When multiple sessions concurrently attempt to add the same coordinator
node using `citus_set_coordinator_host`, there is a potential race
condition. Both sessions may pass the initial metadata check
(`isCoordinatorInMetadata`), but only one will succeed in adding the
node. The other session will fail with an assertion error
(`Assert(!nodeAlreadyExists)`), causing the server to crash. Even though
the `AddNodeMetadata` function takes an exclusive lock, it appears that
the lock is not preventing the race condition before the initial
metadata check.
- **Issue**: The current logic allows concurrent sessions to pass the
check for existing coordinators, leading to an attempt to insert
duplicate nodes, which triggers the assertion failure.
- **Impact**: This race condition leads to crashes during operations
that involve concurrent coordinator additions, as seen in
https://github.com/citusdata/citus/issues/7646.
**Test Plan:**
- Isolation Test Limitation: An isolation test was added to simulate
concurrent additions of the same coordinator node, but due to the
behavior of PostgreSQL locking mechanisms, the test does not trigger the
edge case. The lock applied within the function serializes the
operations, preventing the race condition from occurring in the
isolation test environment.
While the edge case is difficult to reproduce in an isolation test, the
fix addresses the core issue by ensuring concurrency control through
proper locking.
- Existing Tests: All existing tests related to node metadata and
coordinator management have been run to ensure that no regressions were
introduced.
**After the Fix:**
- Concurrent attempts to add the same coordinator node will be
serialized. One session will succeed in adding the node, while the
others will skip the operation without crashing the server.
Co-authored-by: Mehmet YILMAZ <mehmet.yilmaz@microsoft.com>
(cherry picked from commit 4775715691)
This is prep work for successful compilation with PG17
PG17added foreach_ptr, foreach_int and foreach_oid macros
Relevant PG commit
14dd0f27d7cd56ffae9ecdbe324965073d01a9ff
14dd0f27d7
We already have these macros, but they are different with the
PG17 ones because our macros take a DECLARED variable, whereas
the PG16 macros declare a locally-scoped loop variable themselves.
Hence I am renaming our macros to foreach_declared_
I am separating this into its own PR since it touches many files. The
main compilation PR is https://github.com/citusdata/citus/pull/7699
This fixes#7454: master_disable_node() has only two arguments, but
calls citus_disable_node() that tries to read three arguments
Co-authored-by: Karina Litskevich <litskevichkarina@gmail.com>
(cherry picked from commit 683e10ab69)
DESCRIPTION: Presenting citus_pause_node UDF enabling pausing by
node_id.
citus_pause_node takes a node_id parameter and fetches all the shards in
that node and puts AccessExclusiveLock on all the shards inside that
node. With this lock, insert is disabled, until citus_pause_node
transaction is closed.
---------
Co-authored-by: Hanefi Onaldi <Hanefi.Onaldi@microsoft.com>
Citus build with PG16 fails because of the following warnings:
- using char* instead of Datum
- using pointer instead of oid
- candidate function for format attribute
- remove old definition from PG11 compatibility 62bf571ced
This commit fixes the above.
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.
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>
When using multiline strings, we occasionally forget to add a single
space at the end of the first line. When this line is concatenated with
the next one, the resulting string has a missing space.
DESCRIPTION: Allow citus_update_node() to work with nodes from different clusters
citus_update_node(), citus_nodename_for_nodeid(), and citus_nodeport_for_nodeid() functions only checked for nodes in their own clusters and hence last two returned NULLs and the first one showed an error is the nodeId was from a different cluster.
Fixes https://github.com/citusdata/citus/issues/6433
pg_dist_node and pg_dist_colocation have a primary key index, not a replica identity index.
Citus catalog tables are created in public schema, which has replica identity index by default
as primary key index. Later the citus catalog tables are moved to pg_catalog schema.
During pg_upgrade, all tables are recreated, and given that pg_dist_colocation is found in
pg_catalog schema, it is recreated in that schema, and when it is recreated it doesn't
have a replica identity index, because catalog tables have no replica identity.
Further action:
Do we even need to acquire this lock on the primary key index?
Postgres doesn't acquire such locks on indexes before deleting catalog tuples.
Also, catalog tuples don't have replica identities by definition.
Added create_distributed_table_concurrently which is nonblocking variant of create_distributed_table.
It bases on the split API which takes advantage of logical replication to support nonblocking split operations.
