[IDD #VJN-743814]: GEMPAK

Hi Massoud,

re:
> We still have a problem. I copied the revised syslog.conf
> file to /etc.

Presumably you did this as 'root', AND you made sure that the file permissions
were the same before and after the copy?

> Afterwards, I did the following:
> 
> [ldm@wimden ~]$ pwd
> /home/ldm
> [ldm@wimden ~]$ cd ~ldm/logs
> [ldm@wimden logs]$ ls -ltr
> total 76
> -rw-r--r--  1 ldm weather     0 Jun 28 09:35 ldmd.log
> -rw-r--r--  1 ldm weather 70029 Jul  3 16:15 ldm-mcidas.log

> [ldm@wimden logs]$ rm -i ldmd.log
> rm: remove regular empty file `ldmd.log'? y

> [ldm@wimden logs]$ ls -ltr
> total 76
> -rw-r--r--  1 ldm weather 70029 Jul  3 16:15 ldm-mcidas.log

> [ldm@wimden logs]$ touch ldmd.log
> [ldm@wimden logs]$ ls -ltr
> total 76
> -rw-r--r--  1 ldm weather 70029 Jul  3 16:15 ldm-mcidas.log
> -rw-r--r--  1 ldm weather     0 Jul  5 09:22 ldmd.log

> [ldm@wimden logs]$ hupsyslog
> hupsyslog: couldn't open /var/run/syslogd.pid

Interesting.  I believe that syslogd's PID file should be in the same location
under RedHat WS 3/4 as it is under Fedora Core (the Linux variant we use here
in Unidata).

If 'ls -alt /var/run/syslogd.pid' indicates that that the file really doesn't
exist, I would stop and restart 'syslogd':

<as 'root'>
/etc/init.d/syslog stop
/etc/init.d/syslog start

After doing this, check to see if the PID file was created:

ls -alt /var/run/syslogd.pid

If it was not created, contact your system administrator and ask him/her to
look into the problem.

> [ldm@wimden logs]$ find -name syslogd.pid
> [ldm@wimden logs]$ find -name *syslogd.pid*
> [ldm@wimden logs]$ cd ~var/run/
> -bash: cd: ~var/run/: No such file or directory

You should be looking in /var/run, not ~var/run.

> [ldm@wimden logs]$ hupsyslog
> hupsyslog: couldn't open /var/run/syslogd.pid

The HUP signal is being sent to the daemon 'syslogd', but it is apparently
not rereading its configuration file, /etc/syslog.conf, since it can't find
its PID file, /var/run/syslogd.pid.

> [ldm@wimden logs]$ logger -p local0.debug 'test of LDM logging'

Since 'syslogd' has not reread its configuration file, this will not work.

> [ldm@wimden logs]$ ls -ltr
> total 76
> -rw-r--r--  1 ldm weather 70029 Jul  3 16:15 ldm-mcidas.log
> -rw-r--r--  1 ldm weather     0 Jul  5 09:22 ldmd.log
> [ldm@wimden logs]$
> 
> My ~ldm/logs/ldmd.log file doesn't contain 'test of LDM
> logging.'

This is because 'syslogd' never reread its configuration file.
Forcing it to stop and start (see above) will force it to read
its configuration file.  After doing the stop/start of syslogd
(see above), run the logger test again:

logger -p local0.debug 'test of LDM logging'

> Any suggestions on what I can do next?

If this still doesn't work, and assuming that the copy of syslog.conf that
I modified was not altered in any way by me attaching it to my previous
reply and you extracting it from that reply, then your system has a problem
which your system administrator will need to look into.

Cheers,

Tom
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: VJN-743814
Department: Support IDD
Priority: Normal
Status: Closed