-----Original Message----- From: Judy Stobby
[mailto:firstname.lastname@example.org] Sent:Friday, June 07, 2002 To: 'email@example.com' Subject: RE: [nv-l] /var file
system & snmpd.log
Thanks for the response.The message I received from netmon was
just that it had stopped and to restart it.Yesterday the /var file system filled up
again, so I decided to do a shutdown, after I increased the file system
again.When the system came back
up, it the /var file system was only at 10%, and still is this morning.Hidden files somewhere I suspect,
but I really do not know.
The reason I wanted to delete lines is I
wanted to keep the original file in tact, not really necessary I know.
-----Original Message----- From: firstname.lastname@example.org
[mailto:email@example.com] Sent:Friday, June 07, 2002 To: firstname.lastname@example.org Subject: Re: [nv-l] /var file
system & snmpd.log
I am confused here. Take a look at the man page for snmpd.conf.
If you set logging level to zero, then you are essentially turning logging off,
except for failures and exceptions. So there may not be any new entries. There
certainly would not be anything like the volume you would have with level=3.
But I am puzzled as to why you did not just start a new log. There is no need
to edit the old one. Just rename it and restart snmpd -- he'll create a new
one. You can move, archive, or trash the old snmpd log file as you like.
I take it that you know you can stop snmpd on AIX at any time using
stopsrc -s snmpd
and then restart with
startsrc -s snmpd
so this is easy to do.
As snmpd is an OS daemon, AIX's in this case, and not NetView's, I don't know
what he will do if you run out of space in the file system that you have told
him to log to. It is possible that he will just stop, or get stuck in a loop
trying to write an error message to the log that he cannot write to the log. I
don't know. That's just speculation. But if you enable logging, I'd make sure
you note that somewhere and check on your space periodically.
What message did you get from netmon which led to this? That he could not talk
to snmpd? There is no "dependency" on the /var file system in NetView,
but there sure is one for snmpd, and if he isn't working right, then netmon and
most of NetView won't work well either.
Level 3 Support for Tivoli
NetView for UNIX and NT Tivoli Software / IBM Software
06/05/02 05:40 PM
<email@example.com> cc: Subject: [nv-l] /var file system
I’m not sure if the problems I’ve seen
lately are related to netview or aix, but maybe someone has a clue as to what
happened. Our system is running netview 7.1.1 on aix 4.3.3 ML9. One morning
last week netmon daemon stopped running with a message to restart it. I looked
at the file systems and /var and /usr/OV were at 100%. I deleted the log files
in /usr/OV/log and increased the file systems, and restarted netmon. Then the
next day /var was again at 100%, and netmon daemon stopped again. I increased
the file system a little more which took care of it, but I would like to know
why the /var file system filled up. What is the relationship between Netview
and /var? I turned on netmon tracing with netmon M -1, and then looked at the
trace files. I saw a lot of querying HP Jet Direct cards, with public still as
the community, but nothing that seemed to be failing.
I had turned on SNMP logging (level 3) because the
event log was filling up with authentication errors, so I tired to look at the
snmpd.log, which gave me the error too large to fit in file system. So I move
the snmpd.log to /home and change the location is snmpd.conf, also changing the
logging back to level 0.
Since this move, there have not been any entries in
the snmpd.log. I stopped and restart the daemon thinking this would make the
snmpd.conf file be read again. But it is still not logging any entries. I have
not found any documentation regarding moving the location of this file along
with any special commands to restart logging. The file is full of entries that
go way back, is it possible to use the editor to delete old entries out of this