EDB Postgres Distributed 5.8.0 release notes v5.8.0

Add the --summary and --options flags to pgd node show command to filter the output of the pgd node show command. This also maintains symmetry with other show commands.

Add the bdr.lock_table_locking and bdr.truncate_locking GUCs to list of GUCs verfied in pgd cluster verify command.

While commit scope processing does not have any direct affect on non-data nodes, by skipping it altogether we can avoid potentially confusing error messages.

The bdr.galloc_chunk_info() function provides information about the chunk allocation for a given sequence. This function returns the chunk ID, the sequence ID, and the chunk size. This function is useful for debugging and understanding how sequences are allocated in BDR.

Improve the formating of the log messages to be more readable and symmetrical with Postgres log messages.

Add the recommended_value column to the result of the pgd cluster verify --settings command. The column will not be displayed in tabular output but will be displayed in JSON output.

The output for the commands with tabular output will be sorted by the resource name. For the commands that display more than one resource, the output will be sorted by each resource column in order.

If executed with --mode=camo, but the provided test script is not wrapped in an explicit transaction, pgd_bench will not be able to retrieve the expected transaction_id value. Now the emitted error message contains a hint about a possible missing transaction.

An issue, CVE-2025-2506, was discovered in pglogical which is present in later versions of BDR and PGD. The issue could enable a user with CONNECT access to obtain read access to replicated tables.

CDC failover now has additional safety nets to ensure that the consumer does not start replication from a node that is not the creator of the replication slot. This is to prevent data loss or duplicate transactions. The changes also add additional checks to ensure that the consumer does not start replication from a node that does not have the required WAL files to decode the transactions that are missing on the consumer but were included in the initial snapshot that the newly joined node had obtained (physical or logical).

Updated to compare tuples on lookup to ensure it is the same when handling non-unique indexes.

The shared memory information for connection information may get obfuscated when a password is present and become useless. Instead of reading from there, we are now using the primary_conninfo GUC which is now available on all supported PG versions.

The pgd cluster show command would exit with an error regarding clock drift if only one node was up and running in a N node cluster, and not show the associated health and summary information. The command now returns output for, health and summary, while reporting an appropriate error for clock-drift.

The pgd node show command crashed if a non-existent node was given to the command. The command now fails gracefully with an appropriate error message.

The pgd replication show command previously crashed when formatting EPAS timestamps.

The node_name argument to bdr.taskmgr_set_leader is required. The function now throws an appropriate error in case node_name := NULL is passed in.

We have replaced TRUNCATEs with DELETEs from all BDR catalogs on a local node drop. This is to avoid deadlock in bdr_init_physical if the user happens to run monitoring queries during node joining / cleaning unwanted source node data.

Setting nci->min_lsn to XactLastCommitEnd of watermark message Tx to ensure CATCHUP phase finishes on new joiner only after watermark is processed.

Ensure that any custom value set in the deprecated GUC bdr.raft_election_timeoutis applied to its replacement bdr.raft_global_election_timeout.

During subscription reconfiguration, if there is no change required to a subscription, do not enable it since it could have been disabled explicitly by the user. Skip reconfiguring subscriptions if there are no leadership changes.

Subscriber-only nodes will no longer attempt to take a lock on the cluster when running DDL. The DDL will be executed locally and not replicated to other nodes.

Fixed deadlock between bdr_init_physical cleaning unwanted node data and concurrent monitoring queries.

Fixed an issue when a new node joining the cluster finishes CATCHUP phase before getting its replication progress against all data nodes. This could have caused a new node to be out of sync with the cluster.