Adds Support for Single Shard Tables in
`update_distributed_table_colocation`.
This PR changes checks that make sure tables should be hash distributed
table to hash or single shard distributed tables.
Verify Citus UDFs work well with single shard tables
SUPPORTED
* citus_table_size
* citus_total_relation_size
* citus_relation_size
* citus_shard_sizes
* truncate_local_data_after_distributing_table
* create_distributed_function // test function colocated with a single
shard table
* undistribute_table
* alter_table_set_access_method
UNSUPPORTED - error out for single shard tables
* master_create_empty_shard
* create_distributed_table_concurrently
* create_distributed_table
* create_reference_table
* citus_add_local_table_to_metadata
* citus_split_shard_by_split_points
* alter_distributed_table
DESCRIPTION: Adds citus.enable_schema_based_sharding GUC that allows
sharding the database based on schemas when enabled.
* Refactor the logic that automatically creates Citus managed tables
* Refactor CreateSingleShardTable() to allow specifying colocation id
instead
* Add support for schema-based-sharding via a GUC
### What this PR is about:
Add **citus.enable_schema_based_sharding GUC** to enable schema-based
sharding. Each schema created while this GUC is ON will be considered
as a tenant schema. Later on, regardless of whether the GUC is ON or
OFF, any table created in a tenant schema will be converted to a
single shard distributed table (without a shard key). All the tenant
tables that belong to a particular schema will be co-located with each
other and will have a shard count of 1.
We introduce a new metadata table --pg_dist_tenant_schema-- to do the
bookkeeping for tenant schemas:
```sql
psql> \d pg_dist_tenant_schema
Table "pg_catalog.pg_dist_tenant_schema"
┌───────────────┬─────────┬───────────┬──────────┬─────────┐
│ Column │ Type │ Collation │ Nullable │ Default │
├───────────────┼─────────┼───────────┼──────────┼─────────┤
│ schemaid │ oid │ │ not null │ │
│ colocationid │ integer │ │ not null │ │
└───────────────┴─────────┴───────────┴──────────┴─────────┘
Indexes:
"pg_dist_tenant_schema_pkey" PRIMARY KEY, btree (schemaid)
"pg_dist_tenant_schema_unique_colocationid_index" UNIQUE, btree (colocationid)
psql> table pg_dist_tenant_schema;
┌───────────┬───────────────┐
│ schemaid │ colocationid │
├───────────┼───────────────┤
│ 41963 │ 91 │
│ 41962 │ 90 │
└───────────┴───────────────┘
(2 rows)
```
Colocation id column of pg_dist_tenant_schema can never be NULL even
for the tenant schemas that don't have a tenant table yet. This is
because, we assign colocation ids to tenant schemas as soon as they
are created. That way, we can keep associating tenant schemas with
particular colocation groups even if all the tenant tables of a tenant
schema are dropped and recreated later on.
When a tenant schema is dropped, we delete the corresponding row from
pg_dist_tenant_schema. In that case, we delete the corresponding
colocation group from pg_dist_colocation as well.
### Future work for 12.0 release:
We're building schema-based sharding on top of the infrastructure that
adds support for creating distributed tables without a shard key
(https://github.com/citusdata/citus/pull/6867).
However, not all the operations that can be done on distributed tables
without a shard key necessarily make sense (in the same way) in the
context of schema-based sharding. For example, we need to think about
what happens if user attempts altering schema of a tenant table. We
will tackle such scenarios in a future PR.
We will also add a new UDF --citus.schema_tenant_set() or such-- to
allow users to use an existing schema as a tenant schema, and another
one --citus.schema_tenant_unset() or such-- to stop using a schema as
a tenant schema in future PRs.
Enable router planner and a limited version of INSERT .. SELECT planner
for the queries that reference colocated null shard key tables.
* SELECT / UPDATE / DELETE / MERGE is supported as long as it's a router
query.
* INSERT .. SELECT is supported as long as it only references colocated
null shard key tables.
Note that this is not only limited to distributed INSERT .. SELECT but
also
covers a limited set of query types that require pull-to-coordinator,
e.g.,
due to LIMIT clause, generate_series() etc. ...
(Ideally distributed INSERT .. SELECT could handle such queries too,
e.g.,
when we're only referencing tables that don't have a shard key, but
today
this is not the case. See
https://github.com/citusdata/citus/pull/6773#discussion_r1140130562.
Add tests for ddl coverage:
* indexes
* partitioned tables + indexes with long names
* triggers
* foreign keys
* statistics
* grant & revoke statements
* truncate & vacuum
* create/test/drop view that depends on a dist table with no shard key
* policy & rls test
* alter table add/drop/alter_type column (using sequences/different data
types/identity columns)
* alter table add constraint (not null, check, exclusion constraint)
* alter table add column with a default value / set default / drop
default
* alter table set option (autovacuum)
* indexes / constraints without names
* multiple subcommands
Adds support for
* Creating new partitions after distributing (with null key) the parent
table
* Attaching partitions to a distributed table with null distribution key
(and automatically distribute the new partition with null key as well)
* Detaching partitions from it
With this PR, we allow creating distributed tables with without
specifying a shard key via create_distributed_table(). Here are the
the important details about those tables:
* Specifying `shard_count` is not allowed because it is assumed to be 1.
* We mostly call such tables as "null shard-key" table in code /
comments.
* To avoid doing a breaking layout change in create_distributed_table();
instead of throwing an error, it will inform the user that
`distribution_type`
param is ignored unless it's explicitly set to NULL or 'h'.
* `colocate_with` param allows colocating such null shard-key tables to
each other.
