compreal
asked on
syslogd help
I just ran into a problem with my linux box. Here it is: my
/usr/adm/messages file doesnt tell me important system info anymore???
Here is the last few lines in the file:
Mar 2 02:42:20 kabal pppd[2930]: Terminating on signal 2.
Mar 2 02:42:20 kabal pppd[2930]: Connection terminated.
Mar 2 02:42:20 kabal pppd[2930]: Exit.
Mar 2 02:42:54 kabal init: Switching to runlevel: 6
Ever since then, everytime I invoke pppd, this file doesnt tell me.
Normally i run a tail -f /usr/adm/messages in the top right of my Xdisplay so i can c if anyone tries to telnet, ftp, rlogin into my machine, etc.
Now, this file no longer does that. Not only that, but it doesnt tell me
when I su either??? I have no clue why this is happening.
I have run /sbin/runlevel and I am running in runlevel three. Also, when
i ps -auxw | more, the init process is set to [3].
I am running slackware linux kernel 2.0
I'd like to follow up my previous post about my problem with the
/usr/adm/messages file with this:
one of my friends mentionted that i should have klogd and syslogd running
in order for the file to capture sysinfo. well i did a ps for the daemons
and sure enough syslogd is not running... i tried running syslogd manually
from the command line, but when i do a ps after this, this is what i get:
root 163 0.0 1.9 772 292 p2 S 22:48 0:00
/usr/sbin/syslogd
root 164 0.0 0.0 0 0 ? Z 22:48 0:00 (syslogd
<zombie>)
what is this <zombie>? anyway, this is not a fix for my problem since the
messages file still does no reporting...
AND, when i do a ps about 5 minutes after invoking syslogd, the daemon is
no longer running??
i've tried doing a man on syslogd but got no man page on it. one more
point of interest is when i boot up my system and all of my daemons are
invoked, there is an unusually long pause after the syslogd daemon. this
has never happened before and i am at my wits end.
another point of interest is that everytime i run syslogd,
the program creates a core file... i ran "file core" on the
core file and here is what i got:
core: ELF 32-bit LSB core file i386 (386 and up) Version 1
I hope this is enough info for someone to be able to help.
Thanks alot.
serge
/usr/adm/messages file doesnt tell me important system info anymore???
Here is the last few lines in the file:
Mar 2 02:42:20 kabal pppd[2930]: Terminating on signal 2.
Mar 2 02:42:20 kabal pppd[2930]: Connection terminated.
Mar 2 02:42:20 kabal pppd[2930]: Exit.
Mar 2 02:42:54 kabal init: Switching to runlevel: 6
Ever since then, everytime I invoke pppd, this file doesnt tell me.
Normally i run a tail -f /usr/adm/messages in the top right of my Xdisplay so i can c if anyone tries to telnet, ftp, rlogin into my machine, etc.
Now, this file no longer does that. Not only that, but it doesnt tell me
when I su either??? I have no clue why this is happening.
I have run /sbin/runlevel and I am running in runlevel three. Also, when
i ps -auxw | more, the init process is set to [3].
I am running slackware linux kernel 2.0
I'd like to follow up my previous post about my problem with the
/usr/adm/messages file with this:
one of my friends mentionted that i should have klogd and syslogd running
in order for the file to capture sysinfo. well i did a ps for the daemons
and sure enough syslogd is not running... i tried running syslogd manually
from the command line, but when i do a ps after this, this is what i get:
root 163 0.0 1.9 772 292 p2 S 22:48 0:00
/usr/sbin/syslogd
root 164 0.0 0.0 0 0 ? Z 22:48 0:00 (syslogd
<zombie>)
what is this <zombie>? anyway, this is not a fix for my problem since the
messages file still does no reporting...
AND, when i do a ps about 5 minutes after invoking syslogd, the daemon is
no longer running??
i've tried doing a man on syslogd but got no man page on it. one more
point of interest is when i boot up my system and all of my daemons are
invoked, there is an unusually long pause after the syslogd daemon. this
has never happened before and i am at my wits end.
another point of interest is that everytime i run syslogd,
the program creates a core file... i ran "file core" on the
core file and here is what i got:
core: ELF 32-bit LSB core file i386 (386 and up) Version 1
I hope this is enough info for someone to be able to help.
Thanks alot.
serge
ASKER
When I ran syslogd -d, I got a
"segmentation fault: (core dumped)"
I looked at /etc/syslog.conf and the file looks normal to me.
Here is what it says:
# /etc/syslog.conf
# For info about the format of this file, see "man syslog.conf" (the BSD man
# page), and /usr/doc/sysklogd/README.l
#
# NOTE: YOU HAVE TO USE TABS HERE - NOT SPACES.
# I don't know why.
#
*.=info;*.=notice /usr/adm/messages
*.=debug /usr/adm/debug
*.warn /usr/adm/syslog
#
# This might work instead to log on a remote host:
# * @hostname
BTW, I do have man pages on syslogd. The proper syntax is not
"man syslogd" but rather "man 8 sysklogd"
I'm still hacking away trying to resolve this conflict. All help
is very appreciated.
Thanks alot
Serge
PS. You can contact me at my email address at compreal@panix.com
"segmentation fault: (core dumped)"
I looked at /etc/syslog.conf and the file looks normal to me.
Here is what it says:
# /etc/syslog.conf
# For info about the format of this file, see "man syslog.conf" (the BSD man
# page), and /usr/doc/sysklogd/README.l
#
# NOTE: YOU HAVE TO USE TABS HERE - NOT SPACES.
# I don't know why.
#
*.=info;*.=notice /usr/adm/messages
*.=debug /usr/adm/debug
*.warn /usr/adm/syslog
#
# This might work instead to log on a remote host:
# * @hostname
BTW, I do have man pages on syslogd. The proper syntax is not
"man syslogd" but rather "man 8 sysklogd"
I'm still hacking away trying to resolve this conflict. All help
is very appreciated.
Thanks alot
Serge
PS. You can contact me at my email address at compreal@panix.com
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Did you have a disk crash?
Did you change anything in one of your setup files?
The <zombie> means the proces has already died, but its parent hasn't acknowledged this. The zombie should go away after the other syslogd proces ends.
Try running syslogd with the -d flag (for debugging). Look at the file /etc/syslogd.conf to find out if there is something strange. If possible, compare it to your backup.
Also, it's strange that you don't have a syslogd man page. This would indicate an installation failure. It should be in /usr/man/man8 . What does `man 8 sysklogd' (note the k) say?