Hello,
I’ve been recently having problems with my postgresql 9.2.3 database, and I have no directions on what the problem could be.
I have one script that makes a db dump which I make for a good backup routine every day at midnight, but last sunday and yesterday it simply didn’t restart the service,
which means in the morning, when I try to start the programs that use the database they all show me the error message “the database system is shutting down”
The postgresql log keeps on a loop like this until I log in as root and reboot the computer:
…
2016-04-14 00:31:21 BRT FATAL: the database system is shutting down
2016-04-14 00:31:34 BRT FATAL: the database system is shutting down
2016-04-14 00:31:34 BRT FATAL: no PostgreSQL user name specified in startup packet
2016-04-14 00:31:34 BRT FATAL: the database system is shutting down
2016-04-14 00:31:36 BRT FATAL: the database system is shutting down
2016-04-14 00:31:36 BRT FATAL: no PostgreSQL user name specified in startup packet
2016-04-14 00:31:36 BRT FATAL: the database system is shutting down
…
I’m currently using openSUSE 13.2 (Harlequin) (x86_64) on a HP Proliant ML 350p gen8 Server hdd with btrfs (hardware mirrored),
the built-in tools didn’t accuse any hardware errors and I’m suspecting it could be that maybe one or another computer
turned on are constantly trying to connect and this hangs the ‘service postgresql restart’. It is strange, because some
days it works, other days it simply won’t work and hang like this.
I’ve made some search and I’ve discovered that in the script /etc/init.d/postgresql the restart command is
calling pg_ctl stop with the -m fast and the documentation says it disconnect the clients to make the restart,
so I have no idea why this happens. Maybe it is related with this “no PostgreSQL user name specified in startup packet” but
I had no luck searching this.
If you can please help me with some ideas of what the problem is. Also, should I be looking at postgresql forums or a service restart
related problem is here?
Thank you!
~Fernando