* We define this table type, i.e., NULL_SHARD_KEY_TABLE, as a subclass
of
DISTRIBUTED_TABLE because we mostly want to treat them as distributed
tables in terms of SQL / DDL / operation support.
* Metadata for such tables look like:
- distribution method => DISTRIBUTE_BY_NONE
- replication model => REPLICATION_MODEL_STREAMING
- colocation id => **!=** INVALID_COLOCATION_ID (distinguishes from
Citus local tables)
* We assign colocation groups for such tables to different nodes in a
round-robin fashion based on the modulo of "colocation id".
Note that this PR doesn't care about DDL (except CREATE TABLE) / SQL /
operation (i.e., Citus UDFs) support for such tables but adds a
preliminary
API.
We need to break sequence dependency for a table while creating the
table during non-transactional metadata sync to ensure idempotency of
the creation of the table.
**Problem:**
When we send `SELECT
pg_catalog.worker_drop_sequence_dependency(logicalrelid::regclass::text)
FROM pg_dist_partition` to workers during the non-transactional sync,
table might not be in `pg_dist_partition` at worker, and sequence
dependency is not broken at the worker.
**Solution:**
We break sequence dependency via `SELECT
pg_catalog.worker_drop_sequence_dependency(logicalrelid::regclass::text)`
for each table while creating it at the workers. It is safe to send
since the udf is a no-op when there is no sequence dependency.
DESCRIPTION: Fixes a bug related to sequence idempotency at
non-transactional sync.
Fixes https://github.com/citusdata/citus/issues/6888.
DESCRIPTION: Adds control for background task executors involving a node
### Background and motivation
Nonblocking concurrent task execution via background workers was
introduced in [#6459](https://github.com/citusdata/citus/pull/6459), and
concurrent shard moves in the background rebalancer were introduced in
[#6756](https://github.com/citusdata/citus/pull/6756) - with a hard
dependency that limits to 1 shard move per node. As we know, a shard
move consists of a shard moving from a source node to a target node. The
hard dependency was used because the background task runner didn't have
an option to limit the parallel shard moves per node.
With the motivation of controlling the number of concurrent shard
moves that involve a particular node, either as source or target, this
PR introduces a general new GUC
citus.max_background_task_executors_per_node to be used in the
background task runner infrastructure. So, why do we even want to
control and limit the concurrency? Well, it's all about resource
availability: because the moves involve the same nodes, extra
parallelism won’t make the rebalance complete faster if some resource is
already maxed out (usually cpu or disk). Or, if the cluster is being
used in a production setting, the moves might compete for resources with
production queries much more than if they had been executed
sequentially.
### How does it work?
A new column named nodes_involved is added to the catalog table that
keeps track of the scheduled background tasks,
pg_dist_background_task. It is of type integer[] - to store a list
of node ids. It is NULL by default - the column will be filled by the
rebalancer, but we may not care about the nodes involved in other uses
of the background task runner.
Table "pg_catalog.pg_dist_background_task"
Column | Type
============================================
job_id | bigint
task_id | bigint
owner | regrole
pid | integer
status | citus_task_status
command | text
retry_count | integer
not_before | timestamp with time zone
message | text
+nodes_involved | integer[]
A hashtable named ParallelTasksPerNode keeps track of the number of
parallel running background tasks per node. An entry in the hashtable is
as follows:
ParallelTasksPerNodeEntry
{
node_id // The node is used as the hash table key
counter // Number of concurrent background tasks that involve node node_id
// The counter limit is citus.max_background_task_executors_per_node
}
When the background task runner assigns a runnable task to a new
executor, it increments the counter for each of the nodes involved with
that runnable task. The limit of each counter is
citus.max_background_task_executors_per_node. If the limit is reached
for any of the nodes involved, this runnable task is skipped. And then,
later, when the running task finishes, the background task runner
decrements the counter for each of the nodes involved with the done
task. The following functions take care of these increment-decrement
steps:
IncrementParallelTaskCountForNodesInvolved(task)
DecrementParallelTaskCountForNodesInvolved(task)
citus.max_background_task_executors_per_node can be changed in the
fly. In the background rebalancer, we simply give {source_node,
target_node} as the nodesInvolved input to the
ScheduleBackgroundTask function. The rest is taken care of by the
general background task runner infrastructure explained above. Check
background_task_queue_monitor.sql and
background_rebalance_parallel.sql tests for detailed examples.
#### Note
This PR also adds a hard node dependency if a node is first being used
as a source for a move, and then later as a target. The reason this
should be a hard dependency is that the first move might make space for
the second move. So, we could run out of disk space (or at least
overload the node) if we move the second shard to it before the first
one is moved away.
Fixes https://github.com/citusdata/citus/issues/6716
DESCRIPTION: Adds views that monitor statistics on tenant usages
This PR adds `citus_stats_tenants` view that monitors the tenants on the
cluster.
`citus_stats_tenants` shows the node id, colocation id, tenant
attribute, read count in this period and last period, and query count in
this period and last period of the tenant.
Tenant attribute currently is the tenant's distribution column value,
later when schema based sharding is introduced, this meaning might
change.
A period is a time bucket the queries are counted by. Read and query
counts for this period can increase until the current period ends. After
that those counts are moved to last period's counts, which cannot
change. The period length can be set using 'citus.stats_tenants_period'.
`SELECT` queries are counted as _read_ queries, `INSERT`, `UPDATE` and
`DELETE` queries are counted as _write_ queries. So in the view read
counts are `SELECT` counts and query counts are `SELECT`, `INSERT`,
`UPDATE` and `DELETE` count.
The data is stored in shared memory, in a struct named
`MultiTenantMonitor`.
`citus_stats_tenants` shows the data from local tenants.
`citus_stats_tenants` show up to `citus.stats_tenant_limit` number of
tenants.
