Appendix C. FAQ (Frequently Asked Questions)

Table of Contents

C.1. General
C.1.1. What's the difference between the repmgr versions?
C.1.2. What's the advantage of using replication slots?
C.1.3. How many replication slots should I define in max_replication_slots?
C.1.4. Does repmgr support hash indexes?
C.1.5. Can repmgr assist with upgrading a PostgreSQL cluster?
C.1.6. What does this error mean: ERROR: could not access file "$libdir/repmgr"?
C.1.7. How can I obtain old versions of repmgr packages?
C.1.8. Is repmgr required for streaming replication?
C.1.9. Will replication stop working if repmgr is uninstalled?
C.1.10. Does it matter if different repmgr versions are present in the replication cluster?
C.1.11. Should I upgrade repmgr?
C.1.12. Why do I need to specify the data directory location in repmgr.conf?
C.2. repmgr
C.2.1. Can I register an existing PostgreSQL server with repmgr?
C.2.2. Can I use a standby not cloned by repmgr as a repmgr node?
C.2.3. What does repmgr write in recovery.conf, and what options can be set there?
C.2.4. How can a failed primary be re-added as a standby?
C.2.5. Is there an easy way to check my primary server is correctly configured for use with repmgr?
C.2.6. When cloning a standby, how can I get repmgr to copy postgresql.conf and pg_hba.conf from the PostgreSQL configuration directory in /etc?
C.2.7. Do I need to include shared_preload_libraries = 'repmgr' in postgresql.conf if I'm not using repmgrd?
C.2.8. I've provided replication permission for the repmgr user in pg_hba.conf but repmgr/repmgrd complains it can't connect to the server... Why?
C.2.9. When cloning a standby, why do I need to provide the connection parameters for the primary server on the command line, not in the configuration file?
C.2.10. When cloning a standby, how do I ensure the WAL files are placed in a custom directory?
C.2.11. Why is there no foreign key on the node_id column in the table?
C.2.12. Why are some values in recovery.conf surrounded by pairs of single quotes?
C.3. repmgrd
C.3.1. How can I prevent a node from ever being promoted to primary?
C.3.2. Does repmgrd support delayed standbys?
C.3.3. How can I get repmgrd to rotate its logfile?
C.3.4. I've recloned a failed primary as a standby, but repmgrd refuses to start?
C.3.5. repmgrd ignores pg_bindir when executing promote_command or follow_command
C.3.6. repmgrd aborts startup with the error "upstream node must be running before repmgrd can start"