Release notes for EDB Postgres Distributed version 4.3.3 v4

Released: 14 Nov 2023 Updated: 18 Jan 2024

EDB Postgres Distributed version 4.3.3 is a patch release of EDB Postgres Distributed 4, which includes bug fixes for issues identified in previous versions.


This version is required for EDB Postgres Advanced Server versions 12.15, 13.11, 14.10, 15.5 and later.

BDR4.3.3Bug fixEnsure that the WalSender process does not request locks on the partitions, thus avoiding a deadlock with user process waiting on sync commit. (RT97952)
BDR4.3.3Bug fixOnly consider CAMO transactions to be asynchronous when the CAMO setup was degraded to local mode. This solves the split brain problem when deciding fate of transactions that happened during failover. (RT78928)
BDR4.3.3Bug fixHandle partitions with different attribute numbers when batch inserting. (RT99115)
BDR4.3.3Bug fixFix unsafe CAMO decisions in remote_write mode.
BDR4.3.3Bug fixEnsure that the taskmgr process does not ignore SIGINT.
BDR4.3.3Bug fixClean up stale records in bdr.taskmgr_local_work_queue during physical join.
BDR4.3.3Bug fixFix a bug in copying bdr.autopartition_rules during logical join.
BDR4.3.3Bug fixOverride bdr.ddl_replication=off in taskmgr worker.
BDR4.3.3Bug fixAvoid a potential deadlock between bdr.autopartition_wait_for_partitions() and taskmgr
BDR4.3.3Bug fixBlock new EPAS automatic partitioning on PGD cluster.
BDR4.3.3Bug fixAllow existing automatically partitioned tables when cluster is created or upgraded.
BDR4.3.3Bug fixBlock PGD Autopartition on EPAS INTERVAL partitioned table.
BDR4.3.3Bug fixFix memory leak in streaming transaction processing.
BDR4.3.3Bug fixAllow force parting of nodes that are already being parted normally.
BDR4.3.3Bug fixFix nodes getting stuck on PART_CATCHUP due to replication slot conflict.
BDR4.3.3Bug fixFix nodes getting stuck on PART_CATCHUP due to node_catchup_info conflict.
BDR4.3.3Bug fixFix row filter failure for partitions created after a table column was dropped (RT95149)
BDR4.3.3Bug fixEnsure that replication receiver worker reloads configuration correctly when pg_reload_conf() is called.
BDR4.3.3Bug fixPrevent duplicate Raft request ID generation which could break replication.
BDR4.3.3Bug fixFix several rare memory access issues that could potentially cause crashes of workers.
CLI1.1.2Bug fixFixed connect_timeout issue when sslmode=allow or sslmode=prefer by upgrading database driver library version.
Utilities1.2.0Bug fixRemove the pre-upgrade step which sets the CONNECTION LIMIT to 0 at database level in bdr_pg_upgrade