Thread

1. bug

   Sandeep Joshi <sjoshi@zambeel.com> — 2000-12-21T02:19:16Z

   
   
   
   Your name   :            Sandeep Joshi
   Your email address  :  sjoshi@zambeel.com
   
   
   System Configuration
   ---------------------
     Architecture (example: Intel Pentium)   : Intel Pentium
   
     Operating System (example: Linux 2.0.26 ELF)  : Linux 2.2.16-3
   
     PostgreSQL version (example: PostgreSQL-7.0):   PostgreSQL-7.0.3
   
     Compiler used (example:  gcc 2.8.0)   : egcs-2.91.66 19990314/Linux
   (egcs-1.1.2 release)
   
   
   
   Please enter a FULL description of your problem:
   ------------------------------------------------
   
   1. If I kill postmaster then if I try to restart then I get errors
   saying it is
       already running.
   
   pg_ctl: It seems another postmaster is running. Try to start postmaster
   anyway.
   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.
   /work/postgres/bin/postmaster: cannot create UNIX stream port
   pg_ctl: Cannot start postmaster. Is another postmaster is running?
   
   
   
   Please describe a way to repeat the problem.   Please try to provide a
   concise reproducible example, if at all possible:
   ----------------------------------------------------------------------
   
   pg_ctl start
   ps auxww | grep postmaster
   kill -9 $pid
   pg_ctl start
   
   
   
   If you know how this problem might be fixed, list the solution below:
   ---------------------------------------------------------------------
   
   pg_ctl should check whether the process and socket specified does exist
   or not.
   
   
   
   

2. Re: bug

   Chris <linux.fellow@libertysurf.fr> — 2000-12-21T18:37:13Z

   Sandeep Joshi wrote:
   > 1. If I kill postmaster then if I try to restart then I get errors
   > saying it is
   >     already running.
   > 
   > pg_ctl start
   > ps auxww | grep postmaster
   > kill -9 $pid
   > pg_ctl start
   > 
   
   
   It is clearly written in the docs that a server MUST NOT be killed with SIGKILL,
   because there's a risk of loosing datas. If you want to kill the server, use
   'kill $pid' (or if you really want to specify a signal, use SIGTERM: number 15)
   
   FYI, the signal 9 kills an app without giving her an info, so the server cannot
   exit cleanly. The signal 15 was then defined in that way : is asks the app to
   terminate.
   
   
   hope this helped you...