Hi^^
since I am unable to find more Information regarding clustered
services via searching the web/mathias-kettner.de I am resorting to
this Mailinglist.
Our network environment uses check_mk 1.2.6p12 provided by * Open
Monitoring Distribution Version 2.11.20160620-labs-edition* to
monitor uptime of servers/services.
As expected some specific services are shared among several servers
and in general there are two situations which I’d like to display
accordingly:
**Scenario 1)**
Two servers provide the same service comparable to an active-backup
solution. (During normal operation only the primary server handles
requests. In case of failure services will migrate to the secondary
server)
Scenario 2)
Two servers provide the same service comparable to adaptive load
balancing. A common example would be certain implementations of DHCP
in a given network. (Both servers will handle requests at the same
time)
You can specify clusters in the WATO-configuration menu "Hosts"
and the services which should be monitored in ** “Host &
Service Parameters”**. Editing those rulesets allow you to*“make the outcome of the ruleset positive/negative”*.
To my understanding “make the outcome of the ruleset positiv”
means the status is OK as long as one of the cluster nodes
provides the service while * “make the outcome of the ruleset
negative”* will set it to CRIT as soon as a single node
fails.
Assuming my interpretation of this configuration option is
correct this would be perfect to handle the requirements. Scenario
1 matches “make the outcome of the ruleset positiv” ,
Scenario 2 matches “make the outcome of the ruleset negative”
However in our implementation (Scenario 1) it works like this:
ruleset set to “make the outcome of the ruleset positive”:
This will result in a CRIT-notification and status detail “CRIT -
: service ok, : service error” while services
in the inventory of each node will show as expected -> OK for the
first node, CRIT for the second and grouped under * “Already
configured clustered services (located on cluster host)”
-
You can see the services of the cluster in the attached
screenshot
ruleset set to “make the outcome of the ruleset negative”:
Data will be displayed as if there was no cluster defined.
Only the status for each node will be shown.
Services of the cluster are stated as “No entries.”
Am i missing a crucial part of the configuration or is this
feature just not fully implemented yet?
Thanks in advance!
Yours faithfully,
Stefan Erl
···
-- Stefan Erl ANDURAS intranet security AG
Messestraße 3 - 94036 Passau - Germany
Web: Tel:+49 (0)851-4 90 50-42 Fax:+49 (0)851-4 90 50-55
Rechtsform: Aktiengesellschaft Sitz: Passau Amtsgericht: Passau HRB 6032
Vorstand: Dipl.-Inf. Sven Anders, Dipl.-Inf. Marcus Junker
Vorsitzender des Aufsichtsrats: RA Mark Peters