Now that single-document indexing requests are executed on the bulk
thread pool the index thread pool is no longer needed. This commit
removes this thread pool from Elasticsearch.
The test was using a parameter on GET /_cluster/health that older nodes
do not understand. Yet, we do no even need to make this call here, we
can use ensure green for the index.
Binary doc values are retrieved during the DocValueFetchSubPhase through an instance of ScriptDocValues.
Since 6.0 ScriptDocValues instances are not allowed to reuse the object that they return (https://github.com/elastic/elasticsearch/issues/26775) but BinaryScriptDocValues doesn't follow this restriction and reuses instances of BytesRefBuilder among different documents.
This results in `field` values assigned to the wrong document in the response.
This commit fixes this issue by recreating the BytesRef for each value that needs to be returned.
Fixes#29565
When calculating the current load on each ML node during the
node allocation process we should be ignoring failed jobs.
This is because failed jobs do not have a corresponding native
process, so do not consume memory or CPU resources.
relates elastic/x-pack-elasticsearch#4381
Original commit: elastic/x-pack-elasticsearch@1cb0ca973e
These tests test actually watcher 6.x features and are not needed in the
master branch anymore, as there is no way that watcher is running on the
master node only in Elaticsearch 7.x.
Original commit: elastic/x-pack-elasticsearch@ac5b933994
Rolling up indices that contain multiple schemas requires extra
considerations for the user, and in some cases, some limitations. This
commit tries to enumerate the issues so the user is aware.
Original commit: elastic/x-pack-elasticsearch@bf96eeab4e
When comparing doubles, fixed epsilons can fail because the absolute
difference in values may be quite large, even though the relative
difference is tiny (e.g. with two very large numbers).
Instead, we can scale epsilon by the absolute value of the expected
value. This means we are looking for a diff that is epsilon-percent
away from the value, rather than just epsilon.
This is basically checking the relative error using junit's assertEqual.
Closes#29456, unmutes the test
As part of adding support for new API to the high-level REST client,
we added support for the `flat_settings` parameter to some of our
request classes. We added documentation that such flag is only ever
read by the high-level REST client, but the truth is that it doesn't
do anything given that settings are always parsed back into a `Settings`
object, no matter whether they are returned in a flat format or not.
It was a mistake to add support for this flag in the context of the
high-level REST client, hence this commit removes it.
Instead of allowing the `_xpack/monitoring/_bulk` to remain on the same
thread, it should execute on a separate thread to avoid blocking the
http worker thread whenever the exporters get stuck waiting on the
monitoring cluster.
Original commit: elastic/x-pack-elasticsearch@25ce9a4df0
This refactors MapperService so that it wraps a single `DocumentMapper` rather
than a `Map<String, DocumentMapper>`. We will need follow-ups since I haven't
fixed most APIs that still expose collections of types of mappers, but this is
a start...
Currently this fails because the Eclipse configuration splits the main and test
folders into separate projects to avoid circular dependencies.
Relates #29336
Today the translog of an engine is exposed and can be accessed directly.
While this exposure offers much flexibility, it also causes these troubles:
- Inconsistent behavior between translog method and engine method.
For example, rolling a translog generation via an engine also trims
unreferenced files, but translog's method does not.
- An engine does not get notified when critical errors happen in translog
as the access is direct.
This change isolates translog of an engine and enforces all accesses to
translog via the engine.
Updates documentation describing the parameters for optionally signing the
SAML medatata document that the saml-metadata command creates.
Original commit: elastic/x-pack-elasticsearch@0e8146541a
The index thread pool is no longer needed as its primary use-case for
single-document indexing requests has been relieved now that
single-document indexing requests are converted to bulk indexing
requests (with a single document payload).
We want to remove the index thread pool as it is no longer needed since
single-document indexing requests are executed as bulk requests
now. Analyze requests are also executed on the index thread pool though
and they need a thread pool to execute on. The bulk thread does not seem
like the right thread pool, let us keep that thread pool conceptually
for bulk requests and free for bulk requests. None of the existing
thread pools make sense for analyze requests either. The generic thread
pool would be a terrible choice since it has an unbounded queue and that
is a bad idea for user-facing APIs. This commit introduces a small by
default (size=1, queue_size=16) thread pool for analyze requests.
