Active IQ Unified Manager Discussions
Active IQ Unified Manager Discussions
I know I can have the Filer send a trap to a trap host that can trigger an email.
Is there any easier way?
This is a small environment so I will have to set up a TrapHost server for just one Trap! Solarwinds has a freeware version but as soon as you need a second trap or host to be monitored you get to the paid version.
Any other ideas to solve this?
Russ,
DataFabric Manager 3.1 and later includes an SNMP trap listener as part of the server service. The SNMP trap listener listens for SNMP traps from monitored NetApp appliances, if they have been manually configured to send traps to the workstation over UDP port 162. NOTE:it will only recognize traps from NetApp appliances!
So, you would not need to install any other traphost software in your environment. Operations Manager is listening for traps on the "public" community by default. Simply add the hostname/ip address of the Operations Manager server as a traphost on your NetApp controller. An example of how to do this on a controller is shown below:
NTAPPRI01> snmp traphost add opsmgr
NTAPPRI01> snmp traphost
opsmgr (opsmgr) <192.168.134.160>
NTAPPRI01> options snmp.enable on
You can also do this through System Manager if you prefer a Windows-based GUI administration. It can be configured for a host under Configuration -> System Tools -> SNMP. Now your controller is configured to send SNMP traps to the Operations Manager traphost.
In Operations Manager, under Setup -> Alarms, you can configure an alarm based on an SNMP Trap Event. You can have that alarm e-mail you. The events available are:
The key here is knowing what severity the temperature SNMP Trap is sent at. You could also try using a regular expression to match just the temperature trap, but that could be tricky. For more information on configuring SNMP Traps, refer to TR-3608: Data ONTAP SNMP Trap Management.
I hope this helps.
Reid
Reide,
Thanks! I was aware I could use Ops Mgr but in a small environment with modest requirements the licensing cost is a barrier (did Ops Mgr get free when I wasn’t looking?). I was looking for other suggestions such as free SNMP trap hosts that I have yet to come across, that is other than the one I mentioned.
I appreciate your reply ☺
Russ Witt | Senior Storage and Virtualization Engineer | Service Delivery, Storage
405-513-9428 O | | 866-876-6810 F | russ.witt@peakuptime.com
[cid:image003.jpg@01CB44A4.98EC0100]
You can reach our Help Desk by phone at 1-918-669-8000 or through 1-800-797-8388 and selecting “Support - #2” from the options menu. You can E-mail details of your issue to support@peakuptime.com. A case will be created and assigned to an appropriate resource.
Russ,
Sorry, I mis-understood what you were asking.
If you don't want to go through the hassle of setting-up an SNMP trap host, consider something like logwatch. Have it monitor the /etc/messages file of the controller(s) via a CIFS share. I haven't used logwatch in a long time, but I believe it can take action based on keywords it finds in the logs. You could have it e-mail you when it sees "temperature" or something similar.
If you don't want to use CIFS to monitor the logs, configure syslogd on each controller to fowward messages to a centralized log host (i.e. free linux server). Then run logwatch on that single host and you could monitor all the controllers in your environment for temperature warnings.
Reid
Reide,
Thanks! That’s very helpful and just what I was looking for. Someplace to look that I had not considered. ☺
Much appreciated!
Russ Witt | Senior Storage and Virtualization Engineer | Service Delivery, Storage
405-513-9428 O | | 866-876-6810 F | russ.witt@peakuptime.com
[cid:image003.jpg@01CB44A8.91A31650]
You can reach our Help Desk by phone at 1-918-669-8000 or through 1-800-797-8388 and selecting “Support - #2” from the options menu. You can E-mail details of your issue to support@peakuptime.com. A case will be created and assigned to an appropriate resource.
Also you can monitor the SNMP mib entry, .1.3.6.1.4.1.789.1.2.4.1.0 which represents:
envOverTemperature OBJECT-TYPE
SYNTAX INTEGER {
no(1),
yes(2)
}
MAX-ACCESS read-only
STATUS current
DESCRIPTION
"An indication of whether the hardware is
currently operating outside of its recommended
temperature range. The hardware will shutdown
if the temperature exceeds critical thresholds."
::= { environment 1 }
linux> snmpget -v 1 myfiler -c public .1.3.6.1.4.1.789.1.2.4.1.0
SNMPv2-SMI::enterprises.789.1.2.4.1.0 = INTEGER: 1
Regards,
- Rick -
Rick,
Thanks for your input!
So if I am following, I would kick that command off from a cron script on a unix host to periodically read the NetApp SNMP mib, and trigger an smtp message based on parsing the output? Where can I get more information on how to implement this?
I guess there is no way to get the NetApp to throw an email by itself? It’s not like you can control the autosupport smtp mechanism, is it?
Russ Witt | Senior Storage and Virtualization Engineer | Service Delivery, Storage
405-513-9428 O | | 866-876-6810 F | russ.witt@peakuptime.com
[cid:image003.jpg@01CB4500.275835C0]
You can reach our Help Desk by phone at 1-918-669-8000 or through 1-800-797-8388 and selecting “Support - #2” from the options menu. You can E-mail details of your issue to support@peakuptime.com. A case will be created and assigned to an appropriate resource.
Russ -
There is no way for the NetApp storage controller to throw an e-mail. You have two choices: 1. Set an SNMP trap on the MIB and have a custom SNMP trap handler send the e-mail; or 2, use DFM to catch the trap and send the e-mail.
Regards,
- Rick -