Maintain Rulesets for Displaying and Sending Statistical Alerts

Previous  Top  Next

This function allows you to customize events that will trigger or suppress alerts. The difference between this function and Maintain Rulesets for Displaying and Sending Alerts, is that these events are based on information from the Statistical Logpages, rather than sense codes.

Every time data is returned by polling devices, it is filtered through the rule sets. If the rule sets allow the error to pass, then the message is forwarded to reporting logic. Messages can go to a screen pop-up, the E-MAIL engine, or both. A copy of the message will always be recorded in the STATUSHISTORY.LOG file provided you did not define a rule which suppresses reporting of that message.

Rules are evaluated in the order that they are shown on the display with the one at the top being evaluated first.
When you pull up this screen, it will guide you through configuring alerts for specific devices. First click on the top window to select a controller, then device to report on.

Next, the program will build the middle window (Total write errors is highlighted) with log page entries which are valid for this particular make and model of device.

Finally, you select the Threshold
and Action(s) you wish the software to perform if that particular log page entry meets or exceeds the threshold. If you set the threshold to zero, then an alert will always appear. [The value for a zero-threshold is if you wanted to trap something like drive temperature over time, or number of I/Os per second].

Device Settings
The Adapter, ID, and LUN fields correspond to the SCSI disk drive that will be polled. The Description corresponds to the text description of the log page.

Add Rules
Click the Add Rule button, and an empty row will appear. Add information as required.

Delete Rules Fields
Clicking the Delete Rule button will remove the rule set selected by the cursor.

Clicking this button exits the window, and all changes you have made will be discarded.

Clicking this button writes these rule sets to disk, and commits them. Next time the drive is polled, S.M.A.R.T. Disk Monitor will read all of the log pages configured on this screen, for that drive, and submit them to the reporting engine for possible alerting.

1. Each entry in the database will require one I/O per polling period. If your polling period is small, then this can effect performance.
2. IDE disk drives don't have log pages, so this feature won't be configurable for them.