Thread

  1. Problem under FreeBSD 2.2.6

    Dave Schwartz <davenpat@kw.igs.net> — 1998-05-20T00:30:01Z

    I don't know if this is a port bug or what.
    
    I had been happily using Postgres95 (yes 95) with the intention of upgrading
    to Postgresql 6.something when I got a 'round tuit'. Well my hand was forced
    this weekend when the host machine died a horrible death. I got the latest
    release (6.3.2) and it compiles just fine under FreeBSD 2.2.6. However, it
    won't run nearly as well. The install and initdb seem to work fine and the
    postmaster will start. As soon as I try to 'psql template1', the postmaster
    crashes with a 'bad system call' dump. When debug level 3, the following
    output is received (sorry for the length) and postmaster dies (the lock file
    is still in /tmp).
    
    Any ideas??
    
    FindBackend: searching PATH ...
    ValidateBackend: can't stat "/usr/home/postgres/bin/postgres"
    ValidateBackend: can't stat "/sbin/postgres"
    ValidateBackend: can't stat "/usr/sbin/postgres"
    ValidateBackend: can't stat "/bin/postgres"
    ValidateBackend: can't stat "/usr/bin/postgres"
    ValidateBackend: can't stat "/usr/contrib/bin/postgres"
    ValidateBackend: can't stat "/usr/contrib/mh/bin/postgres"
    FindBackend: found "/usr/local/pgsql/bin/postgres" using PATH
    binding ShmemCreate(key=52e2c1, size=758344)
    postmaster: ServerLoop:         handling reading 4
    postmaster: ServerLoop:         handling reading 4
    postmaster: ServerLoop:         handling writing 4
    postmaster: BackendStartup: environ dump:
    -----------------------------------------
            TERM=ansi
    
    PATH=/usr/home/postgres/bin:/sbin:/usr/sbin:/bin:/usr/bin:/usr/contrib/b
    in:/usr/contrib/mh/bin:/usr/local/pgsql/bin:/usr/local/bin:/usr/local/pgsql/
    bin:
    .
            SHELL=/usr/local/bin/tcsh
            HOME=/usr/home/postgres
            LOGNAME=postgres
            USER=postgres
            HOSTTYPE=FreeBSD
            VENDOR=intel
            OSTYPE=FreeBSD
            MACHTYPE=i386
            SHLVL=1
            PWD=/usr/home/postgres/postgres95/src
            GROUP=user
            HOST=web.kw.igs.net
            REMOTEHOST=ttyB14.kw.igs.net
            EDITOR=vi
            EXINIT=set autoindent
            PAGER=more
            BLOCKSIZE=1k
            NNTPSERVER=news.kw.igs.net
            RNINIT=/usr/home/postgres/.rninit
            WWW_HOME=http://www.kw.igs.net/
            MANPATH=/usr/local
            PGLIB=/usr/local/pgsql/lib
            PGDATA=/usr/local/pgsql/data
            LD_LIBRARY_PATH=/usr/local/pgsql/lib
            POSTPORT=5432
            POSTID=2147483647
            POSTID=2147483647
            PG_USER=postgres
            IPC_KEY=5432000
    -----------------------------------------
    postmaster child[0]: execv(/usr/local/pgsql/bin/postgres, -p, -d3, -P4, -v
    65536
    , template1, )
    postmaster: BackendStartup: pid 27750 user postgres db template1 socket 4
    FindBackend: found "/usr/local/pgsql/bin/postgres" using argv[0]
            ---debug info---
            Quiet =        f
            Noversion =    f
            timings   =    f
            dates     =    Normal
            bufsize   =    64
            sortmem   =    512
            query echo =   f
            DatabaseName = [template1]
            ----------------
    
            InitPostgres()..
    postmaster: reaping dead processes...
    postmaster: CleanupProc: pid 27750 exited with status 140
    postmaster: CleanupProc: reinitializing shared memory and semaphores
    Bad system call (core dumped)
    
    
    
  2. Re: [GENERAL] Problem under FreeBSD 2.2.6

    Marc G. Fournier <scrappy@hub.org> — 1998-05-20T02:56:40Z

    On Tue, 19 May 1998, Dave Schwartz wrote:
    
    > I don't know if this is a port bug or what.
    > 
    > I had been happily using Postgres95 (yes 95) with the intention of upgrading
    > to Postgresql 6.something when I got a 'round tuit'. Well my hand was forced
    > this weekend when the host machine died a horrible death. I got the latest
    > release (6.3.2) and it compiles just fine under FreeBSD 2.2.6. However, it
    > won't run nearly as well. The install and initdb seem to work fine and the
    > postmaster will start. As soon as I try to 'psql template1', the postmaster
    > crashes with a 'bad system call' dump. When debug level 3, the following
    
    	Bad System Call *generally* indicates that you haven't enabled the
    SYSV* shared memory options in your kernel...
    
    Marc G. Fournier                                
    Systems Administrator @ hub.org 
    primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org 
    
    
    
  3. Re: [GENERAL] Problem under FreeBSD 2.2.6

    Herouth Maoz <herouth@oumail.openu.ac.il> — 1998-05-20T07:18:38Z

    At 5:56 +0300 on 20/5/98, The Hermit Hacker wrote:
    
    
    > 	Bad System Call *generally* indicates that you haven't enabled the
    > SYSV* shared memory options in your kernel...
    
    By the way, this question (and the answer) are in the FAQ:
    
       http://www.postgreSQL.org/docs/faq-english.shtml#2.5
    
    I seem to recall that in addition to shared memory it also requires
    semaphores to be enabled in the kernel. Right or wrong? If right, should be
    added to the FAQ.
    
    Herouth
    
    --
    Herouth Maoz, Internet developer.
    Open University of Israel - Telem project
    http://telem.openu.ac.il/~herutma