Technical recipes for frequently and infrequently recurring problems
We monitor the systems we run with nagios. It sends out alerts if a system is down, and measures uptime. For some clients, it sends our uptime reports. For some of our contracts we have committed to maintaining a certain level of uptime, so keeping accurate statistics is important.
See also:
The main page is http://nagios.curationexperts.com/nagios/. You will need nagios credentials. If you need an account, ask Bess or Mark to create one for you.
Quick links:
The server is nagios.curationexperts.com. Nagios is installed in /usr/local/nagios. Configuration is in /usr/local/nagios/etc. Please note, this directory is under local version control. If you make any changes, please identify yourself and commit your changes with a meaningful commit message. E.g.:
$ git config user.name "Bess Sadler"
$ git config user.email "bess@curationexperts.com"
$ git commit -a -m "Adding an entry for BarBaz University hosting contract"
We send out a weekly email report summarizing uptime for our hosted clients. Right now, our only client who gets these is Emory.
/usr/local/nagios/bin/email_pdf_reports.sh/opt/EmoryNagiosReportscd /usr/local/nagios/etcgit status to ensure the directory is in a clean, known statehtpasswd /usr/local/nagios/etc/htpasswd.users NEW_USER_NAME. Provide a password when prompted.cgi.cfg, everywhere that user needs access.Git commit -a -m "Adding NEW_USER_NAME to nagios config"$ sudo /etc/init.d/nagios reload