Internal catalogs and views v5

Catalogs and views are presented here in alphabetical order.


An internal catalog table holding state per DDL epoch.

bdr.ddl_epoch columns

ddl_epochint8Monotonically increasing epoch number
origin_node_idoidInternal node ID of the node that requested creation of this epoch
epoch_consume_timeouttimestamptzTimeout of this epoch
epoch_consumedbooleanSwitches to true as soon as the local node has fully processed the epoch
epoch_consumed_lsnbooleanLSN at which the local node has processed the epoch


Internal catalog table that tracks cluster membership events for a given PGD node. Specifically, it tracks:

bdr.event_history columns

event_node_idoidThe ID of the node to which the event refers to
event_typeintThe type of the event (a node, raft or worker related event)
event_sub_typeintThe sub-type of the event, i.e. if it's a join, a state change or an error
event_sourcetextThe name of the worker process where the event was sourced
event_timetimestamptzThe timestamp at which the event occurred
event_texttextA textual representation of the event (e.g. the error of the worker)
event_detailtextA more detailed description of the event (for now, only relevant for worker errors)


A view of the bdr.event_history catalog that display the information in a more human-friendly format. Specifically, it displays the event types and sub-types as textual representations, rather than integers.


An internal catalog table with per node configuration options.

bdr.node_config columns

node_idoidThe node id
node_route_priorityintPriority assigned to this node
node_route_fencebooleanSwitch to fence this node
node_route_writesbooleanSwitch to allow writes
node_route_readsbooleanSwitch to allow reads
node_route_dsntextThe interface of this node


An internal catalog table with per node group configuration options.

bdr.node_group_config columns

node_group_idoidThe node group id
route_writer_max_lagbigintMaximum write lag accepted
route_reader_max_lagbigintMaximum read lag accepted
route_writer_wait_flushbooleanSwitch if we need to wait for the flush


Per node group routing configuration options.


An internal catalog table holding current routing information for a proxy.

bdr.node_group_routing_info columns

node_group_idoidThe node group id tha this proxy uses
write_node_idoidCurrent write node
prev_write_node_idoidPrevious write node
read_node_idsoid[]List of read nodes IDs


A view of bdr.node_group_routing_info catalog which shows the information in more friendly way


Per node routing configuration options.


An internal catalog table holding proxy specific configurations.

bdr.proxy_config columns

proxy_namenameThe name of the proxy
node_group_idoidThe node group id that this proxy uses
listen_portintPort that the proxy uses
max_client_connintNumber of maximum client connections that the proxy accepts
max_server_connintNumber of maximum connections that the server accepts
server_conn_timeoutintervalTimeout for the server connections
server_conn_keepaliveintervalInterval between the server connection keep alive
fallback_group_timeoutintervalTimeout needed for the fallback
fallback_group_idsoid[]List of group IDs to be used for the fallback
listen_addrstext[]Listen address


Per proxy instance specific configuration options.


An internal state table storing the type of each non-local sequence. We recommend the view bdr.sequences for diagnostic purposes.

bdr.sequence_kind columns

seqidoidInternal OID of the sequence
seqkindcharInternal sequence kind ('l'=local,'t'=timeshard,'s'=snowflakeid,'g'=galloc)