The tenants are scored based on the number of queries they run and the
recency of those queries. Every query ran increases the score of tenant
by `ONE_QUERY_SCORE`, and after every period ends the scores are halved.
Halving is done lazily.
To retain information a longer the monitor keeps up to 3 times
`citus.stats_tenant_limit` tenants. When the tenant count hits `3 *
citus.stats_tenant_limit`, last `citus.stats_tenant_limit` tenants are
removed. To see all stored tenants you can use
`citus_stats_tenants(return_all_tenants := true)`
- [x] Create collector view that gets data from all nodes. #6761
- [x] Add monitoring log #6762
- [x] Create enable/disable GUC #6769
- [x] Parse the annotation string correctly #6796
- [x] Add local queries and prepared statements #6797
- [x] Rename to citus_stat_statements #6821
- [x] Run pgbench
- [x] Fix role permissions #6812
---------
Co-authored-by: Gokhan Gulbiz <ggulbiz@gmail.com>
Co-authored-by: Jelte Fennema <github-tech@jeltef.nl>
DESCRIPTION:
- The CDC decoder is refacroted into a seperate extension that can be used loaded dynamically without having to reload citus.
- CDC decoder code can be compiled using DECODER flag to work with different decoders like pgoutput and wal2json.
by default the base decode is "pgoutput".
- the dynamic_library_path config is adjusted dynamically to prefer the decoders in cdc_decoders directory in citus init
so that the users can use the replication subscription commands without having to make any config changes.
DESCRIPTION: Refactor and unify shard move and copy functions
Shard move and copy functions share a lot of code in common. This PR
unifies these functions into one, along with some helper functions. To
preserve the current behavior, we'll introduce and use an enum
parameter, and hardcoded strings for producing error/warning messages.
Add new metadata sync methods which uses MemorySyncContext api so that during the sync we can
- free memory to prevent OOM,
- use either transactional or nontransactional modes according to the GUC .
- Create MetadataSyncContext api to encapsulate
both transactional and nontransactional modes,
- Add a GUC to switch between metadata sync transaction modes.
This pull request proposes a change to the logic used for propagating
identity columns to worker nodes in citus. Instead of creating a
dependent sequence for each identity column and changing its default
value to `nextval(seq)/worker_nextval(seq)`, this update will pass the
identity columns as-is to the worker nodes.
Please note that there are a few limitations to this change.
1. Only bigint identity columns will be allowed in distributed tables to
ensure compatibility with the DDL from any node functionality. Our
current distributed sequence implementation only allows insert
statements from all nodes for bigint sequences.
2. `alter_distributed_table` and `undistribute_table` operations will
not be allowed for tables with identity columns. This is because we do
not have a proper way of keeping sequence states consistent across the
cluster.
DESCRIPTION: Prevents using identity columns on data types other than
`bigint` on distributed tables
DESCRIPTION: Prevents using `alter_distributed_table` and
`undistribute_table` UDFs when a table has identity columns
DESCRIPTION: Fixes a bug that prevents enforcing identity column
restrictions on worker nodes
Depends on #6740Fixes#6694
Description:
Implementing CDC changes using Logical Replication to avoid
re-publishing events multiple times by setting up replication origin
session, which will add "DoNotReplicateId" to every WAL entry.
- shard splits
- shard moves
- create distributed table
- undistribute table
- alter distributed tables (for some cases)
- reference table operations
The citus decoder which will be decoding WAL events for CDC clients,
ignores any WAL entry with replication origin that is not zero.
It also maps the shard names to distributed table names.
Today we allow planning the queries that reference non-colocated tables
if the shards that query targets are placed on the same node. However,
this may not be the case, e.g., after rebalancing shards because it's
not guaranteed to have those shards on the same node anymore.
This commit adds citus.enable_non_colocated_router_query_pushdown GUC
that can be used to disallow planning such queries via router planner,
when it's set to false. Note that the default value for this GUC will be
"true" for 11.3, but we will alter it to "false" on 12.0 to not
introduce
a breaking change in a minor release.
Closes#692.
Even more, allowing such queries to go through router planner also
causes
generating an incorrect plan for the DML queries that reference
distributed
tables that are sharded based on different replication factor settings.
For
this reason, #6779 can be closed after altering the default value for
this
GUC to "false", hence not now.
DESCRIPTION: Adds `citus.enable_non_colocated_router_query_pushdown` GUC
to ensure generating a consistent distributed plan for the queries that
reference non-colocated distributed tables (when set to "false", the
default is "true").
Because they're only interested in distributed tables. Even more,
this replaces HasDistributionKey() check with
IsCitusTableType(DISTRIBUTED_TABLE) because this doesn't make a
difference on main and sounds slightly more intuitive. Plus, this
would also allow safely using this function in
https://github.com/citusdata/citus/pull/6773.
Fixes#6672
2) Move all MERGE related routines to a new file merge_planner.c
3) Make ConjunctionContainsColumnFilter() static again, and rearrange the code in MergeQuerySupported()
4) Restore the original format in the comments section.
5) Add big serial test. Implement latest set of comments
This implements the phase - II of MERGE sql support
Support routable query where all the tables in the merge-sql are distributed, co-located, and both the source and
target relations are joined on the distribution column with a constant qual. This should be a Citus single-task
query. Below is an example.
SELECT create_distributed_table('t1', 'id');
SELECT create_distributed_table('s1', 'id', colocate_with => ‘t1’);
MERGE INTO t1
USING s1 ON t1.id = s1.id AND t1.id = 100
WHEN MATCHED THEN
UPDATE SET val = s1.val + 10
WHEN MATCHED THEN
DELETE
WHEN NOT MATCHED THEN
INSERT (id, val, src) VALUES (s1.id, s1.val, s1.src)
Basically, MERGE checks to see if
There are a minimum of two distributed tables (source and a target).
