only change recovery throttling to slow down recoveries. The recovery file chunk size updates are not picked up by ongoing recoveries. That cause the recovery to take too long even after the default settings are restored.
Also - change document creation to reuse field names in order to speed up the test.
We don't rely upon GC to cleanup mappedbytebuffers, we unmap them
explicitly on close in lucene. But the JDK has crazy loops with
explicit GCs in exceptional cases to try to force unmapping.
In general we don't want any of our code or library code calling
this method: so its banned in forbidden-apis as well.
We clone RateLimitedIndexOutput from lucene just to collect pausing
statistics we can do this in a more straight forward way in a delegating
RateLimiter.
Closes#6625
Thread rejection should return too many requests status code, and not 503, which is used to also show that the cluster is not available
relates to #6627, but only for rejections for now
closes#6629
We want to make sure recycling will not fail for any reason while trying to send a response back that is caused by a failure, for example, if we have circuit breaker on it (at one point), sending an error back will not be affected by it.
closes#6631
the test failed but couldn't repro (yet), at the very least, make sure we have the exception message as the reason, can help to track down the failure itself when it happens again
If the match query with cutoff_frequency encounters stacked tokens,
like synonyms in the same position, it returns a boolean query instead
of a common terms query. However, if the original operator was set
to "and", it was ignoring that and resetting the operator to "or".
In fact, if operator is "and" then there is little benefit in using
a common terms query as a must query is already
executed efficiently.
Waiting for ongoing recoveries was not good enough as it can run before the master finishing processing the started events of primary shards, causing the recovery response to be erroneously empty
Made AckedClusterStateUpdateTask an abstract class instead of an interface, which contains the common methods.
Also introduced the AckedRequest interface to mark both AcknowledgedRequest & ClusterStateUpdateRequest so that the different ways of updating the cluster state (with or without a MetaData*Service) can share the same code.
Removed ClusterStateUpdateListener as we can just use its base class ActionListener instead.
Closes#6559
Sandboxes the groovy scripting language with multiple configurable
whitelists:
`script.groovy.sandbox.receiver_whitelist`: comma-separated list of string
classes for objects that may have methods invoked.
`script.groovy.sandbox.package_whitelist`: comma-separated list of
packages under which new objects may be constructed.
`script.groovy.sandbox.class_whitelist` comma-separated list of classes
that are allowed to be constructed.
As well as a method blacklist:
`script.groovy.sandbox.method_blacklist`: comma-separated list of
methods that are never allowed to be invoked, regardless of target
object.
The sandbox can be entirely disabled by setting:
`script.groovy.sandbox.enabled: false`
Introduced the use of the FilterClient in all of the REST actions, which delegates all of the operations to the internal Client, but makes sure that the headers are properly copied if needed from REST requests to TransportRequest(s) when it comes to executing them.
Added new abstract handleRequest method to BaseRestHandler with additional Client argument and made private the client instance member (was protected before) to force the use of the client received as argument.
The list of headers to be copied over is by default empty but can be extended via plugins.
Closes#6513
After #6517 we ended up registering all of the actions (included admin ones) to the NodeClient.
Made sure that only the proper type of Action instances are registered to each client type.
Also fixed some compiler warnings: unused members, imports and non matching generic types.
Closes#6563
Mid-term we should switch from `BytesValues` to Lucene's doc values APIs, in
particular the `SortedSetDocValues` class. While `BytesValues.WithOrdinals` and
SortedSetDocValues expose the same functionality, `BytesValues.WithOrdinals`
exposes its ordinals via a different `Ordinals.Docs` object while
`SortedSetDocValues` exposes them on the same object as the one that holds the
values. This commit merges ordinals into `BytesValues.WithOrdinals` in order to
make both classes even closer.
Global ordinals were a bit tricky to migrate so I just changed them to use
Lucene's OrdinalMap that will soon (LUCENE-5767, scheduled for 4.9) have the
same optimizations as our global ordinals.
Close#6524
The `exists` and `missing` filters need to merge postings lists of all existing
terms, which can be very costly, especially on high-cardinality fields. This
commit indexes the field names of a document under `_field_names` and reuses it
to speed up the `exists` and `missing` filters.
This is only enabled for indices that are created on or after Elasticsearch
1.3.0.
Close#5659
VersionFieldMapper.defaultDocValuesFormat claims that the default is `disk`.
This is not used to choose the DV format in the index but for mappings
serialization in order to know when the _version doc values format is
different from the default format. This made it impossible to use the `disk`
doc values format since mappings would never retain that information at
serialization time.
Close#6523
The TTL, size, timestamp and index meta properties could be lost on an
update of a single field mapping due to a wrong comparison in the
merge method (which was caused by a wrong initialization, which marked
an update as explicitely disabled instead of unset.
Closes#5053
Boaz Leskes
Simon Willnauer
Robert Muir
Shay Banon
Robert Muir
Lee Hinman
mahdeto
Clinton Gormley
Andrew Raines
Alexander Reelsen
Martijn van Groningen
javanna
Adrien Grand