Re: LDM Startup error, any suggestions?

Chris Novy wrote:
> 
> I'm receiving the following error upon starting LDM (5.1.2):
> 
>     hupsyslog: couldn't open /etc/syslog.pid
> 
> It started after a bad shutdown the other day.  LDM is still processing
> files OK but I also receive an error message from the newlog routine:
> 
>     Your "cron" job on delyte
>     /usr/LDM/ldm/bin/ldmadmin newlog
> 
>     produced the following output:
> 
>     hupsyslog: couldn't open /etc/syslog.pid
> 
> Any idea of what got erased or made inaccessible?
> 
> ..Chris..


Jeff Masters wrote:
> 
> Check to see if syslogd is running. If it is not running, there will be no
> /etc/syslog.pid file.
> 
> Jeff
> ------------------------------------------------------------------------------
>  Dr. Jeff Masters (address@hidden)                           (  )
>  Chief Meteorologist                              /\ Home of the       (    )
>  The Weather Underground, Inc.               /\  /  \  /\       /\    (      )
>  300 N Fifth Ave #240                       /  \/    \/  \ /\  /  \    ------
>  Ann Arbor, MI 48104                 ______/              /  \/    \_   \\\\\
>  734-994-8824 (voice)                           Weather Underground      \`\`\
>  734-994-8919 (fax)                         http://www.wunderground.com
> 


> Chris,
> 
> We had the same problem a while back. Either your syslog daemon
> is not running, or it is but the /etc/syslog.pid does not exist.
> 
> You'll need to do a "ps" and pipe the output through "grep syslogd".
> If it is running, you'll need to kill it and restart it. If it
> isn't running, just restart it.
> 
> ----------------------------------------------------------------
> Stephen D. Adams
> Vice President - Research and Development
> Agricultural Weather Information Service, Inc.
> ph: (334) 826-2149 ex103        email: address@hidden
>     (334) 826-2152 (FAX)        http://www.awis.com
> ----------------------------------------------------------------


Chris Novy wrote:
> 
> 
> Jeff:
> 
> It's not running!  Might I have killed it during the bad shutdown?   Is it
> something I can simply restart (as root) or do I need the sysadmin to yo
> yo  the whole system?
> 
> ..Chris..


Jeff Masters wrote:
> 
> Just restart it from the command line as root. You can put a check to see
> if syslogd is running in cron and restart it, too, I have a script that
> does it one one Solaris machine I have that has trouble with syslogd
> crashes.
> 
> Jeff


Chris Novy wrote:
> 
> Jeff, perhaps you could send me that script.  I'm running Solaris.
> 
> ..Chris..


Jeff Masters wrote:
> 
> Here's a script that will restart ldm and syslogd on a Soalaris machine.
> If syslogd dies, no error messages will get written to ldmd.log, so you
> need to restart ldm when restarting syslogd.
> 
> #!/bin/sh
> y=`ps -ef| grep /usr/sbin/syslogd | grep -v grep | wc -l`
> if [ $y = 0 ]
>   then
>   echo "/usr/sbin/syslogd not running, restarting syslogd and ldm..."
>   /usr/sbin/syslogd
>   /bin/su - ldm -c "/usr/local/ldm/bin/ldmadmin stop"
>   /bin/su - ldm -c "/usr/local/ldm/bin/ldmadmin newlog"
>   sleep 30
>   /bin/su - ldm -c "/usr/local/ldm/bin/ldmadmin start"
> fi
> 
> Jeff


***************************************************
Anne Wilson                     UCAR Unidata Program            
address@hidden                 P.O. Box 3000
                                  Boulder, CO  80307
----------------------------------------------------
Unidata WWW server       http://www.unidata.ucar.edu/
****************************************************