See #3125 for details on each item.
* Remove real-time/router executor tests-1
These are the ones which doesn't have '_%d' in the test
output files.
* Remove real-time/router executor tests-2
These are the ones which has in the test
output files.
* Move the tests outputs to correct place
* Make sure that single shard commits use 2PC on adaptive executor
It looks like we've messed the tests in #2891. Fixing back.
* Use adaptive executor for all router queries
This becomes important because when task-tracker is picked, we
used to pick router executor, which doesn't make sense.
* Remove explicit references to real-time/router executors in the tests
* JobExecutorType never picks real-time/router executors
* Make sure to go incremental in test output numbers
* Even users cannot pick real-time anymore
* Do not use real-time/router custom scans
* Get rid of unnecessary normalizations
* Reflect unneeded normalizations
* Get rid of unnecessary test output file
DESCRIPTION: Provide a GUC to turn of the new dependency propagation functionality
In the case the dependency propagation functionality introduced in 9.0 causes issues to a cluster of a user they can turn it off almost completely. The only dependency that will still be propagated and kept track of is the schema to emulate the old behaviour.
GUC to change is `citus.enable_object_propagation`. When set to `false` the functionality will be mostly turned off. Be aware that objects marked as distributed in `pg_dist_object` will still be kept in the catalog as a distributed object. Alter statements to these objects will not be propagated to workers and may cause desynchronisation.
DESCRIPTION: Add feature flag to turn off create type propagation
When `citus.enable_create_type_propagation` is set to `false` citus will not propagate `CREATE TYPE` statements to the workers. Types are still distributed when tables that depend on these types are distributed.
/*
* local_executor.c
*
* The scope of the local execution is locally executing the queries on the
* shards. In other words, local execution does not deal with any local tables
* that are not shards on the node that the query is being executed. In that sense,
* the local executor is only triggered if the node has both the metadata and the
* shards (e.g., only Citus MX worker nodes).
*
* The goal of the local execution is to skip the unnecessary network round-trip
* happening on the node itself. Instead, identify the locally executable tasks and
* simply call PostgreSQL's planner and executor.
*
* The local executor is an extension of the adaptive executor. So, the executor uses
* adaptive executor's custom scan nodes.
*
* One thing to note that Citus MX is only supported with replication factor = 1, so
* keep that in mind while continuing the comments below.
*
* On the high level, there are 3 slightly different ways of utilizing local execution:
*
* (1) Execution of local single shard queries of a distributed table
*
* This is the simplest case. The executor kicks at the start of the adaptive
* executor, and since the query is only a single task the execution finishes
* without going to the network at all.
*
* Even if there is a transaction block (or recursively planned CTEs), as long
* as the queries hit the shards on the same, the local execution will kick in.
*
* (2) Execution of local single queries and remote multi-shard queries
*
* The rule is simple. If a transaction block starts with a local query execution,
* all the other queries in the same transaction block that touch any local shard
* have to use the local execution. Although this sounds restrictive, we prefer to
* implement in this way, otherwise we'd end-up with as complex scenarious as we
* have in the connection managements due to foreign keys.
*
* See the following example:
* BEGIN;
* -- assume that the query is executed locally
* SELECT count(*) FROM test WHERE key = 1;
*
* -- at this point, all the shards that reside on the
* -- node is executed locally one-by-one. After those finishes
* -- the remaining tasks are handled by adaptive executor
* SELECT count(*) FROM test;
*
*
* (3) Modifications of reference tables
*
* Modifications to reference tables have to be executed on all nodes. So, after the
* local execution, the adaptive executor keeps continuing the execution on the other
* nodes.
*
* Note that for read-only queries, after the local execution, there is no need to
* kick in adaptive executor.
*
* There are also few limitations/trade-offs that is worth mentioning. First, the
* local execution on multiple shards might be slow because the execution has to
* happen one task at a time (e.g., no parallelism). Second, if a transaction
* block/CTE starts with a multi-shard command, we do not use local query execution
* since local execution is sequential. Basically, we do not want to lose parallelism
* across local tasks by switching to local execution. Third, the local execution
* currently only supports queries. In other words, any utility commands like TRUNCATE,
* fails if the command is executed after a local execution inside a transaction block.
* Forth, the local execution cannot be mixed with the executors other than adaptive,
* namely task-tracker, real-time and router executors. Finally, related with the
* previous item, COPY command cannot be mixed with local execution in a transaction.
* The implication of that any part of INSERT..SELECT via coordinator cannot happen
* via the local execution.
*/
With this commit, we're introducing the Adaptive Executor.
The commit message consists of two distinct sections. The first part explains
how the executor works. The second part consists of the commit messages of
the individual smaller commits that resulted in this commit. The readers
can search for the each of the smaller commit messages on
https://github.com/citusdata/citus and can learn more about the history
of the change.
/*-------------------------------------------------------------------------
*
* adaptive_executor.c
*
* The adaptive executor executes a list of tasks (queries on shards) over
* a connection pool per worker node. The results of the queries, if any,
* are written to a tuple store.
*
* The concepts in the executor are modelled in a set of structs:
*
* - DistributedExecution:
* Execution of a Task list over a set of WorkerPools.
* - WorkerPool
* Pool of WorkerSessions for the same worker which opportunistically
* executes "unassigned" tasks from a queue.
