ense - your other tool must have been
reporting deltas. I'm not sure what controls the update rate - I
usually work with things that track 5 minute samples and haven't worried
much about smaller granularity. Since you are counting
;Again, what specific MIB elements changed observed behavior between
> different OS versions?
And it still doesn't make any sense for counters to go back to 0 after
being read. Maybe there is something wrong with the polling client. Do
you see this if you use snmpwalk to read tha
ver-increasing counter that the client
converts to a rate based on the poll interval. It shouldn't be going
back to zero no matter how often you read it.
--
Les Mikesell
lesmikes...@gmail.com
--
Start unco
f you don't mind running a
large java program, OpenNMS (http://www.opennms.org) will accept snmp
traps (and do a lot of other stuff...). I'm sure there are other
sim
iant with Net-SNMP?
>
> Also, is there any simple NMS solution which makes use of SNMP querries
> for polling?
>
I'm not sure it qualifies as simple, but the java based OpenNMS
(http://www.opennms.org) has pretty useful defaults so you can start using it
without too much effort
Mike Ayers wrote:
>> From: Dave Shield [mailto:d.t.shi...@liverpool.ac.uk]
>> Sent: Friday, December 04, 2009 12:35 AM
>
>> 2009/12/3 Les Mikesell :
>>> And is there some particular reason you'd want to do this [SNMP-over-
>> TCP]?
>>> Seems lik
nk you'd have to use the socks-proxy mode of ssh/putty and do
something to make the snmp client socks-aware to make this work.
Something like openvpn might be an easier approach to tunnel udp.
--
Les Mikesell
lesmikes...@gmail.com
use:
snmpwalk.exe -v1 -c public TCP:localhost:4711
on the client side? (per "man snmpcmd").
And is there some particular reason you'd want to do this? Seems like a
bad idea if your network is unreliable and not needed if it is.
--
Les Mikesell
lesmikes...@gmail.com
---
using the C (or perl) API calls.
>
> But it'll be a lot easier to code this once you've got a firm grasp
> of exactly what operations are needed.
I've always been surprised that this commonly-wanted operation was so
conceptually d
Dave Shield wrote:
> 2008/12/30 Les Mikesell :
>> On Linux, how should a monitoring program scale the load average and cpu
>> use values to determine when a machine has reached capacity? Is there a
>> way to get the number of CPUs from linux net-snmp, or per-cpu usage value
On Linux, how should a monitoring program scale the load average and cpu
use values to determine when a machine has reached capacity? Is there a
way to get the number of CPUs from linux net-snmp, or per-cpu usage values?
--
Les Mikesell
lesmikes...@gmail.com
ckaged
versions. I use Centos5 and using yum to install from their 'unstable'
repository works fine - and includes the otherwise missing Sun Java that
it needs. Regardless of how you install (or update) you do need to run
their install script that creates or updates the postgresql d
/www.opennms.org) does about everything you could want in
terms of monitoring, recording, and notifications.
--
Les Mikesell
[EMAIL PROTECTED]
-
This SF.Net email is sponsored by the Moblin Your Move Developer'
t more info?
>
http://www.opennms.org has a large open source java network management
program. Maps are sort of an afterthought in the current version but it
is great at monitoring large numbers of things and sending notifications
about outages.
--
Les Mikes
do it. I think it gets the MAC table from the switch and
matches it up with the arp table and/or the mac/IP addresses from the
connected devices.
--
Les Mikesell
[EMAIL PROTECTED]
-
Check out the new SourceForge.net Mar
eb interface language to eventually
determine setting?
--
Les Mikesell
[EMAIL PROTECTED]
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search lo
OURCES-MIB::hrSWRunPerfMem.1350 = INTEGER: 400 KBytes
etc., but of course the indexes are likely to change frequently.
Are there any higher level tools that can monitor per-process
CPU/memory, etc., given a process name?
--
Les Mikesell
[EMAIL
7;d want to monitor have some switch device on one
side and a windows or linux box on the other. Maybe there will be
enough errors on the side that choses half duplex to deduce the problem.
--
Les Mikesell
[EMAIL PROTECTED]
--
Dave Shield wrote:
> On 25/10/2007, Les Mikesell <[EMAIL PROTECTED]> wrote:
>>> To monitor the setting, you might have a look at the Etherlike-MIB
>>> http://net-snmp.sourceforge.net/docs/mibs/EtherLike-MIB.txt.
>> Does that mean it is not in the stock Cisco/Lin
Roth, Gabrielle wrote:
> Les Mikesell asked:
>
>> Is there a common way to monitor the duplex setting of a network
> device,
> in particular to discover mismatches on connected ports?
>
> To monitor the setting, you might have a look at the Etherlike-MIB
> http://ne
Is there a common way to monitor the duplex setting of a network device,
in particular to discover mismatches on connected ports?
--
Les Mikesell
[EMAIL PROTECTED]
-
This SF.net email is sponsored by: Splunk Inc
dependencies that I'd
> have to fix as a result).
>
The RPM version on sourceforge dropped in OK on a CentOS4 box. I'd
expect it to work on RHEL unless something else isn't up to date.
--
Les Mikesell
[EMAIL PROTECTED]
--
garding my machine:-
> ==
> #uname -a
> Linux lxpra24337 2.4.21-47.ELsmp #1 SMP Wed Jul 5 20:38:41 EDT 2006 i686
> i686 i386 GNU/Linux
> and machine has this version of beecrypto library :-
>
> beecrypt-3.0.1-0.20
Les Mikesell wrote:
> I'm trying to use opennms to collect data from some CentOS3.x x86_64
> boxes but I am getting an error "Integer too large: cannot decode"
> that http://bugzilla.opennms.org/cgi-bin/bugzilla/show_bug.cgi?id=1316
> says is a bug in net-snmp
er,
when I try to build the 5.4.1 src.rpm from sourceforge I get
"/usr/lib/libpopt.so: could not read symbols: Invalid operation"
although it builds OK on a 32 bit system. Is there another version
that would work better?
--
Les Mik
nity.html. Anyone can
then download the free player at
http://www.vmware.com/products/player/ and run it. I think
the images are typically built with NAT network support and
it might be easier to test snmp with bridged, but it is
possible to make that change after the fact in the .vmx file
if neces
less this needs really low-latency response, I'd just set
up ssh keys for passwordless access and use scripted ssh
commands. You probably need that to keep the machines
updated anyway.
--
Les Mikesell
[EMAIL PROTECTED]
---
SF.Net email
ux box with an x86_64 2.4 kernel? I'm getting:
/usr/lib/libpopt.so: could not read symbols: Invalid operation
--
Les Mikesell
[EMAIL PROTECTED]
---
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register f
On Mon, 2005-10-24 at 03:10, Dave Shield wrote:
> On Fri, 2005-10-21 at 17:32 -0500, Les Mikesell wrote:
> > What's supposed to happen when the value doesn't fit in a Counter32?
>
> The Counter32 value will report the bottom 32-bits
> (with the higher bits simply be
3.0 Mb)
This is an older version but with RH based systems it's hard to tell
what fixes have been backported to old version numbers.
--
Les Mikesell
[EMAIL PROTECTED]
---
This SF.Net email is sponsored by:
Power Archite
, then go crazy, showing
100 Meg traffic (the interface limit) most of the time.
--
Les Mikesell
[EMAIL PROTECTED]
---
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more
31 matches
Mail list logo
