email notifications are not being delivered

[ostendorf16]

NagiosXI will not send email notifications when my host or service is in a down state. Strange how I can send a test message from within the Nagiosxi web interface. Am I missing a configuration?

[egalstad]

Make sure you check on the following:

1. Ensure the XI machine can properly resolve DNS names

2. Make sure outbound SMTP connections aren't being blocked by a firewall/filter somewhere on your network

3. In your personal account settings, make sure you enable notifications (global) and enable email notifications. You must also select the types of notifications you would like to receive (host down, service recovery, etc.). Once you set these options, use the "Send Test Notifications" option to test delivery to your email.

4. Lastly, when configuring new hosts/services through the configuration wizard, make sure you enable notifications and select the proper notification targets (e.g. yourself).

[Eciatasa]

I did all that, test emails work but still no notifications if a host/service goes down.

[Box293]

I'm probably repeating egalstad but here are some tips I've found when troubleshooting notifications.

One thing I found is all the little help ? [question marks] are poplulated however some of them are missing vital information to help you understand some concepts.

1) First things first do the test email, make sure you are actually receiving the emails in your mailbox.
Click the Admin link at the top
click Manage Email Settings
Click the Send A Test Email link
Click the Send Test Email button

If you receive this then we know emails are getting to your mailbox.

2) Now send a test from a host
Goto the Host Status Detail page for the host you are monitoring
Click the Advanced tab
Under the Commands window click Send custom notification link
Tick the box Forced
Type a comment
Click Commit

If you receive this notficiation then ALL the requirements are met for notifications to work.

3) If you did not receive that notification then it most likely comes down to all the settings in the host (or service) itself.
These settings *should* be all that is required to make notifications work
Goto Core Configuration Manager
Goto Hosts
Click the modify icon for the host
Click the Check Settings tab

• Retry interval = 1 (this is 1 minute)
  check interval = 1 (this is 1 minute, when testing, set the check interval to a low number)
  Max. check attempts = 3
  Alert checks enabled = on

Click the Alert Settings tab

• Contacts = Click the Modify group icon
  A window will appear
  Select your name in the left and pane and click the arrow button pointing to the right
  Click the Save button
  Now your name should appear selected in this list.
  Under this list select the radio button Standard
  Notification options = tick them all ? will explain what they all are but just select them all now for testing purposes
  Notification interval = 20 (this is 20 minutes)
  Notification enabled = on

Click the Save button
Answer Yes to the warning
Click Apply Configuration

Once this is successful then go and follow the steps in Point 2 to send a test notification. You should receive this notification. You can also test by disconnecting the network cable to this host.

Note: When editing a host or service have a look at the Additional templates section on the Common Settings tab.
Under here is a list of templates that are currently being applied. These templates may or may not contain check and alert settings. If there is more than 1 template in the list, it will read the bottom one first, then the next one and apply any settings defined in the second template (which may overwrite some of the settings from the first template) and so on.
The key to using templates is the function skip or null. I've found myself that using skip seems to always work for the items I am not interested in applying in this template. This way, as each template is processed and evaluated the settings will combine to provide the functionality you need.
For example I have 3 template types I commonly use, a)Check template b)Alert template c)Service/Host Templates. This way I can have multiple alert and check templates that apply to different levels of monitoring. Being able to apply templates in this way makes monitoring very flexible in nagios. Service templates are great for defining checks with all the variables applied. Using templates means if you need to make a change you only need to do it in one place and it applies to everything using that template.
Go and have a look at how some of the wizards apply templates to your hosts and services and follow the trail back to the begining and you can see how they make it work.

If you are still not receiving a notification for a host down event then it may not be turned on on your user account.
Click the Configure tab
Click the Change Your Account Settings link
On the left hand side Click Notification Preferences
Under notfication types make sure they are all selected
Make sure your notification times are 00:00 to 24:00


I hope this helps you get your notifications working, it took me a little while to work it out.

[pzygadlo]

Hello Folks,

I am experiencing this issue as well. It seems to have started after upgrading to 2009R1.1G. In the sendmail maillog I can see the test message going out, but no automatic notifications, even manual ones that are done by following the previous posters suggestions. Any thoughts on where to start after following all the steps from the previous poster?

[gary_ford]

i would concur that this happened with the most recent update r1.1g - worked fine for me before.

[mmestnik]

I'd open a ticket, but I'm already sure that our Development crew is fully aware of this problem.

[gary_ford]

great - i will stop trying to fix it then..

[mmestnik]

I would not assume that it's something outside of the normal system configuration issue. As you stated upgrading to 1.1G likely changed some thing that effects mail processing. This does not automatically mean that it's code related.

Check your system thoroughly to ensure every part functions as it should, we believe that it's most likely a problem with the settings or DNS, ect, ect. That is the upgrade changed something, but to fix this we need more information.

[gary_ford]

I will provide whatever you ask for. I have already done the testing as per a few posts back, i can send a test email from my account settings but a test form a host does not sent a notification, but further than gui based testing i am at a loss as to what is happening.

I am not a linux person, i come from a wintel/vmware background, hence my reason for suggesting to my director to go with nagiosxi - I have used nagios as a tool in the past but never set it up from scratch and dont posess the skills to do so.

For me to fault find an error of this nature would be to time consuming as i would be learning new skills along the way, if future releases are going to break major functinality (and i am not saying the recent release did) then i for one have a major issue.

Now if the recent release is the root cause then i would suggest that your RC testing process needs some consideration if you want this product to be serious contender.

A lesson learnt for me is to snapshot the vm before i upgrade and make sure i fully test, tis may seam like common sense but i did some other testing around getting further plugins working and removed my rc snapshot.