Our Docker build uses a multi-stage Docker build. This requires Docker
version 17.05 or greater. Without an explicit check here, the build
fails in a mysterious way such as "invalid reference format" that is
hard to track down (Google searches for "Docker invalid reference
format" do not turn up anything useful). This commit refactors our
existing Docker checks, and adds a new one for the minimum Docker
version.
This commit changes the Docker assemble tasks so that they attempt to
build Docker if the Docker binaries exist, of if build.docker is set to
true. If the Docker binaries do not exist, or if build.docker is set to
false, then no attempt is made to build the Docker images.
Moves all remaining (rolling-upgrade and mixed-version) REST tests to use Zen2. To avoid adding
extra configuration, it relies on Zen2 being set as the default discovery type. This required a few
smaller changes in other tests. I've removed AzureMinimumMasterNodesTests which tests Zen1
functionality and dates from a time where host providers were not configurable and each cloud
plugin had its own discovery.type, subclassing the ZenDiscovery class. I've also adapted a few tests
which were unnecessarily adding addTestZenDiscovery = false for the same legacy reasons. Finally,
this also moves the unconfigured-node-name REST test to Zen2, testing the auto-bootstrapping
functionality in development mode when no discovery configuration is provided.
Closes#34820
With this change we allow for no tests being ran in randomized testing
task, and forbid empty testing tasks from the testing conventions task.
We will no longer have to disable the task if all tests are muted.
* Improve logged exec output readability
- Split error and out streams and log them separately
- Log everything in a single call to prevent interference from
other log messages
Includes:
LUCENE-8594: DV update are broken for updates on new field
LUCENE-8590: Optimize DocValues update datastructures
LUCENE-8593: Specialize single value numeric DV updates
Relates #36286
This commit moves back to use explicit dependsOn for test tasks on
check. Not all tasks extending RandomizedTestingTask should be run by
check directly.
This commit fixes an oops when pushing a change to add the building of
the Docker images. A change that was made for testing was accidentally
left behind. This commit addresses that.
This commit introduces the building of the Docker images as bonafide
packaging formats alongside our existing archive and packaging
distributions. This build is migrated from a dedicated repository, and
converted to Gradle in the process.
Currently we use Math.random() in a few places in the tests which makes these
tests not reproducable with the random seed mechanism that comes with
ESTestCase. The change removes those instances.
Closes#35435
- make it easier to add additional testing tasks with the proper configuration and add some where they were missing.
- mute or fix failing tests
- add a check as part of testing conventions to find classes not included in any testing task.
The logic in the dockerComposeSupported method currently returns false
even when docker and docker compose are available on the build machine.
This change updates the check to see if docker compose is available in
one of the two paths and allows the `tests.fixture.enabled` property to
disable the tests even if docker compose is available.
This commit upgrades netty. This will close#35360. Netty started
throwing an IllegalArgumentException if a CompositeByteBuf is
created with < 2 components. Netty4Utils was updated to reflect this
change.
This commit removes padding from the java version in the output of
compiler/runtime/gradle java versions. This value is only ever 1 or 2
digits, and the later information (hotspot/jvm vendor info) is separated
by the java home path from the other versions, so there isn't a visual
reason for needing exact vertical alignment.
Some times the test fixtures plugin did not correctly disable tasks
from the build plugin as it should.
The plugin manager and tasks both use domain name collections so
the previus conde should have worked.
I did not have trime to track it down, but suspect there's some race
condition in Gradle causing this. The plugin manager is still incubating.
Since the tasks are on the cp even if the plugin is not applyed, we
don't really need to involve the plugin at all.
Closes#36041
Our `REPRODUCE WITH` line wasn't working for tests with `(` or `)` in
the method name. This isn't super common, but happens sometimes for our
yml based rest tests. This is because randomized runner's globbing
mechanism dind't escape `(` or `)`. I filed this upstream at
https://github.com/randomizedtesting/randomizedtesting/issues/271
and Dawid kindly fixed the issue. This upgrades to pick up the fix.
Closes#35692
In #35259 we switched the default number of VMs to fork for unit tests to
the number of physical CPU cores. But because we could only get an accurate
count on machines with a normal `/proc` filesystem, macOS machine did not
pick up the new default. Given that macOS is a huge portion of developer
machines, we'd like to get the right default there. This does that.
It also moves the default-finding process from happening once per testing
task to happening once at startup. This seems like a good choice in general,
but a very good choice for macOS because we have to run a command to list
the count.
