After update of nagios to 4.4.3, DNS checks all fail

[raskolnikov88]

I'd been using nagios on centos 7 for a couple of years and everything worked as expected. Yesterday it updated from 4.3.4 to 4.4.3 and DNS checks all began failing. All of the monitored DNS servers are working fine, but nagios always returns an error of the form:

DNS CRITICAL - query type of -querytype=A was not found for 192.168.111.20

Is it a known bug with 4.4.3? or did the configuration requirements change somehow?

[scottwilkerson]

This wouldn't be a nagios core problem, but a nagios-plugins problem

Did you update plugins as well?

Can you show the command that is running for check_dns along with the version of the plugin you are running?

Did you upgrade from RPM or source install?

I also noted a similar bug reported here:
https://bugzilla.redhat.com/show_bug.cgi?id=1525609

[tamar]

Well, it looks like more of us are going to come out of the woodwork to report this being an issue, as it's begun to happen to me as well. I did get a Nagios plugins update.

I assume this is coming from v2.2.2-1, and this is the version I am using:
nagios-plugins-dns-2.2.2-1.20190919git00cff01.fc30.x86_64

I do a regular dnf update, so it's not coming from the source.

I'm simply running check_dns without any further arguments:

Code: Select all

]$ /usr/lib64/nagios/plugins/check_dns -H 127.0.0.1
DNS CRITICAL - query type of -querytype=A was not found for 127.0.0.1

That never used to happen.

Any ideas on what a possible workaround might be for this?

[scottwilkerson]

As mentioned above we don't manage the RPM builds, and to be honest, nagios-plugins 2.2.2 hasn't even been released so I'm not sure why it would be released in fedora 30, do you have a testing repo enabled?

[tamar]

Nope, not using any testing repos from what I see:

Code: Select all

# sudo dnf repolist
Last metadata expiration check: 0:54:08 ago on Wed 02 Oct 2019 03:21:19 PM EDT.
repo id                                                    repo name                                                                       status
*fedora                                                    Fedora 30 - x86_64                                                              56,582
*fedora-modular                                            Fedora Modular 30 - x86_64                                                         135
*updates                                                   Fedora 30 - x86_64 - Updates                                                    13,143
*updates-modular                                           Fedora Modular 30 - x86_64 - Up

I'm not sure then how this got there, but I'm open to a workaround if it's known -- and yes, I understand you don't manage these RPMs. I do wonder how it got there then especially if this is some testing build though. Clearly others are likely to encounter this issue if they haven't yet, so I'd be curious what args I could potentially use that will return the old data that shows DNS is OK.

[scottwilkerson]

I think you can do the following

Code: Select all

dnf downgrade nagios-plugins-2.2.1

[tamar]

Thanks Scott. Just curious - I am not adverse to downgrading, but am curious if the current version is simply buggy or requires a few new args. Any insights?

[scottwilkerson]

I don't even know what they release because we have not released 2.2.2 of the plugins... I would guess they accidentally released an RPM based on untested development code.

2.2.1 is the latest release
https://github.com/nagios-plugins/nagio ... s/releases

[tamar]

Got it, thanks.

[scottwilkerson]

Got it, thanks.

no problem