
As it stands no one should be allowed to block any messages in the
security context, not even for development purposes.
Bug: 26178938
Change-Id: Ibdc76bc0fe29ba05be168b623af1c9f41d7edbd2
Add __android_log_is_loggable() checking for all buffers except
LOG_ID_SECURITY. Return -EPERM if blocked. Since we are sniffing
the log tag, check validity and return -EINVAL.
NB: Try not to call __android_log_is_loggable() in native code within
a signal handler. Both here, and in the system properties, there
are locking paths that are not guaranteed to play well in that
environment. This has also been the case for the log writer path
even before this change. All attempts have been made to use trylock,
and to use a more expensive code path when contention occurs rather
than lead to deadlock.
Bug: 19544788
Bug: 26178938
Change-Id: I98738c662f6328189a6703251eb8721a05e956f9
Use 1972 as a right delineation. Otherwise use half way point
between the monotonic and realtime. Treat correction factor as
unsigned, ensure that any wrapping to a negative value is
dropped or set to EPOCH. Acknowledge that we can get a more
accurate time track by acquiring the time rather than relying on
healthd timestamp.
Bug: 26331432
Change-Id: I09075fca58676a30cf7d87baf2d4b0f53795abaa
- Add a new statistic that reports per pid and log_id for AID_SYSTEM
- Add a new pruning filter ~1000/! boolean
- Use this new statistic to prune on worst pid within AID_SYSTEM
Bug: 26029733
Bug: 21615139
Bug: 22855208
Change-Id: Iab5dd28f807dcf03d276372853883f3b6afa8294
* Added new kernel GID named "wakelock" (AID_WAKELOCK = 3010)
* Changed the group access for /sys/power/wake_lock and
/sys/power/wake_unlock from "system" to "wakelock"
* Added "wakelock" to the list of groups for the healthd process/service
Bug: 25864142
Change-Id: Ieabee9964cccec3107971a361a43aa9805164aa9
Some compilers erroneously see uninitialized use, even despite all
accesses being behind !not_locked. Confirmed initialization does not
affect expected performance.
Bug: 26178938
Bug: 26029733
Bug: 17760225
Change-Id: Ib36ed8dd2c4b196ca84ef79a9531625dcee77e15
Bertrand Simonnet