* - WorkerSession:
* Connection to a worker that is used to execute "assigned" tasks
* from a queue and may execute unasssigned tasks from the WorkerPool.
* - ShardCommandExecution:
* Execution of a Task across a list of placements.
* - TaskPlacementExecution:
* Execution of a Task on a specific placement.
* Used in the WorkerPool and WorkerSession queues.
*
* Every connection pool (WorkerPool) and every connection (WorkerSession)
* have a queue of tasks that are ready to execute (readyTaskQueue) and a
* queue/set of pending tasks that may become ready later in the execution
* (pendingTaskQueue). The tasks are wrapped in a ShardCommandExecution,
* which keeps track of the state of execution and is referenced from a
* TaskPlacementExecution, which is the data structure that is actually
* added to the queues and describes the state of the execution of a task
* on a particular worker node.
*
* When the task list is part of a bigger distributed transaction, the
* shards that are accessed or modified by the task may have already been
* accessed earlier in the transaction. We need to make sure we use the
* same connection since it may hold relevant locks or have uncommitted
* writes. In that case we "assign" the task to a connection by adding
* it to the task queue of specific connection (in
* AssignTasksToConnections). Otherwise we consider the task unassigned
* and add it to the task queue of a worker pool, which means that it
* can be executed over any connection in the pool.
*
* A task may be executed on multiple placements in case of a reference
* table or a replicated distributed table. Depending on the type of
* task, it may not be ready to be executed on a worker node immediately.
* For instance, INSERTs on a reference table are executed serially across
* placements to avoid deadlocks when concurrent INSERTs take conflicting
* locks. At the beginning, only the "first" placement is ready to execute
* and therefore added to the readyTaskQueue in the pool or connection.
* The remaining placements are added to the pendingTaskQueue. Once
* execution on the first placement is done the second placement moves
* from pendingTaskQueue to readyTaskQueue. The same approach is used to
* fail over read-only tasks to another placement.
*
* Once all the tasks are added to a queue, the main loop in
* RunDistributedExecution repeatedly does the following:
*
* For each pool:
* - ManageWorkPool evaluates whether to open additional connections
* based on the number unassigned tasks that are ready to execute
* and the targetPoolSize of the execution.
*
* Poll all connections:
* - We use a WaitEventSet that contains all (non-failed) connections
* and is rebuilt whenever the set of active connections or any of
* their wait flags change.
*
* We almost always check for WL_SOCKET_READABLE because a session
* can emit notices at any time during execution, but it will only
* wake up WaitEventSetWait when there are actual bytes to read.
*
* We check for WL_SOCKET_WRITEABLE just after sending bytes in case
* there is not enough space in the TCP buffer. Since a socket is
* almost always writable we also use WL_SOCKET_WRITEABLE as a
* mechanism to wake up WaitEventSetWait for non-I/O events, e.g.
* when a task moves from pending to ready.
*
* For each connection that is ready:
* - ConnectionStateMachine handles connection establishment and failure
* as well as command execution via TransactionStateMachine.
*
* When a connection is ready to execute a new task, it first checks its
* own readyTaskQueue and otherwise takes a task from the worker pool's
* readyTaskQueue (on a first-come-first-serve basis).
*
* In cases where the tasks finish quickly (e.g. <1ms), a single
* connection will often be sufficient to finish all tasks. It is
* therefore not necessary that all connections are established
* successfully or open a transaction (which may be blocked by an
* intermediate pgbouncer in transaction pooling mode). It is therefore
* essential that we take a task from the queue only after opening a
* transaction block.
*
* When a command on a worker finishes or the connection is lost, we call
* PlacementExecutionDone, which then updates the state of the task
* based on whether we need to run it on other placements. When a
* connection fails or all connections to a worker fail, we also call
* PlacementExecutionDone for all queued tasks to try the next placement
* and, if necessary, mark shard placements as inactive. If a task fails
* to execute on all placements, the execution fails and the distributed
* transaction rolls back.
*
* For multi-row INSERTs, tasks are executed sequentially by
* SequentialRunDistributedExecution instead of in parallel, which allows
* a high degree of concurrency without high risk of deadlocks.
* Conversely, multi-row UPDATE/DELETE/DDL commands take aggressive locks
* which forbids concurrency, but allows parallelism without high risk
* of deadlocks. Note that this is unrelated to SEQUENTIAL_CONNECTION,
* which indicates that we should use at most one connection per node, but
* can run tasks in parallel across nodes. This is used when there are
* writes to a reference table that has foreign keys from a distributed
* table.
*
* Execution finishes when all tasks are done, the query errors out, or
* the user cancels the query.
*
*-------------------------------------------------------------------------
*/
All the commits involved here:
* Initial unified executor prototype
* Latest changes
* Fix rebase conflicts to master branch
* Add missing variable for assertion
* Ensure that master_modify_multiple_shards() returns the affectedTupleCount
* Adjust intermediate result sizes
The real-time executor uses COPY command to get the results
from the worker nodes. Unified executor avoids that which
results in less data transfer. Simply adjust the tests to lower
sizes.
* Force one connection per placement (or co-located placements) when requested
The existing executors (real-time and router) always open 1 connection per
placement when parallel execution is requested.
That might be useful under certain circumstances:
(a) User wants to utilize as much as CPUs on the workers per
distributed query
(b) User has a transaction block which involves COPY command
Also, lots of regression tests rely on this execution semantics.
So, we'd enable few of the tests with this change as well.
