A.118. Release 3.7.1

A.118.1. Bug fixes

• Improve Makefiles. (Bo Peng)

  Patch provided by Tomoaki Sato.

• Doc: Fix document typo and mistakes. (Bo Peng)

• Replace /bin/ed with /bin/sed in pgpool_setup, because /bin/sed is included in most distribution's base packages. (Tatsuo Ishii)

• Change the pgpool.service and sysconfig files to output Pgpool-II log. (Bo Peng)

  Removing "Type=forking" and add OPTS=" -n" to run Pgpool-II with non-daemon mode, because we need to redirect logs. Using "journalctl" command to see Pgpool-II systemd log.

• Add documentation "Compiling and installing documents" for SGML document build. (Tatsuo Ishii)

• Fix per node health check parameters ignored. (bug 371) (Tatsuo Ishii)

  Also pgpool_setup is modified to add appropriate per node health check parameters to pgpool.conf.

• Fix health checking process death and forking forever. (Tatsuo Ishii)

  When failed to read from backend socket (this could happen when wrong health_check_user is specified), the health check process raises a FATAL error in pool_read(), which causes death of health check process. And Pgpool-II main forks off a new health check process. This repeats forever.

• Fix timestamp data inconsistency by replication mode. (Bo Peng)

  From PostgreSQL10 the column default value such as 'CURRENT_DATE' changes, Pgpool-II didn't rewrite timestamp by the added default values. This caused data inconsistency.

• Doc: Fix watchdog_setup doc. (Tatsuo Ishii)

  It lacked to mention that it supports logical replication mode.

• Downgrade a log message to debug message. (Tatsuo Ishii)

  That was mistaken left while last development cycle.

• Test: Add test data for bug 370. (Tatsuo Ishii)

• Fix for re-sync logic in reading packet from backend. (Tatsuo Ishii)

  read_kind_from_backend(), which reads message kind from backend, re-syncs backend nodes when a ready for query message is received. Unfortunately it forgot to call pool_pending_message_pull_out() to delete sync pending message. This leads to random stuck while reading packets from backend. Fix this to call pool_pending_message_pull_out().

• Fix Pgpool-II hangs. (bug 370) (Tatsuo Ishii)

  If an erroneous query is sent to primary and without a sync message the next query that requires a catalog cache look up is send, Pgpool-II hangs in do_query().

• Fix returning transaction state when "ready for query" message received. (Tatsuo Ishii)

  We return primary or master node state of ready for query message to frontend. In most cases this is good. However if other than primary node or master node returns an error state (this could happen if load balance node is other than primary or master node and the query is an erroneous SELECT), this should be returned to frontend, because the frontend already received an error.

• Test: Fix bug with extended-query-test test driver. (Tatsuo Ishii)

• Doc: Enhance document "Running mode of Pgpool-II". (Tatsuo Ishii)