This inserts newlines in order to reduce line lengths in the
o.e.action.admin.cluster package to 140 characters or less. This
also remves the checkstyle suppressions for affected files.
Relates #34884, #34923
- The current version was hard coded in the test, causing it to fail as
we removed the qualifier
- also applied the base plugin to the root project to get the `clean`
task to work as expected. This was preventing the failure from
reproducing locally.
* Manage dependencies for test clusters
Create a configuration and add the distribution to it automatically.
A task is created and added as a dependency to any task that uses a test
cluster.
The task extracts all the zip archives ( only zip support for now )
in the configuration.
We do this only once because most tests mostly use the same distribution
and thus we can avoid extracting it multiple times.
With this we will be able to start the node from the same files which
will most of the time live in OS caches or COW if the
configuration requires it.
* Swithc to jcenter
Jcenter suposedly operates on a CDN.
It should be faster and more reliable.
It's the default in Andorid Studio currently
( it switched from mavenCentral ).
This change is safe because jcenter is a superset of mavenCentral.
* update comment
* DISCOVERY: 0s Initial State Timeout in Tests
* Don't wait for initial state even with a single node, otherwise the loop writing the discovery file causes that single node to wait
for its own transport.ports file for 30s.
* Closes#35456
* DISCOVERY: Fix RollingUpgradeTests
* Don't manually manage min master nodes if not necessary
* Remove some dead code
* Allow for manually supplying list of seed nodes
* Closes#35178
With this change, `Version` no longer carries information about the qualifier,
we still need a way to show the "display version" that does have both
qualifier and snapshot. This is now stored by the build and red from `META-INF`.
* Add missing up-to-date configuration
The source properties file and the dynamic elasticsearch version
(set based on properties ) were missing from task outputs leading
to the task being incorrectly considered up to date.
Closes#35204
* Introduce property to set version qualifier
- VersionProperties.elasticsearch is now a string which can have qualifier
and snapshot too
- The Version class in the build no longer cares about snapshot and
qualifier.
This further applies the pattern set in #34125 to reduce copy-and-paste
in the multi-document CRUD portion of the High Level REST Client docs.
It also adds line wraps to snippets that are too wide to fit into the box
when rendered in the docs, following up on the work started in #34163.
* DISCOVERY: Use Realistic Num. of Min Master Nodes
* With all 3 nodes starting in parallel 2 nodes can win a master election and
start waiting for nodes to join when started in parallel. We wait for 30s on the
rest tests for the cluster health endpoint to show 3 nodes which breaks if
one of the master nodes itself waits for 30s for joining nodes, waiting for 5 seconds
fixes the issue and allows us to run with `minimum master nodes < node count`
* relates #33675
* Fix linelength suppressions in index.fielddata
* Some lines that were too long were dead code => Removed them and all code that became dead because of it
* Relates #34884
- we already require Java 11 to build, yet we target the minimum
supported version in build-tools ( currently 8 )
- this is because we have some checks that are executed in a new JVM
which could be running the minimum version.
- For everything else it would be nice to be able to use new features,
like the new process API.
With this change, we selectively compile the few classes that need an
older target version and move everything over to Java 10.
Unfortunately the current Gradle version does not support 11 as a target
version yet.
With this change, we apply the common test config automatically to all
newly created tasks instead of opting in specifically.
For plugin authors using the plugin externally this means that the
configuration will be applied to their RandomizedTestingTasks as well.
The purpose of the task is to simplify setup and make it easier to
change projects that use the `test` task but actually run integration
tests to use a task called `integTest` for clarity, but also because
we may want to configure and run them differently.
E.x. using different levels of concurrency.
- Restrict visibility of Aggregators and Factories
- Move PipelineAggregatorBuilders up a level so it is consistent with
AggregatorBuilders
- Checkstyle line length fixes for a few classes
- Minor odds/ends (swapping to method references, formatting, etc)
This commit adds the ability for docs tests to add a tear down
snippet. This snippet will be converted to a tear down section of the
generated REST tests.
In the docs tests, we have pre-defined setups in the build.gradle file,
and we can also define test setup sections within the doc page
itself. Alas, these two are incompatible in that if you try to use a
pre-defined setup alongside a test setup section, the pre-defined setup
will be silently ignored. This commit enables pre-defined setup sections
to be used together with test setup sections. The ordering here is that
pre-defined setup sections will be executed first, followed by the test
setup section.