* For parameters to be resolved before using them
For the details, see PostgreSQL's copyParamList()
* Unified executor sorts the returning output
* Ensure that unified executor doesn't ignore sequential execution of DDLJob's
Certain DDL commands, mainly creating foreign keys to reference tables,
should be executed sequentially. Otherwise, we'd end up with a self
distributed deadlock.
To overcome this situaiton, we set a flag `DDLJob->executeSequentially`
and execute it sequentially. Note that we have to do this because
the command might not be called within a transaction block, and
we cannot call `SetLocalMultiShardModifyModeToSequential()`.
This fixes at least two test: multi_insert_select_on_conflit.sql and
multi_foreign_key.sql
Also, I wouldn't mind scattering local `targetPoolSize` variables within
the code. The reason is that we'll soon have a GUC (or a global
variable based on a GUC) that'd set the pool size. In that case, we'd
simply replace `targetPoolSize` with the global variables.
* Fix 2PC conditions for DDL tasks
* Improve closing connections that are not fully established in unified execution
* Support foreign keys to reference tables in unified executor
The idea for supporting foreign keys to reference tables is simple:
Keep track of the relation accesses within a transaction block.
- If a parallel access happens on a distributed table which
has a foreign key to a reference table, one cannot modify
the reference table in the same transaction. Otherwise,
we're very likely to end-up with a self-distributed deadlock.
- If an access to a reference table happens, and then a parallel
access to a distributed table (which has a fkey to the reference
table) happens, we switch to sequential mode.
Unified executor misses the function calls that marks the relation
accesses during the execution. Thus, simply add the necessary calls
and let the logic kick in.
* Make sure to close the failed connections after the execution
* Improve comments
* Fix savepoints in unified executor.
* Rebuild the WaitEventSet only when necessary
* Unclaim connections on all errors.
* Improve failure handling for unified executor
- Implement the notion of errorOnAnyFailure. This is similar to
Critical Connections that the connection managament APIs provide
- If the nodes inside a modifying transaction expand, activate 2PC
- Fix few bugs related to wait event sets
- Mark placement INACTIVE during the execution as much as possible
as opposed to we do in the COMMIT handler
- Fix few bugs related to scheduling next placement executions
- Improve decision on when to use 2PC
Improve the logic to start a transaction block for distributed transactions
- Make sure that only reference table modifications are always
executed with distributed transactions
- Make sure that stored procedures and functions are executed
with distributed transactions
* Move waitEventSet to DistributedExecution
This could also be local to RunDistributedExecution(), but in that case
we had to mark it as "volatile" to avoid PG_TRY()/PG_CATCH() issues, and
cast it to non-volatile when doing WaitEventSetFree(). We thought that
would make code a bit harder to read than making this non-local, so we
move it here. See comments for PG_TRY() in postgres/src/include/elog.h
and "man 3 siglongjmp" for more context.
* Fix multi_insert_select test outputs
Two things:
1) One complex transaction block is now supported. Simply update
the test output
2) Due to dynamic nature of the unified executor, the orders of
the errors coming from the shards might change (e.g., all of
the queries on the shards would fail, but which one appears
on the error message?). To fix that, we simply added it to
our shardId normalization tool which happens just before diff.
* Fix subeury_and_cte test
The error message is updated from:
failed to execute task
To:
more than one row returned by a subquery or an expression
which is a lot clearer to the user.
* Fix intermediate_results test outputs
Simply update the error message from:
could not receive query results
to
result "squares" does not exist
which makes a lot more sense.
* Fix multi_function_in_join test
The error messages update from:
Failed to execute task XXX
To:
function f(..) does not exist
* Fix multi_query_directory_cleanup test
The unified executor does not create any intermediate files.
* Fix with_transactions test
A test case that just started to work fine
* Fix multi_router_planner test outputs
The error message is update from:
Could not receive query results
To:
Relation does not exists
which is a lot more clearer for the users
* Fix multi_router_planner_fast_path test
The error message is update from:
Could not receive query results
To:
Relation does not exists
which is a lot more clearer for the users
* Fix isolation_copy_placement_vs_modification by disabling select_opens_transaction_block
* Fix ordering in isolation_multi_shard_modify_vs_all
* Add executor locks to unified executor
* Make sure to allocate enought WaitEvents
The previous code was missing the waitEvents for the latch and
postmaster death.
* Fix rebase conflicts for master rebase
* Make sure that TRUNCATE relies on unified executor
* Implement true sequential execution for multi-row INSERTS
Execute the individual tasks executed one by one. Note that this is different than
MultiShardConnectionType == SEQUENTIAL_CONNECTION case (e.g., sequential execution
mode). In that case, running the tasks across the nodes in parallel is acceptable
and implemented in that way.
However, the executions that are qualified here would perform poorly if the
tasks across the workers are executed in parallel. We currently qualify only
one class of distributed queries here, multi-row INSERTs. If we do not enforce
true sequential execution, concurrent multi-row upserts could easily form
a distributed deadlock when the upserts touch the same rows.
* Remove SESSION_LIFESPAN flag in unified_executor
* Apply failure test updates
We've changed the failure behaviour a bit, and also the error messages
that show up to the user. This PR covers majority of the updates.
* Unified executor honors citus.node_connection_timeout
With this commit, unified executor errors out if even
a single connection cannot be established within
citus.node_connection_timeout.