All the distributed tables are indeed colocated.
MERGE relations are joined on the distribution column
MERGE .. USING .. ON target.dist_key = source.dist_key
The query should touch only a single shard i.e. JOIN AND with a constant qual
MERGE .. USING .. ON target.dist_key = source.dist_key AND target.dist_key = <>
If any of the conditions are not met, it raises an exception.
(cherry picked from commit 44c387b978)
This implements MERGE phase3
Support pushdown query where all the tables in the merge-sql are Citus-distributed, co-located, and both
the source and target relations are joined on the distribution column. This will generate multiple tasks
which execute independently after pushdown.
SELECT create_distributed_table('t1', 'id');
SELECT create_distributed_table('s1', 'id', colocate_with => ‘t1’);
MERGE INTO t1
USING s1
ON t1.id = s1.id
WHEN MATCHED THEN
UPDATE SET val = s1.val + 10
WHEN MATCHED THEN
DELETE
WHEN NOT MATCHED THEN
INSERT (id, val, src) VALUES (s1.id, s1.val, s1.src)
*The only exception for both the phases II and III is, UPDATEs and INSERTs must be done on the same shard-group
as the joined key; for example, below scenarios are NOT supported as the key-value to be inserted/updated is not
guaranteed to be on the same node as the id distribution-column.
MERGE INTO target t
USING source s ON (t.customer_id = s.customer_id)
WHEN NOT MATCHED THEN - -
INSERT(customer_id, …) VALUES (<non-local-constant-key-value>, ……);
OR this scenario where we update the distribution column itself
MERGE INTO target t
USING source s On (t.customer_id = s.customer_id)
WHEN MATCHED THEN
UPDATE SET customer_id = 100;
(cherry picked from commit fa7b8949a8)
Now that we will soon add another table type having DISTRIBUTE_BY_NONE
as distribution method and that we want the code to interpret such
tables mostly as distributed tables, let's make the definition of those
other two table types more strict by removing
CITUS_TABLE_WITH_NO_DIST_KEY
macro.
And instead, use HasDistributionKey() check in the places where the
logic applies to all table types that have / don't have a distribution
key. In future PRs, we might want to convert some of those
HasDistributionKey() checks if logic only applies to Citus local /
reference tables, not the others.
And adding HasDistributionKey() also allows us to consider having
DISTRIBUTE_BY_NONE as the distribution method as a "table attribute"
that can apply to distributed tables too, rather something that
determines the table type.
Split the main logic that allows creating a Citus table into the
internal function CreateCitusTable().
Old CreateDistributedTable() function was assuming that it's creating
a reference table when the distribution method is DISTRIBUTE_BY_NONE.
However, soon this won't be the case when adding support for creating
single-shard distributed tables because their distribution method would
also be the same.
Now the internal method CreateCitusTable() doesn't make any assumptions
about table's replication model or such. Instead, it expects callers to
properly set all such metadata bits.
Even more, some of the parameters the old CreateDistributedTable() takes
--such as the shard count-- were not meaningful for a reference table,
and would be the same as for new table type.
DESCRIPTION: Fixes a bug in shard copy operations.
For copying shards in both shard move and shard split operations, Citus
uses the COPY statement.
A COPY all statement in the following form
` COPY target_shard FROM STDIN;`
throws an error when there is a GENERATED column in the shard table.
In order to fix this issue, we need to exclude the GENERATED columns in
the COPY and the matching SELECT statements. Hence this fix converts the
COPY and SELECT all statements to the following form:
```
COPY target_shard (col1, col2, ..., coln) FROM STDIN;
SELECT (col1, col2, ..., coln) FROM source_shard;
```
where (col1, col2, ..., coln) does not include a GENERATED column.
GENERATED column values are created in the target_shard as the values
are inserted.
Fixes#6705.
---------
Co-authored-by: Teja Mupparti <temuppar@microsoft.com>
Co-authored-by: aykut-bozkurt <51649454+aykut-bozkurt@users.noreply.github.com>
Co-authored-by: Jelte Fennema <jelte.fennema@microsoft.com>
Co-authored-by: Gürkan İndibay <gindibay@microsoft.com>
Postgres got minor updates this starts using the images with the latest
version for our tests.
These new Postgres versions caused a compilation issue in PG14 and PG13
due to some function being backported that we had already backported
ourselves. Due this backport being a static inline function it doesn't
matter who provides this and there will be no linkage errors when either
running old Citus packages on new PG versions or the other way around.
In #6314 I refactored the connection cleanup to be simpler to
understand and use. However, by doing so I introduced a use-after-free
possibility (that valgrind luckily picked up):
In the `ShouldShutdownConnection` path of
`AfterXactHostConnectionHandling`
we free connections without removing the `transactionNode` from the
dlist that it might be part of. Before the refactoring this wasn't a
problem, because the dlist would be completely reset quickly after in
`ResetGlobalVariables` (without reading or writing the dlist entries).
The refactoring changed this by moving the `dlist_delete` call to
`ResetRemoteTransaction`, which in turn was called in the
`!ShouldShutdownConnection` path of `AfterXactHostConnectionHandling`.
Thus this `!ShouldShutdownConnection` path would now delete from the
`dlist`, but the `ShouldShutdownConnection` path would not. Thus to
remove itself the deleting path would sometimes update nodes in the list
that were freed right before.
There's two ways of fixing this:
1. Call `dlist_delete` from **both** of paths.
2. Call `dlist_delete` from **neither** of the paths.
This commit implements the second approach, and #6684 implements the
first. We need to choose which approach we prefer.
