The 5.3.0 version, released yesterday includes a new check to monitor the EMS(Event Management System)-Logfile. This allows several new checks among them one to send an alarm, if an unusal high number of autogrow-events take place.
Let’s have some examples:
# check myfiler for autoSize events and alarm according to the defaults check_netapp_ems -H myfiler --name=wafl.vol.autoSize.done # same as above, but check only on node 01 check_netapp_ems -H myfiler --node=myfiler-01 --name=wafl.vol.autoSize.done # check any autoSize events (even the failed ones) check_netapp_ems -H myfiler --name="~^wafl\.vol\.autoSize\..*" # warn if more than 50 events per hour occur within the last 2 hours check_netapp_ems -H myfiler --name=wafl.vol.autoSize.done -w 50 -c 70 --lookbehind=2h --rate=per_hour
As you can see in the third example above, the
--name filter understands regex-patterns as well. A value starting with a tilde (
~) will be interpreted as a regular expression.
--lookbehind period tells the check how long into the past ist should look when calculating the rate. The longer this period is, the smoother the “curve” of the rate will be. For higher lookbehind-values the getter has to collect data for some time. During this period of insufficient data the check will interpolate the existing data.
New getter and compatibility
As already mentioned, this check requires the collector
get_netapp_ems to collect the data. This new getter contacts the quiet new RESTful API available on ONTAP 9.6 or later. Therefore this check is not supported for ONTAP versions smaller than 9.6. Please also take into consideration, that this check is in an alpha state and its UI may change in one of the next minor releases.
Of course the
--name argument can filter on arbirary any other event-name as well, which makes this check extremly versatile. We are looking forward to hear from other use cases our customers will find for it.