Nagios Log Server not sending alert Emails

[rgignac]

Fist of all is there anyway to send a test email similar to nagios XI to test the mail settings?

Second I added an alert for a query but despite it being in critical at over 717 out of 590:600 and even now at a warning, I've not received any emails.
Any ideas on where to look.

[abrist]

Can you check the smtp relay server for any auth errors in the logs?

[scottwilkerson]

I will add that I agree we should add a test button to test the settings, and have created a feature request for this.

[networkeng]

We are having the same issue. Using the same SMTP settings that are working on Nagios XI.

Are there any logs in Nagios Log Server that we can check?


Suggest that a "Test Mail Server" button be added ASAP.

[slansing]

The feature request has been added as per Scott's reply, have you talked with your mail team to see if they can pull anything out of the relay logs? Please show the output of the following, ( you may want to redact some of it):

Code: Select all

tail -50 /var/log/maillog

[networkeng]

Here is an entry (sanitized) from the Nagios Log Server maillog. It looks like it is not using the SMTP server IP we specified in the mail settings (Administration -> Mail Settings -> SMTP Server Address). Our SMTP server IP is 10.10x.xx.xx and I cannot find it anywhere in the maillog.

Jan 5 01:40:50 nagiosls postfix/smtp[8842]: 9E2CB43261: to=<[email protected]>, relay=none, delay=407900, delays=407870/0.02/30/0, dsn=4.4.1, status=deferred (connect to maxxxxxxp.com[5x.6x.2xx.1]:25: Connection timed out)

Looking forward to the 'Test Mail Server' feature.

[sreinhardt]

Just to be 100% sure, you currently have log server set to use the internal IP address, not a resolvable hostname correct? If you were using a hostname for m****p.com and internal to your network, it does resolve to an external IP, this would be fully expected. If not, and you do have an IP set, I would definitely agree that it seems to be just using the domain from the email and ignoring or minimizing the use of smtp settings that log server has defined.

[networkeng]

Yes, we have the SMTP server set to an internal ip address, 10.x.x.x.

In Nagios-XI we use the same configuration and can receive email alerts. However, in the Nagios-XI /var/log/maillog I can see the same errors with it trying to resolve our email domain to an external IP with the connection timing out.

In the Nagios-XI maillog I can not find any entries for the successful emails that were sent through our internal SMTP server...no mention of the IP of our internal SMTP server. So, I'm not sure if the maillog stores this information???

[lgroschen]

Have you tried preceding the hostname with either of the following?:
ssl://
tls:// (we are adding this to some products ie Reactor, Incident Manager, but this may not work with TLS until we find a workaround)

Could simply be missing the encryption protocol. This also could be added as a dropdown in the future.

Let us know if this helps at all.

[networkeng]

Thanks. I will try this tomorrow time permitting. Been very busy lately as we are coming out of a month and a half change freeze. Testing it is kind of a pain in the a$$ because there is no "test mail server" button so I have to set up alerts and then generate them.

Any idea on when the test mail server feature will be implemented?

Also, is there a good write up on how to integrate these Nagios Log Server alerts into Nagios-XI so we can just bypass this Log Server email issue altogether?