And, as a side effect this fixes failure_connection_establishment
test.
* Properly increment/decrement pool size variables
Before this commit, the idle and active connection
counts were not properly calculated.
* insert_select_executor goes through unified executor.
* Add missing file for task tracker
* Modify ExecuteTaskListExtended()'s signature
* Sort output of INSERT ... SELECT ... RETURNING
* Take partition locks correctly in unified executor
* Alternative implementation for force_max_query_parallelization
* Fix compile warnings in unified executor
* Fix style issues
* Decrement idleConnectionCount when idle connection is lost
* Always rebuild the wait event sets
In the previous implementation, on waitFlag changes, we were only
modifying the wait events. However, we've realized that it might
be an over optimization since (a) we couldn't see any performance
benefits (b) we see some errors on failures and because of (a)
we prefer to disable it now.
* Make sure to allocate enough sized waitEventSet
With multi-row INSERTs, we might have more sessions than
task*workerCount after few calls of RunDistributedExecution()
because the previous sessions would also be alive.
Instead, re-allocate events when the connectino set changes.
* Implement SELECT FOR UPDATE on reference tables
On master branch, we do two extra things on SELECT FOR UPDATE
queries on reference tables:
- Acquire executor locks
- Execute the query on all replicas
With this commit, we're implementing the same logic on the
new executor.
* SELECT FOR UPDATE opens transaction block even if SelectOpensTransactionBlock disabled
Otherwise, users would be very confused and their logic is very likely
to break.
* Fix build error
* Fix the newConnectionCount calculation in ManageWorkerPool
* Fix rebase conflicts
* Fix minor test output differences
* Fix citus indent
* Remove duplicate sorts that is added with rebase
* Create distributed table via executor
* Fix wait flags in CheckConnectionReady
* failure_savepoints output for unified executor.
* failure_vacuum output (pg 10) for unified executor.
* Fix WaitEventSetWait timeout in unified executor
* Stabilize failure_truncate test output
* Add an ORDER BY to multi_upsert
* Fix regression test outputs after rebase to master
* Add executor.c comment
* Rename executor.c to adaptive_executor.c
* Do not schedule tasks if the failed placement is not ready to execute
Before the commit, we were blindly scheduling the next placement executions
even if the failed placement is not on the ready queue. Now, we're ensuring
that if failed placement execution is on a failed pool or session where the
execution is on the pendingQueue, we do not schedule the next task. Because
the other placement execution should be already running.
* Implement a proper custom scan node for adaptive executor
- Switch between the executors, add GUC to set the pool size
- Add non-adaptive regression test suites
- Enable CIRCLE CI for non-adaptive tests
- Adjust test output files
* Add slow start interval to the executor
* Expose max_cached_connection_per_worker to user
* Do not start slow when there are cached connections
* Consider ExecutorSlowStartInterval in NextEventTimeout
* Fix memory issues with ReceiveResults().
* Disable executor via TaskExecutorType
* Make sure to execute the tests with the other executor
* Use task_executor_type to enable-disable adaptive executor
* Remove useless code
* Adjust the regression tests
* Add slow start regression test
* Rebase to master
* Fix test failures in adaptive executor.
* Rebase to master - 2
* Improve comments & debug messages
* Set force_max_query_parallelization in isolation_citus_dist_activity
* Force max parallelization for creating shards when asked to use exclusive connection.
* Adjust the default pool size
* Expand description of max_adaptive_executor_pool_size GUC
* Update warnings in FinishRemoteTransactionCommit()
* Improve session clean up at the end of execution
Explicitly list all the states that the execution might end,
otherwise warn.
* Remove MULTI_CONNECTION_WAIT_RETRY which is not used at all
* Add more ORDER BYs to multi_mx_partitioning
Adds support for propagation of SET LOCAL commands to all workers
involved in a query. For now, SET SESSION (i.e. plain SET) is not
supported whatsoever, though this code is intended as somewhat of a
base for implementing such support in the future.
As SET LOCAL modifications are scoped to the body of a BEGIN/END xact
block, queries wishing to use SET LOCAL propagation must be within such
a block. In addition, subsequent modifications after e.g. any SAVEPOINT
or ROLLBACK statements will correspondingly push or pop variable mod-
ifications onto an internal stack such that the behavior of changed
values across the cluster will be identical to such behavior on e.g.
single-node PostgreSQL (or equivalently, what values are visible to
the end user by running SHOW on such variables on the coordinator).
If nodes enter the set of participants at some point after SET LOCAL
modifications (or SAVEPOINT, ROLLBACK, etc.) have occurred, the SET
variable state is eagerly propagated to them upon their entrance (this
is identical to, and indeed just augments, the existing logic for the
propagation of the SAVEPOINT "stack").
A new GUC (citus.propagate_set_commands) has been added to control this
behavior. Though the code suggests the valid settings are 'none', 'local',
'session', and 'all', only 'none' (the default) and 'local' are presently
implemented: attempting to use other values will result in an error.
The feature is only intended for getting consistent outputs for the regression tests.
RETURNING does not have any ordering gurantees and with unified executor, the ordering
of query executions on the shards are also becoming unpredictable. Thus, we're enforcing
ordering when a GUC is set.
We implicitly add an `ORDER BY` something equivalent of
`
RETURNING expr1, expr2, .. ,exprN
ORDER BY expr1, expr2, .. ,exprN
`
As described in the code comments as well, this is probably not the most
performant approach we could implement. However, since we're only
targeting regression tests, I don't see any issues with that. If we
decide to expand this to a feature to users, we should revisit the
implementation and improve the performance.