This commit add the `include_type_name` option to the `index`, `update`,
`delete`, `get`, `bulk` and `search` APIs. When set to `false`, the response
will omit the `_type` in the response. This option doesn't work if the endpoint
contains a type. For instance, the following call would succeed:
```
GET index/_doc/1?include_type_name=false
```
But the following one would fail:
```
GET index/some_type/1?include_type_name=false
```
Relates #15613
Previously "certutil" would generate a zip file if there were multiple
certificates.
However, this means that if the user specified "-multiple" or "-in"
then the output format will vary based on whether they entered
multiple instance names (-multiple) or whether the input file
contained multiple instance records (-in).
It is better if the output format is predictable based on the command
line options, so this change forces zip output whenever any of the
following command line options are supplied:
-pem
-keep-ca-key
-multiple
-in
Original commit: elastic/x-pack-elasticsearch@344baa5f17
The handling of the "--pass" option has complex rules due to the
different requirements of different output formats, and the way
joptsimple handles specs with optional arguments.
A standard unix shell will parse
certutil --pass=""
as being identical to
certutil --pass=
and joptsimple treats that as being "no argument", i.e. the same as
certutil --pass
The only way to pass a blank password is
certutil --pass ""
(though this is shell dependent)
Original commit: elastic/x-pack-elasticsearch@bc09d0f298
OpenSAML requires that the context classloader be set to the
classloader that contains the opensaml jar files in order to
initialise itself. However, sometimesrthis seems to cause the
classloading for slf4j/log4j to take place while that context
classloader is set, which fails.
This change forces slf4j/log4j to be initialised prior to setting
the context classloader.
Original commit: elastic/x-pack-elasticsearch@85ed8eb488
This test is failing because of an addition of a call to GET
/_cluster/health with the parameter wait_for_no_initializing_shards set
to true. As older versions of Elasticsearch do not understand this
parameter, this request fails and the test fails. This commit marks this
test as awaiting a fix.
This commit fixes plugin warning confirmation to include native
controller confirmation when no security policy exists. The case was
already covered for meta plugins, but not for normal plugins. Tests are
also added for all cases.
With the move long ago to execute all single-document indexing requests
as bulk indexing request, the method
PipelineExecutionService#executeIndexRequest is unused and will never be
used in production code. This commit removes this method and cuts over
all tests to use PipelineExecutionService#executeBulkRequest.
This ignores data collection when the cluster is not ready, in addition to
the existing check that ignores when the cluster state's version is
unknown.
Original commit: elastic/x-pack-elasticsearch@54257d7e6f
Rollup was using a historical `rollup` doc type, when we should be using
`_doc` so that it is forward-compatible with the removal in 8.0
Original commit: elastic/x-pack-elasticsearch@4b3188e6c8
This is required so the Logstash Centralized Configuration Management UI in Kibana may make the GET / request to Elasticsearch and retrieve the cluster UUID. It then uses this cluster UUID to make a call to a Kibana Monitoring API to retrieve a list of pipelines from Monitoring. In order for the Kibana Monitoring API request to succeed, the logged-in user needs to have the built-in monitoring_user role anyway, so we give this role the cluster:monitor/main privilege.
Original commit: elastic/x-pack-elasticsearch@bf6ad5c1df
This changes the behaviour of AbstractWatcherIntegrationTestCase and its
startWatcher/stopWatcher methods. Instead of checking for the target
state and just starting or stopping if it does not match, the methods
now wait for certain states to be reached before starting or stopping.
This will fix test failures where a failure is started instead of
stopped or vice versa.
Original commit: elastic/x-pack-elasticsearch@f0b0954803
Fix bug in checking whether the expression properties has been
modified or not.
Change test to move from SIN to SQRT (as the former seem to return
different values (extra digit) across JDK versions - 8 vs 10)
Fixelastic/x-pack-elasticsearch#4335
Original commit: elastic/x-pack-elasticsearch@8f672c455d
Jason Tedor
Jim Ferenczi
David Roberts
Julie Tibshirani
Lisa Cawley
Alexander Reelsen
Zachary Tong
Jack Conradson
Dimitris Athanasiou
Luca Cavanna
Adrien Grand
Chris Earle
Igor Motov
Nhat Nguyen
Ioannis Kakavas
Tim Vernum
olcbean
Ryan Ernst
Zachary Tong
Shaunak Kashyap
Costin Leau