My specific problem is that we are unable to restart our postgreSQL database.
We have an ubuntu server (release 18.04) with PostgreSQL 10 installed. I commonly revise my IP address in the pg_hba.conf file to allow connections from pgAdmin on my home machine. When changes are made to the config file, this requires a restart of postgresql to take effect. I did a stop and start: sudo /etc/init.d/postgresql stop, followed by sudo /etc/init.d/postgresql start
Unfortunately, my colleague was running some processing when I did that. I can't be exactly sure what his procedures were doing (I'm working on figuring that out), but his last issued commands before I stopped the database were:
(root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1)
(root) CMD ( cd / && run-parts --report /etc/cron.hourly)
Upon restart we get some serious errors in the log (below). I've tried to start the database multiple times, to no avail. I've also rebooted the server; we still get these primary and secondary checkpoint errors, which I understand to be unfixable and requiring a database to be restored.
I wanted to check and see if anyone here has any workarounds that don't require a database restore. In other words, how can we get this database restarted without having to revert to a backup? Should a database restore be required, any recommendations for best practices with these specific errors in mind are welcome. Thanks!
2023-10-26 07:17:29.628 PDT [30714] LOG: received smart shutdown request
2023-10-26 07:17:29.629 PDT [13239] redacted@redacted FATAL: canceling authentication due to timeout
2023-10-26 07:17:29.632 PDT [28456] redacted@redacted FATAL: terminating connection due to administrator command
2023-10-26 07:17:29.640 PDT [29087] redacted@redacted FATAL: terminating connection due to administrator command
2023-10-26 07:17:29.807 PDT [30714] LOG: worker process: logical replication launcher (PID 30723) exited with exit code 1
2023-10-26 07:17:29.824 PDT [30716] LOG: shutting down
2023-10-26 07:17:31.751 PDT [30714] LOG: database system is shut down
2023-10-26 07:17:35.675 PDT [13261] LOG: listening on IPv4 address "0.0.0.0", port 5432
2023-10-26 07:17:35.675 PDT [13261] LOG: listening on IPv6 address "::", port 5432
2023-10-26 07:17:37.826 PDT [13261] LOG: listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"
2023-10-26 07:17:42.643 PDT [13262] LOG: database system was interrupted; last known up at 2023-09-18 16:02:09 PDT
2023-10-26 07:17:44.553 PDT [13262] LOG: invalid primary checkpoint record
2023-10-26 07:17:44.553 PDT [13262] LOG: invalid secondary checkpoint record
2023-10-26 07:17:44.553 PDT [13262] PANIC: could not locate a valid checkpoint record
pg_ctl: server did not start in time