This commit has two goals:
(a) Ensure to access both edges of the allocated stack
(b) Ensure that any compiler optimizations to prevent the
function optimized away.
Stack size after the patch:
sudo grep -A 1 stack /proc/2119/smaps
7ffe305a6000-7ffe307a9000 rw-p 00000000 00:00 0 [stack]
Size: 2060 kB
Stack size before the patch:
sudo grep -A 1 stack /proc/3610/smaps
7fff09957000-7fff09978000 rw-p 00000000 00:00 0 [stack]
Size: 132 kB
In this context, we define "Fast Path Planning for SELECT" as trivial
queries where Citus can skip relying on the standard_planner() and
handle all the planning.
For router planner, standard_planner() is mostly important to generate
the necessary restriction information. Later, the restriction information
generated by the standard_planner is used to decide whether all the shards
that a distributed query touches reside on a single worker node. However,
standard_planner() does a lot of extra things such as cost estimation and
execution path generations which are completely unnecessary in the context
of distributed planning.
There are certain types of queries where Citus could skip relying on
standard_planner() to generate the restriction information. For queries
in the following format, Citus does not need any information that the
standard_planner() generates:
SELECT ... FROM single_table WHERE distribution_key = X; or
DELETE FROM single_table WHERE distribution_key = X; or
UPDATE single_table SET value_1 = value_2 + 1 WHERE distribution_key = X;
Note that the queries might not be as simple as the above such that
GROUP BY, WINDOW FUNCIONS, ORDER BY or HAVING etc. are all acceptable. The
only rule is that the query is on a single distributed (or reference) table
and there is a "distribution_key = X;" in the WHERE clause. With that, we
could use to decide the shard that a distributed query touches reside on
a worker node.
The file handling the utility functions (DDL) for citus organically grew over time and became unreasonably large. This refactor takes that file and refactored the functionality into separate files per command. Initially modeled after the directory and file layout that can be found in postgres.
Although the size of the change is quite big there are barely any code changes. Only one two functions have been added for readability purposes:
- PostProcessIndexStmt which is extracted from PostProcessUtility
- PostProcessAlterTableStmt which is extracted from multi_ProcessUtility
A README.md has been added to `src/backend/distributed/commands` describing the contents of the module and every file in the module.
We need more documentation around the overloading of the COPY command, for now the boilerplate has been added for people with better knowledge to fill out.
Each PostgreSQL backend starts with a predefined amount of stack and this stack
size can be increased if there is a need. However, stack size increase during
high memory load may cause unexpected crashes, because if there is not enough
memory for stack size increase, there is nothing to do for process apart from
crashing. An interesting thing is; the process would get OOM error instead of
crash, if the process had an explicit memory request (with palloc) for example.
However, in the case of stack size increase, there is no system call to get OOM
error, so the process simply crashes.
With this change, we are increasing the stack size explicitly by requesting extra
memory from the stack, so that, even if there is not memory, we can at least get
an OOM instead of a crash.
This commit by default enables hiding shard names on MX workers
by simple replacing `pg_table_is_visible()` calls with
`citus_table_is_visible()` calls on the MX worker nodes. The latter
function filters out tables that are known to be shards.
The main motivation of this change is a better UX. The functionality
can be opted out via a GUC.
We also added two views, namely citus_shards_on_worker and
citus_shard_indexes_on_worker such that users can query
them to see the shards and their corresponding indexes.
We also added debug messages such that the filtered tables can
be interactively seen by setting the level to DEBUG1.
When a hash distributed table have a foreign key to a reference
table, there are few restrictions we have to apply in order to
prevent distributed deadlocks or reading wrong results.
The necessity to apply the restrictions arise from cascading
nature of foreign keys. When a foreign key on a reference table
cascades to a distributed table, a single operation over a single
connection can acquire locks on multiple shards of the distributed
table. Thus, any parallel operation on that distributed table, in the
same transaction should not open parallel connections to the shards.
Otherwise, we'd either end-up with a self-distributed deadlock or
read wrong results.
As briefly described above, the restrictions that we apply is done
by tracking the distributed/reference relation accesses inside
transaction blocks, and act accordingly when necessary.
The two main rules are as follows:
- Whenever a parallel distributed relation access conflicts
with a consecutive reference relation access, Citus errors
out
- Whenever a reference relation access is followed by a
conflicting parallel relation access, the execution mode
is switched to sequential mode.
There are also some other notes to mention:
- If the user does SET LOCAL citus.multi_shard_modify_mode
TO 'sequential';, all the queries should simply work with
using one connection per worker and sequentially executing
the commands. That's obviously a slower approach than Citus'
usual parallel execution. However, we've at least have a way
to run all commands successfully.
- If an unrelated parallel query executed on any distributed
table, we cannot switch to sequential mode. Because, the essense
of sequential mode is using one connection per worker. However,
in the presence of a parallel connection, the connection manager
picks those connections to execute the commands. That contradicts
with our purpose, thus we error out.
- COPY to a distributed table cannot be executed in sequential mode.
Thus, if we switch to sequential mode and COPY is executed, the
operation fails and there is currently no way of implementing that.
Note that, when the local table is not empty and create_distributed_table
is used, citus uses COPY internally. Thus, in those cases,
create_distributed_table() will also fail.
