Network ports of a filer are not always in use. Up to now, the NetPort check would send an alarm for every port that was not marled as up. Of course, this is not useful for ports that have been deactivated on purpose. Previously, these ports could be excluded from being monitored using the switch exclude and --include …
Nicht immer sind alle Netzwerk Ports eines Filers in Verwendung. Der Check NetPort sorgte bisher dafür, dass alle Ports die nicht up waren einen Alarm verursachten. Bei bewusst deaktivieren Ports ist das natürlich nicht sinnvoll. Hier gab es bisher schon die Möglichkeit, diese über die Schalter --exclude und --include …
Raidstatus monitors the redundancy or the RAID status of one or more aggregates. It is important to make sure that the patterns viewed as OK are as accurate as possible. An example for an OK pattern is the raid_dp status, not OK would be raid_dp, degraded. However, the following configuration --ok_state=raid_dp is …
When the shelfs of a single system are being inspected, the current getter for the shelf-bay object is missing one of the two port lists. This is not an error, if the NetApp system is a single-system (not an HA). For instance, a FAS2520 system with an integrated shelf does not have a second filer, therefore the B …
When the status of an aggregate changes to _degraded, _it does not always trigger the Health-State of the cdot to change as well. If you would like to receive warning messages as soon as the status of an aggregate changes to _degraded, _you have to configure the RaidStatus check.
The SnapMirrorMetrics check monitors the lag-time (and other values) in a snap-mirror-relation. To this point, the thresholds were set in seconds: -w 10800 -c 259200 Soon, this will be a lot easier: -w 3 -c 72 --factor=h The output will also be a lot more readable: NETAPP_PRO SNAPMIRRORMETRICS OK - 1 snapmirror …
Shorty, the Snapshots check will receive a new switch: ‑‑snapmirror_label_matches This switch can be used to analyze file sizes and the number of snapshots based on the snap-mirror-lable. Therefore, it is also possible to search for snapshots that don’t have a snap-mirror-lable. Here are a few examples: $ …
Unfortunately, for some performance checks it can occur that the performance data for multiple instances has been given the exact same label. The data is then rendered unusable for tools that create charts (PNP,…). Checks affected by this problem include, for example: NVRAM or FlashCache . This will be resolved …
The check for flash cache performance (FlashCache ) will receive three additional counters in the next release (3.3.0):
disk_reads_replaced: Estimated number of disk reads per second replaced by cache evicts: Number of blocks evicted from the external cache to make room for new blocks evicts_ref: Number of blocks …
The snapshots check can be usedin combination with** ‑‑metric=number **in order to make sure that snapshot copies are created regularly. If you would like to receive warning messages in steps (WARNING if no copies have been created in the last 24h, CRITICAL if no copies in the last 48h), two checks have to be …