To make calling `dlist_delete` from both paths actually work, we also need
to use a slightly different check to determine if we need to call dlist_delete.
Various regression tests showed that there can be cases where the
`transactionState` is something else than `REMOTE_TRANS_NOT_STARTED`
but the connection was not added to the `InProgressTransactions` list
One example of such a case is when running `TransactionStateMachine`
without calling `StartRemoteTransactionBegin` beforehand. In those
cases the connection won't be added to `InProgressTransactions`, but
the `transactionState` is changed to `REMOTE_TRANS_SENT_COMMAND`.
Sidenote: This bug already existed in 11.1, but valgrind didn't catch it
back then. My guess is that this happened because #6314 was merged after
the initial release branch was cut.
Fixes#6638
We should disallow dropping table_name option if foreign table is in
metadata. Otherwise, we get table not found error which contains
shardid.
DESCRIPTION: Fixes an unexpected foreign table error by disallowing to drop the table_name option.
Fixes#6663
Recursive planner should handle all the tree from bottom to top at
single pass. i.e. It should have already recursively planned all
required parts in its first pass. Otherwise, this means we have bug at
recursive planner, which needs to be handled. We add a check here and
return error.
DESCRIPTION: Fixes wrong results by throwing error in case recursive
planner multipass the query.
We found 3 different cases which causes recursive planner passes the
query multiple times.
1. Sublink in WHERE clause is planned at second pass after we
recursively planned a distributed table at the first pass. Fixed by PR
#6657.
2. Local-distributed joins are recursively planned at both the first and
the second pass. Issue #6659.
3. Some parts of the query is considered to be noncolocated at the
second pass as we do not generate attribute equivalances between
nondistributed and distributed tables. Issue #6653
DESCRIPTION: Fix foreign key validation skip at the end of shard move
In eadc88a we started completely skipping foreign key constraint
validation at the end of a non blocking shard move, instead of only for
foreign keys to reference tables. However, it turns out that this didn't
work at all because of a hard to notice bug: By resetting the
SkipConstraintValidation flag at the end of our utility hook, we
actually make the SET command that sets it a no-op.
This fixes that bug by removing the code that resets it. This is fine
because #6543 removed the only place where we set the flag in C code. So
the resetting of the flag has no purpose anymore. This PR also adds a
regression test, because it turned out we didn't have any otherwise we
would have caught that the feature was completely broken.
It also moves the constraint validation skipping to the utility hook.
The reason is that #6550 showed us that this is the better place to skip
it, because it will also skip the planning phase and not just the
execution.
DESCRIPTION: Fix regression in allowed foreign keys on distributed
tables
In commit eadc88a we changed how we skip foreign key validation. The
goal was to skip it in more cases. However, one change had the
unintended regression of introducing failures when trying to create
certain foreign keys. This reverts that part of the change.
The way of skipping validation of foreign keys that was introduced in
eadc88a was skipping validation during execution. The reason that
this caused this regression was because some foreign key validation
queries already fail during planning. In those cases it never gets to
the execution step where it would later be skipped.
Fixes#6543
This implements the phase - II of MERGE sql support
Support routable query where all the tables in the merge-sql are distributed, co-located, and both the source and
target relations are joined on the distribution column with a constant qual. This should be a Citus single-task
query. Below is an example.
SELECT create_distributed_table('t1', 'id');
SELECT create_distributed_table('s1', 'id', colocate_with => ‘t1’);
MERGE INTO t1
USING s1 ON t1.id = s1.id AND t1.id = 100
WHEN MATCHED THEN
UPDATE SET val = s1.val + 10
WHEN MATCHED THEN
DELETE
WHEN NOT MATCHED THEN
INSERT (id, val, src) VALUES (s1.id, s1.val, s1.src)
Basically, MERGE checks to see if
There are a minimum of two distributed tables (source and a target).
All the distributed tables are indeed colocated.
MERGE relations are joined on the distribution column
MERGE .. USING .. ON target.dist_key = source.dist_key
The query should touch only a single shard i.e. JOIN AND with a constant qual
MERGE .. USING .. ON target.dist_key = source.dist_key AND target.dist_key = <>
If any of the conditions are not met, it raises an exception.
The original implementation of GPIDs didn't work correctly when using
`pg_dist_poolinfo` together with PgBouncer. The reason is that it
assumed that once a connection was made to a worker, the originating
GPID should stay the same for ever. But when pg_dist_poolinfo is used
this isn't the case, because the same connection on the worker might be
used by different backends of the coordinator.
This fixes that issue by updating the GPID whenever a new application
name is set on a connection. This is the only thing that's needed,
because PgBouncer already sets the application name correctly on the
server connection whenever a client is updated.
DESCRIPTION: Enable adding CHECK constraints on distributed tables
without the client having to provide a constraint name.
This PR enables the following command syntax for adding check
constraints to distributed tables.
ALTER TABLE ... ADD CHECK ...
by creating a default constraint name and transforming the command into
the below syntax before sending it to workers.
ALTER TABLE ... ADD CONSTRAINT \<conname> CHECK ...
In #6412 I made a change to not re-assign the global PID if it was
already set. This inadvertently introduced a regression where `userId`
and `databaseId` would not be set on the backend data when the global
PID was assigned in the authentication hook.
This fixes it by doing two things:
1. Removing `userId` from `BackendData`, since it's not used anywhere
anyway.
2. Move assignment of `databaseId` to dedicated
`SetBackendDataDatabaseId` function, that isn't a no-op when global
pid is already set.
Since #6412 is not released yet this does not need a description.
DESCRIPTION: Adds support for creating table constraints UNIQUE and
EXCLUDE via ALTER TABLE command without client having to specify a name.
