* NN optimization for hdfs data segments.
* HdfsDataSegmentKiller, HdfsDataSegment finder changes to use new storage
format.Docs update.
* Common utility function in DataSegmentPusherUtil.
* new static method `makeSegmentOutputPathUptoVersionForHdfs` in JobHelper
* reuse getHdfsStorageDirUptoVersion in
DataSegmentPusherUtil.getHdfsStorageDir()
* Addressed comments.
* Review comments.
* HdfsDataSegmentKiller requested changes.
* extra newline
* Add maprfs.
* No more singleton. Reduce iterations
* Granularities
* Fix the delay in the test
* Add license header
* Remove unused imports
* Lot more unused imports from all the rearranging
* CR feedback
* Move javadoc to constructor
* Refactor Segment Granularity
* Beginning of one granularity
* Copy the fix for custom periods in segment-grunalrity over here.
* Remove the custom serialization for now.
* Compilation cleanup
* Reformat code
* Fixing unit tests
* Unify to use a single iterable
* Backward compatibility for rolling upgrade
* Minor check style. Cosmetic changes.
* Rename length and millis to duration
* CR feedback
* Minor changes.
* auto reset option for Kafka Indexing service in case message at the offset being fetched is not present anymore at kafka brokers
* review comments
* review comments
* reverted last change
* review comments
* review comments
* fix typo
* Fix#3795 (Java 7 compatibility).
Also introduce Animal Sniffer checks during build, which would
have caught the original problems.
* Add Animal Sniffer on caffeine-cache for JDK8.
* Fix PathChildrenCache's executorService leak in Announcer, CuratorInventoryManager and RemoteTaskRunner
* Use a single ExecutorService for all workerStatusPathChildrenCaches in RemoteTaskRunner
* Blacklist workers if they fail for too many times
* Adding documentation
* Changing to timeout to period and updating docs
* 1. Add configurable maxPercentageBlacklistWorkers
2. Rename variable
* Change maxPercentageBlacklistWorkers to double
* Remove thread.sleep
* Update emitter dependency to 0.4.0 and emit "version" dimension for all druid metrics, not only query metrics
* Remove unused imports
* Use empty string instead of "testing-version" as a version placeholder
* ability to not rollup at index time, make pre aggregation an option
* rename getRowIndexForRollup to getPriorIndex
* fix doc misspelling
* test query using no-rollup indexes
* fix benchmark fail due to jmh bug
Fixes inconsistent metric handling between the two implementations. Formerly,
RealtimePlumber only emitted query/segmentAndCache/time and query/wait and
Appenderator only emitted query/partial/time and query/wait (all per sink).
Now they both do the same thing:
- query/segmentAndCache/time, query/segment/time are the time spent per sink.
- query/cpu/time is the CPU time spent per query.
- query/wait/time is the executor waiting time per sink.
These generally match historical metrics, except segmentAndCache & segment
mean the same thing here, because one Sink may be partially cached and
partially uncached and we aren't splitting that out.
* validate X-Druid-Task-Id header in request and add header to response
* modify KafkaIndexTaskClient to take a TaskLocationProvider as the TaskLocation may not remain constant
* Datasource as lookup tier
* Adds an option to let indexing service tasks pull their lookup tier from the datasource they are working for.
* Fix bad docs for lookups lookupTier
* Add Datasource name holder
* Move task and datasource to be pulled from Task file
* Make LookupModule pull from bound dataSource
* Fix test
* Fix code style on imports
* Fix formatting
* Make naming better
* Address code comments about naming
- Introduce `AuthorizationInfo` interface, specific implementations of which would be provided by extensions
- If the `druid.auth.enabled` is set to `true` then the `isAuthorized` method of `AuthorizationInfo` will be called to perform authorization checks
- `AuthorizationInfo` object will be created in the servlet filters of specific extension and will be passed as a request attribute with attribute name as `AuthConfig.DRUID_AUTH_TOKEN`
- As per the scope of this PR, all resources that needs to be secured are divided into 3 types - `DATASOURCE`, `CONFIG` and `STATE`. For any type of resource, possible actions are - `READ` or `WRITE`
- Specific ResourceFilters are used to perform auth checks for all endpoints that corresponds to a specific resource type. This prevents duplication of logic and need to inject HttpServletRequest inside each endpoint. For example
- `DatasourceResourceFilter` is used for endpoints where the datasource information is present after "datasources" segment in the request Path such as `/druid/coordinator/v1/datasources/`, `/druid/coordinator/v1/metadata/datasources/`, `/druid/v2/datasources/`
- `RulesResourceFilter` is used where the datasource information is present after "rules" segment in the request Path such as `/druid/coordinator/v1/rules/`
- `TaskResourceFilter` is used for endpoints is used where the datasource information is present after "task" segment in the request Path such as `druid/indexer/v1/task`
- `ConfigResourceFilter` is used for endpoints like `/druid/coordinator/v1/config`, `/druid/indexer/v1/worker`, `/druid/worker/v1` etc
- `StateResourceFilter` is used for endpoints like `/druid/broker/v1/loadstatus`, `/druid/coordinator/v1/leader`, `/druid/coordinator/v1/loadqueue`, `/druid/coordinator/v1/rules` etc
- For endpoints where a list of resources is returned like `/druid/coordinator/v1/datasources`, `/druid/indexer/v1/completeTasks` etc. the list is filtered to return only the resources to which the requested user has access. In these cases, `HttpServletRequest` instance needs to be injected in the endpoint method.