Co-authored-by: Marco Slot <marco.slot@gmail.com>
Co-authored-by: aykutbozkurt <aykut.bozkurt1995@gmail.com>
Do not obtain AccessShareLock before acquiring the distributed locks.
Acquiring an AccessShareLock ensures that the relations which we are trying to get a distributed lock on will not be dropped in the time between when the LOCK command is issued and the LOCK commands are send to the worker. However, this also leads to distributed deadlocks in such scenarios:
```sql
-- for dist lock acquiring order coor, w1, w2
-- on w2
LOCK t1 IN ACCESS EXLUSIVE MODE;
-- acquire AccessShareLock locally on t1 to ensure it is not dropped while we get ready to distribute the lock
-- concurrently on w1
LOCK t1 IN ACCESS EXLUSIVE MODE;
-- acquire AccessShareLock locally on t1 to ensure it is not dropped while we get ready to distribute the lock
-- acquire dist lock on coor, w1, gets blocked on local AccessShareLock on w2
-- on w2 continuation of the execution above
-- starts to acquire dist locks and gets blocked on the coor by the lock acquired by w1
-- distributed deadlock
```
We opt for avoiding such deadlocks with the cost of the possibility of running into errors when the relations on which we are trying to acquire locks on get dropped.
It is often useful to be able to sync the metadata in parallel
across nodes.
Also citus_finalize_upgrade_to_citus11() uses
start_metadata_sync_to_primary_nodes() after this commit.
Note that this commit does not parallelize all pieces of node
activation or metadata syncing. Instead, it tries to parallelize
potenially large parts of metadata, which is the objects and
distributed tables (in general Citus tables).
In the future, it would be nice to sync the reference tables
in parallel across nodes.
Create ~720 distributed tables / ~23450 shards
```SQL
-- declaratively partitioned table
CREATE TABLE github_events_looooooooooooooong_name (
event_id bigint,
event_type text,
event_public boolean,
repo_id bigint,
payload jsonb,
repo jsonb,
actor jsonb,
org jsonb,
created_at timestamp
) PARTITION BY RANGE (created_at);
SELECT create_time_partitions(
table_name := 'github_events_looooooooooooooong_name',
partition_interval := '1 day',
end_at := now() + '24 months'
);
CREATE INDEX ON github_events_looooooooooooooong_name USING btree (event_id, event_type, event_public, repo_id);
SELECT create_distributed_table('github_events_looooooooooooooong_name', 'repo_id');
SET client_min_messages TO ERROR;
```
across 1 node: almost same as expected
```SQL
SELECT start_metadata_sync_to_primary_nodes();
Time: 15664.418 ms (00:15.664)
select start_metadata_sync_to_node(nodename,nodeport) from pg_dist_node;
Time: 14284.069 ms (00:14.284)
```
across 7 nodes: ~3.5x improvement
```SQL
SELECT start_metadata_sync_to_primary_nodes();
┌──────────────────────────────────────┐
│ start_metadata_sync_to_primary_nodes │
├──────────────────────────────────────┤
│ t │
└──────────────────────────────────────┘
(1 row)
Time: 25711.192 ms (00:25.711)
-- across 7 nodes
select start_metadata_sync_to_node(nodename,nodeport) from pg_dist_node;
Time: 82126.075 ms (01:22.126)
```
In the past (pre-11), we allowed removing worker nodes
that had active placements for replicated distributed
table, without even checking if there are any other
replicas of the same placement.
However, with #5469, we prevent disabling nodes via a hard
error when there is the last active placement of shard, as we
do for reference tables. Note that otherwise, we'd allow
users to lose data.
As of today, the NOTICE is completely irrelevant.
First worker node has a special meaning for modifications on the replicated tables
It is used to acquire a remote lock, such that the modifications are serialized.
With this commit, we make sure that we do not let any distributed query to see a
different 'first worker node' while first worker node is disabled.
Note that, maybe implicitly mentioned above, when first worker node is disabled,
the first worker node changes, that's why we have to handle the situation.
Before this commit, we had:
```SQL
SELECT citus_disable_node(nodename, nodeport, force boolean DEFAULT false)
```
Where, we allow forcing to disable first worker node with
`force:=true`. However, it entails the risk for losing
data / diverging placement data etc.