The `term` and `phrase` suggesters have different options to filter candidates
based on their frequencies. The `popular` mode for instance filters candidate
terms that occur in less docs than the original term. However when we compute this threshold
we use the total term frequency of a term instead of the document frequency. This is not inline
with the actual filtering which is always based on the document frequency. This change fixes
this discrepancy and clarifies the meaning of the different frequencies in use in the suggesters.
It also ensures that the threshold doesn't overflow the maximum allowed value (Integer.MAX_VALUE).
Closes#34282
Applies our line length guidance for all classes in the server in `lucene`
directories *except* `XMoreLikeThis`. The only long line in
`XMoreLikeThis` says "remove this when we upgrade to Lucene 5. Given
that we're on Lucene 8, this is a little terrifying and deserves another
look.
This drops checkstyle suppressions that refer to files that don't exist
since those suppressions don't do anything other than make us feel bad.
It also updates some suppressions to more closely match the path to the
file that they suppress. These suppressions are still needed but didn't
pass the "the file exists" test because they weren't precise. It is just
easier on future-me if they are precise.
We generate tests from our documentation, including assertions about the
responses returned by a particular API. But sometimes we *can't* assert
that the response is correct because of some defficiency in our tooling.
Previously we marked the response `// NOTCONSOLE` to skip it, but this
is kind of odd because `// NOTCONSOLE` is really to mark snippets that
are json but aren't requests or responses. This introduces a new
construct to skip response assertions:
```
// TESTRESPONSE[skip:reason we skipped this]
```
Now that JDK 11 is GA, we would switch our 6.x and master branches to
the JDK 11 compiler. This commit makes this change, as well as removes
JDK 10 from the CI configuration.
This slightly reworks the expert script plugin example so it fits on the
page when the docs are rendered. The box in which it is rendered is not
very wide so it took a bit of twisting to make it readable.
Mainly this fixes a warning by replacing the unchecked `new ActionListener`
with the checked `new ActionListener<Response>`, and it also fixes the line
length violations in this class.
We use wrap code in `// tag` and `//end` to include it in our docs. Our
current docs style wraps code snippets in a box that is only wide enough
for 76 characters and adds a horizontal scroll bar for wider snippets
which makes the snippet much harder to read. This adds a checkstyle check
that looks for java code that is included in the docs and is wider than
that 76 characters so all snippets fit into the box. It solves many of
the failures that this catches but suppresses many more. I will clean
those up in a follow up change.
This change adds the OneStatementPerLineCheck to our checkstyle precommit
checks. This rule restricts the number of statements per line to one. The
resoning behind this is that it is very difficult to read multiple statements on
one line. People seem to mostly use it in short lambdas and switch statements in
our code base, but just going through the changes already uncovered some actual
problems in randomization in test code, so I think its worth it.
Wraps all lines in our test framework at 140 characters because that is
our standard line length and removes all of the checkstyle suppressions
for the test framework.
Drops most of `ModuleTestCase` because it isn't used and we're moving
away from using guice in the way that it wants to test anyway. Also
switches a few classes that extend it but don't use it to extend
`ESTestCase` instead.
Gradle can sometimes emit mixed log lines due to how we spawn things in
separate processes. This commit changes the jarhell integ test to only
look for the exception and message, instead of including the outer part
about the exception in thread main.
closes#33774
Make sure that all java files have a package declaration and that all of
the package declarations line up with the directory structure. This
would have caught the bug that I caused in
190ea9a6de and fixed in
b6d68bd805.
This commit changes the sanity check which ensures the test task was
properly replaced with randomized testing to have a per project check,
isntead of a global one. The previous global check assumed all test
tasks within the root project and below should be randomized testing,
but that is not the case for a multi project in which only one project
is an elasticsearch plugin. While the new check is not able to emit all
of the failed replacements in one error message, the efficacy of the
check remains.
This commit switches the joda time backcompat in scripting to use
augmentation over ZonedDateTime. The augmentation methods provide
compatibility with the missing methods between joda's DateTime and
java's ZonedDateTime. Due to getDayOfWeek returning an enum in the java
API, ZonedDateTime is wrapped so that the method can return int like the
joda time does. The java time api version is renamed to
getDayOfWeekEnum, which will be kept through 7.x for compatibility while
users switch back to getDayOfWeek once joda compatibility is removed.
* LeafCollector.setScorer() now takes a Scorable
* Scorers may not have null Weights
* IndexWriter.getFlushingBytes() reports how much memory is being used by IW threads writing to disk
Today the FilterRoutingTests take the belt-and-braces approach of excluding
some node attribute values and including some others. This means that we don't
really test that both inclusion and exclusion work correctly: as long as one of
them works as expected then the test will pass. This change improves these
tests by only using one approach at once, demonstrating that both do indeed
work, and adds tests for various other scenarios too.
