We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
1. snmp polling - method 2. snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here.
Hi, there are two ways. You can either just snmp-poll the HP iLO-interface of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably wont need a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com:
We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
- snmp polling - method
- snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here. _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Thanks for reply,
We are linux shop we only need basic info about hardware status, i should disable other MIBs so to lighten the load but only worry is how big DB machine i need to handle 1000 machine, currently i have 8 core CPU and 8G memory Dell server.
Does observium has option to use external poller? so i can setup extra machine for just polling data and other machine for GUI/DB.
On Wed, Nov 8, 2017 at 1:55 PM, Markus Klock markus@best-practice.se wrote:
Hi, there are two ways. You can either just snmp-poll the HP iLO-interface of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably wont need a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com:
We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
- snmp polling - method
- snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here. _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi, That *might* be enough, you might need a little more memory. But yes, you can totally run the database on another machine, you also can run distributed pollers on multiple machines, I wrote a guide on the subject here: http://blog.best-practice.se/2017/10/scaling-observium-horizontally.html
/Markus
2017-11-08 20:43 GMT+01:00 Satish Patel satish.txt@gmail.com:
Thanks for reply,
We are linux shop we only need basic info about hardware status, i should disable other MIBs so to lighten the load but only worry is how big DB machine i need to handle 1000 machine, currently i have 8 core CPU and 8G memory Dell server.
Does observium has option to use external poller? so i can setup extra machine for just polling data and other machine for GUI/DB.
On Wed, Nov 8, 2017 at 1:55 PM, Markus Klock markus@best-practice.se wrote:
Hi, there are two ways. You can either just snmp-poll the HP iLO-interface of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably wont
need
a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com:
We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
- snmp polling - method
- snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here. _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
very nice article. Thanks
On Wed, Nov 8, 2017 at 2:49 PM, Markus Klock markus@best-practice.se wrote:
Hi, That *might* be enough, you might need a little more memory. But yes, you can totally run the database on another machine, you also can run distributed pollers on multiple machines, I wrote a guide on the subject here: http://blog.best-practice.se/2017/10/scaling-observium-horizontally.html
/Markus
2017-11-08 20:43 GMT+01:00 Satish Patel satish.txt@gmail.com:
Thanks for reply,
We are linux shop we only need basic info about hardware status, i should disable other MIBs so to lighten the load but only worry is how big DB machine i need to handle 1000 machine, currently i have 8 core CPU and 8G memory Dell server.
Does observium has option to use external poller? so i can setup extra machine for just polling data and other machine for GUI/DB.
On Wed, Nov 8, 2017 at 1:55 PM, Markus Klock markus@best-practice.se wrote:
Hi, there are two ways. You can either just snmp-poll the HP iLO-interface of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably wont need a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com:
We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
- snmp polling - method
- snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here. _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
now question is how to add those 1000 device? Does Observium has any kind of auto discovery and add device itself?
On Wed, Nov 8, 2017 at 2:55 PM, Satish Patel satish.txt@gmail.com wrote:
very nice article. Thanks
On Wed, Nov 8, 2017 at 2:49 PM, Markus Klock markus@best-practice.se wrote:
Hi, That *might* be enough, you might need a little more memory. But yes, you can totally run the database on another machine, you also can run distributed pollers on multiple machines, I wrote a guide on the subject here: http://blog.best-practice.se/2017/10/scaling-observium-horizontally.html
/Markus
2017-11-08 20:43 GMT+01:00 Satish Patel satish.txt@gmail.com:
Thanks for reply,
We are linux shop we only need basic info about hardware status, i should disable other MIBs so to lighten the load but only worry is how big DB machine i need to handle 1000 machine, currently i have 8 core CPU and 8G memory Dell server.
Does observium has option to use external poller? so i can setup extra machine for just polling data and other machine for GUI/DB.
On Wed, Nov 8, 2017 at 1:55 PM, Markus Klock markus@best-practice.se wrote:
Hi, there are two ways. You can either just snmp-poll the HP iLO-interface of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably wont need a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com:
We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
- snmp polling - method
- snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here. _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
if you have a list of the hostnames the easiest way would be to just use add_device.php and feed it a textfile of all your servers. Otherwise you could use LLDP-discovery for example if all your servers are talking LLDP to the switches and you already have the switches in Observium. /Markus
2017-11-08 21:23 GMT+01:00 Satish Patel satish.txt@gmail.com:
now question is how to add those 1000 device? Does Observium has any kind of auto discovery and add device itself?
