Received: from sss.pgh.pa.us (sss.pgh.pa.us [209.114.132.154]) by mail.postgresql.org (8.11.1/8.11.1) with ESMTP id f0470v762306 for ; Thu, 4 Jan 2001 02:00:57 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.11.1/8.11.1) with ESMTP id f0471PB20544; Thu, 4 Jan 2001 02:01:25 -0500 (EST) To: Markus Wagner cc: "pgsql-sql@postgresql.org" Subject: Re: postmaster: address already in use In-reply-to: <3A5426D1.4D000AC5@imsd.uni-mainz.de> References: <3A5426D1.4D000AC5@imsd.uni-mainz.de> Comments: In-reply-to Markus Wagner message dated "Thu, 04 Jan 2001 08:31:29 +0100" Date: Thu, 04 Jan 2001 02:01:24 -0500 Message-ID: <20541.978591684@sss.pgh.pa.us> From: Tom Lane X-Archive-Number: 200101/42 X-Sequence-Number: 703 Markus Wagner writes: > my postmaster suddenly begins to make problems. It was running fine, but > then: > FATAL: StreamServerPort: bind() failed: Address already in use > Is another postmaster already running on that port? > If not, remove socket node (/tmp/.s.PGSQL.5432) and retry. > /sys/svr/pgs/bin/postmaster: cannot create UNIX stream port > There is no other postmaster running and there is no socket node file in > /tmp That's really weird. Is it possible that there are still some old clients holding open connections on that socket address, even though you shut down the old postmaster? For INET sockets, there's a timeout in many kernels, such that the address that had been used for an INET socket is considered "in use" for a few minutes after you close the socket. I've never heard of any similar behavior for Unix sockets, however. Ideas anyone? regards, tom lane