Change the logging infrastructure to handle when the node name isn't
available in `elasticsearch.yml`. In that case the node name is not
available until long after logging is configured. The biggest change is
that the node name logging no longer fixed at pattern build time.
Instead it is read from a `SetOnce` on every print. If it is unset it is
printed as `unknown` so we have something that fits in the pattern.
On normal startup we don't log anything until the node name is available
so we never see the `unknown`s.
This change collapses all metrics aggregations classes into a single package `org.elasticsearch.aggregations.metrics`.
It also restricts the visibility of some classes (aggregators and factories) that should not be used outside of the package.
Relates #22868
The main benefit of the upgrade for users is the search optimization for top scored documents when the total hit count is not needed. However this optimization is not activated in this change, there is another issue opened to discuss how it should be integrated smoothly.
Some comments about the change:
* Tests that can produce negative scores have been adapted but we need to forbid them completely: #33309Closes#32899
Solves all of the xpack line length suppressions and then merges the
remainder of the xpack checkstyle_suppressions.xml file into the core
checkstyle_suppressions.xml file. At this point that just means the
antlr generated files for sql.
It also adds an exclusion to the line length tests for javadocs that
are just a URL. We have one such javadoc and breaking up the line would
make the link difficult to use.
This commit introduces the formal notion of a private setting. This
enables us to register some settings that we had previously not
registered as fully-fledged settings to avoid them being exposed via
APIs such as the create index API. For example, we had hacks in the
codebase to allow index.version.created to be passed around inside of
settings objects, but was not registered as a setting so that if a user
tried to use the setting on any API then they would get an
exception. This prevented users from setting index.version.created on
index creation, or updating it via the index settings API. By
introducing private settings, we can continue to reject these attempts,
yet now we can represent these settings as actual settings. In this
change, we register index.version.created as an actual setting. We do
not cutover all settings that we had been treating as private in this
pull request, it is already quite large due to moving some tests around
to account for the fact that some tests need to be able to set the
index.version.created. This can be done in a follow-up change.
Exclude classes meant for newer versions than what we are auditing against, those classes won't be found. There's no reason to exclude JDK classes from newer versions, with this PR, we will not extract them in the first place.
The new implementation is functional equivalent with the old, ant based one.
It parses task standard error to get the missing classes and violations in the same way.
I considered re-using ForbiddenApisCliTask but Gradle makes it hard to build inheritance with tasks that have task actions , since the order of the task actions can't be controlled.
This inheritance isn't dully desired either as the third party audit task is much more opinionated and we don't want to expose some of the configuration.
We could probably extract a common base class without any task actions, but probably more trouble than it's worth.
Closes#31715
This commit removes the setting of the fork options maximum memory size
in our build plugin and instead adds the value in the gradle.properties
file to be alongside the value set in jvmArgs.
This change is necessary when using parallel compilation as 512m is not
sufficient for parallel compilation on some machines.
This reworks how we configure the `shadow` plugin in the build. The major
change is that we no longer bundle dependencies in the `compile` configuration,
instead we bundle dependencies in the new `bundle` configuration. This feels
more right because it is a little more "opt in" rather than "opt out" and the
name of the `bundle` configuration is a little more obvious.
As an neat side effect of this, the `runtimeElements` configuration used when
one project depends on another now contains exactly the dependencies needed
to run the project so you no longer need to reference projects that use the
shadow plugin like this:
```
testCompile project(path: ':client:rest-high-level', configuration: 'shadow')
```
You can instead use the much more normal:
```
testCompile "org.elasticsearch.client:elasticsearch-rest-high-level-client:${version}"
```
Nhat Nguyen
Jason Tedor
James Baiera
Christian Schneider
Yannick Welsch
Alpar Torok
David Roberts
Armin Braun
Ryan Ernst
Adrien Grand
Christoph Büscher
Alan Woodward
Jay Modi
Tim Brooks
Nik Everett
Hendrik Muhs
Marios Trivyzas
Jim Ferenczi
Gordon Brown
Martijn van Groningen
Lee Hinman
Tom Callahan
Lukáš Vlček
Nick Knize
Vladimir Dolzhenko
Mark Tozzi
vboulaye
Igor Motov
Jake Landis
Jack Conradson
Zachary Tong
Kazuhiro Sera
David Turner
Tanguy Leroux
Jonathan Little