* Broker: Await initialization before finishing startup.
In particular, hold off on announcing the service and starting the
HTTP server until the server view and SQL metadata cache are finished
initializing. This closes a window of time where a Broker could return
partial results shortly after startup.
As part of this, some simplification of server-lifecycle service
announcements. This helps ensure that the two different kinds of
announcements we do (legacy and new-style) stay in sync.
* Remove unused imports.
* Fix NPE in ServerRunnable.
* FilteredRequestLogger: Fix start/stop, invalid delegate behavior.
Fixes two bugs:
1) FilteredRequestLogger did not start/stop the delegate.
2) FilteredRequestLogger would ignore an invalid delegate type, and
instead silently substitute the "noop" logger. This was due to a larger
problem with RequestLoggerProvider setup in general; the fix here is
to remove "defaultImpl" from the RequestLoggerProvider interface, and
instead have JsonConfigurator be responsible for creating the
default implementations. It is stricter about things than the old system
was, and is only willing to make a noop logger if it doesn't see any
request logger configs. Otherwise, it'll raise a provision error.
* Remove unneeded annotations.
* autosize processing buffers based on direct memory sizing
* remove oops, more test
* max 1gb autosize buffers, test, start of docs
* fix oops
* revert accidental change
* print buffer size in exception
* change the things
* Broker backpressure.
Adds a new property "druid.broker.http.maxQueuedBytes" and a new context
parameter "maxQueuedBytes". Both represent a maximum number of bytes queued
per query before exerting backpressure on the channel to the data server.
Fixes#4933.
* Fix query context doc.
* Rename io.druid to org.apache.druid.
* Fix META-INF files and remove some benchmark results.
* MonitorsConfig update for metrics package migration.
* Reorder some dimensions in inner queries for some reason.
* Fix protobuf tests.
* Cache: Add maxEntrySize config.
The idea is this makes it more feasible to cache query types that
can potentially generate large result sets, like groupBy and select,
without fear of writing too much to the cache per query.
Includes a refactor of cache population code in CachingQueryRunner and
CachingClusteredClient, such that they now use the same CachePopulator
interface with two implementations: one for foreground and one for
background.
The main reason for splitting the foreground / background impls is
that the foreground impl can have a more effective implementation of
maxEntrySize. It can stop retaining subvalues for the cache early.
* Add CachePopulatorStats.
* Fix whitespace.
* Fix docs.
* Fix various tests.
* Add tests.
* Fix tests.
* Better tests
* Remove conflict markers.
* Fix licenses.
- Setting druid.request.logging.delegate has no effect.
- The provider is injected based on a type parameter & this looks to be scoped to delegate for filtered loggers
* Anonymous authenticator that authenticates all requests and then directs them to an authorizer.
* Adding documentation
* Removed some fields from class AnonymousAuthenticator
* Updating docs
* fix freeSpacePercent in segmentCache.locations
* the check should probably test the other way around
* documentation should put the option in the right place
* examples have a superfluous backslash
* add test to verify correct behavior
* switch to Path and test with jimfs
Path allows to use different filesystems.
Jimfs provides an actual (in memory) filesystem.
This also allows more complex test scenarios.
The behavior should be unchanged by this commit.
* Revert "switch to Path and test with jimfs"
This reverts commit 8b9a418d65.
* add default caffeine cache size based on runtime Xmx or max 1GB
* update docs for caffeine cache
* fix formatting
* test caffeine size should never be less than 0
* set caffeine max default size to 1G not 1M
* fix caffeine cache tests
* Update defaultHadoopCoordinates in documentation.
To match changes applied in #5382.
* Remove a parameter with defaults from example configuration file.
If it has reasonable defaults, then why would it be in an example config file?
Also, it is yet another place that has been forgotten to be updated and will be forgotten in the future.
Also, if someone is running different hadoop version, then there's much more work to be done than just changing this property, so why give users false hopes?
* Fix typo in documentation.
* Add config to allow setting up custom unsecured paths for druid nodes.
* return all resources for Unsecured paths
* review comment - Add test
* fix tests
* fix test
* Add graceful shutdown timeout
* Handle interruptedException
* Incorporate code review comments
* Address code review comments
* Poll for activeConnections to be zero
* Use statistics handler to get active requests
* Use native jetty shutdown gracefully
* Move log line back to where it was
* Add unannounce wait time
* Make the default retain prior behavior
* Update docs with new config defaults
* Make duration handling on jetty shutdown more consistent
* StatisticsHandler is a wrapper
* Move jetty lifecycle error logging to error
Jihoon Son
Mingming Qiu
Gian Merlino
David Lim
Vadim Ogievetsky
Clint Wylie
seoeun
Caroline1000
Dayue Gao
Jonathan Wei
Nishant Bangarwa
Dylan Wylie
varaga
awelsh93
Atul Mohan
Michael Schnupp
Stuart McLean
Jakub Kukul
Joseph Glanville
Charles Allen