| 1 |
I know people probably have seen dicussions about this elsewhere but I |
| 2 |
wanted to bring it to this list to get a broader sharing of ideas. |
| 3 |
Lately, I've been wanting to do more with cacti and nagios both, but it |
| 4 |
just seems as though there should be a way to tie the two together to |
| 5 |
make things better. The biggest complaint I have thus far is polling the |
| 6 |
same data twice using two different methods. I've been reading several |
| 7 |
places of how people have combined this. Here are some of the variants |
| 8 |
I've heard and I'm curious what you all think. |
| 9 |
|
| 10 |
1) Putting all the scripts into net-snmp and having both cacti and |
| 11 |
nagios making snmpget calls to get the information. |
| 12 |
|
| 13 |
This seems like an interesting idea. To me, snmp seems to be a rather |
| 14 |
effiecient method of gathering information, but this method would |
| 15 |
require some ideas on making this happen. I have yet to learn how to |
| 16 |
extend net-snmp, but I know its rather easy. I'm thinking that it might |
| 17 |
be possibly to have net-snmp directly calling nagios plugins and putting |
| 18 |
it as another OID. The only other option with this idea is creating |
| 19 |
custom scripts/MIB's for specific things. |
| 20 |
|
| 21 |
2) Having Cacti directly calling nrpe to gather data. |
| 22 |
|
| 23 |
I don't really like this idea personally because I feel that nrpe is |
| 24 |
much slower than snmp calls. This would also require both programs to |
| 25 |
access it doubling the polling which I don't like. |
| 26 |
|
| 27 |
3) Having nagios polling the rrd files that cacti makes. |
| 28 |
|
| 29 |
I don't think this is the best way to handle monitoring information for |
| 30 |
nagios. You can't trust the information will be precisely correct all |
| 31 |
the time. |
| 32 |
|
| 33 |
4) Have a script that parses the performance data log nagios produces |
| 34 |
and have cacti poll that. |
| 35 |
|
| 36 |
I'm not sure how well this could work, but it was an idea that popped in |
| 37 |
my head just now. Nagios can pull detailed peformance data that cacti |
| 38 |
could use. I'm not sure how well this would work for making graphs. I |
| 39 |
would fear having gaps in the graphs. |
| 40 |
|
| 41 |
|
| 42 |
Have any of you found a good way to integrate both of these excellent |
| 43 |
tools to a more effecient manner, if so, PLEASE SHARE! :) |
| 44 |
-- |
| 45 |
Lance Albertson <ramereth@g.o> |
| 46 |
Gentoo Infrastructure |
| 47 |
|
| 48 |
--- |
| 49 |
GPG Public Key: <http://www.ramereth.net/lance.asc> |
| 50 |
Key fingerprint: 0423 92F3 544A 1282 5AB1 4D07 416F A15D 27F4 B742 |
| 51 |
|
| 52 |
ramereth/irc.freenode.net |
Archives