ALTER TABLE ... ADD CONSTRAINT <conname> UNIQUE ...
ALTER TABLE ... ADD CONSTRAINT <conname> EXCLUDE ...
commands require the client to provide an explicit constraint name.
However, in postgres it is possible for clients not to provide a name
and let the postgres generate it using the following commands
ALTER TABLE ... ADD UNIQUE ...
ALTER TABLE ... ADD EXCLUDE ...
This PR enables the same functionality for citus tables.
DESCRIPTION: Drop `SHARD_STATE_TO_DELETE` and use the cleanup records
instead
Drops the shard state that is used to mark shards as orphaned. Now we
insert cleanup records into `pg_dist_cleanup` so "orphaned" shards will
be dropped either by maintenance daemon or internal cleanup calls. With
this PR, we make the "cleanup orphaned shards" functions to be no-op, as
they would not be needed anymore.
This PR includes some naming changes about placement functions. We don't
need functions that filter orphaned shards, as there will be no orphaned
shards anymore.
We will also be introducing a small script with this PR, for users with
orphaned shards. We'll basically delete the orphaned shard entries from
`pg_dist_placement` and insert cleanup records into `pg_dist_cleanup`
for each one of them, during Citus upgrade.
We also have a lot of flakiness fixes in this PR.
Co-authored-by: Jelte Fennema <github-tech@jeltef.nl>
DESCRIPTION: Defers cleanup after a failure in shard move or split
We don't need to do a cleanup in case of failure on a shard transfer or
split anymore. Because,
* Maintenance daemon will clean them up anyway.
* We trigger a cleanup at the beginning of shard transfers/splits.
* The cleanup on failure logic also can fail sometimes and instead of
the original error, we throw the error that is raised by the cleanup
procedure, and it causes confusion.
DESCRIPTION: Cleanup the shard on the target node in case of a
failed/aborted shard move
Inserts a cleanup record for the moved shard placement on the target
node. If the move operation succeeds, the record will be deleted. If
not, it will remain there to be cleaned up later.
fixes: #6580
All the tables (target, source or any CTE present) in the SQL statement are local i.e. a merge-sql with a combination of Citus local and
Non-Citus tables (regular Postgres tables) should work and give the same result as Postgres MERGE on regular tables. Catch and throw an
exception (not-yet-supported) for all other scenarios during Citus-planning phase.
Removes unused job boundary tag `SUBQUERY_MAP_MERGE_JOB`.
Only usage is at `BuildMapMergeJob`, which is only called when the
boundary = `JOIN_MAP_MERGE_JOB`. Hence, it should be safe to remove.
We already have citus_job_wait to wait until the job reaches the desired
state. That PR adds waiting on task state to allow more granular
waiting. It can be used for Citus operations. Moreover, it is also
useful for testing purposes. (wait until a task reaches specified state)
Related to #6459.
DESCRIPTION: Create replication artifacts with unique names
We're creating replication objects with generic names. This disallows us
to enable parallel shard moves, as two operations might use the same
objects. With this PR, we'll create below objects with operation
specific names, by appending OparationId to the names.
* Subscriptions
* Publications
* Replication Slots
* Users created for subscriptions
Adds signal handlers for graceful termination, cancellation of
task executors and detecting config updates. Related to PR #6459.
#### How to handle termination signal?
Monitor need to gracefully terminate all running task executors before
terminating. Hence, we have sigterm handler for the monitor.
#### How to handle cancellation signal?
Monitor need to gracefully cancel all running task executors before
terminating. Hence, we have sigint handler for the monitor.
#### How to detect configuration changes?
Monitor has SIGHUP handler to reflect configuration changes while
executing tasks.
DESCRIPTION: Extend cleanup process for replication artifacts
This PR adds new cleanup record types for:
* Subscriptions
* Replication slots
* Publications
* Users created for subscriptions
We add records for these object types, to `pg_dist_cleanup` during
creation phase. Once the operation is done, in case of success or
failure, we iterate those records and drop the objects. With this PR we
will not be dropping any of these objects during the operation. In
short, we will always be deferring the drop.
One thing that's worth mentioning is that we sort cleanup records before
processing (dropping) them, because of dependency relations among those
objects, e.g a subscription might depend on a publication. Therefore, we
always drop subscriptions before publications.
We have some renames in this PR:
* `TryDropOrphanedShards` -> `TryDropOrphanedResources`
* `DropOrphanedShardsForCleanup` -> `DropOrphanedResourcesForCleanup`
* `run_try_drop_marked_shards` -> `run_try_drop_marked_resources`
as these functions now process replication artifacts as well.
This PR drops function `DropAllLogicalReplicationLeftovers` and its all
usages, since now we rely on the deferring drop mechanism.
Improvement on our background task monitoring API (PR #6296) to support
concurrent and nonblocking task execution.
Mainly we have a queue monitor background process which forks task
executors for `Runnable` tasks and then monitors their status by
fetching messages from shared memory queue in nonblocking way.
When debugging issues it's quite useful to see the originating gpid in
the application_name of a query on a worker. This already happens for
most queries, but not for queries created by the rebalancer or by
run_command_on_worker. This adds a gpid to those two application_names
too.
Note, that if the GPID of the new application_names is different than
the current GPID of the backend the backend will continue to keep
the old gpid as its actual GPID. This PR is just meant to make sure
that the application_name is as useful as it can be for users to
look at. Updating of gpids will be done in a follow-up PR, and
adding gpids to all internal connections will make this easier.
DESCRIPTION: Makes sure to disallow triggers that depend on extensions
We were already doing so for `ALTER trigger DEPENDS ON EXTENSION`
commands. However, we also need to disallow creating Citus tables
having such triggers already, so this PR fixes that.
increasing logical clock. Clock guarantees to never go back in value after restarts,
and makes best attempt to keep the value close to unix epoch time in milliseconds.