Note -
JAX-RS specification provides an interface called `SecurityContext`. However, we did not use this but provided our own interface `AuthorizationInfo` mainly because it provides more flexibility. For example, `SecurityContext` has a method called `isUserInRole(String role)` which would be used for auth checks and if used then the mapping of what roles can access what resource needs to be modeled inside Druid either using some convention or some other means which is not very flexible as Druid has dynamic resources like datasources. Fixes#2355 with PR #2424
* Add back FilteredServerView removed in a32906c7fd to reduce memory usage using watched tiers.
* Add functionality to specify "druid.broker.segment.watchedDataSources"
segment creation deterministic.
This means that each segment will contain data from just one Kafka
partition. So, users will probably not want to have a super high number
of Kafka partitions...
Fixes#2703.
Avoids the following message from being printed on Overlord startup:
WARNING: Parameter 1 of type io.druid.indexing.common.actions.TaskActionHolder<T> from
public <T> javax.ws.rs.core.Response io.druid.indexing.overlord.http.OverlordResource.doAction
(io.druid.indexing.common.actions.TaskActionHolder<T>) is not resolvable to a concrete type
This PR changes the retry of task actions to be a bit more aggressive
by reducing the maxWait. Current defaults were 1 min to 10 mins, which
lead to a very delayed recovery in case there are any transient network
issues between the overlord and the peons.
doc changes.
- It's okay to suppress InterruptedException during graceful shutdown, as
tasks may use it to accelerate their own shutdown.
- It's okay to ignore return statuses during graceful shutdown (which may
be FAILED!) because it actually doesn't matter what they are.
Geared towards supporting transactional inserts of new segments. This involves an
interface "DataSourceMetadata" that allows combining of partially specified metadata
(useful for partitioned ingestion).
DataSource metadata is stored in a new "dataSource" table.
Appenderators are a way of getting more control over the ingestion process
than a Plumber allows. The idea is that existing Plumbers could be implemented
using Appenderators, but you could also implement things that Plumbers can't do.
FiniteAppenderatorDrivers help simplify indexing a finite stream of data.
Also:
- Sink: Ability to consider itself "finished" vs "still writable".
- Sink: Ability to return the number of rows contained within the sink.
The incremental indexes handle that now so it's not necessary.
Also, add debug logging and more detailed exceptions to the incremental
indexes for the case where there are parse exceptions during aggregation.
To bring consistency to docs and source this commit changes the default
values for maxRowsInMemory and rowFlushBoundary to 75000 after
discussion in PR https://github.com/druid-io/druid/pull/2457.
The previous default was 500000 and it's lower now on the grounds that
it's better for a default to be somewhat less efficient, and work,
than to reach for the stars and possibly result in
"OutOfMemoryError: java heap space" errors.
- Add TaskLocation class
- Add registerListener to TaskRunner
- Add getLocation to TaskRunnerWorkItem
- Implement location tracking in existing TaskRunners
- Rework WorkerTaskMonitor to do management out of a single thread so it can
handle status and location updates more simply.
* Moves last run task state information to Worker
* Makes WorkerTaskRunner a TaskRunner which has interfaces to help with getting information about a Worker
Two changes:
- Allow IncrementalIndex to suppress ParseExceptions on "aggregate".
- Add "reportParseExceptions" option to realtime tuning configs. By default this is "false".
Behavior of the counters should now be:
- processed: Number of rows indexed, including rows where some fields could be parsed and some could not.
- thrownAway: Number of rows thrown away due to rejection policy.
- unparseable: Number of rows thrown away due to being completely unparseable (no fields salvageable at all).
If "reportParseExceptions" is true then "unparseable" will always be zero (because a parse error would
cause an exception to be thrown). In addition, "processed" will only include fully parseable rows
(because even partial parse failures will cause exceptions to be thrown).
Fixes#2510.
* Don't put druid****selfcontained.jar at the end of the hadoop isolated classpath
* Add `<scope>provided</scope>` to prevent repeated dependency inclusion in the extension directories
- Throw most exceptions rather than suppressing them, which should help
detect problems. Continue suppressing exceptions that make sense to
suppress.
- Handle payload length checks consistently, and improve error message.
- Remove unused WorkerCuratorCoordinator.announceTaskAnnouncement method.