- There is a GUC called citus.enforce_foreign_key_restrictions
to disable all the checks. We added that GUC since the restrictions
we apply is sometimes a bit more restrictive than its necessary.
The user might want to relax those. Similarly, if you don't have
CASCADEing reference tables, you might consider disabling all the
checks.
To support more flexible (i.e. not at compile-time) specification of
libpq connection parameters, this change adds a new GUC, node_conninfo,
which must be a space-separated string of key-value pairs suitable for
parsing by libpq's connection establishment methods.
To avoid rebuilding and parsing these values at connection time, this
change also adds a cache in front of the configuration params to permit
immediate use of any previously-calculated parameters.
* Change worker_hash_partition_table() such that the
divergence between Citus planner's hashing and
worker_hash_partition_table() becomes the same.
* Rename single partitioning to single range partitioning.
* Add single hash repartitioning. Basically, logical planner
treats single hash and range partitioning almost equally.
Physical planner, on the other hand, treats single hash and
dual hash repartitioning almost equally (except for JoinPruning).
* Add a new GUC to enable this feature
After this change all the logic related to shard data fetch logic
will be removed. Planner won't plan any ShardFetchTask anymore.
Shard fetch related steps in real time executor and task-tracker
executor have been removed.
This commit introduces a new GUC to limit the intermediate
result size which we handle when we use read_intermediate_result
function for CTEs and complex subqueries.
Adds ```citus.enable_statistics_collection``` GUC variable, which ```true``` by default, unless built without libcurl. If statistics collection is enabled, sends basic usage data to Citus servers every 24 hours.
The data that is collected consists of:
- Citus version
- OS name & release
- Hardware Id
- Number of tables, rounded to next power of 2
- Size of data, rounded to next power of 2
- Number of workers
We added a new GUC citus.log_distributed_deadlock_detection
which is off by default. When set to on, we log some debug messages
related to the distributed deadlock to the server logs.
With this commit, the maintenance deamon starts to check for
distributed deadlocks.
We also introduced a GUC variable (distributed_deadlock_detection_factor)
whose value is multiplied with Postgres' deadlock_timeout. Setting
it to -1 disables the distributed deadlock detection.
We send SIGINT to a backend that is cancelled due to a deadlock. That
approach ends up being a very confusing error message.
With this commit we intercept the error messages and show a more
meaningful error message to the user.
Now that we already have the necessary infrastructure for detecting
distributed deadlocks. Thus, we don't need enable_deadlock_prevention
which is purely intended for preventing some forms of distributed
deadlocks.
This GUC has two settings, 'always' and 'never'. When it's set to
'never' all behavior stays exactly as it was prior to this commit. When
it's set to 'always' only SELECT queries are allowed to run, and only
secondary nodes are used when processing those queries.
Add some helper functions:
- WorkerNodeIsSecondary(), checks the noderole of the worker node
- WorkerNodeIsReadable(), returns whether we're currently allowed to
read from this node
- ActiveReadableNodeList(), some functions (namely, the ones on the
SELECT path) don't require working with Primary Nodes. They should call
this function instead of ActivePrimaryNodeList(), because the latter
will error out in contexts where we're not allowed to write to nodes.
- ActiveReadableNodeCount(), like the above, replaces
ActivePrimaryNodeCount().
- EnsureModificationsCanRun(), error out if we're not currently allowed
to run queries which modify data. (Either we're in read-only mode or
use_secondary_nodes is set)
Some parts of the code were switched over to use readable nodes instead
of primary nodes:
- Deadlock detection
- DistributedTableSize,
- the router, real-time, and task tracker executors
- ShardPlacement resolution
maxTaskStringSize determines the size of worker query string.
It was originally hard coded to a specific value. This has caused
issues at some users. Since it determines initial shared memory
allocation, we did not want to set it to an arbitrary higher number.
Instead made it configurable.
This commit introduces a new GUC variable max_task_string_size
Changes in this variable requires restart to be in effect.
This commit adds distributed transaction id infrastructure in
the scope of distributed deadlock detection.
In general, the distributed transaction id consists of a tuple
in the form of: `(databaseId, initiatorNodeIdentifier, transactionId,
timestamp)`.
Briefly, we add a shared memory block on each node, which holds some
information per backend (i.e., an array `BackendData backends[MaxBackends]`).
Later, on each coordinated transaction, Citus sends
`SELECT assign_distributed_transaction_id()` right after `BEGIN`.
For that backend on the worker, the distributed transaction id is set to
the values assigned via the function call.
The aim of the above is to correlate the transactions on the coordinator
to the transactions on the worker nodes.
Adds support for PostgreSQL 10 by copying in the requisite ruleutils
and updating all API usages to conform with changes in PostgreSQL 10.
Most changes are fairly minor but they are numerous. One particular
obstacle was the change in \d behavior in PostgreSQL 10's psql; I had
to add SQL implementations (views, mostly) to mimic the pre-10 output.
In this PR, we aim to deduce whether each of the RTE_RELATION
is joined with at least on another RTE_RELATION on their partition keys. If each
RTE_RELATION follows the above rule, we can conclude that all RTE_RELATIONs are
joined on their partition keys.
In order to do that, we invented a new equivalence class namely:
AttributeEquivalenceClass. In very simple words, a AttributeEquivalenceClass is
identified by an unique id and consists of a list of AttributeEquivalenceMembers.
