A.139. Release 3.7.4

A.139.1. Bug fixes

• Fix Pgpool-II hung if replication delay is too much, when query cache enabled in extended query mode. (Tatsuo Ishii)

  See [pgpool-general-jp: 1534] for more details.

• Doc: Fix document typo of PCP commands option "-U". (Bo Peng)

• Delete some debug code. (Bo Peng)

• In extended query mode, do not set writing tx flag with SET TRANSACTION READ ONLY. (Tatsuo Ishii)

• Fix wrong parameter %P (old primary node id) passed to failover script in 3.7.3 and 3.7.2. (Tatsuo Ishii)

• Doc: Clarify that failover_require_consensus requires that health check is enabled. (Tatsuo Ishii)

• Doc: Update outdated pcp_proc_info manual. (Tatsuo Ishii)

• Test: Fix test.sh in extended_query_test. (Tatsuo Ishii)

• Add missing health_check_timeout in pgpool_setup. (Tatsuo Ishii)

• Doc: Enhance online recovery document to Clarify that recovery_2nd_stage_command is only required in native replication mode. (Tatsuo Ishii)

• Prevent pcp_recovery_node from recovering "unused" status node. (Tatsuo Ishii)

  This allowed to try to recovery a node without configuration data, which leads to variety of problems.

  See [pgpool-general: 5963] for more details.

  Also I fixed pgpool_recovery function so that it quotes an empty string argument with double quotes.