Switch port monitoring - up/down status as an error

[psonenberk]

Hello all,
could you tell me simple way how monitor router or switch
port? I mean case when some device in some switch port
runs down, Zenoss doesn't notice error. Port has still status UP.
I mean case when that device (i.e. device in switch port)
isn't supervised by Zenoss. And I want to know about this
event, for example as an standard error on that switch.
There are Administrative status and Operational status. Is it
possibility to use it for this?
 
Thank you very much for your help.
Have a nice day.

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57283#57283





--end--

[guyverix]

Depending on the switch you should have options to turn on SNMP trapping.  This will give you the information that you are looking for.  As for the router, I cannot think of any that do not have the ability to send snmp traps.  Just turn it on, and set your trap destination as your Zenoss IP address..

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57298#57298





--end--

[nilie]

Or you can use syslog if your device supports it. Using this with Cisco devices, you have the bonus to be able to turn it off on a per interface basis like you don't want to be bothered by PC starting-up/shutting-down but uplinks and servers are of your concern.
 
A lot of us were asking on this forum for the port status monitoring, it is the single most important feature missing from Zenoss and it is still not on the roadmap for the future versions of Zenoss. For us, syslog is doing a fine job alerting us when a port goes up/down but there're no way in hell you could tell the current status of a port.

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57288#57288





--end--

[quentin.gillet]

Hi,
 
Here http://community.zenoss.org/docs/DOC-2494
It can be useful

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57327#57327





--end--

[nilie]

Thank you for pointing me to that document. Unfortunately we tried that and it didn't work quite well for us. Maybe we were not doing it properly but having more than 45000 interfaces to monitor we found that the amount of events generated by ports going up or down was unacceptable.

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57331#57331





--end--

[aletatone]

I did with Syslog...
I configured Cisco devices with:
 
logging trap debugging
logging +zenossipaddress+
 
than I used Transform to change alarm severity based on inteface name
 
Cheers
Ale

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57319#57319





--end--

[psonenberk]

Hi Quentin,
Thank you very much for your advice. But there is something strange ........
Command snmpwalk (for example to Cisco 2800) gets me answer:
zen:/#snmpwalk -c public -v 2c 10.1.1.220 .1.3.6.1.2.1.2.2.1.8
IF-MIB::ifOperStatus.1 = INTEGER: up(1)
IF-MIB::ifOperStatus.2 = INTEGER: down(2)
IF-MIB::ifOperStatus.3 = INTEGER: up(1)
IF-MIB::ifOperStatus.4 = INTEGER: up(1)
 
And return value isn't "clear" integer.
By the way zenoss in this case returns this error message:
 
10.1.1.220|ifOperStatus|/Perf/Snmp||1|Error reading value for \"ifOperStatus\" on 10.1.1.220 (oid .1.3.6.1.2.1.2.2.1.8 is bad)
 
I've tried the way via script (it's bellow). Script returns figure 100
when port is UP, and 0 when port is down.
This cript I've added as a Local Template, Data Sources, View and Edit Details,
Data Sources with name "PORT2VALUE", Threshold.
Threshold has configuration: MinMax, Severity: Error, Minimum value: 50, Maximum value 100.
When script returns value 0 Zenoss enounces error.
It works fine, but this way isn't so good for switch with 24 ports. And when there is more
than 1 switch in Zenoss, it's a little unpractical. It's necessary to prepare 24 scripts
and 24 declaration for any ports of the switch.
 
I would be grateful for some better solution.
Thank you. Have a nice day.
P.Sonenberk
 

#!/bin/bash
# version for Linux - /usr/local/zenoss/common/libexec
# Script for port no. 2
 
TYPE="up(1)"
 
PROM=`/usr/bin/snmpget -c public -v2c www.xxx.yyy.zzz 1.3.6.1.2.1.2.2.1.8.2 | /bin/cut -f4 -d' '`
if [ "$TYPE" = "$PROM" ]
then
# VALUE=100, port is up
  /bin/echo "VALUE OK|PORT2VALUE="100
else
# VALUE=0, port is down
  /bin/echo "VALUE OK|PORT2VALUE="0
fi
exit 0
#
#end of script

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57396#57396





--end--

[psonenberk]

Hello,
Thank you for your advice. Yes, it's possible to use a trap.
I've tried it. (For example this configuration in Cisco
router: #snmp-server host www.xxx.yyy.zzz traps SNMPv2c event-manager)
But in this case Zenoss up/down port information describes
only as an "information". This event hasn't error level, it has
"blue" level. I need "red" or "orange" level - critical or error.
Does exist some easy way how to do it? I don't know.....
(It's necessary to send an e-mail about this event).
 
I would be grateful for some fine solution.
Thank you. Have a nice day.
P.Sonenberk

--------------------------------------------------------------

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/57525#57525





--end--