repmgrd: activate inactive node record at startup

If a PostgreSQL instance was shut down while repmgrd was running, and
repmgrd was subsequently restarted (this chain of events could occur
during e.g. a server reboot), the node record will have been set to
"inactive". Previously, in this case repmgrd would refuse to start up.
However, as we can determine the node is running, it should normally
be no problem to automatically set the node record to "active".

The old behaviour can be restored by setting the new parameter
"repmgrd_exit_on_inactive_node" to "true".

RM19604.
This commit is contained in:
Ian Barwick
2021-07-12 17:38:40 +09:00
parent f64f498afb
commit 79d1f005db
10 changed files with 90 additions and 3 deletions

View File

@@ -135,6 +135,7 @@
#define DEFAULT_ASYNC_QUERY_TIMEOUT 60 /* seconds */
#define DEFAULT_PRIMARY_NOTIFICATION_TIMEOUT 60 /* seconds */
#define DEFAULT_REPMGRD_STANDBY_STARTUP_TIMEOUT -1 /*seconds */
#define DEFAULT_REPMGRD_EXIT_ON_INACTIVE_NODE false,
#define DEFAULT_STANDBY_DISCONNECT_ON_FAILOVER false
#define DEFAULT_SIBLING_NODES_DISCONNECT_TIMEOUT 30 /* seconds */
#define DEFAULT_CONNECTION_CHECK_TYPE CHECK_PING