With `force` flag, we control disabling the first worker node,
and with `async` flag we control whether the changes are done
via bg worker or immediately.
```SQL
SELECT citus_disable_node(nodename, nodeport, force boolean DEFAULT false, sync boolean DEFAULT false)
```
Where we can achieve all the following:
| Mode | Data loss possibility | Can run in 2PC | Handle multiple node failures | Immediately effective |
| --- |--- |--- |--- |--- |
| force:false, sync: false | false | true | true | false |
| force:false, sync: true | false | false | false | true |
| force:true, sync: false | true | true | true | false |
| force:true, sync: true | false | false | false | true |
If a worker node is being added, a command is sent to get the server_id of the worker from the pg_dist_node_metadata table. If the worker's id is the same as the node executing the code, we will know the node is trying to add itself. If the node tries to add itself without specifying `groupid:=0` the operation will result in an error.
CitusInitiatedBackend was a pre-mature implemenation of the whole
GlobalPID infrastructure. We used it to track whether any individual
query is triggered by Citus or not.
As of now, after GlobalPID is already in place, we don't need
CitusInitiatedBackend, in fact it could even be wrong.
Replaces citus.enable_object_propagation with citus.enable_metadata_sync
Also, within Citus 11 release cycle, we added citus.enable_metadata_sync_by_default,
that is also replaced with citus.enable_metadata_sync.
In essence, when citus.enable_metadata_sync is set to true, all the objects
and the metadata is send to the remote node.
We strongly advice that the users never changes the value of
this GUC.
With this commit we've started to propagate sequences and shell
tables within the object dependency resolution. So, ensuring any
dependencies for any object will consider shell tables and sequences
as well. Separate logics for both shell tables and sequences have
been removed.
Since both shell tables and sequences logic were implemented as a
part of the metadata handling before that logic, we were propagating
them while syncing table metadata. With this commit we've divided
metadata (which means anything except shards thereafter) syncing
logic into multiple parts and implemented it either as a part of
ActivateNode. You can check the functions called in ActivateNode
to check definition of different metadata.
Definitions of start_metadata_sync_to_node and citus_activate_node
have also been updated. citus_activate_node will basically create
an active node with all metadata and reference table shards.
start_metadata_sync_to_node will be same with citus_activate_node
except replicating reference tables. stop_metadata_sync_to_node
will remove all the metadata. All of those UDFs need to be called
by superuser.
* Require superuser while activating a node
With this change, we require ActiveNode() (hence citus_add_node(),
citus_activate_node()) explicitly require for a superuser.
Before this commit, these functions were designed to work with
non-superuser roles with the relevent GRANTs given.
However, that is not a widely used way for calling the functions
above.
Due to possibility of non-super user calling the UDFs, they were
designed in a way that some commands were using some additional
short-lived superuser connections. That is:
(a) breaking transactional behavior (e.g., ROLLBACK
wouldn't fully rollback the whole transaction)
(b) Making it very complicated to reason about which
parts of the node activation goes over which connections,
and becoming vulnerable to deadlocks / visibility issues.
We prefer the background daemon to only sync node metadata. That's
why we move placement metadata changes from disable node to
activate node. With that, we can make sure that disable node
only changes node metadata, whereas activate node syncs all
the metadata changes. In essence, we already expect all
nodes to be up when a node is activated. So, this does not change
the behavior much.
Before this commit, Citus was triggering metadata syncing
in the background when a function is distributed. However,
with Citus 11, we expect all clusters to have metadata synced
enabled. So, we do not expect any nodes not to have the metadata.
This change:
(a) pro: simplifies the code and opens up possibilities
to simplify futher by reducing the scope of
bg worker to only sync node metadata
(b) pro: explicitly asks users to sync the metadata such that
any unforseen impact can be easily detected
(c) con: For distributed functions without distribution
argument, we do not necessarily require the metadata
sycned. However, for completeness and simplicity, we
do so.
Mehmet YILMAZ
Naisila Puka
Karina
Jelte Fennema-Nio
Gürkan İndibay
Teja Mupparti
aykut-bozkurt
aykutbozkurt
Onur Tirtir
Marco Slot
Önder Kalacı
Ahmet Gedemenli
Hanefi Onaldi
Philip Dubé
Alexander Kukushkin
Gledis Zeneli
Marco Slot
Halil Ozan Akgul
Burak Velioglu
Önder Kalacı