| Age | Commit message (Collapse) | Author |
|---|
|
oracle10 is failing during setup with:
The command "bash install-jdk.sh -F 10 -L BCL --target $JAVA_HOME --workspace ${TRAVIS_HOME}/.cache/install-jdk" failed and exited with 8 during .
We should be adding support for Java 11 and dropping Java 9-10 anyway,
since Java 9-10 are unsupported.
|
|
Throw away Travis-CI's custom Maven settings, because they are causing
massive CI failures when Maven slows to a crawl/hangs because of
failures contaicting repository.apache.org.
Travis-CI's settings includes repo.maven.apache.org, oss.sonatype
(releases and snapshots), and repository.apache.org (releases and
snapshots). Now we will just be using Maven's default, which may just
be repo.maven.apache.org.
|
|
Wildcard apparently doesn't work... silently. Move versions into their
own directory so we can use a hard-coded string in the travis
configuration.
|
|
Previously it was just erroring:
[[: command not found
This has been broken since #3638
|
|
For jdk10: disable errorprone and fix javadoc warnings
|
|
This fixes a bug that always hard codes jdk8.
Fixing this revealed an issue for jdk10, so let's remove jdk10
for now and revisit it in a separate PR.
|
|
We always want to use a consistent version of protobuf; avoid the need
for the caller (which may be a person running the script) to specify the
version.
|
|
|
|
ARCH can be '32' or '64'. If it is not set then default to '64'.
make_dependencies.sh should do the symlinking
|
|
|
|
protoc-3.5.0-linux-x86_64 introduced GLIBC_2.14 dependency and broke
gRPC release process (https://github.com/google/protobuf/issues/4138).
3.5.1-1 is the proper re-build.
|
|
|
|
With #3608 Kokoro is now able to handle OS X. I'm not removing the
OS X-specific parts of the .travis.yml in case we need to revert back to
using Travis.
Fixes #3466
|
|
Thrift support is moved to https://github.com/grpc-ecosystem/grift
|
|
|
|
|
|
|
|
ErrorProne provides static analysis for common issues, including
misused variables GuardedBy locks.
This increases build time by 60% for parallel builds and 30% for
non-parallel, so I've provided a way to disable the check. It is on by
default though and will be run in our CI environments.
|
|
|
|
Commit 80f73a4a was in response to OS X receiving caching support.
Caches began being saved shortly after and they've been working since.
The official announcment was
https://blog.travis-ci.com/2016-05-03-caches-are-coming-to-everyone
|
|
The xcode7 image is being retired on Nov 28th[1]. The default has been
7+ for a while now[2]. I assume 7+ all have JDK 8, so using the default
seems safe.
1. https://blog.travis-ci.com/2016-11-17-retiring-some-osx-images/
2. https://blog.travis-ci.com/2016-09-15-new-default-osx-image-coming/
|
|
Also removed warnings about protoc version matching runtime, since this
is no longer supposed to be a problem (starting with 3.0.0-beta-4) and
all our tests ran fine when using protoc 3.0.2 with protobuf runtime
3.1.0.
Fixes #2316
|
|
protoc no longer builds in 3.0.0 because auto-download of the gmock zip
now fails. 3.0.2 has a fix to autogen:
https://github.com/google/protobuf/commit/bba446bbf2ac7b0b9923d4eb07d5acd0665a8cf0
All that was strictly necessary was to update .travis.yml and
buildscripts/, but it helps our sanity to keep the rest of the protobuf
versions in sync. Lite is left on its existing version, because it did
not see a bump of neither the java library nor the protoc plugin.
|
|
Fixes #2086
|
|
The examples are no longer part of the normal build, although they are
built with Travis. The examples now include their own copy of the gradle
wrapper to ease usage from IDEs which can now properly detect the
correct version of gradle to use.
The build files were generated using "gradle init" and "mvn
archetype:generate" and then modified following our README.
Fixes #1414
|
|
|
|
This allows us to play with zero-copy and proto3 support for lite.
Unfortunately, it introduced some warnings, so deprecated warnings are
now ignored for benchmarks and interop-testing.
|
|
$HOME can be different between different platforms/configurations, so
using /tmp means the path is consistent and be shared in the caches.
|
|
This is a remanent from when we were building Netty from master. Gradle
uses ~/.gradle/caches/modules-2/files-2.1/ for its Maven Central cache.
|
|
It doesn't appear that OS X is saving its cache yet, so this may not be
needed in the future, but for now we need OS X to avoid attempting to
use the Linux binaries.
|
|
When running :grcp-interop-testing:test, Travis has been hanging on OS X
or flaking on Linux with:
> Process 'Gradle Test Executor 4' finished with non-zero exit value 137
Exit code 137 indicates SIGKILL (128 + 9) and is most likely caused by
the JVM being killed by the kernel's OOM killer.
The limit in .travis.yml is 2x what was necessary to do a parallel
build. The main test memory limit in build.gradle is well above 16m
which is necessary for the tests. Interop-testing is well above 64m
which is necessary for interop-testing, but we use 1.5g to help prevent
timeouts on Travis.
Protobuf and protobuf-nano each have one tests that decodes a proto >64M
in size, which prevents them from running with less than 512m and 768m,
respectively.
|
|
Codecov.io provides patch-based code coverage, so you can easily know
how many of the added lines are covered. It also has a more useful UI.
Unfortunately, the percentage it reports does not include partially-
covered lines--those with uncovered conditions. Thus the reported
percentage is about 6% lower than the coverage we've been looking at
previously. Because of this alone, I don't expect to remove coveralls
support soon.
Use the bash script instead of python module since pip isn't available
by default on Travis OS X.
jacocoTestReport uses mustRunAfter (contrary to the docs; see
https://issues.gradle.org/browse/GRADLE-2960) to make sure it runs after
all tests, only if testing is taking place. We would like
:grpc-all:jacocoTestReport to behave the same way. Without it, we would
need two separate invocations of gradle (adding ~1m to Travis run) in
order to prevent getting random results depending on what tests just so
happened to have been run.
|
|
|
|
Update option name that disables protobuf tests.
|
|
The issue is closed and our latest runs on Travis are only needing to
run brew update once.
|
|
|
|
Since some of the tests would need only minor modification to work for
normal proto, normal proto got a few more tests as well.
|
|
|
|
|
|
This reverts commit d6dc790f05c90ffba2c7a4356121ae4d586e467e.
|
|
|
|
|
|
|
|
It's only working today because alpha2 is still in the cache.
|
|
We are using branches for other flows of development, so we really want
it running everywhere.
|
|
This doesn't catch 100% of such problems, as it runs before tests have
been run. However, almost all of our protos are used in 'main' so this
limitation shouldn't be a big deal.
If it does become a problem, we change it from 'before_script' to
'script', but then we'll also need to include gradle build instructions.
|
|
|
|
Gradle 2.3 is using a -bin.zip name instead of a -all.zip name like 2.2.
|
|
Artifacts that Gradle caches from Maven Central go in
~/.gradle/caches/modules-2, not ~/.m2. We only need .m2 for the
locally-installed Netty.
|
|
We were seeing errors on Travis like:
> Process 'Gradle Test Executor 2' finished with non-zero exit value 137
That doesn't make much sense, other than maybe the OOM killer killing
our processes. Turning off parallel execution seemed to fix the problem,
so we'll just assume memory was the actual problem and doing fewer
things in parallel reduces our maximum memory usage.
Travis documentation seems to agree with that being a likely cause:
http://docs.travis-ci.com/user/common-build-problems/#My-build-script-is-killed-without-any-error
|