- Max znode length should be int, not long.
- Add tests.
* Defaults the thread priority to java.util.Thread.NORM_PRIORITY in io.druid.indexing.common.task.AbstractTask
* Each exec service has its own Task Factory which is assigned a priority for spawned task. Therefore each priority class has a unique exec service
* Added priority to tasks as taskPriority in the task context. <0 means low, 0 means take default, >0 means high. It is up to any particular implementation to determine how to handle these numbers
* Add options to ForkingTaskRunner
* Add "-XX:+UseThreadPriorities" default option
* Add "-XX:ThreadPriorityPolicy=42" default option
* AbstractTask - Removed unneded @JsonIgnore on priority
* Added priority to RealtimePlumber executors. All sub-executors (non query runners) get Thread.MIN_PRIORITY
* Add persistThreadPriority and mergeThreadPriority to realtime tuning config
Historical will drop a segment that shouldn't be dropped in the following scenario:
Historical node tried to load segmentA, but failed with SegmentLoadingException,
then ZkCoordinator called removeSegment(segmentA, blah) to schedule a runnable that would drop segmentA by deleting its files. Now, before that runnable executed, another LOAD request was sent to this historical, this time historical actually succeeded on loading segmentA and announced it. But later on, the scheduled drop-of-segment runnable started executing and removed the segment files, while historical is still announcing segmentA.
- Shedding locks at startup is bad, we actually want to keep them. Stop doing that.
- stopGracefully now interrupts the run thread if had started running finishJob. This avoids
waiting for handoff unnecessarily.
- fixes#1970
- extracted out segment handoff callbacks in SegmentHandoffNotifier
which is responsible for tracking segment handoffs and doing callbacks
when handoff is complete.
- Coordinator now maintains a view of segments in the cluster, this
will affect the jam heap requirements for the overlord for large
clusters.
realtime index task and nodes now use HTTP end points exposed by the
coordinator to get serverView
review comment
fix realtime node guide injection
review comments
make test not rely on scheduled exec
fix compilation
fix import
review comment
introduce immutableSegmentLoadInfo
fix son reading
remove unnecessary logging
This is done by killing and respawning the jvms rather than reconnecting to existing
jvms, for a couple reasons. One is that it lets you restore tasks after server reboots
too, and another is that it lets you upgrade all the software on a box at once by just
restarting everything.
The main changes are,
1) Add "canRestore" and "stopGracefully" methods to Tasks that say if a task can
stop gracefully, and actually do a graceful stop. RealtimeIndexTask is the only
one that currently implements this.
2) Add "stop" method to TaskRunners that attempts to do an orderly shutdown.
ThreadPoolTaskRunner- call stopGracefully on restorable tasks, wait for exit
ForkingTaskRunner- close output stream to restorable tasks, wait for exit
RemoteTaskRunner- do nothing special, we actually don't want to shutdown
3) Add "restore" method to TaskRunners that attempts to bootstrap tasks from last run.
Only ForkingTaskRunner does anything here. It maintains a "restore.json" file with
a list of restorable tasks.
4) Have the CliPeon's ExecutorLifecycle lock the task base directory to avoid a restored
task and a zombie old task from stomping on each other.
Deserialization of Optionals does not work quite right- they come back as actual
nulls, rather than absent Optionals. So these probably only ever worked for the local
task action client.
This is a feature meant to allow realtime tasks to work without being told upfront
what shardSpec they should use (so we can potentially publish a variable number
of segments per interval).
The idea is that there is a "pendingSegments" table in the metadata store that
tracks allocated segments. Each one has a segment id (the same segment id we know
and love) and is also part of a sequence.
The sequences are an idea from @cheddar that offers a way of doing replication.
If there are N tasks reading exactly the same data with exactly the same logic
(think Kafka tasks reading a fixed range of offsets) then you can place them
in the same sequence, and they will generate the same sequence of segments.
In #933 the ForkingTaskRunner's logging was changed to buffered from
unbuffered. This means that the last few KB of the logs are generally
not visible while a task is running, which makes debugging running
tasks difficult.
This is accomplished by making sure that scheduleTasksCleanupForWorker is only called
from the PathChildrenCache event thread, having it cancel existing cleanup tasks when
it adds a new one, and having tasks check on finish that the thing they are removing
from the task list is actually themselves.
Akash Dwivedi
praveev
kaijianding
Parag Jain
David Lim
Himanshu
Gian Merlino
DaimonPl
Jihoon Son
Charles Allen
Roman Leventov
Niketh Sabbineni
Dave Li
Nishant
Hyukjin Kwon
Xavier Léauté
jon-wei
Fangjin Yang
binlijin
Björn Zettergren
Bingkun Guo
Andrés Gomez
Parag Jain
Charles Allen
Jonathan Wei
Kurt Young
navis.ryu
Xavier Léauté
Bartosz Ługowski
zhxiaog