Home / server / Questions / 867910
Accepted
Sethos II
Asked: 2017-08-11 01:51:50 +0800 CST

How to monitor services with SNMP?

  • 772

I want to monitor the status of several services via SNMP.

The manpage says:

proc NAME [MAX [MIN]]
              monitors the number of processes called NAME (as reported by "/bin/ps -e") running on the local system.

So I added these lines to /etc/snmp/snmpd.conf:

proc smbd
proc nfsd

but the corresponding OIDs still return zero:

iso.3.6.1.4.1.2021.2.1.100.4 = INTEGER: 0
iso.3.6.1.4.1.2021.2.1.100.5 = INTEGER: 0

allthough ps -e shows smbd and nfsd processes:

> ps -e | grep "smbd\|nfsd"
 8267 ?        00:00:01 smbd
10179 ?        00:00:04 smbd
11068 ?        00:00:00 nfsd4_callbacks
11072 ?        00:12:53 nfsd
11073 ?        00:13:33 nfsd
11074 ?        00:14:35 nfsd
11075 ?        00:16:23 nfsd
11076 ?        00:21:52 nfsd
11077 ?        00:33:34 nfsd
11078 ?        00:48:10 nfsd
11079 ?        01:02:44 nfsd
15135 ?        00:00:03 smbd
15137 ?        00:00:17 smbd
15141 ?        00:00:05 smbd
20603 ?        00:00:00 smbd

I'm also able to get the list of running processes with OID 1.3.6.1.2.1.25.4.2.1.2 which also contains smbd and nfsd processes:

> snmpwalk -v 2c -c secret host 1.3.6.1.2.1.25.4.2.1.2 | grep "smbd\|nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11068 = STRING: "nfsd4_callbacks"
iso.3.6.1.2.1.25.4.2.1.2.11072 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11073 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11074 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11075 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11076 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11077 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11078 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11079 = STRING: "nfsd"
iso.3.6.1.2.1.25.4.2.1.2.11724 = STRING: "smbd"
iso.3.6.1.2.1.25.4.2.1.2.12367 = STRING: "smbd"
iso.3.6.1.2.1.25.4.2.1.2.15135 = STRING: "smbd"
iso.3.6.1.2.1.25.4.2.1.2.15137 = STRING: "smbd"
iso.3.6.1.2.1.25.4.2.1.2.15141 = STRING: "smbd"
iso.3.6.1.2.1.25.4.2.1.2.29897 = STRING: "smbd"

Is it supposed to work like this and is monitoring services with the proc directive of SNMP a good idea or should I use extends (already got them working for ZFS and SLURM status)?

Environement:

  • NET-SNMP 5.7.3
  • Ubuntu Server 16.04
linux

1 Answers

  1. Best Answer

    The zeroes reported in OIDs below 1.3.6.1.4.1.2021.2.1.100 actually mean that everything is alright and non-zero would indicate an error.

    For future reference: OIDs below 1.3.6.1.4.1.2021.2.1.101 give further information in case of an error.

    • 0