#Cisco Switch monitoring fails with "poller has no port"

1 messages · Page 1 of 1 (latest)

somber owl
#

Just added, finally, some cisco switches to our nabox system. The havrest container fails to restart. The last log messages are several instances of this error: source=havrest.go:83 msg="poller has no port". Any suggestions as to where I have gone wrong?

#

Is it necessary to manually specify a port number for each (new) Poller instance?

#

( i.e. with a "port: nnnn" entry in harvest.yml? )

gilded plinth
#

@maiden jackal This seems like related to NABox.

somber owl
#

A followup question could be ... how to choose a good port number? I.e. one that won't conflict with those automagically generated by nabox

somber owl
#

FYI: Adding a port number e.g. "port: 13001" allows the container to get started, but it seems as if no CiscoRest poller processes get started

#

Also not if I add "enabled: true"

#

Aha. I also had to add "use_insecure_tls: true" ... Now they run!

frigid saffron
#

hi @somber owl excellent. I was just getting ready to say that it works on our local NABox instance

somber owl
#

News: software developer says "It works on my system" - lit! 😎

#

So I just manually picked some ports 13001 - 13004 for this test. It would still be good to know if there is a better or "more correct" solution

#

( for the nabox solution )

frigid saffron
#

those ports are fine - the "correct" solution would be for @maiden jackal to publish a range of ports that he promises not to use for any internal service. That way you know what range is available. Until then, it looks like you are free to use anything except for these

somber owl
#

In the havrest container nabox auto-configures the pollers with ports starting from 12000 (for the unix platform), then 12001, 12002, and so on. I figure 13001 leaves room for a few Filers 🙂