Each AttributeEquivalenceMember is designed to identify attributes uniquely within the
whole query. The necessity of this arise since varno attributes are defined within
a single level of a query. Instead, here we want to identify each RTE_RELATION uniquely
and try to find equality among each RTE_RELATION's partition key.
Whenever we find an equality clause A = B, where both A and B originates from
relation attributes (i.e., not random expressions), we create an
AttributeEquivalenceClass to record this knowledge. If we later find another
equivalence B = C, we create another AttributeEquivalenceClass. Finally, we can
apply transitity rules and generate a new AttributeEquivalenceClass which includes
A, B and C.
Note that equality among the members are identified by the varattno and rteIdentity.
Each equality among RTE_RELATION is saved using an AttributeEquivalenceClass where
each member attribute is identified by a AttributeEquivalenceMember. In the final
step, we try generate a common attribute equivalence class that holds as much as
AttributeEquivalenceMembers whose attributes are a partition keys.
With this change, we start to error out if loaded citus binaries does not match
the available major version or installed citus extension version. In this case
we force user to restart the server or run ALTER EXTENSION depending on the
situation
Custom Scan is a node in the planned statement which helps external providers
to abstract data scan not just for foreign data wrappers but also for regular
relations so you can benefit your version of caching or hardware optimizations.
This sounds like only an abstraction on the data scan layer, but we can use it
as an abstraction for our distributed queries. The only thing we need to do is
to find distributable parts of the query, plan for them and replace them with
a Citus Custom Scan. Then, whenever PostgreSQL hits this custom scan node in
its Vulcano style execution, it will call our callback functions which run
distributed plan and provides tuples to the upper node as it scans a regular
relation. This means fewer code changes, fewer bugs and more supported features
for us!
First, in the distributed query planner phase, we create a Custom Scan which
wraps the distributed plan. For real-time and task-tracker executors, we add
this custom plan under the master query plan. For router executor, we directly
pass the custom plan because there is not any master query. Then, we simply let
the PostgreSQL executor run this plan. When it hits the custom scan node, we
call the related executor parts for distributed plan, fill the tuple store in
the custom scan and return results to PostgreSQL executor in Vulcano style,
a tuple per XXX_ExecScan() call.
* Modify planner to utilize Custom Scan node.
* Create different scan methods for different executors.
* Use native PostgreSQL Explain for master part of queries.
This adds a replication_model GUC which is used as the replication
model for any new distributed table that is not a reference table.
With this change, tables with replication factor 1 are no longer
implicitly MX tables.
The GUC is similarly respected during empty shard creation for e.g.
existing append-partitioned tables. If the model is set to streaming
while replication factor is greater than one, table and shard creation
routines will error until this invalid combination is corrected.
Changing this parameter requires superuser permissions.
They make fixing explain for prepared statement harder, and they don't
really fit into EXPLAIN in the first place. Additionally they're
currently not exercised in any tests.
Currently there are several places in citus that map placements to
connections and that manage placement health. Centralize this
knowledge. Because of the centralized knowledge about which
connection has previously been used for which shard/placement, this
also provides the basis for relaxing restrictions around combining
various forms of DDL/DML.
Connections for a placement can now be acquired using
GetPlacementConnection(). If the connection is used for DML or DDL the
FOR_DDL/DML flags should be used respectively. If an individual
remote transaction fails (but the transaction on the master succeeds)
and FOR_DDL/DML have been specified, the placement is marked as
invalid, unless that'd mark all placements for a shard as invalid.
We used to disable router planner and executor
when task executor is set to task-tracker.
This change enables router planning and execution
at all times regardless of task execution mode.
We are introducing a hidden flag enable_router_execution
to enable/disable router execution. Its default value is
true. User may disable router planning by setting it to false.
One less place managing remote transactions. It also makes it fairly
easy to use 2PC for certain modifications (e.g. reference tables). Just
issue a CoordinatedTransactionUse2PC(). If every placement failure
should cause the whole transaction to abort, additionally mark the
relevant transactions as critical.
This includes basic infrastructure for logging of commands sent to
remote/worker nodes. Note that this has no effect as of yet, since no
callers are converted to the new infrastructure.
Connections are tracked and released by integrating into postgres'
transaction handling. That allows to to use connections without having
to resort to having to disable interrupts or using PG_TRY/CATCH blocks
to avoid leaking connections.
This is intended to eventually replace multi_client_executor.c and
connection_cache.c, and to provide the basis of a centralized
transaction management.
The newly introduced transaction hook should, in the future, be the only
one in citus, to allow for proper ordering between operations. For now
this central handler is responsible for releasing connections and
resetting XactModificationLevel after a transaction.
Adds support for PostgreSQL 9.6 by copying in the requisite ruleutils
file and refactoring the out/readfuncs code to flexibly support the
old-style copy/pasted out/readfuncs (prior to 9.6) or use extensible
node APIs (in 9.6 and higher).
Most version-specific code within this change is only needed to set new
fields in the AggRef nodes we build for aggregations. Version-specific
test output files were added in certain cases, though in most they were
not necessary. Each such file begins by e.g. printing the major version
in order to clarify its purpose.
The comment atop citus_nodes.h details how to add support for new nodes
for when that becomes necessary.
Related to #786
This change adds the `pg_dist_node` table that contains the information
about the workers in the cluster, replacing the previously used
`pg_worker_list.conf` file (or the one specified with `citus.worker_list_file`).
