= Monitoring =

You can monitor nagios alerts in https://monitor.mayfirst.org

The [wiki:bender bender] irc daemon also copies nagios alerts to the #mayfirst [wiki:faq/chat irc channel]. 

== How to handle Nagios alerts ==

This page is a list of alerts and the recommended actions you can take to address them. 

''List of alerts taken from /etc/nagios3/conf.d/services_nagios2.cfg''

 HTTP::       
   edit
 HTTPS::      
   edit
 SSH::       
   edit
 SMTP::
   edit
 Secure IMAP::      
   edit
 Secure POP::      
   edit
 Disk Space::      
   edit
 Upgrade::       
   edit
 Mail Status::      
   This alert indicates that the postfix mail queue for the host has accumulated enough deferred messages to reach either a WARNING or CRITICAL threshold. The values for these threshholds are hardcoded into in the mf-monitor-mailq script {{{puppet/modules/mayfirst/files/monitor-utils/mf-monitor-mailq}}}  but can be overridden by local environment variables.
   Evaluation: Using the {{{mf-mailq-summary}}} is a good place to start. The script will show a two column output with number of mails corresponding to a particular sending address. This will allow you to quickly determine whether the majority of accumulated mail is originating from one particular user, address or domain. If you can identify a particular address of interest try getting a sample of queue id numbers from that address {{{mailq | grep $somepattern | tail}}} . Use the queue id number to examine the headers and other details {{{postcat -q $queueid}}}.
 Status of critical services::    
   edit
 Monitor backup servers for backup failures::  
   edit
 Monitor all servers for backup errors::  
   edit
 Mail Injector Status::     
   edit
 Oom Killer Status::     
   edit
 Relayers Status::      
   See [wiki:email-deliverability] , run `mf-check-relay-mail-users` on the indicated server to see who is being reported.
 Blocklist Status::      
   See suggestions in wiki [wiki:email-deliverability] also see examples in ticket #13288
 MySQL error Status::     
   edit
 Check for high number of email abuse complaints::
   Check for high number of email abuse complaints to a given IP. See the [wiki:mail-abuse Mail abuse page].
 PHP::       
   edit
 Ensure Email goes to Inbox::   
   See the [wiki:filter-check filter check page]
 fedcam::       
   review the [https://fedcam.mayfirst.org Fedcam web interface]. See [wiki:fedcam fedcam wiki]
 sks-pool::       
   edit
 keys.mayfirst.org::       
   edit
 roundcube.mayfirst.org::       
   edit
 roundcube.dev.mayfirst.org::
   edit

== Creating new nagios monitors ==

https://support.mayfirst.org/wiki/how-to/servers/puppet/setup-nagios-monitor