On Wed, Nov 8, 2017 at 2:55 PM, Satish Patel satish.txt@gmail.com wrote:
very nice article. Thanks
On Wed, Nov 8, 2017 at 2:49 PM, Markus Klock markus@best-practice.se
wrote:
Hi, That *might* be enough, you might need a little more memory. But yes, you can totally run the database on another machine, you also
can
run distributed pollers on multiple machines, I wrote a guide on the
subject
here: http://blog.best-practice.se/2017/10/scaling-observium-
horizontally.html
/Markus
2017-11-08 20:43 GMT+01:00 Satish Patel satish.txt@gmail.com:
Thanks for reply,
We are linux shop we only need basic info about hardware status, i should disable other MIBs so to lighten the load but only worry is how big DB machine i need to handle 1000 machine, currently i have 8 core CPU and 8G memory Dell server.
Does observium has option to use external poller? so i can setup extra machine for just polling data and other machine for GUI/DB.
On Wed, Nov 8, 2017 at 1:55 PM, Markus Klock markus@best-practice.se wrote:
Hi, there are two ways. You can either just snmp-poll the HP
iLO-interface
of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably
wont
need a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com:
We have currently 100 network gear we are monitoring but now we have 1000 HP servers and i want to monitor its health using HP ILO snmp poll. I want i get alerted when system hardware failed or FAN or powersupply whatever.
so i have two option:
- snmp polling - method
- snmp trap - method
How big machine i need to poll data from 1000 HP ILO snmp data? looking for good advice here. _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Another note, only HP iLO4 actually have a real SNMP-stack, iLO1-3 only passthrough to SNMP-querys to the operating system running on the server. /Markus
2017-11-08 21:32 GMT+01:00 Markus Klock markus@best-practice.se:
if you have a list of the hostnames the easiest way would be to just use add_device.php and feed it a textfile of all your servers. Otherwise you could use LLDP-discovery for example if all your servers are talking LLDP to the switches and you already have the switches in Observium. /Markus
2017-11-08 21:23 GMT+01:00 Satish Patel satish.txt@gmail.com:
now question is how to add those 1000 device? Does Observium has any kind of auto discovery and add device itself?
On Wed, Nov 8, 2017 at 2:55 PM, Satish Patel satish.txt@gmail.com wrote:
very nice article. Thanks
On Wed, Nov 8, 2017 at 2:49 PM, Markus Klock markus@best-practice.se
wrote:
Hi, That *might* be enough, you might need a little more memory. But yes, you can totally run the database on another machine, you also
can
run distributed pollers on multiple machines, I wrote a guide on the
subject
here: http://blog.best-practice.se/2017/10/scaling-observium-horiz
ontally.html
/Markus
2017-11-08 20:43 GMT+01:00 Satish Patel satish.txt@gmail.com:
Thanks for reply,
We are linux shop we only need basic info about hardware status, i should disable other MIBs so to lighten the load but only worry is how big DB machine i need to handle 1000 machine, currently i have 8 core CPU and 8G memory Dell server.
Does observium has option to use external poller? so i can setup extra machine for just polling data and other machine for GUI/DB.
On Wed, Nov 8, 2017 at 1:55 PM, Markus Klock <markus@best-practice.se
wrote:
Hi, there are two ways. You can either just snmp-poll the HP
iLO-interface
of the server to get basic hardware-stuff or if you run linux on the servers you can easily install the hp-snmp-agent on them to have them expose additional hardware via snmpd directly. I do this on my servers and they answer very fast so you probably
wont
need a very big machine to make it work. Observium can only do snmp polling, no snmp traps supported.
/Markus
2017-11-08 16:45 GMT+01:00 Satish Patel satish.txt@gmail.com: > > We have currently 100 network gear we are monitoring but now we
have
> 1000 HP servers and i want to monitor its health using HP ILO snmp > poll. I want i get alerted when system hardware failed or FAN or > powersupply whatever. > > so i have two option: > > 1. snmp polling - method > 2. snmp trap - method > > > How big machine i need to poll data from 1000 HP ILO snmp data? > looking for good advice here. > _______________________________________________ > observium mailing list > observium@observium.org > http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
participants (2)
-
Markus Klock -
Satish Patel