Also, introduces a new GUC "citus.enable_cluster_clock", when true, every
distributed transaction is stamped with logical causal clock and persisted
in a catalog pg_dist_commit_transaction.
DESCRIPTION: Drops GUC defer_drop_after_shard_split
DESCRIPTION: Drops GUC defer_drop_after_shard_move
Drop GUCs and related parts from the code.
Delete tests that specifically added for the GUCs.
Keep tests that can be used without the GUCs.
Update test output changes.
The motivation for this PR is to have an "always deferring" mechanism.
These two GUCs provide an option to not deferring dropping objects
during a shard move/split, and dropping them immediately. With this PR,
we will be always deferring dropping orphaned shards and other types of
objects.
We will have a separate PR to extend the deferred cleanup operation, so
that we would create records for deferred drop, for Subscriptions,
Publications, Replication Slots etc. This will make us be able to keep
track of created objects that needs to be dropped, during a shard
move/split. We will have objects created specifically for the current
operation; and those objects will be dropped at the end.
We have an issue (a draft roadmap) for enabling parallel shard moves.
For details please see: https://github.com/citusdata/citus/issues/6437
DESCRIPTION: Adds failure test for shard move
DESCRIPTION: Remove function `WaitForAllSubscriptionsToBecomeReady` and
related tests
Adding some failure tests for shard moves.
Dropping the not-needed-anymore function
`WaitForAllSubscriptionsToBecomeReady`, as the subscriptions now start
as ready from the beginning because we don't use logical replication
table sync workers anymore.
fixes: #6260
To be able to test non-blocking shard moves we take an advisory lock, so
we can pause the shard move at an interesting moment. Originally this
was during the logical replication catch up phase. But when I added
tests for the rebalancer progress I moved this lock before the initial
data copy. This allowed testing of the rebalance progress, but
inadvertently made our non-blocking tests not actually test if we held
unintended locks during logical replication catch up.
This fixes that by creating two types of advisory locks, one before the
copy and one after. This causes the tests to actually test their
intended scenario again.
Furthermore it starts using one of these locks for blocking shard moves
too. Which allowed me to reduce the complexity of the rebalance progress
test suite quite a bit. It also allowed enabling some flaky tests again,
because this stopped them from being flaky. And finally it allowed
testing of rebalance progress for blocking shard copy operations as
well.
In passing it fixes a flaky test during parallel blocking shard moves by
ordering the output.
DESCRIPTION: Adds status column to get_rebalance_progress()
Introduces a new column named `status` for the function
`get_rebalance_progress()`. For each ongoing shard move, this column
will reveal information about that shard move operation's current
status.
For now, candidate status messages could be one of the below.
* Not Started
* Setting Up
* Copying Data
* Catching Up
* Creating Constraints
* Final Catchup
* Creating Foreign Keys
* Completing
* Completed
PostgreSQL 15 exposes WL_SOCKET_CLOSED in WaitEventSet API, which is
useful for detecting closed remote sockets. In this patch, we use this
new event and try to detect closed remote sockets in the executor.
When a closed socket is detected, the executor now has the ability to
retry the connection establishment. Note that, the executor can retry
connection establishments only for the connection that has not been
used. Basically, this patch is mostly useful for preventing the executor
to fail if a cached connection is closed because of the worker node
restart (or worker failover).
In other words, the executor cannot retry connection establishment if we
are in a distributed transaction AND any command has been sent over the
connection. That requires more sophisticated retry mechanisms. For now,
fixing the above use case is enough.
Fixes#5538
Earlier discussions: #5908, #6259 and #6283
### Summary of the current approach regards to earlier trials
As noted, we explored some alternatives before getting into this.
https://github.com/citusdata/citus/pull/6283 is simple, but lacks an
important property. We should be checking for `WL_SOCKET_CLOSED`
_before_ sending anything over the wire. Otherwise, it becomes very
tricky to understand which connection is actually safe to retry. For
example, in the current patch, we can safely check
`transaction->transactionState == REMOTE_TRANS_NOT_STARTED` before
restarting a connection.
#6259 does what we intent here (e.g., check for sending any command).
However, as @marcocitus noted, it is very tricky to handle
`WaitEventSets` in multiple places. And, the executor is designed such
that it reacts to the events. So, adding anything `pre-executor` seemed
too ugly.
In the end, I converged into this patch. This patch relies on the
simplicity of #6283 and also does a very limited handling of
`WaitEventSets`, just for our purpose. Just before we add any connection
to the execution, we check if the remote session has already closed.
With that, we do a brief interaction of multiple wait event processing,
but with different purposes. The new wait event processing we added does
not even consider cancellations. We let that handled by the main event
processing loop.
Co-authored-by: Marco Slot <marco.slot@gmail.com>
PG15 introduced a function called ReplicationSlotName that causes
conflicts with our function with the same name. I solved this issue by
renaming our function to ReplicationSlotNameForNodeAndOwner
Relevant PG commit:
c3b5992b91
DESCRIPTION: Fixes a bug that prevents retaining columnar table options after a table-rewrite
A fix for this issue: Columnar: options ignored during ALTER TABLE
rewrite #5927
The OID for the temporary table created during ALTER TABLE was not the
same as the original table's OID so the columnar options were not being
applied during rewrite.
The change is that I applied the original table's columnar options to
the new table so that it has the correct options during write. I also
added a test.
In Split, Logical replication logic and ShardCleaner we call
`SendCommandListToWorkerOutsideTransaction` and
`SendOptionalCommandListToWorkerOutsideTransaction` frequently. This
opens new connection for each of those calls, even though we already
have a perfectly good connection lying around.
