Commit Graph

13 Commits (7a25ebe2575b4e0fbbb31ead879ec5098e93519f)

Author SHA1 Message Date
Marco Slot 35dbacdb69 Do not reinitialise MyBackendData 2017-12-19 15:56:26 +01:00
Marco Slot af201a2f6d Allow intermediate results to be used in parallel workers 2017-12-18 19:05:08 +01:00
Andres Freund d063658d6d Protect some initializations from being called during backend startup.
On EXEC_BACKEND builds these functions shouldn't be called at every
backend start.
2017-11-20 15:29:51 -08:00
Marco Slot ae47df01ea Observe prepared xacts twice in RecoverWorkerTransactions to avoid race condition 2017-11-20 11:44:08 +01:00
Marco Slot 2410c2e450 Rewrite recover_prepared_transactions to be fast, non-blocking 2017-11-20 11:27:40 +01:00
Onder Kalaci 5b48de7430 Improve deadlock detection for MX
We added a new field to the transaction id that is set to true only
for the transactions initialized on the coordinator. This is only
useful for MX in order to distinguish the transaction that started
the distributed transaction on the coordinator where we could
 have the same transactions' worker queries on the same node.
2017-08-12 13:28:37 +03:00
Onder Kalaci a333c9f16c Add infrastructure for distributed deadlock detection
This commit adds all the necessary pieces to do the distributed
deadlock detection.

Each distributed transaction is already assigned with distributed
transaction ids introduced with
3369f3486f. The dependency among the
distributed transactions are gathered with
80ea233ec1.

With this commit, we implement a DFS (depth first seach) on the
dependency graph and search for cycles. Finding a cycle reveals
a distributed deadlock.

Once we find the deadlock, we examine the path that the cycle exists
and cancel the youngest distributed transaction.

Note that, we're not yet enabling the deadlock detection by default
with this commit.
2017-08-12 13:28:37 +03:00
Onder Kalaci b5ea3ab6a3 Improve locking semantics for backend management
We use the backend shared memory lock for preventing
new backends to be part of a new distributed transaction
or an existing backend to leave a distributed transaction
while we're reading the all backends' data.

The primary goal is to provide consistent view of the
current distributed transactions while doing the
deadlock detection.
2017-08-09 17:17:12 +03:00
Marco Slot 81198a1d02 Add function for dumping local wait edges 2017-07-25 16:52:32 +02:00
Onder Kalaci 58faffa42b Fix bug on error check for assigning distributed transaction id
to a backend that has already been assigned a transaction.
2017-07-25 14:58:07 +03:00
Marco Slot 601b17d544 Use distributed transaction number in 2PC identifiers 2017-07-21 17:36:33 +02:00
Marco Slot 18a6e478af Fix typo in GetCurrentDistributedTransctionId 2017-07-21 17:36:33 +02:00
Onder Kalaci 3369f3486f Introduce distributed transaction ids
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.
2017-07-18 15:01:42 +03:00