Upon update, `pg_worker_list.conf` file is read and `pg_dist_node` table is
populated with the file's content. After that, `pg_worker_list.conf` file
is renamed to `pg_worker_list.conf.obsolete`
For adding and removing nodes, the change also includes two new UDFs:
`master_add_node` and `master_remove_node`, which require superuser
permissions.
'citus.worker_list_file' guc is kept for update purposes but not used after the
update is finished.
I had changed these callbacks to use the same method I chose for the
router executor (for consistency), but as that method is flawed, we now
want to ensure we directly register them from PG_init as well.
Not entirely sure why we went with the shared memory hook approach, but
it causes problems (multiple registration) during crashes. Changing to
a simple direct registration call from PG_init.
Three changes here to get to true multi-statement, multi-relation DDL
transactions (same functionality pre-5.2, with benefits of atomicity):
1. Changed the multi-shard utility hook to always run (consistency
with router executor hook, removes ad-hoc "installed" boolean)
2. Change the global connection list in multi_shard_transaction to
instead be a hash; update related functions to operate on global
hash instead of local hash/global list
3. Remove check within DDL code to prevent subsequent DDL commands;
place unset/reset guard around call to ConnectToNode to permit
connecting to additional nodes after DDL transaction has begun
In addition, code has been added to raise an error if a ROLLBACK TO
SAVEPOINT is attempted (similar to router executor), and comprehensive
tests execute all multi-DDL scenarios (full success, user ROLLBACK, any
actual errors (say, duplicate index), partial failure (duplicate index
on one node but not others), partial COMMIT (one node fails), and 2PC
partial PREPARE (one node fails)). Interleavings with other commands
(DML, \copy) are similarly all covered.
We can now support richer set of queries in router planner.
This allow us to support CTEs, joins, window function, subqueries
if they are known to be executed at a single worker with a single
task (all tables are filtered down to a single shard and a single
worker contains all table shards referenced in the query).
Fixes : #501
Allows the use of modification commands (INSERT/UPDATE/DELETE) within
transaction blocks (delimited by BEGIN and ROLLBACK/COMMIT), so long as
all modifications hit a subset of nodes involved in the first such com-
mand in the transaction. This does not circumvent the requirement that
each individual modification command must still target a single shard.
For instance, after sending BEGIN, a user might INSERT some rows to a
shard replicated on two nodes. Subsequent modifications can hit other
shards, so long as they are on one or both of these nodes.
SAVEPOINTs are supported, though if the user actually attempts to send
a ROLLBACK command that specifies a SAVEPOINT they will receive an
ERROR at the end of the topmost transaction.
Placements are only marked inactive if at least one replica succeeds
in a transaction where others fail. Non-atomic behavior is possible if
the shard targeted by the initial modification within a transaction has
a higher replication factor than another shard within the same block
and a node with the latter shard has a failure during the COMMIT phase.
Other methods of denoting transaction blocks (multi-statement commands
sent all at once and functions written in e.g. PL/pgSQL or other such
languages) are not presently supported; their treatment remains the
same as before.
Fixes#513
This change modifies the DDL Propagation logic so that DDL queries
are propagated via 2-Phase Commit protocol. This way, failures during
the execution of distributed DDL commands will not leave the table in
an intermediate state and the pending prepared transactions can be
commited manually.
DDL commands are not allowed inside other transaction blocks or functions.
DDL commands are performed with 2PC regardless of the value of
`citus.multi_shard_commit_protocol` parameter.
The workflow of the successful case is this:
1. Open individual connections to all shard placements and send `BEGIN`
2. Send `SELECT worker_apply_shard_ddl_command(<shardId>, <DDL Command>)`
to all connections, one by one, in a serial manner.
3. Send `PREPARE TRANSCATION <transaction_id>` to all connections.
4. Sedn `COMMIT` to all connections.
Failure cases:
- If a worker problem occurs before sending of all DDL commands is finished, then
all changes are rolled back.
- If a worker problem occurs after all DDL commands are sent but not after
`PREPARE TRANSACTION` commands are finished, then all changes are rolled back.
However, if a worker node is failed, then the prepared transactions in that worker
should be rolled back manually.
- If a worker problem occurs during `COMMIT PREPARED` statements are being sent,
then the prepared transactions on the failed workers should be commited manually.
- If master fails before the first 'PREPARE TRANSACTION' is sent, then nothing is
changed on workers.
- If master fails during `PREPARE TRANSACTION` commands are being sent, then the
prepared transactions on workers should be rolled back manually.
- If master fails during `COMMIT PREPARED` or `ROLLBACK PREPARED` commands are being
sent, then the remaining prepared transactions on the workers should be handled manually.
This change also helps with #480, since failed DDL changes no longer mark
failed placements as inactive.
Now, master_create_empty_shard() will create shards according to the
value of citus.shard_placement_policy which also makes default round-robin
instead of random.
SaitTalhaNisanci
Önder Kalacı
Hadi Moshayedi
Nils Dijk
Philip Dubé
Jason Petersen
Marco Slot
Jason Petersen
Burak Yucesoy
velioglu
Brian Cloutier
Murat Tuncer
Jason Petersen
Metin Doslu
mehmet furkan şahin
Andres Freund
Hadi Moshayedi
Murat Tuncer
Eren Basak
Eren