Feature Request - Disable Nagios exit codes if Alert Providers used

[varuzam]

Feature Request

add args like --nagios-mode

Feature Description

it is nice to have such behavior:
when the script is executed with --nagios-mode argument it returns exit code 1 for alerts with warning severity and 2 - critical
without the argument it returns non-zero exit code only if any exceptions or critical errors happen

Use Cases

I run the script from cronjob and need to monitor that it is working properly without any exceptions and any errors. I use alert provider for getting alerts. Such approach gives me ability to distinguish alerts about failure of execution the script and alerts about AWS limits.

Testing Assistance

I can help in testing it

Installation Method

I noticed that it behaves such way only in official docker image and, likely, when installed from the repo. When installed from pip it returns exit code 0 even when any AWS limits exceeded

[jantman]

@varuzam Hmm... this one is really interesting to me. I think there might be two separate issues here...

First, just to make sure I'm understanding you correctly, what you're asking for is an option so that when an Alert Provider is used, awslimitchecker only exits non-zero if there's some sort of exception? As opposed to its current behavior, where Alert Providers are in addition to the 0/1/2 exit code (ok/warning/critical)?

Secondly, I'm a bit confused/surprised about what you said in the "Installation Method" section. You're saying that you've installed awslimitchecker via pip but it's returning zero even when there are warnings and criticals? Could you please provide a debug-level output example of this, by running awslimitchecker -vv [any other options you specify]; echo $?

[varuzam]

@jantman you understand me correctly

sorry, my bad. installed via pip it works as expected. returning the 0/1/2 exit code

[jantman]

@varuzam I just want to make sure I understand the current state of this issue:

1. awslimitchecker is returning 0/1/2 properly based on ok/warning/critical.
2. You'd like to request a feature/option where, if Alert Providers are used, awslimitchecker will not exit non-zero just because of warnings/criticals, but only because of an exception or other program error.

Is this correct?

[varuzam]

Correct

[jantman]

This has been fixed in 9.0.0, which is now live on PyPI and on the Docker Hub. Thank you so much!