This PR adds two new APIs
`SendCommandListToWorkerOutsideTransactionWithConnection` and
`SendOptionalCommandListToWorkerOutsideTransactionWithConnection` that
allow sending a list of queries in a transaction over an existing
connection. We also update the callers (Split, ShardCleaner, Logical
Replication) to use these new APIs instead.
Co-authored-by: Nitish Upreti <niupre@microsoft.com>
Co-authored-by: Onder Kalaci <onderkalaci@gmail.com>
As we did for GENERATED STORED columns in #4613, we should not drop
column
default expressions that are not based on sequences from shard relation
since
such expressions need to exist e.g. for foreign key actions.
For the column default expressions that are based on sequences we cannot
do much, so we need to disallow having ON DELETE SET DEFAULT actions on
such columns in a separate PR, see #6339.
Fixes#6318.
DESCRIPTION: Fixes a bug that might cause inserting incorrect DEFAULT
values when applying foreign key actions
DESCRIPTION: Improve logging during shard split and resource cleanup
### DESCRIPTION
This PR makes logging improvements to Shard Split :
1. Update confusing logging to fix#6312
2. Added new `ereport(LOG` to make debugging easier as part of telemetry review.
Comment from the code is clear on this:
/*
* The statistics objects of the distributed table are not relevant
* for the distributed planning, so we can override it.
*
* Normally, we should not need this. However, the combination of
* Postgres commit 269b532aef55a579ae02a3e8e8df14101570dfd9 and
* Citus function AdjustPartitioningForDistributedPlanning()
* forces us to do this. The commit expects statistics objects
* of partitions to have "inh" flag set properly. Whereas, the
* function overrides "inh" flag. To avoid Postgres to throw error,
* we override statlist such that Postgres does not try to process
* any statistics objects during the standard_planner() on the
* coordinator. In the end, we do not need the standard_planner()
* on the coordinator to generate an optimized plan. We call
* into standard_planner() for other purposes, such as generating the
* relationRestrictionContext here.
*
* AdjustPartitioningForDistributedPlanning() is a hack that we use
* to prevent Postgres' standard_planner() to expand all the partitions
* for the distributed planning when a distributed partitioned table
* is queried. It is required for both correctness and performance
* reasons. Although we can eliminate the use of the function for
* the correctness (e.g., make sure that rest of the planner can handle
* partitions), it's performance implication is hard to avoid. Certain
* planning logic of Citus (such as router or query pushdown) relies
* heavily on the relationRestrictionList. If
* AdjustPartitioningForDistributedPlanning() is removed, all the
* partitions show up in the, causing high planning times for
* such queries.
*/
DESCRIPTION: Show citus_copy_shard_placement progress in
get_rebalance_progress
When rebalancing to a new node that does not have reference tables yet
the rebalancer will first copy the reference tables to the nodes.
Depending on the size of the reference tables, this might take a long
time. However, there's no indication of what's happening at this stage
of the rebalance.
This PR improves this situation by also showing the progress of any
citus_copy_shard_placement calls when calling get_rebalance_progress.
We can now do the following:
- Distribute sequence with logged/unlogged option
- ALTER TABLE my_sequence SET LOGGED/UNLOGGED
- ALTER SEQUENCE my_sequence SET LOGGED/UNLOGGED
Relevant PG commit
344d62fb9a
DESCRIPTION: Add a rebalancer that uses background tasks for its
execution
Based on the baclground jobs and tasks introduced in #6296 we implement
a new rebalancer on top of the primitives of background execution. This
allows the user to initiate a rebalance and let Citus execute the long
running steps in the background until completion.
Users can invoke the new background rebalancer with `SELECT
citus_rebalance_start();`. It will output information on its job id and
how to track progress. Also it returns its job id for automation
purposes. If you simply want to wait till the rebalance is done you can
use `SELECT citus_rebalance_wait();`
A running rebalance can be canelled/stopped with `SELECT
citus_rebalance_stop();`.
The logical replication catchup part for shard splits and shard moves is
very similar. This abstracts most of that similarity away into a single
function. This also improves the logic for non blocking shard splits a
bit, by using faster foreign key creation. It also parallelizes index creation
which shard moves were already doing, but shard splits did not.
DESCRIPTION: Add infrastructure to run long running management operations in background
This infrastructure introduces the primitives of jobs and tasks.
A task consists of a sql statement and an owner. Tasks belong to a
Job and can depend on other tasks from the same job.
When there are either runnable or running tasks we would like to
make sure a bacgrkound task queue monitor process is running. A Task
could be in running state while there is actually no monitor present
due to a database restart or failover. Once the monitor starts it
will reset any running task to its runnable state.
To make sure only one background task queue monitor is ever running
at once it will acquire an advisory lock that self conflicts.
Once a task is done it will find all tasks depending on this task.
After checking that the task doesn't have unmet dependencies it will
transition the task from blocked to runnable state for the task to
be picked up on a subsequent task start.
Currently only one task can be running at a time. This can be
improved upon in later releases without changes to the higher level
API.
The initial goal for this background tasks is to allow a rebalance
to run in the background. This will be implemented in a subsequent PR.
Halil Ozan Akgül
Ahmet Gedemenli
Onur Tirtir
Naisila Puka
Teja Mupparti
aykut-bozkurt
rajeshkt78
Marco Slot
aykutbozkurt
Gokhan Gulbiz
Marco Slot
Emel Şimşek
Jelte Fennema
Önder Kalacı
songjinzhou
Hanefi Onaldi
Ying Xu
Nitish Upreti
Jelte Fennema
Nils Dijk
naisila