# alert me when things start to look fishy
check system xxx.com
if loadavg (1min) > 4 then alert
if loadavg (5min) > 4 then alert
if memory usage > 87% then alert
if cpu usage (user) > 70% then alert
if cpu usage (system) > 40% then alert
if cpu usage (wait) > 45% then alert
# Check the disk usage on root
check device rootfs with path /
if space usage > 90% then alert
if space usage > 99% then exec "/srv/web/... .../maintenance_redirect offline"
# make sure there is log space
check device san_log_mnt with path /mnt/san-logs/
if space usage > 90% then alert
if space usage > 99% then exec "/srv/web/... .../maintenance_redirect offline"
# zeo should always be running
check process zeo0 with pidfile /srv/web/... .../var/ZEO.pid
start program = "/etc/init.d/zeo0 start" with timeout 80 seconds
stop program = "/etc/init.d/zeo0 stop"
if failed host 127.0.1.1 port 8101 timeout 60 seconds then restart
check process zope0 with pidfile /var/run/zope0.pid
start program = "/etc/init.d/zope0 start" with timeout 80 seconds
stop program = "/etc/init.d/zope0 stop"
if failed host 127.0.1.1 port 8080 protocol HTTP timeout 120 seconds then alert
if failed host 127.0.1.1 port 8080 protocol HTTP timeout 300 seconds then restart
if cpu > 70% for 3 cycles then alert
if totalmem > 2000.0 MB for 5 cycles then alert
if loadavg(5min) greater than 10 for 8 cycles then alert
if 3 restarts within 5 cycles then timeout
check process varnishd with pidfile /mnt/san-logs/... .../varnish/varnish.pid
start program = "/etc/init.d/varnishd start" with timeout 60 seconds
stop program = "/etc/init.d/varnishd stop"
if cpu > 60% for 2 cycles then alert
if cpu > 90% for 5 cycles then alert
if totalmem > 500.0 MB for 5 cycles then alert
if children > 250 then alert
if loadavg(5min) greater than 10 for 8 cycles then alert
check process httpd with pidfile /var/run/httpd.pid
start program = "/etc/init.d/httpd start" with timeout 60 seconds
stop program = "/etc/init.d/httpd stop"
if cpu > 60% for 2 cycles then alert
if cpu > 90% for 5 cycles then alert
if totalmem > 500.0 MB for 5 cycles then alert
if children > 250 then alert
if loadavg(5min) greater than 10 for 8 cycles then alert
Comments (0)
You don't have permission to comment on this page.