postgresql and sockets

Krzysztof Duleba krzysan@skrzynka.pl
Mon May 2 23:27:00 GMT 2005


Hi

I am using Cygwin postgresql as my db engine for high load jboss/Hibernate
web app. It works fine, but after some 10k transactions socket seems
broken and I get this message:

$ psql
psql: could not connect to server: Bad file descriptor
        Is the server running locally and accepting
        connections on Unix domain socket "/tmp/.s.PGSQL.5432"?

$ cat /tmp/.s.PGSQL.5432
!<socket >1304 s D872D6B4-84B51718-5811C571-3E0793C6

However, `psql -h localhost -p 5432' succeeds. So in fact the socket is
still there, but the link stored in /tmp/.s.PGSQL.5432 is broken.

If I restart postmaster service, psql works again. Now /tmp/.s.PGSQL.5432
has different content:

$ cat /tmp/.s.PGSQL.5432
!<socket >1661 s A50B48C0-681AB839-08CFE772-54D105A7

I use the default postgresql.conf, only the following line was changed:
tcpip_socket = true

Any ideas is the problem related to Cygwin? If yes then I will try to
provide more info. I know it's not too specific, but a month ago
postgresql worked more stable (it wasn't as loaded as it is now though).

Regards
Krzysztof Duleba
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cygcheck.out
Type: application/octet-stream
Size: 49479 bytes
Desc: not available
URL: <http://cygwin.com/pipermail/cygwin/attachments/20050502/08fddd9b/attachment.obj>
-------------- next part --------------
--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


More information about the Cygwin mailing list